Model User Guide for Generic Renewable Energy

System Models

3002014083

14496639
14496639
Model User Guide for Generic Renewable Energy
System Models

3002014083
Technical Update, July 2018

EPRI Project Manager

E. Farantatos

ELECTRIC POWER RESEARCH INSTITUTE

3420 Hillview Avenue, Palo Alto, California 94304-1338  PO Box 10412, Palo Alto, California 94303-0813  USA
14496639800.313.3774  650.855.2121  [email protected]  www.epri.com
DISCLAIMER OF WARRANTIES AND LIMITATION OF LIABILITIES
THIS DOCUMENT WAS PREPARED BY THE ORGANIZATION(S) NAMED BELOW AS AN ACCOUNT OF
WORK SPONSORED OR COSPONSORED BY THE ELECTRIC POWER RESEARCH INSTITUTE, INC. (EPRI).
NEITHER EPRI, ANY MEMBER OF EPRI, ANY COSPONSOR, THE ORGANIZATION(S) BELOW, NOR ANY
PERSON ACTING ON BEHALF OF ANY OF THEM:
(A) MAKES ANY WARRANTY OR REPRESENTATION WHATSOEVER, EXPRESS OR IMPLIED, (I) WITH
RESPECT TO THE USE OF ANY INFORMATION, APPARATUS, METHOD, PROCESS, OR SIMILAR ITEM
DISCLOSED IN THIS DOCUMENT, INCLUDING MERCHANTABILITY AND FITNESS FOR A PARTICULAR
PURPOSE, OR (II) THAT SUCH USE DOES NOT INFRINGE ON OR INTERFERE WITH PRIVATELY OWNED
RIGHTS, INCLUDING ANY PARTY'S INTELLECTUAL PROPERTY, OR (III) THAT THIS DOCUMENT IS
SUITABLE TO ANY PARTICULAR USER'S CIRCUMSTANCE; OR
(B) ASSUMES RESPONSIBILITY FOR ANY DAMAGES OR OTHER LIABILITY WHATSOEVER (INCLUDING
ANY CONSEQUENTIAL DAMAGES, EVEN IF EPRI OR ANY EPRI REPRESENTATIVE HAS BEEN ADVISED OF
THE POSSIBILITY OF SUCH DAMAGES) RESULTING FROM YOUR SELECTION OR USE OF THIS
DOCUMENT OR ANY INFORMATION, APPARATUS, METHOD, PROCESS, OR SIMILAR ITEM DISCLOSED IN
THIS DOCUMENT.
REFERENCE HEREIN TO ANY SPECIFIC COMMERCIAL PRODUCT, PROCESS, OR SERVICE BY ITS TRADE
NAME, TRADEMARK, MANUFACTURER, OR OTHERWISE, DOES NOT NECESSARILY CONSTITUTE OR
IMPLY ITS ENDORSEMENT, RECOMMENDATION, OR FAVORING BY EPRI.
THE FOLLOWING ORGANIZATION, UNDER CONTRACT TO EPRI, PREPARED THIS REPORT:
Power and Energy, Analysis, Consulting and Education, PLLC (PEACE)

This is an EPRI Technical Update report. A Technical Update report is intended as an informal report of
continuing research, a meeting, or a topical study. It is not a final EPRI technical report.

NOTE
For further information about EPRI, call the EPRI Customer Assistance Center at 800.313.3774 or
e-mail [email protected].

Electric Power Research Institute, EPRI, and TOGETHER…SHAPING THE FUTURE OF ELECTRICITY
are registered service marks of the Electric Power Research Institute, Inc.
Copyright © 2018 Electric Power Research Institute, Inc. All rights reserved.

14496639
 ACKNOWLEDGMENTS
The following organization, under contract to the Electric Power Research Institute (EPRI),
prepared this report:
Power and Energy, Analysis, Consulting and Education, PLLC (PEACE)
2221 Justin Rd #119-414
Flower Mound, TX 75028
Principal Investigator
P. Pourbeik

This report describes research sponsored by EPRI.

This publication is a corporate document that should be cited in the literature in the following
manner:
Model User Guide for Generic Renewable Energy System Models. EPRI, Palo Alto, CA: 2018.
3002014083.

14496639 iii
14496639
 ABSTRACT
In the last several years, the Electric Power Research Institute (EPRI) has led the technical
development, through a broad industry wide effort, of a new set of generic and public models for
renewable energy systems. The work was done primarily within the Western Electricity
Coordinating Council (WECC) Renewable Energy Modeling Task Force, and the models were
adopted by several of the major commercial software vendors in North America and approved by
WECC in 2013/2014. This brief report is a guide for the use of these models. References are
provided to other publicly available documents that contain the detailed specification of the
models. This second revision of the report replaces an earlier version. It is substantially the same
with some updates and corrections, based on further development of the renewable energy
systems models in the years since the 2015 version was published.
Keywords
Wind turbine generators
Photovoltaic generation
Energy storage
Dynamic models

14496639 v
14496639
 CONTENTS
ABSTRACT ................................................................................................................................v
1 INTRODUCTION ..................................................................................................................1-1
2 THE SECOND GENERATION GENERIC RENEWABLE ENERGY MODELS .....................2-1
2.1 The Model Library ........................................................................................................2-1
2.2 Naming of the Models in Various Software Platforms ...................................................2-2
2.3 Modeling Wind Turbine Generator (WTG) Plants .........................................................2-2
2.4 Modeling Type 1 WTG..................................................................................................2-3
2.5 Modeling Type 2 WTG..................................................................................................2-5
2.6 Modeling Type 3 WTG..................................................................................................2-6
2.7 Modeling Type 4 WTG..................................................................................................2-8
2.8 Modeling Photovoltaics (PV).......................................................................................2-10
2.9 Modeling Battery Energy Storage Systems (BESS) ....................................................2-11
2.10 Limitations of the Generic Positive Sequence Stability Models and On-going
Work to Continue to Improve Them ..................................................................................2-12
2.11 Tabular Summary of the RES Models ......................................................................2-16
3 CONTROL STRATEGY OPTIONS USING THE NEW RES MODELS .................................3-1
3.1 Overview ......................................................................................................................3-1
3.2 The Generator/Converter Model regc_a .......................................................................3-2
3.3 The Renewable Energy Electrical Controls Models ......................................................3-3
3.4 The Renewable Energy Plant Controller Model ............................................................3-9
3.5 Low and High Voltage and Frequency Rde-Through ..................................................3-13
4 VALIDATION OF THE MODELS ..........................................................................................4-1
5 CONCLUSION AND SUMMARY ..........................................................................................5-1
6 REFERENCES .....................................................................................................................6-1
A MODEL NAMES IN THE MAJOR SOFTWARE PLATFORMS USED IN NORTH
AMERICA ............................................................................................................................... A-1
B CURRENT LIMIT LOGIC .................................................................................................... B-1
C BLOCK DIAGRAMS FOR THE WIND TURBINE RELATED MODELS .............................. C-1
D BLOCK DIAGRAMS FOR RE*** MODELS ......................................................................... D-1
E CONVERSION BETWEEN THE 1ST AND 2ND GENERATION WIND TURBINE
GENERATOR MODELS ......................................................................................................... E-1
E.1 Type 1 and Type 2WTG .............................................................................................. E-1
E.2 Type 3 WTG ............................................................................................................... E-3
E.3 Type 4 WTG ............................................................................................................... E-7
F PARAMETER LIST FOR THE 2ND GENERATION GENERIC MODELS ............................. F-1

14496639 vii
14496639
 LIST OF FIGURES
Figure 1-1 The four many wind turbine technologies ................................................................1-2
Figure 2-1 Simple aggregated model for a WTG power plant ...................................................2-2
Figure 2-2 Complex plant aggregate model .............................................................................2-3
Figure 2-3 Type 1 WTG model .................................................................................................2-4
Figure 2-4 Type 2 WTG model .................................................................................................2-5
Figure 2-5 Type 3 WTG model .................................................................................................2-7
Figure 2-6 Type 4 WTG model .................................................................................................2-9
Figure 2-7 PV plant model .....................................................................................................2-11
Figure 2-8 Simple BESS model..............................................................................................2-12
Figure 2-9 Block diagram of proposed REGC_B model which shows the voltage-source
interface ...........................................................................................................................2-14
Figure 2-10 A revised proposal for REGC_B model, based on work done by EPRI in [28] .....2-15
Figure 3-1 Options for the reactive power control path in the reec_a model .............................3-6
Figure 3-2 Renewable energy electrical control model state transition diagram for
the (reec_a) ........................................................................................................................3-6
Figure 3-3 Current limit for reec_a model .................................................................................3-7
Figure 3-4 Extra part of the reec_c model for simulating charging and discharging of a
storage mechanism ............................................................................................................3-9
Figure 3-5 The plant controller model repc_a .........................................................................3-11
Figure 3-6 The plant controller model repc_b .........................................................................3-13
Figure C-1 Wind turbine generator drive-train model (wtgt_a) ................................................. C-1
Figure C-2 Wind turbine generator aero-dynamic model (wtgar_a) ......................................... C-1
Figure C-3 Wind turbine generator pitch-controller model (wtgpt_a)........................................ C-2
Figure C-4 Wind turbine generation torque model (wtgtrq_a) .................................................. C-3
Figure D-1 Renewable energy electrical controls model a (reec_a)......................................... D-1
Figure D-2 Renewable energy electrical controls model B (reec_b) ........................................ D-2
Figure D-3 Renewable energy electrical controls model C (reec_c) ........................................ D-3
Figure D-4 Renewable energy generator/converter model A (regc_a) ..................................... D-4
Figure E-1 Model wt1p which was part of the 1st generation generic WTG models and
is no longer recommended for use. ................................................................................... E-2
Figure E-2 Model wt1p_b, which is the new pitch-controller for the type 1 and 2 WTGs
and the recommended model for use with these turbines. ................................................. E-2

14496639 ix
14496639
 LIST OF TABLES
Table 2-1 List of RES models.................................................................................................2-16
Table 2-2 Combinations of the models for modeling various RES ..........................................2-16
Table 3-1 Reactive power control modes for the reec_a model ................................................3-5
Table 3-2 Reactive power control modes for the reec_* + repc_a models ..............................3-10
Table A-1 Model names across four main commercial software tools used in
North America ................................................................................................................... A-1

14496639 xi
14496639
1
INTRODUCTION
In recent years, at the culmination of extensive research and development, the Electric Power
Research Institute (EPRI), working together with many stakeholders within the industry, helped
to develop the second generation of generic stability models for wind generators and
photovoltaic generation. These models were developed deliberately in a modular format to
facilitate the ability to add to the library of these models, new features and functions without
significant effort. For example, in the past couple of years a generic battery energy storage
model, and a complex plant controller were developed as modules that were added to the library
of models. Presently, work is continuing to develop further modules such as a voltage-source
converter representation. Although EPRI has lead much of the technical development and
testing of the models, the effort has been a broad industry effort with true collaboration among
many stakeholders including several commercial power system simulation software vendors,
equipment manufacturers, utilities, two national laboratories (NREL and Sandia) and many
others. The collaborative community of stakeholders has worked under the Western Electricity
Coordinating Council’s (WECC) Renewable Energy Modeling Task Force (REMTF). Thus, at
the culmination of the work, the models were WECC approved and have found their way into
several of the major commercial software platforms, namely, Siemens PTI PSS®E, GE PSLF,
PowerWorld Simulator and PowerTech Labs TSAT™.
The detailed model specifications may be found in [1], which is the WECC approved document
and definitive model specification. References [2], [3], [4], [5] and [6] provide other details and
the documentation of the gradual development of the models, as well as testing and validation
results. We will not repeat any significant portion of these materials in this report, as all these
documents are publicly available.
It should be noted that the International Electrotechnical Commission (IEC) Technical
Committee (TC) 88, Working Group (WG) 27 is also working on developing specifications for
an international standard on generic models for wind turbine generators. The development of the
wind turbine generators model definitions has been completed 1, however, the work on the
development of plant controller models is still in progress. EPRI has also been engaged in this
work, particularly in the early stages of the work (2010 – 2013). The IEC models are for the
most part very similar to the WECC approved models, but they do have some differences,
particularly with respect to the type 3 wind turbine generators (see [4], [5] and [27]). It is outside
the scope of this document to further discuss these issues. The WECC approved models have
been deployed in many commercial software platforms, have been tested by multiple entities and
validated against many cases of field data, and are already in use. The IEC models are still under
development. Therefore, the WECC models are the subject of this document.

1
IEC 61400-27-1 Ed.1: Wind turbines - Part 27-1: Electrical simulation models - Wind turbines 9/30/2016.

14496639 1-1
It should be noted that as of January 2018, at the WECC Modeling Validation Working Group
(MVWG) an earnest dialogue has started to look at further revisions and additions to these
second-generation generic models, and as such some of the details in the IEC models may yet be
introduced into the models developed here in North America.
For those who may be unfamiliar with the four-main wind turbine generator technologies, they
are shown pictorially in Figure 1-1. Today, to our knowledge, all the newly developed wind
power plants, both in North America and overseas, are of the type 3 and 4 wind turbine
generators (WTG). However, many type 1 and 2 WTG plants do still exist and so they too need
to be modeled.

Type 2 – Variable Rotor-

Type 1 – Conventional Resistance Induction
Induction Generator Generator

To Grid
To Grid

Type 3 – Doubly-Fed Type 4 – Full-Converter

Asynchronous Generator Unit
To Grid

To Grid
ac/dc dc/ac

Can be gearless;
generator can be induction,
synchronous or permanent
ac/dc dc/ac magnet generator

Figure 1-1
The four many wind turbine technologies

These so-called second generation generic models were developed with two goals in mind:
1. They were developed in a modular format to allow for easy implementation of new modules
to build on the existing components to allow for the ever growing and changing technology,
and

14496639 1-2
2. They were made significantly more flexible than the so-called first generation generic models
to allow for emulation of a wider range of control philosophies and thus the ability to be
parameterized for representing a wide range of equipment.
In June 2014, these models were officially released. Since that time several additions have also
been made that will be discussed in this revised report. Several industry workshops, at WECC,
NERC and other forums have also been held since that time to discuss the models. The goal of
this document is to provide a user guide to allow for greater understanding of these models and
more easy adoption.
The document is organized as follows:
Section 2 – gives an overview of the model library and how the various pieces can be put
together to form the various types of renewable energy systems (RES).
Section 3 – describes in more detail the various options for control strategies associated with
type 3 and 4 WTGs, PV plants and energy storage.
Section 4 – gives a brief statement, with references to other public reports, about the validations
done with these models.
Section 5 – provides a very brief conclusion to the report.
Section 6 – is a list of the various references.
Appendices – provide various supporting information. Appendix E provides a detailed one to one
translation of the 1st generation WTG models to the 2nd generation WTG models. It is hoped that
this section might facilitate the great adoption of these models and the slow phasing out of the 1st
generation models. The tables presented in appendix E, for converting from the older 1st
generation models to the newer 2nd generation models, may in due course be adopted by the
commercial software vendors to be made a part of an automated conversion process within their
tools. Appendix F provides tables with indicative range of values for each parameter used in the
2nd generation renewable energy models.

14496639 1-3
14496639
2
THE SECOND GENERATION GENERIC RENEWABLE
ENERGY MODELS
2.1 The Model Library
The second generation of generic renewable energy systems models consists, currently, of a
library of ten models:
REGC_A
REEC_A
REEC_B
REEC_C
REPC_A
REPC_B
WTGT_A
WTGAR_A
WTGPT_A
WTGTRQ_A
WT1P_B 2
The majority of these were developed between 2010 to 2013 [1], and since mid-2014 have been
adopted by several commercial software vendors, including Siemens PTI PSS®E, GE PSLF,
PowerWorld Simulator and PowerTech Labs TSATTM. The complex plant controller model
(REPC_B) was developed latter, and the model specification can be found in reference [8].
In addition to the above models the following models, that have existed since the first-generation
models were developed years ago, are still valid and useful:
WT1G
WT2G
WT2E
LHVRT
LHFRT
Collectively, these models may be used to model:
Type 1 WTG wind turbines or plants
Type 2 WTG wind turbines or plants
Type 3 WTG wind turbines or plants
Type 4 WTG wind turbines or plants
Photovoltaic (PV) plants
Battery Energy Storage Systems (BESS)

2
This model has not been adopted yet by all the major software vendors, as of the writing of this report. It does,
however, exist in a few of the major commercial tools.

14496639 2-1
In the following sections a description is given on how to develop these various WTG, PV and
BESS devices and plants.

2.2 Naming of the Models in Various Software Platforms

Since the 2nd generation generic RES models were developed in a unified way within the broad
collaborative effort hosted within the WECC REMTF, much effort was put forth by the REMTF
to benchmark and verify the implementation of these models across the various major
commercial software platforms in use in North America. Nonetheless, there are some slight
differences in the naming convention of the models within the major software platforms, owing
simply to the naming conventions uses by each software vendor. Appendix A provides a list of
the model names across the four main commercial software tools used in North America.

2.3 Modeling Wind Turbine Generator (WTG) Plants

For bulk power system stability analysis, where the main concern is the dynamic behavior of the
power plant at the point of common coupling (PCC), based on current industry practice, a simple
model structure such as shown in Figure 2-1 may be used [10], [11]. A few additional comments
are pertinent:
1. If the WTG plant is based on type 1 or 2 WTGs, then many of these technologies utilize
switched shunt capacitor banks at the terminals of the WTGs. Thus, this will require an
explicit model of the shunt capacitor banks at either the low-voltage (LV) or the medium-
voltage (MV) bus at the turbine terminals. This must be explicitly modelled.
2. In many different plants there are switched or controller shunt compensation deployed at
the MV bus at the substation. Again, these should be appropriately modeled.
If the shunt compensation devices are controlled, the dynamic behavior needs to be separately
modeled and is not part of the generic models presented here. For example, SVCs and
STATCOMs can be modeled using the generic models developed for Static Var Systems [9].
LV MV PCC
GSU LV
Single equivalent
MV
feeder
EHV

Substation
Single Transformer
Aggregated WTG model

Figure 2-1
Simple aggregated model for a WTG power plant

Finally, it should be noted that more complex plants are possible and often more detailed models
may be necessary, for example as shown in Figure 2-2. To model such complex plants the
complex plant controller model may need to be used (REPC_B).

14496639 2-2
 GSU Single equivalent
feeder
Type 3

SVC or Substation
Single STATCOM Transformer
Aggregated WTG model

GSU Single equivalent

feeder
Type 4 MSC

Single
Aggregated WTG model

Figure 2-2
Complex plant aggregate model

In all cases, unless otherwise stated, it is assumed that a wind or PV power plant is modeled as
shown in Figure 2-1. The components are modeled in power flow as follows:
1. The substation transformer is modeled as usual using the transformer nameplate data.
When modeling an existing wind/PV power plant, it is imperative to properly model the
taps, and fixed-tap settings, as set in the field at the plant.
2. The equivalent single feeder model is calculated from the detailed collector system data
using the National Renewable Energy Laboratory (NREL) methodology to reduce it to a
single equivalent feeder model [12]. If the plant is a planned future plant with no present
collector system data, a reasonable assumption might be R = 0.011 pu, X = 0.027 pu and
B = 0.069 pu on 100 MVA base (this is an average value taken from the typical
parameters [13]).
3. The generator step-up (GSU) transformer is based on the transformer name plate data,
and the models MVA rating is simply scaled up by the number of turbines. For example,
if a single GSU is 0.06 pu on 1.5 MVA and there are 100 turbines in the plant, then the
GSU is modeled as X = 0.06 pu on 150 MVA.
4. The single aggregated WTG (or PV) is modeled with the appropriate parameters for the
specific equipment and the models MVA rating is again scaled up by the number of
turbines in service. For example, if a single WTG is rated at 1.65 MVA and there are 100
turbines in the plant, then the aggregated unit is modeled with the same parameters as the
single WTG on 165 MVA.

2.4 Modeling Type 1 WTG

The simple aggregated WTG plant model is shown in Figure 2-1. For a type 1 WTG plant, the
aggregated turbine model is represented as shown in Figure 2-3. As shown in the figure a type 1
WTG, which is a conventional induction generator, is developed using three models combined:

14496639 2-3
 1. wt1g – this is the electrical model of the induction generator. This is the standard
machine equations for a single (or double) cage induction machine and is available in
most commercial software tools. The model parameters are the standard parameters of
an induction machine, namely the electrical machine impedances (Ls, Lp, Lpp, Ll), time
constants (Tpo, Tppo), armature resistance (Ra) and saturation parameters (Se1, Se2).
The actual parameter names may vary between software platforms.
2. wt1t – the model of the turbine generator shaft, which may be modeled as either a single
lumped mass or two masses representing the generator and turbine assembly. The
parameters are the combined total shaft inertia (H), the fraction of the inertia that
represents the turbine assembly (Hfrac), the frequency of the first torsional mode
(Freq1) and the mechanical damping coefficients (D, Dshaft). For cases where one
wishes to model a single equivalent mass, only H and D need to be specified.
3. wt1p_b – the emulation of the active pitch controller. See Appendix E.1 for a more
detailed description of this module.
Vt

wt1g

Id

Generator

Iq

ω
wt1p_b Pm wt1t
Pitch-Control Drive-Train

wt1g Ls Lp Lpp Ll Ra Tpo Tppo Se1 Se2

wt1t H D Htfrac Freq1 Dshaft
wt1p_b Tr To rmax rmin Pmin Pset vt1 t1 vt2 t2 vt3 t3 vt4 t4

Figure 2-3
Type 1 WTG model

A few items should be noted:

• Typically, smaller and older type 1 WTGs (i.e. < 1 MW) where designed without active pitch
control. These so-called “stall” design turbines have fixed blade pitch. In these cases, the
only models needed are the wt1g and wt1t. Larger designs of type 1 WTGs employ active
stall and will have associated pitch control, for which the wt1p_b model may be used. For an
explanation of these control philosophies and their implications for system dynamic
performance see, chapter 3 of [14].
• The first generation generic WTG models that were released many years ago had associated
with them a pitch controller model called wt1p. This model was later identified as providing
potentially erroneous response particularly for simulations resulting in system frequency
events. Thus, it is recommended that this model should either be replaced with the wt1p_b

14496639 2-4
 model or at least removed and not used. Another issue with the wt1p model is that it does not
represent the fast ramp-down of mechanical power effected by the active-stall controls during
nearby voltage dips. This is explained further in Appendix E.
• The model parameters, and an explanation of each of the parameters of the new active pitch
controller model (wt1p_b) can be found in [1]. It has been shown that the wt1p_b model
does reasonably emulate the behavior of actual type 1 and 2 WTG performance [6].

2.5 Modeling Type 2 WTG

The simple aggregated WTG plant model is shown in Figure 2-1. For a type 2 WTG plant, the
aggregated turbine model is represented as shown in Figure 2-4. As shown in the figure, a type 2
WTG, which is a wound rotor winding induction generator with an externally controller rotor
resistance, is developed using four models combined:
1. wt2g – this is the electrical model of an induction generator with a wound rotor winding
that is externally accessible. The model parameters are electrical machine impedances
(Ls, Lp, Ll), time constant (Tpo), armature resistance (Ra), saturation parameters (Se1,
Se2) and the initial rotor speed. The actual parameter names may vary between software
platforms.
2. wt2t – the model of the turbine generator shaft as for the type 1 WTG.
3. wt2e – this is the model of the controller external rotor resistance. This is part of the 1st
generation models and was never changed. The user’s manual of the respective
commercial software tool should be consulted for the description of the parameters.
4. wt1p_b – the emulation of the active pitch controller, as with the type 1 WTG.
Vt

wt2g

wt2e Id
External Rext
Resistor Generator
Control
Iq

ω
wt1p_b Pm wt1t
Pitch-Control Drive-Train

wt2g Ls Lp Ll Ra Tpo Se1 Se2 spdrot

wt1t H D Htfrac Freq1 Dshaft
wt2e Tw Kw Tp Kp Kpp Kip Rmax Rmin Slip1 Slip2 Slip3 Slip4 Slip5 Powr1 Powr2 Powr3 Powr4 Powr5
wt1p_b Tr To rmax rmin Pmin Pset vt1 t1 vt2 t2 vt3 t3 vt4 t4

Figure 2-4
Type 2 WTG model

14496639 2-5
A few items should be noted:
• All type 2 WTGs will have some form of pitch control. As for the type 1 WTG, the new
wt1p_b model is recommended. The older pitch controller model (wt1p) should not be used.
For an explanation of the control philosophy of the type 2 WTG and its implications for
system dynamic performance see, chapter 3 of [14].
• The model parameters, and an explanation of each of the parameters, of the wt2e model can
be found in the user’s manual of the commercial software platforms.

2.6 Modeling Type 3 WTG

The simple aggregated WTG plant model is shown in Figure 2-1. For a type 3 WTG plant, the
aggregated turbine model is represented as shown in Figure 2-5. As shown in the figure a type
3 WTG, which is a doubly-fed asynchronous generator, is developed using seven models, all of
which are among the second-generation generic models:
1. regc_a – which is the renewable energy generator/converter model and has inputs of real
(Ipcmd) and reactive (Iqcmd) current command and outputs of real (Ip) and reactive (Iq)
current injection into the grid model.
2. reec_a – which is the renewable energy electrical controls model a, and has inputs of
real power reference (Pref) that can be externally controlled, reactive power reference 3
(Qref) that can be externally controlled, and feedback of the reactive power generated
(Qgen). The outputs of this model are the real (Ipcmd) and reactive (Iqcmd) current
command.
3. wtgt_a – which is the emulation of the drive-train oscillations. The output of this model
is speed (spd). In this case speed is assumed to be a vector spd = [ωt ωg], where ωt is the
turbine speed and ωg the generator speed. The inputs to the model are mechanical and
electrical power.
4. wtgar_a – which is a simple linear model of the turbine aero-dynamics. This is based on
reference [15]. The input to the model is pitch-angle (θ), and its output is mechanical
power (Pm).
5. wtgpt_a – which is a simple pitch-control model. The inputs to the model are electrical
power order (Pord), power reference (Pref0), speed reference (ωref) and speed (spd).
The output is pitch-angle (θ).
6. wtgtrq_a – which is a simple torque controller. The inputs to the model are speed (spd),
power reference (Pref0), electrical power (Pe) and the outputs are speed reference (ωref)
and electrical power reference (Pref).
7. repc_a – which is the renewable energy plant controller model a. This model has inputs
of either voltage reference (Vref) and measured/regulated voltage (Vreg) at the plant
level, or reactive power reference (Qref) and measured (Qgen) at the plant level. The
output of the repc_a model is a reactive power command that connects to Qref in the

3
The reactive power reference can also be an external voltage reference coming from the plant controller, depending
on the selection of the various control options. This is explained later in the report.

14496639 2-6
 reec_a model. In addition, the model can also emulate primary frequency response
based on the measured total plant real power output at the point of common coupling and
measured system frequency.
Vref/Vreg or
Qref/Qgen Freq_ref/Freq and
At plant level Plant_pref/Pgen

repc_a

Plant Level Control

Vt

reec_a
regc_a
Qref
Iqcmd’ Iqcmd
(or Qext) Q Control Iq
Qgen Current Generator/
Limit Converter
wtgtrq_a
Pref0 Logic Model
Torque Pref Ipcmd’ Ipcmd
P Control Ip
Control (or PExt)
Pe
Pord
spd
ωref

Pqflag
wtgar_a Pm wtgt_a = 1 (P priority)
wtgpt_a
Pitch-Control
θ Aero Drive-Train = 0 (Q priority)

regc_a lvplsw rrpwr brkpt zerox lvpl1 vtmax lvpnt1 lvpnt0 qmin tg tfltr iqrmax iqrmin
reec_a Vdip Vup Trv dbd1 dbd2 Kqv Iqh1 Iql1 Vrefo Iqfrz Thld Thld2 pfaref Tp Qmax Qmin Vmax Vmin
Kqp Kqi Kvp Kvi Vref1 Tiq dPmax dPmin Pmax Pmin Imax PfFlag Vflag Qflag Pqflag VDL1 VDL2 Tpord
wtgt_a Ht Hg Dshaft Kshaft
wtgar_a Ka θo
wtgpt_a Kiw Kpw Kic Kpc Kcc Tθ θmax θmin dθmax dθmin
wtgtrq_a Kip Kpp Tp Tωref Temax Temin Tflag f(Pe)
repc_a Tfltr Kp Kic Tft Tfv RefFlag Vfrz Rc Xc Kc VcompFlag emax emin dbd Qmax Qmin Kpg Kig
Tp fdbd1 fdbd2 femax femin Pmax Pmin Tlag Ddn Dup Pgen_ref Freq_ref vbus branch Freq_flag

Figure 2-5
Type 3 WTG model

In the next section of the report a more detailed account is given of the various control strategies
and functionalities of the type 3 WTG model. Reference [14] provides an account of the
dynamic performance of type 3 WTGs and the various designs of these turbines. A detailed
specification of the above models can be found in [1], together with an explanation of each
parameter. However, for the sake of completeness, in Appendix F of this report an explanation
is provided of all the parameters associated with each of the models and the typical expected
range of values, as a guide to help avoid setting inappropriate values.

14496639 2-7
2.7 Modeling Type 4 WTG
The simple aggregated WTG plant model is shown in Figure 2-1. For a type 4 WTG plant, the
aggregated turbine model is represented as shown in Figure 2-6. As shown in the figure a type
4 WTG, which is a full-converter interface asynchronous generator, is developed using either
three or four models, all of which are among the second-generation generic models:
1. regc_a – which is the renewable energy generator/converter model and has inputs of real
(Ipcmd) and reactive (Iqcmd) current command and outputs of real (Ip) and reactive (Iq)
current injection into the grid model.
2. reec_a – which is the renewable energy electrical controls model a, and has inputs of
real power reference (Pref) that can be externally controlled, reactive power reference
(Qref) that can be externally controlled, and feedback of the reactive power generated
(Qgen). The outputs of this model are the real (Ipcmd) and reactive (Iqcmd) current
command.
3. wtgt_a – which is the emulation of the drive-train oscillations. The output of this model
is speed (spd). In this case speed is assumed to be a vector spd = [ωt ωg], where ωt is the
turbine speed and ωg the generator speed. The inputs to the model are mechanical and
electrical power. This model may be used for type 4 A WTG plants where the torsional
response of the turbine-generator assembly is observable in the electrical power output
of the unit/plant.
4. repc_a – which is the renewable energy plant controller model a. This model has inputs
of either voltage reference (Vref) and measured/regulated voltage (Vreg) at the plant
level, or reactive power reference (Qref) and measured (Qgen) at the plant level. The
output of the repc_a model is a reactive power command that connects to Qref to the
reec_a model. In addition, the model can also emulate primary frequency response base
on the measured total plant real power output at the point of common coupling and
measured system frequency.

14496639 2-8
 Vref/Vreg or
Qref/Qgen Freq_ref/Freq and
At plant level Plant_pref/Pgen

repc_a

Plant Level Control

Vt

reec_a
regc_a
Qref
Iqcmd’ Iqcmd
(or Qext) Q Control Iq
Qgen Current Generator/
Limit Converter
Logic Model
Ipcmd’ Ipcmd
Pref P Control Ip
(or TExt)

spd
wtgt_a
Drive-Train
Pqflag
= 1 (P priority)
= 0 (Q priority)

Type 4 A
Vref/Vreg or
Qref/Qgen Freq_ref/Freq and
At plant level Plant_pref/Pgen

repc_a

Plant Level Control

Vt

reec_a
regc_a
Qref
Iqcmd’ Iqcmd
(or Qext) Q Control Iq
Qgen Current Generator/
Limit Converter
Logic Model
Ipcmd’ Ipcmd
Pref P Control Ip
(or TExt)

Pqflag
= 1 (P priority)
= 0 (Q priority)

Type 4 B
regc_a lvplsw rrpwr brkpt zerox lvpl1 vtmax lvpnt1 lvpnt0 qmin tg tfltr iqrmax iqrmin
reec_a Vdip Vup Trv dbd1 dbd2 Kqv Iqh1 Iql1 Vrefo Iqfrz Thld Thld2 pfaref Tp Qmax Qmin Vmax Vmin
Kqp Kqi Kvp Kvi Vref1 Tiq dPmax dPmin Pmax Pmin Imax PfFlag Vflag Qflag Pqflag VDL1 VDL2 Tpord
wtgt_a Ht Hg Dshaft Kshaft (optional)
repc_a Tfltr Kp Kic Tft Tfv RefFlag Vfrz Rc Xc Kc VcompFlag emax emin dbd Qmax Qmin Kpg Kig
Tp fdbd1 fdbd2 femax femin Pmax Pmin Tlag Ddn Dup Pgen_ref Freq_ref vbus branch Freq_flag

Figure 2-6
Type 4 WTG model

14496639 2-9
In the next section of the report a more detailed account is given of the various control strategies
and functionalities of the type 4 WTG model. Reference [14] provides an account of the
dynamic performance of type 4 WTGs and the various designs of these turbines. A detailed
specification of the above models can be found in [1], together with an explanation of each
parameter. Again, for completeness, Appendix F provides a list and explanation for all the
parameters of the models and a range of typical values.

2.8 Modeling Photovoltaics (PV)

The simple aggregated PV plant model is also as shown in Figure 2-1. For a PV plant, the
aggregated PV array model is represented as shown in Figure 2-7. As shown in the figure a PV
plant is developed using three of the second-generation generic models:
1. regc_a – which is the renewable energy generator/converter model and has inputs of real
(Ipcmd) and reactive (Iqcmd) current command and outputs of real (Ip) and reactive (Iq)
current injection into the grid model.
2. reec_a 4 – which is the renewable energy electrical controls model a, and has inputs of
real power reference (Pref) that can be externally controlled, reactive power reference
(Qref) that can be externally controlled, and feedback of the reactive power generated
(Qgen). The outputs of this model are the real (Ipcmd) and reactive (Iqcmd) current
command.
3. repc_a – which is the renewable energy plant controller model a. This model has inputs
of either voltage reference (Vref) and measured/regulated voltage (Vreg) at the plant
level, or reactive power reference (Qref) and measured (Qgen) at the plant level. The
output of the repc_a model is a reactive power command that connects to Qref to the
reec_a model. In addition, the model can also emulate primary frequency response base
on the measured total plant real power output at the point of common coupling and
measured system frequency.

4
We recommend using the reec_a model as it is more sophisticated and able to provide a means of modeling
“momentary cessation” – see the recent NERC modeling notice:
https://www.nerc.com/comm/PC/NERCModelingNotifications/Modeling_Notification_-
_Modeling_Momentary_Cessation_-_2018-02-27.pdf. Originally, many PV plants were modeled using the reec_b
model, but we recommend not using that model.

14496639 2-10
 Vref/Vreg or
Qref/Qgen Freq_ref/Freq and
At plant level Plant_pref/Pgen

repc_a

Plant Level Control

Vt

reec_a
regc_a
Qref
Iqcmd’ Iqcmd
(or Qext) Q Control Iq
Qgen Current Generator/
Limit Converter
Logic Model
Ipcmd’ Ipcmd
Pref P Control Ip
(or TExt)

Pqflag
= 1 (P priority)
= 0 (Q priority)

regc_a lvplsw rrpwr brkpt zerox lvpl1 vtmax lvpnt1 lvpnt0 qmin tg tfltr iqrmax iqrmin
reec_a Vdip Vup Trv dbd1 dbd2 Kqv Iqh1 Iql1 Vrefo Iqfrz Thld Thld2 pfaref Tp Qmax Qmin Vmax Vmin
Kqp Kqi Kvp Kvi Vref1 Tiq dPmax dPmin Pmax Pmin Imax PfFlag Vflag Qflag Pqflag VDL1 VDL2
repc_a Tfltr Kp Kic Tft Tfv RefFlag Vfrz Rc Xc Kc VcompFlag emax emin dbd Qmax Qmin Kpg Kig
Tp fdbd1 fdbd2 femax femin Pmax Pmin Tlag Ddn Dup Pgen_ref Freq_ref vbus branch Freq_flag

Figure 2-7
PV plant model

In the next section of the report a more detailed account is given of the various control strategies
and functionalities of the PV model. The user may also consult [16].

2.9 Modeling Battery Energy Storage Systems (BESS)

For a BESS, the device model is represented as shown in Figure 2-8. As shown in the figure a
BESS unit is modeled using two of the second-generation generic models:
1. regc_a – which is the renewable energy generator/converter model and has inputs of real
(Ipcmd) and reactive (Iqcmd) current command and outputs of real (Ip) and reactive (Iq)
current injection into the grid model. This represents the inverter interface for the BESS
unit.
2. reec_c – which is the renewable energy electrical controls model c, and has inputs of real
power reference (Pref) that can be externally controlled, reactive power reference (Qref)
that can be externally controlled, and feedback of the reactive power generated (Qgen).

14496639 2-11
 The outputs of this model are the real (Ipcmd) and reactive (Iqcmd) current command.
This represents the BESS inverter controls and includes a basic representation of the
charging/discharging dynamics.
In addition to the above models, a repc_a (renewable energy plant controller model a) model
may also be used together with this configuration to allow for voltage and frequency control at a
point of common coupling (see for example reference [23]). A detailed description of the reec_c
model is given in [7].
Vt

reec_c
regc_a
connected to the REPC_A

Iqcmd’ Iqcmd
Pref and Qref can be

Qref Q Control Iq
controller models
or REPC_B plant

Current Generator/
Limit Converter
Logic Model
Ipcmd’ Ipcmd
Pref P Control Ip

Pgen
Pqflag
= 1 (P priority)
= 0 (Q priority)

regc_a lvplsw rrpwr brkpt zerox lvpl1 vtmax lvpnt1 lvpnt0 qmin tg tfltr iqrmax iqrmin
reec_c Vdip Vup Trv dbd1 dbd2 Kqv Iqh1 Iql1 Vrefo SOCini SOCmax SOCmin pfaref Tp Qmax Qmin Vmax Vmin
Kqp Kqi Kvp Kvi Tpord Tiq dPmax dPmin Pmax Pmin Imax PfFlag Vflag Qflag Pqflag VDL1 VDL2 T

Figure 2-8
Simple BESS model

In the next section of the report a more detailed account is given of the various control strategies
and functionalities of the BESS model.

2.10 Limitations of the Generic Positive Sequence Stability Models and On-going
Work to Continue to Improve Them
A detailed discussion on the limitations of the generic positive sequence stability models
developed for modeling renewable energy systems (RES) in time-domain stability analysis is
outside the scope of this document. A more detailed account can be found in reference [22].
However, it is pertinent to identify some key aspects of the limitations of the models presented
here for the general information of the reader. The models presented here were developed
primarily for public use and benefit, and to eliminate the long-standing issues around many
vendor specific models being proprietary and thus neither publicly available nor easily
disseminated among the many stakeholders. Furthermore, using multiple user-defined non-
standard models within large interconnection studies, in many cases, presented huge challenges

14496639 2-12
and problems with effectively and efficiently running the simulations. Therefore, the intended
use of these models is for positive-sequence large interconnected power system stability
simulations. These models may be adequate for other uses as well, but the user must understand
the context of his/her study and use engineering judgement in applying the models.
These models are in general not applicable for the following:
1. They cannot adequately represent the detailed behavior of the equipment for nearby
unbalanced faults, since by their very nature these models are positive sequence models
and developed for use with positive sequence simulation tools. This is even more so in
the case of these RES models because many of the RES technologies interface with the
power system using power electronic converters. To analyze the behavior of power
electronic converters in detail, to unbalanced faults, in many cases will require three-
phase modeling with a thorough understanding of the converter control strategy.
2. These generic models are not adequate for modeling the behavior of the RES
technologies where they are interconnected to a very weak grid – that is, typically a short
circuit ratio of 2 to 3 or less. It is difficult to provide an exact short circuit ratio below
which the models are not applicable, the numbers presented here are only a guide.
3. Although these models may be used to adequately emulate low and high voltage ride-
through for large interconnected studies, when used in conjunction with the lhvrt relay
models, they are not adequate for use in designing the low/high voltage ride-through
systems. The representation is a simple emulation, based on vendor supplied
information, on the expected behavior of the equipment.
4. These models are not adequate for special studies for frequency phenomena outside of
the typical range of dynamics studied in transient stability analysis. For example, they
would not be adequate for the analysis of subsynchronous torsional interactions.
It is important at this point to highlight a developing aspect in relation to item 2. There are two
issues at hand with respect to item 2 above. Firstly, the positive sequence models do not have
the details of the converter inner-current control loops and synchronization circuitry (i.e. phase-
lock loop) represented. Nor can these aspects be represented in detail in a positive sequence
program, as they are outside of the band-width of positive-sequence tools. Therefore, these
aspects cannot be captured in detail, and in weak-grid applications proper tuning of these
controls will be necessary – see reference [24]. The second issue in positive-sequence tools is
one of numerical stability. Both the first and second generation generic models use a “current
source” model approach for modeling the algebraic interface between the network model and the
dynamic RES model. A current source interface is inherently numerically unstable when applied
in a low-short circuit node of the network. This can be easily understood because low-short
circuit implies high impedance. Thus, when one changes the current being injected into the
high-impedance node by a small amount, the voltage will change by a large amount. Thus,
solving the boundary conditions for voltage can become problematic. To resolve this numerical
issue, work is presently being done on a voltage-source model for representing the inverter
interface. This new model, called regc_b (see Figure 2-9), is still in the process of being
developed and tested among the software vendors through a similar collaborative effort in
WECC MVWG [25]. The concept is based on reference [26]. Once developed, the hope is that
this model will offer an alternative interface for full-converter type RES to mitigate some of the
numerical stability issues. Nonetheless, it must be understood that even with this model in use

14496639 2-13
the limitations of the positive-sequence models are still present, and detailed analysis of control
interactions and weak-grid tuning of converter controls will still need close collaboration with
the vendors and possibly initial investigation and simulation work in 3-phase EMT type vendor
specific models, with then appropriately parameterizing the positive-sequence models to provide
adequate performance for grid-wide simulations.
Furthermore, to the discussion above, at the latest WECC modeling and validation working
group meeting, EPRI presented an alternative model for REGC_B, as shown in Figure 2-10 [28].
The idea behind this proposal is to introduce a simple representation of both the inner-current
control loops and the PLL, in order to hopefully increase the fidelity of the voltage-source model
even further by making it somewhat mimic the actual controls of inverter-based generation. This
is still under study by EPRI and WECC, and may be adopted in time if proven to be a fruitful
exercise. It comes, of course, with all the caveats mentioned earlier, that we cannot truly
represent the full details of these controls in a positive-sequence program due to the bandwidth
difference between the actual controls and the simulation platform. None-the-less, initial
simulations have shown that with some manipulation of the parameters of this simplified
representation, there may be some benefits derived therefrom [28].

Figure 2-9
Block diagram of proposed REGC_B model which shows the voltage-source interface

14496639 2-14
Figure 2-10
A revised proposal for REGC_B model, based on work done by EPRI in [28]

In addition to the above, there has been significant discussion at the WECC modeling and
validation working group [29] to continue to develop new modules and improvements to
the 2nd generation generic models. This can be found in [29], but for the sake of completeness is
summarized here in bullet format:
• Development of the new generator/converter model (REGC_B mentioned above)
• Expanded features for converter electrical controls models (REEC_*), such as:
- More points in the VDL tables
- Allowing for delays in reactive current recovery (to model momentary cessation)
- Allowing for droop control at the local-voltage control loop level
- Etc.
• Adjustments to the torque-controller and pitch-controller per suggestions by equipment
vendors.
• Expanded features for the plant controller, e.g. to allow for coordinated shunt compensation
switching at the point of interconnection, and simulating ramps on reactive power, etc.
• Other modules such as, inertia-based fast-frequency response.
EPRI is engaged in these developments and will continue to work with the industry on these
technical updates.

14496639 2-15
2.11 Tabular Summary of the RES Models
The following two tables summarize the list of RES models and how they are combined to make
up the various RES.
Table 2-1
List of RES models

Model Function 1st or 2nd Generation

regc_a RES Generator/Converter Model (current source) 2nd
regc_b RES Generator/Converter Model (voltage source) 2nd (in development)
reec_a RES Electrical Controls Model A 2nd
reec_b RES Electrical Controls Model B 2nd
reec_c RES Electrical Controls Model C 2nd
repc_a RES Plant Controls Model A 2nd
repc_b RES Plant Controls Model B 2nd
wtgt_a WTG Turbine Shaft Model A 2nd
wtgar_a WTG Aero-dynamice Model A 2nd
wtgpt_a WTG Pitch Controller Model A 2nd
wtgtrq_a WTG Torque Controller Model A 2nd
wt1p_b Pitch Controller for type 1 WTG Model B 2nd
wt1g Type 1 WTG generator model 1st
wt1t Type 1 WTG turbine shaft model 1st
wt2g Type 2 WTG generator model 1st
wt2e Type 2 WTG variable external rotor resistance controller 1st
wt2t Type 2 WTG turbine shaft model 1st
lhvrt low/high voltage ride-through relay model 1st
lhfrt low/high frequency ride-through relay model 1st

Table 2-2
Combinations of the models for modeling various RES

RES Model Combination

Type 1 WTG wt1g, wt1t, wt1p_b
Type 2 WTG wt2g, wt2e, wt2t, wt1p_b
Type 3 WTG regc_a, reec_a, repc_a, wtgt_a, wtgar_a, wtgpt_a, wtgtrq_a
Type 4 WTG regc_a, reec_a, repc_a (optional: wtgt_a)
PV plant regc_a, reec_a, repc_a
BESS regc_a, reec_c (optional: repc_a)
1. For PV plant modeling the reec_a model is the recommended model, since it is able to give some
representation of inverter blocking or so-called “momentary cessation” (see NERC modeling notice
https://www.nerc.com/comm/PC/NERCModelingNotifications/Modeling_Notification_-
_Modeling_Momentary_Cessation_-_2018-02-27.pdf).
2. The lhvrt and lhfrt models may be used with any of the wind or PV plant representation to emulate
low/high voltage and frequency ride-through characteristics, as specified by the equipment vendor.

14496639 2-16
3
CONTROL STRATEGY OPTIONS USING THE NEW
RES MODELS
3.1 Overview
The new renewable energy system (RES) models incorporate at their heart three core systems:
1. The regc_a model which represents the interfacing electrical generator/converter,
2. the reec_* models which model the local P/Q controls associated with the power
converter interface of the RES 5, and
3. the repc_* models which model the plant level controls associated with one or more
active devices.
If we consider the bulk of the rest of the model library associated with the new RES models, one
will see that they are relatively self-explanatory. For completeness, let us first examine those in
brief.
WTGT_A – this is a simple two mass equivalent emulation of the turbine-generator shaft used to
emulate the first and dominant torsional mode of the drive-train. The parameters are Ht (turbine
inertia), Hg (inertia of the generator), Kshaft (spring constant of the shaft) and Dshaft (damping
coefficient). The block diagram in shown in Figure C-1, Appendix C. It should be noted that [5]
the shaft damping coefficient (Dshaft) is a fitted parameter to capture the net damping of the
torsional mode seen in the post fault electrical power response of the machine. In the actual
equipment, the drive train oscillations are damped through filtered signals and active damping
controllers, which obviously are significantly different from a simple generic two mass drive
train model. However, for the purposes of large scale power system simulations, the collective
decision of the Western Electricity Coordinating Council’s, Modeling and Validation Working
Group (which approved these models) was that the added complexity to try to model active drive
train damping was not warranted [4]. A detailed analysis of the torsional model, and particularly
if one wishes to perform small-signal stability analysis, requires a different modeling approach.
See for example, Appendix A of reference [14].
WTGAR_A – this is a very simple linear approximation of the aero-dynamic behavior of the wind
turbine based on [15]. There is only one parameter Ka, which represents the linear relationship
between a change in pitch-angle (in degrees) and the change in mechanical power (in pu). The
block diagram is shown in Figure C-2, Appendix C. A typical value for Ka is 0.007 pu/degree.
WTGPT_A – this is a model of the pitch-controller. It is the same as that used in the first-
generation generic wind turbine generator models with one exception, the addition of the cross
term (Kcc) for increased flexibility. The block diagram is shown in Figure C-3, Appendix C.

5
This is used whether the converter is a full converter (e.g. type 4 WTG or PV) or a partially rated converter (e.g. type
3 WTG).

14496639 3-1
WTGTRQ_A – this is a model of the emulation of torque control. It is similar to that in the first-
generation wind turbine generator generic models but has an additional feature of allowing
torque to be regulated based on torque error as well as speed error. The block diagram is shown
in Figure C-4, Appendix C.
WT1P_B – this is the improved pitch-controller for a type 1 or 2 WTG using active-stall [6]. The
block diagram is shown in Figure C-5, Appendix C. Reference [6] gives a detailed explanation
of the models features.
With this brief introduction, let us now turn to the details of the substantially new models, the
re**** family of models, in order to understand their features and usage.

3.2 The Generator/Converter Model regc_a

The regc_a model is shown in Figure D-4 (Appendix D). The two blocks shown on the right
“high-voltage reactive-current management” and “low-voltage active-current management” are
mainly for numerical reasons. In Appendix A of reference [1] flow-charts are provided for both
blocks. The actual detailed implementation may vary slightly among various commercial
software platforms. The purpose of these blocks is to provide for a smooth and reasonable
transition between the dynamic emulation of the current controls and the algebraic network
equations. The “high voltage reactive power logic” performs the action of limiting the reactive
current injected into the network equations in such a way as to prevent the terminal voltage of
the machine from exceeding a given limit. The “low voltage active power logic” is designed to
capture the characteristic of active power under very low voltages, that is, it reduces active
current in a linear fashion as voltage drops to very low levels. These two blocks are of a
numerical nature. They do not exactly represent a physical element in the controls. Their main
function is to minimize numerical issues that arise due to the approximation by a simple model
of what are essentially high bandwidth hardware components. In physical reality the power
converter is an extremely fast control device with switching speeds typically in the kilo-hertz.
These high-frequency phenomena cannot be adequately modelled with large scale positive
sequence simulation tools for many reasons. The foremost reason is that the network equations
are modeled with a static admittance matrix 6, thus the inherent assumption here is that the
phenomena being modeled and studied do not drift more than a hertz or so from fundamental
network frequency, for if they do then the network model is significantly deficient. Thus,
exposing the network model to high-frequency phenomena will yield grossly erroneous results.
The time constant Tg is an emulation of the delay in the power converter switching process. In
reality, this is a pure transport delay in the range of a few milliseconds. Again, modeling such
detail is typically unnecessary for large scale power system simulations, particularly when the
integration time step of typical simulations of this nature tend to be in the range of ¼ to ½ of a
cycle (i.e. 4.16 to 8.3 ms). Typically, Tg is set to a value of between 0.01 to 0.02 s.
The time constant Tfltr is simply an emulation of the filtering time constant associated with
measuring terminal voltage. A typical value would be in the range of 0.01 to 0.02 s.

6
It is fully understood that the admittance matrix changes during re-factorizations where lines/elements are switched
in or out. By static and lumped model here is meant that the reactances and capacitances are represented by fixed
lumped impedances as seen at fundamental network frequency.

14496639 3-2
The reactive current command (Iqcmd) passes through not only the lag time constant Tg, but also
the rate-limits Iqrmax and Iqrmin. These rate limits on reactive power were used by only one of
the original equipment manufacturers (OEMs) that we studied [5]. Furthermore, in that one case
they were invoked when the turbine is operating in local constant Q control for this vendor. That
is, the turbine is holding a constant reactive power output. In this case these limits are imposed
post fault. Iqrmax is active if the initial reactive output of the unit was above zero, and Iqrmin is
active if the initial reactive output is negative. The purpose of the rate limits is to limit rate of
recovery of the reactive power to its initial value after fault clearing. The user is cautioned not to
use this feature unless instructed by the OEM or if they are certain of what they wish to
accomplish.
The Low Voltage Power Limiter (LVPL) logic is used to emulate in a simple way the tendency
of some vendors to limit the active power output of the converter at low voltages. In fact, in
almost all cases when the voltage is depressed to extreme values (i.e. below 5% residual voltage)
the converters ability to produce active power will be severely limited.
Finally, the parameter rrpwr can be used to emulate the rate of rise in active power output
following a grid disturbance. This parameter acts on active current, so it is not an exact
representation of a rate limit on active power output.

3.3 The Renewable Energy Electrical Controls Models

There are presently three reec_* models:
1. reec_a – the most complex, used typically with type 3 and 4 WTGs, and recommended
for PV,
2. reec_b – the simpler version, originally used with PV plants is no longer recommended
to be used since it lacks some of the needed features of the reec_a model, and
3. reec_c – specifically built to represent a battery energy storage system (BESS).
The block diagram for all three are shown in Appendix D. Note that reec_b is a subset of
reec_a. Furthermore, reec_c is also a subset of reec_a, with the addition of a few extra
parameters around the active power portion which facilitates modeling energy storage. In
general, based on more recent experience (see NERC modeling notice
https://www.nerc.com/comm/PC/NERCModelingNotifications/Modeling_Notification_-
_Modeling_Momentary_Cessation_-_2018-02-27.pdf) the use of reec_b is not recommended
any longer.
First consider reec_a, the most general model. There are three parts to the model:
• active current controls which develop the active current command Ipcmd
• reactive current controls which develop the reactive current command Iqcmd
• the converter current limit logic which limits the active and reactive current to within the
ratings of the converter 7

7
The current limit logic is shown in Appendix B. It assumes a full-converter unit. For the sake of simplicity, it was
decided, during the development of this model, by the WECC MVWG to not make a distinction between stator current
limits for the type 3 WTG and converter current limits for the type 4 WTG.

14496639 3-3
Active Power Control: Let us first look at the active power control. In this part there are two
options PFlag = 1, or PFlag = 0. For type 3 WTGs PFlag must equal 1, because the power
developed by the turbine is modulated by perturbations in the shaft speed since the electrical
generator is directly coupled to the grid. In this case, the wtgtrq_a model develops electrical
torque, and so torque times speed yields power. EXTREMELY IMPORTATN NOTE: in most,
if not all, commercial software platforms the implementation of the models is slightly different
than that shown in the original model specifications [1], and thus also shown here. That is,
the output of the torque controller model (e.g. wtgq_a model in GE PSLF) is already equal to
power, since the multiplication of torque and speed is done within the model. Therefore,
PFlag must be set to 0 in reec_a for a type 3 WTG, in programs like GE PSLF, Siemens PTI
PSS®E, etc., for otherwise one would be multiplying torque by speed squared which is of
course incorrect. This is still the exact same model, it is just that some software vendors for
internal software reasons, decided to place the torque × speed calculation in the torque model.
SO PLEASE DOUBLE CHECK THIS IN THE SOFTWARE USER’S MANUAL.
For type 4 WTGs either option may be valid depending on the type of unit. For a type 4A WTG
the electrical power output of the unit is perturbed by the torsional oscillations of the turbine-
generator shaft (see [5] for example validation cases) and so PFlag can be set to 1 and the wtgt_a
model used to approximately emulate this behavior 8. For type 4B WTGs there is no appreciable
observation of electrical power perturbation due to torsional oscillations in the turbine generator
shaft, due to the converter design, and so PFlag is set to zero and no wtgt_a model used.
The rest of the parameters associated with the active power control are the maximum and
minimum power ratings of the unit (Pmax/Pmin), the maximum and minimum rate of change of
power reference (dPmax/dPmin) and the time constant associated with the controls (Tpord).
Since this model is for use with WTGs (or PV) the minimum active current command is clearly
zero (Ipmin = 0), and this is not changeable by the user.
The maximum active current command limit is determined by the current limit logic (Ipmax).
Reactive Power Control: There are several options for reactive power control. This is shown
diagrammatically in Figure 3-1. These are:
• Local constant Q control – PfFlag = 0 and QFlag = 0; VFlag = 1 or 0 (irrelevant)
• Local constant power factor (pf) control – PfFlag = 1 and QFlag = 0; VFlag = 1 or 0
(irrelevant)
• Local terminal voltage control – PfFlag = 0, VFlag = 0 and QFlag = 1
• Local coordinated Q/V control – PfFlag = 0, VFlag = 1 and QFlag = 1
In addition to all this, there is a separate proportional, with deadband, current injection control
which can be used either as proportional voltage control during a voltage dip (deadband set to
zero) or a proportional current injection with deadband during a voltage dip. To disable this

8
The wtgt_a model when used with the type 4 WTG is intended solely for emulating the observed torsional
oscillations post-fault in some type 4 equipment, where this occurs. In these cases, the mechanical power (Pm) of
the wtgt_a model is assumed to be constant. Therefore, this model should not be used for cases where the primary
frequency response feature is used or other external models that change Pref, since keeping Pm constant while Pe is
being change is not realistic.

14496639 3-4
path, Kqv can be set to zero, and Vup and Vdip set to 2 and 0, respectively to disable the voltage
dip logic (see Appendix F). The parameters Iqfrz and Thld can be used in association with this
current injection loop to create various state transitions, as shown in Figure 3-2. These state
transitions were implemented to accommodate various original equipment manufacturer (OEM)
requests during the model development process. The user should use these only as instructed by
OEMs or if he/she clearly understands their implications, as shown in the Figure 3-2. There is
also a parameter, Thld2, which when set to a non-zero value will hold the active current
command (Ipcmd) at the value it has been frozen at during a voltage dip, after the fault clears.
That is, when Voltage_dip = 1, the active current command will be frozen to a given value. If
Thld2 were set to for example 0.1 s, then once the disturbance is over and Voltage_dip =0, the
value of the active current command remains at its frozen value for another 0.1s, before being
released.
For the local voltage control option, the user-defined reference bias Vref1 is typically set to the
default value of zero (0). This bias was provided again at the request of one OEM. It should
only be used if so instructed, otherwise we advise always setting it to zero.
Thus, the various control options are summarized in the table below.
Table 3-1
Reactive power control modes for the reec_a model

Control Mode PfFlag VFlag QFlag

Local constant Q control 0 0 or 1 0
Local constant power factor (pf) contorl 1 0 or 1 0
Local voltage control 0 0 1
Local coordinated Q/V control 0 1 1

14496639 3-5
 dbd1, dbd2
_ Verr
Vt 1 Iqv
Kqv
1 + s Trv
Vt_filt

Iql1

Iqh1
s0 +
Vref0 (user defined) 2 1 0
Iqfrz 0

pfaref tan Qmax Vmax Iqmax

Vmax Iqinj Iqmax
VFlag QFlag +
1 1 + + Kqp + Kqi 1 + Kvp + Kvi 1 +
Pe s
1 + s Tp _ s Iqcmd
_
s1 Qext s2 s3 0
0 0 Vmin Iqmin
Qmin Vmin Iqmin
PfFlag
Qgen Vt_filt
(s0)
Vref1 (user defined)
+ + VDL1
1
1 + s Tiq
Current

0.01
s4
Local Q/Voltage Control Limit Pqflag
VDL2 Logic
Constant Local Voltage Control 0 – Q priority
Constant pf Control 1 – P priority
Vt_filt
Constant Q Control (s0)

Reactive Current Injection During Voltage Dips

Vt_filt
ωg (s0)

0.01
Pmax Thld2
dPmax Ipmax
PFlag
s6 1 1 Pord
Pref
1 + s Tpord Ipcmd
dPmin s5
0 Ipmin = 0
Pmin

Figure 3-1
Options for the reactive power control path in the reec_a model

Figure 3-2
Renewable energy electrical control model state transition diagram for the (reec_a)

Current Limit Logic: The current limit logic implementation is given in Appendix B. In its most
basic form the current limit is a semi-circle around quadrants 1 and 4, as shown in Figure 3-3.
That is, only positive active current is allowed (Ipmin = 0) since this is a model for a generator,
and the total current must be less than or equal to Imax. The selection of the Pqflag determines
whether priority is given to active or reactive current. The VDL1 and VDL2 tables are two look-
up tables with four pairs of numbers that define a piece-wise linear curve. These tables define
the reactive and active current limits, respectively, as a function of voltage. Therefore, in
addition to the basic current limit shown in Figure 3-3, the voltage-dependent limit (VDL) tables

14496639 3-6
can be used to effect further limits on either active or reactive current as a function of voltage.
The values of these tables need to come either directly from the OEM or based on fitting the
values from factory (or field tests) that clearly show the reactive and active power output of a
single WTG (or PV) as a function of various voltage dips. To disable these tables (or if data
is not available) then simply set all the values to Imax for four different voltage settings, e.g.
Vq1 = 0, Iq1 = Imax; Vq2 = 0.2; Iq2 = Imax; Vq3 = 0.5, Iq3 = Imax and Vq4 = 1.0, Iq4 = Imax
etc. The tables could also be used to effect limiting (reducing) reactive output at high voltage
levels. Since the original release of these models, it has become apparent that there would be
great benefit to having a significantly larger number of point in these tables (more than four). As
such, this is currently an item under discussion, to be considered in revisions of the model in the
commercial software platforms – see section 2.10. The VDL tables are critical to representing
so-called “momentary-cessation”, and in general the voltage dependent current characteristics of
the inverter, therefore for actual power plant installations it is important that these tables be
reasonably parameterized rather than being disabled or ignored.
Reactive Current
+ve
Imax

Active Current
-ve +ve

-ve

Figure 3-3
Current limit for reec_a model

REEC_B and REEC_C Models: Consider the reec_b model shown in Figure D-2 in Appendix D.
The differences between this model and the reec_a model are as follows:
1. It does not contain the VDL1 and VDL2 tables, thus the current limit logic is defined
entirely by a semi-circle as shown in Figure 3-3.
2. It does not have the parameters and functionality of the reec_a model associated with the
state-transitions around the reactive current injection loop shown in Figure 3-2.
3. The active power path cannot be modulated by speed and so this model cannot be used
with the wtgt_a model.
Other than the above points, the model is identical to reec_a. It is thus a simpler version of the
reec_a model. These changes were decided by majority vote at one of the WECC REMTF
meetings for the sake of creating a simpler alternative to the reec_a model for use for PV plants.

14496639 3-7
However, given recent developments around the so-called momentary cessation 9 issue it is not
recommended that this model be used for PV plants, but rather the reec_a model still be used
even for PV plants.
Consider the reec_c model shown in Figure D-3 in Appendix D. The differences between this
model and the reec_a model are as follows:
1. It does not have the parameters and functionality of the reec_a model associated with the
state-transitions around the reactive current injection loop shown in Figure 3-2.
2. The active power path cannot be modulated by speed and so this model cannot be used
with the wtgt_a model.
3. It contains an additional path with a simple representation for a charging/discharging
mechanism (energy storage).
4. The minimum active current (Ipmin) is equal to –Ipmax; that is, the model allows power to
be both generated and absorbed, and therefore can be used to model energy storage.
Other than the above points, the model is identical to reec_a. Let us consider in a little more
detail the additional features of this model. The additional part of the model is shown in Figure
3-4. This added feature has the following key aspects:
1. A user defined parameter which specifies the initial state of charge (SOC) of the battery.
This tells the model how much charge the battery has prior to starting the simulation.
2. A representation of the maximum and minimum allowable state of charge (shown as
SOCmax and SOCmin). Most battery OEMs recommend that the battery not be left in a
state of full-charge or full-discharge in order to preserve the battery’s longevity and
performance. The model simulates this through the user specified values for the
maximum (SOCmax) and minimum (SOCmin) allowed state of charge during operation.
Many vendors recommend operating the batteries within a range of 20% to 80% state of
charge.
3. The simple integrator block, with the time constant T, represents the process of charging
and discharging. The level of charge in the battery is proportional to stored energy.
Energy is the time integral of power since power is specified in units of watts = joules
(energy) per second. Thus, by integrating the power coming out of (or going into when
charging) the device, we get a representation of the state of charge.
4. The logic block at the end of the model represents the action of collapsing the output of
the converter (i.e. forcing its active current output) to zero once the maximum or
minimum state of charge has been reached. So, for example, if the SOC is greater than
the allowable SOCmax, then Ipmin is forced to zero, meaning that the battery cannot
absorb/store any more electrical energy.
Consider a simple example of how the reec_c model might be parameterized to represent a
BESS. Assume we have a BESS that is rated at 40 MVA, with an energy rating of 30 MW for 4
hours (120MWh). Also, let us assume that when in operation the BESS is required by the

9
https://www.nerc.com/comm/PC/NERCModelingNotifications/Modeling_Notification_-
_Modeling_Momentary_Cessation_-_2018-02-27.pdf

14496639 3-8
vendor to always be in a state of charge between 20% to 80%, with the same charging rate
(i.e. 4 hours).
Then,
SOCmax = 0.8
SOCmin = 0.2
The total energy of the device = 30×4 = 120 MWh, thus in operation it can go from 0.8×120
(96 MWh) to 0.2×120 (24 MWh), which means that the maximum output would be (96 – 24)/4 =
18 MWh for 4 hours.
Therefore,
T = ((18/30) × (60×60×4)) / (0.8 – 0.2) = 14,400
Pmax = 18/30 = 0.6
Pmin = - Pmax = -0.6
Imax = 40/30 = 1.33
The model MVA = 30 MVA. All other parameters would be set per the OEM data.

Figure 3-4
Extra part of the reec_c model for simulating charging and discharging of a storage mechanism

A more detailed account of modeling battery energy storage with these generic models, and a
comparison of the performance of the generic models to detailed proprietary 3-phase models can
be found in [23].

3.4 The Renewable Energy Plant Controller Model

There are two renewable energy plant controller models: repc_a and repc_b.
The renewable energy plant controller model a (repc_a) is shown in Figure 3-5. The model
connects to the reec_a, reec_b or reec_c models. There are two separate and independent paths
in the model, the reactive power control path that culminates in the variable Qext, and the active
power control path that culminates in the variable Pref. These variables, Qext and Pref, then
connect to the reec_* models and appropriately adjust the inputs to those models. If PfFlag is set
to 1 in the reec_* models to effect local power factor control, then the output of the repc_a
model does not in any way influence the reactive power of the reec_* model.
Now consider the reactive power control path of the repc_a model. There are two general
options:
1. Voltage control: by setting RefFlag = 1 the voltage at a remote bus (Vreg) can be
regulated, typically the voltage at the point of common coupling, which is commonly

14496639 3-9
 either the high- or low-voltage side of the plants substation transformer. Furthermore,
either line drop compensation (using Rc, Xc) can be used with VcompFlag set to 1, or
reactive droop (Kc) can be used with VcompFlag set to 0.
2. Constant Q control: by setting RefFlag = 0 the reactive power through a branch can be
controlled, typically the reactive power through the substation transformer which
represents the reactive power output of the plant.
The proper selection of the deadband (dbd), input (emax/emin) and output (Qmax/Qmin) limits
and gains (Kp/Ki) of this controller is critical to having stable and proper operation of the
controls. The time constants Tft and Tfv can be used to represent any intentional phase lead (Tft)
– typically none – or lag/delay (Tfv) in the communication process between the plant controller
and the turbines. The table below provides a summary of the reactive power control
possibilities.
IMPORTANT NOTE: depending on the settings within the reec_* model down stream of the
plant controller, Qext (the output of repc_a) can be either a “Q-command” or “Voltage-
command”. Therefore, the values of Qmax/Qmin must be set appropriately to respect the
nature of the output signal. For example, if in the downstream reec_* model Pfflag = 0, Vflag
=0 and Qflag = 1 (see Table 3-1) then Qext will be a voltage-set-point and so Qmx/Qmin in the
repc_a model need to be set to values such as 1.1/0.9 (maximum and minimum voltage ste-
point values). While, if Pfflag = 0, Vflag = 1 or 0, and Qflag = 0 (see Table 3-1) then Qext will
be a Q-reference and so Qmx/Qmin in the repc_a model need to be set to values such as 0.3/-
0.3 (maximum and minimum Q-reference).
Table 3-2
Reactive power control modes for the reec_* + repc_a models

reec_* model repc_a model

Control Mode PfFlag VFlag QFlag RefFlag
Plant level Q control 0 0 or 1 0 0
Plant level Vcontrol 0 0 or 1 0 1
Plant level V Control + coordinated local Q/V control 0 1 1 1
Plant level Q Control + coordinated local Q/V control 0 1 1 0
The active power control loop can be used to simulate primary frequency response. This loop
can be enabled by setting Freq_flag = 1, or disabled by setting the flag to 0. The upward (Dup)
and downward (Ddn) regulation droop settings can be different, as well as the deadband on either
side. In addition, a plant power reference Plant_ref is accessible by the user, which can be used
to ramp the plant or controlled by other external models (e.g. AGC).

14496639 3-10
 Freeze state s2 if Vreg < Vfrz

VcompFlag
Vref Reactive Power
+
Vreg
(from bus vbus)
|Vreg – (Rc+jXc)Ibranch|
1 1
_ Control Path
Ibranch 1 + s Tfltr
(current through a define branch, 0 s0
e.g. substation transformer) +
Qbranch Qmax
Kc +
(reactive power through a define branch, e.g. dbd emax
substation transformer)
1 Voltage Control
Kp + Ki 1 + s Tft
1 _ 1 + s Tfv Qext
Qbranch s
(from aggregate 1 + s Tfltr 0 Constant Q Control
turbine model or
emin s2 s3
collection point of s1 + RefFlag
wind plant) Qmin
Qref

Plant_pref
Pmax Freq_flag
+ femax 0
Pbranch 1 _
Kpg + Kig 1 1 Pref
(from aggregate turbine model terminal
or collection point of wind plant)
1 + s Tp s 1 + s Tlag
s4 + femin s5 s6

Ddn
0 Pmin
Active Power
fdbd1,fdbd2
_
+
Control Path
Freq
(from aggregate turbine model terminal
or collection point of wind plant)
+ +
Dup
Freq_ref 0

Figure 3-5
The plant controller model repc_a
Warning: Care must be taken not to simulate up-regulation (i.e. increasing plant output with
decreasing frequency) where it is not physically meaningful – e.g. when the plant is converting
the available incident wind energy (or solar energy for PV) to electrical power, which is
certainly the typical operating condition of a wind power plant in most of North America
today. In these cases, the primary frequency response (PFR) should be disabled in simulation
or Dup or Pmax, or other appropriate parameters, adjusted to not allow for PFR. Our
recommendation is to set Pmax equal to the initial output of the plant model, from power flow,
which would then not allow further increase of the plant output for under-frequency events,
which better emulates reality.
Important Note: The actual implementation of these models in software may require subtle
adjustment to accommodate the way the models need to be initialized in commercial tools. So,
the user should also carefully read the respective commercial software user’s manual for the
usage of the repc_a model.
The plant controller model b (repc_b) is shown in Figure 3-6 [8]. The differences between
model b an a are high-lighted in red. These differences are as follows:
1. The repc_b model allows also for power factor control of the plant. By setting RefFlag
to 2, the user may select a constant power factor control mode. Upon initialization, the
model sets the parameter pfaref = arctan(Qbranch/Pbranch) (i.e. the initial power factor
angle from the power flow solution). In hind-sight, this feature should also have been
made available perhaps in repc_a, but it is not. Consideration, among other things, are
presently been given to new modular components that may include updates such as this to
the repc_a model.
2. Three auxiliary inputs (Vaux, Qaux and Paux) have been added, which should be
accessible to the user and always initialized to zero. These can be used for adding user-

14496639 3-11
 written components by users for introducing any specialized controllers that may be
unique to specific installations.
3. The output of the model can link to up to fifty (50) other dynamic devices to allow for
coordinate control of multiple active devices in a plant. A few pertinent comments are
warranted about these outputs. For a detailed account, including initialization of the
model, see reference [8].
a. For each leg (reactive and active power) the output can interface with up to fifty
(50) different devices. Each pair of outputs, e.g. Po1 and Wo1 go to the same
device (defined by a bus number and device id). When the output is going to a
reactive device (e.g. SVC, STATCOM or synchronous condenser) then Poi is
ignored and Kzi is set to 0. The reactive outputs are designated “Woi” since they
can be either a reactive power bias (e.g. going into an aggregate wind/PV unit so
configured or a voltage reference bias (e.g. going into a SVC, STATCOM or
exciter summing junction).
b. The model’s Woi outputs can connect to Qref on reec_a, reec_b and reec_c, or the
Vsig of svsmo1, svsmo2 and svsmo3, or Vsig on exciter models on esst1a, esst4b,
esac7b and esac8b on a synchronous condenser.
c. The model’s Poi outputs can connect to Pref on reec_a, reec_b and reec_c.
d. It is assumed that the time lag associated with the signals is the same for the
reactive and active power signals. Therefore, one set of up to fifty-time constants
is needed (Tw1 to Tw50).
e. The weights Kw1 to Kw50 and Kz1 to Kz50 are not normalized – the user may of
course normalize them him/herself prior to entering the numbers. These gains are
not renormalized, changed or reconfigured automatically by the model after
initialization, for example if during a simulation a device trips and thus one of the
outputs is no longer controlling anything, then it simply floats, and the remaining
gains also remain unchanged. In practice some vendors may renormalize the gains
and set the gain of the tripped device to zero, but this is too complicated for the
present intended model.
It is our understanding, that the implementation of this model is slightly different in one of the
commercial tools (Siemens PTI PSS®E) such that in that tool, the commercial vendor decided to
split the core plant controller and the lag blocks (shown in red). Thus, the number of
controllable devices is not limited to fifty. Also, clearly the user would need to instantiate
several models to implement the so-called repc_b model. Thus, users should also closely consult
the user’s manual of the commercial tool they are using to understand the instantiation of the
model.

14496639 3-12
 Freeze state s2 if Vreg < Vfrz

VcompFla Vref Vaux

Vreg g +
(from bus vbus) _ +
Ibranch 1 1
|Vreg – (Rc+jXc)Ibranch|
(current through a define branch, e.g. 1 + s Tfltr
0 s0
substation transformer) +
Qmax
Qbranch Kc + Qaux
dbd emax
(reactive power through a define branch, + 1 Wext
Kp + Ki 1 + s Tft
e.g. substation transformer) 1 _
Qbranch s 1 + s Tfv
(from aggregate turbine model 1 + s Tfltr 0 emin s2 s3
or collection point of wind plant) s1 + RefFlag
2 Qmin
Qref

pfaref tan _

wref1 + + +
wref2 …….. wref50

Plant_pref Paux
Pmax Freq_flag
Pbranch +
+ femax 0
(from aggregate 1 _ Kw1 Kw2 Kw50
Kpg + Kig 1 1
turbine model terminal Pext
1 + s Tp s 1 + s Tlag
or collection point of +
s4 femin s5 s6
wind plant)
0 Pmin
fdbd1,fdbd2 Ddn 1 1 1
Freq +
(from aggregate _ 1 + sTw1 1 + sTw2 1 + sTw50
turbine model terminal
or collection point of +
+ Dup Wo1 Wo2 Wo50
wind plant)
Freq_ref (BUS1, ID1) (BUS2, ID2) (BUS50, ID50)
0

+ + +
pref1 pref2 …….. pref50

Kz1 Kz2 Kz50

1 1 1
1 + sTw1 1 + sTw2 1 + sTw50

Po1 Po2 Po50

(BUS1, ID1) (BUS2, ID2) (BUS50, ID50)

Figure 3-6
The plant controller model repc_b

3.5 Low and High Voltage and Frequency Rde-Through

The protection models associated with the wind turbine generator (i.e. low/high voltage and
low/high frequency tripping) have not been addressed in this document since the existing generic
protection models (lhvrt and lhfrt) that exist in GE PSLF (and similar models in Siemens PTI
PSS®E, PowerWorld Simulator, PowerTech Labs TSAT™ etc.) are adequate for application
with these generic models.

14496639 3-13
14496639
4
VALIDATION OF THE MODELS
The intent of this document is to present a guide on the usage of these models and not the
process of validation. NERC is presently undertaking, in a broad collaborative way, in preparing
a guide on model validation of RES models. This report should be publicly available in the next
few months.
Many of the reference documents provided in section 6 provide examples of simulations
performed by EPRI using the EPRI validation software tools for single wind turbine generators
and in one case a wind power plant with these new generic models. Furthermore, the models
have been used by solar energy developers and a few wind turbine OEMs and validations shown
and reported at various WECC and IEEE meetings. EPRI has also used the model in validating
PV inverters with data from one OEM. All this work has shown the models to be useful and
applicable for large scale interconnected power system stability studies. The reader can consult
references [1], [2], [3], [4], [5], [7] and [18] which show many such validation results.
Since the first version of this report was released in 2015, much recent experience has been
gained with performing model validation of large wind and PV power plants using these generic
models, including validation of primary frequency response. Examples of such results may be
found in references [19], [20], [21], [22] and [23].
There are certainly additional improvements to be made to the models to capture other
functionalities of the ever-evolving renewable energy technologies. However, given the modular
format of these models this should hopefully be an achievable task for continued research and
development.

14496639 4-1
14496639
5
CONCLUSION AND SUMMARY
This document provides a concise guide to the use of the second-generation generic, and public,
renewable energy system models for use in positive sequence stability simulation programs.
Further work continues to be done on adding new features and extending some of the existing
features of the models, as briefly discussed in section 2.10. Thus, it is anticipated that this
document may need to be revised again in due course.

14496639 5-1
14496639
6
REFERENCES
[1] WECC Second Generation Wind Turbine Models, January 23, 2014
https://www.wecc.biz/Reliability/WECC-Second-Generation-Wind-Turbine-Models-
012314.pdf
[2] Technical Update: Generic Models and Model Validation for Wind Turbine Generators and
Photovoltaic Generation, EPRI, Palo Alto, CA, 2013, Product ID: 3002001002
(http://www.epri.com/abstracts/Pages/ProductAbstract.aspx?ProductId=00000000300200100
2)
[3] Generic Models and Model Validation for Wind and Solar PV Generation: Technical Update.
EPRI, Palo Alto, CA: 2011, 1021763.
(http://www.epri.com/abstracts/Pages/ProductAbstract.aspx?ProductId=00000000000102176
3)
[4] P. Pourbeik, “Proposed Changes to the WECC WT3 Generic Model for Type 3 Wind
Turbine Generators”, Prepared under Subcontract No. NFT-1-11342-01 with NREL, Issued:
03/26/12 (revised 6/11/12, 7/3/12, 8/16/12, 8/17/12, 8/29/12, 1/15/13, 1/23/13, 9/27/13)
https://www.wecc.biz/Reliability/WECC-Type-3-Wind-Turbine-Generator-Model-Phase-II-
012314.pdf
[5] P. Pourbeik, “Proposed Changes to the WECC WT4 Generic Model for Type 4 Wind
Turbine Generators”, Prepared under Subcontract No. NFT-1-11342-01 with NREL, Issued:
12/16/11 (revised 3/21/12, 4/13/12, 6/19/12, 7/3/12, 8/16/12, 8/17/12, 8/29/12, 1/15/13,
1/23/13) https://www.wecc.biz/Reliability/WECC-Type-4-Wind-Turbine-Generator-Model-
Phase-II-012313.pdf
[6] P. Pourbeik, “Proposed Changes to the WECC WT1 Generic Model for Type 1 Wind
Turbine Generators”, Prepared under Subcontract No. NFT-1-11342-01 with NREL, Issued:
1/21/13
[7] P. Pourbeik, “Simple Model Specification for Battery Energy Storage System”, memo issued
to WECC REMTF, MVWG and EPRI 173.003; 1/15/15 (REVISED 2/25/15; 3/6/15;
3/18/15) https://www.wecc.biz/Reliability/REEC_C_031815_rev3%20Model%20Spec.pdf
[8] P. Pourbeik, “Model Specification for High-Level Plant Controller”, memo issued to WECC
REMTF, MVWG and EPRI 173.003; 11/25/14 (REVISED 1/6/15; 1/20/15; 11/5/15)
https://www.wecc.biz/Reliability/Memo-REPC-B-110515.pdf
[9] Generic Static Var System Models for the Western Electricity Coordinating Council, April
18, 2011 https://www.wecc.biz/Reliability/WECC-Static-Var-System-Modeling-Aug-
2011.pdf
[10] J. Brochu, C. Larose and R. Gagnon, “Validation of single- and multiple-machine
equivalents for modeling wind power plants,” IEEE Transactions on Energy Conversion,
Vol. 26, No. 2, June 2011, pp 532-541.

14496639 6-1
[11] IEEE Task Force Report, Blackout Experiences and Lessons, Best Practices for System
Dynamic Performance, and the Role of New Technologies, IEEE Special Publication
07TP190, July 2007 (http://resourcecenter.ieee-pes.org/publications/blackout-experiences-
lessons-best-practices-system-dynamic-performance-role-new-technologies-pdf/)
[12] E. Muljadi, C. P. Butterfield, A. Ellis, J. Mechenbier, J. Hochheimer, R.Young, N. Miller,
R. Delmerico, R. Zavadil and J.C. Smith, “Equivalencing the collector system of a large wind
power plant,” presented at the IEEE Power Engineering Society General Meeting, Montreal,
QC, June 2006.
[13] WECC Wind Plant Dynamic Modeling Guidelines, May 2014
https://www.wecc.biz/Reliability/WECC%20Wind%20Plant%20Dynamic%20Modeling%20
Guidelines.pdf
[14] CIGRE Technical Brochure 328, Modeling and Dynamic Behavior of Wind Generation as it
Relates to Power System Control and Dynamic Performance, August 2007 (http://www.e-
cigre.org/Search/download.asp?ID=328.pdf)
[15] W. W. Price and J.J. Sanchez-Gasca, “Simplified Wind Turbine Generator Aerodynamic
Models for Transient Stability Studies” Proc. IEEE PES 2006 Power Systems Conference
and Exposition (PSCE), Oct. 29-Nov. 1, 2006, Atlanta, GA, pp. 986-992
[16] WECC Solar Plant Dynamic Modeling Guidelines, April 2014
https://www.wecc.biz/Reliability/WECC%20Solar%20Plant%20Dynamic%20Modeling%20
Guidelines.pdf
[17] Pseudo Governor Model for Type 1 and 2 Generic Turbines, Report by Enernex, October
2012.
[18] Technical Update on Generic Wind and Solar PV Model Development and Validation,
Product ID: 3002003351, EPRI, Palo Alto, CA, December 2014
http://www.epri.com/abstracts/Pages/ProductAbstract.aspx?ProductId=00000000300200335
1
[19] M. P. Richwine, J. J. Sanchez-Gasca and N. W. Miller, “Validation of a second-generation
Type 3 generic wind model”, Proceedings of the IEEE PES General Meeting, 2014.
http://ieeexplore.ieee.org/document/6939017/
[20] P. Pourbeik, N. Etzel and S. Wang, “Model Validation of Large Wind Power Plants
Through Field Testing”, to be published in the IEEE Transactions on Sustainable Energy,
2018 (http://ieeexplore.ieee.org/document/8118170/)
[21] P. Pourbeik, S. Soni, A. Gaikwad and V. Chadliev, “Providing Primary Frequency Response
from Photovoltaic Power Plants”, CIGRE Symposium 2017, Dublin, Ireland, May 2017.
[22] P. Pourbeik, J. Sanchez-Gasca, J. Senthil, J. Weber, P. Zadehkhost, Y. Kazachkov, S. Tacke
and J. Wen, “Generic Dynamic Models for Modeling Wind Power Plants and other
Renewable Technologies in Large Scale Power System Studies”, IEEE Trans. on Energy
Conversion, September 2017, http://ieeexplore.ieee.org/document/7782402/
[23] P. Pourbeik and J. K. Petter, “Modeling and validation of battery energy storage systems
using simple generic models for power system stability studies”, CIGRE Science and
Engineering, October 2017, pp. 63-72 (available for free download at https://e-cigre.org/).

14496639 6-2
[24] EPRI Report, Renewable Energy Systems Modeling Considerations for Weak Grid Studies:
Limitations of Positive-Sequence Models in Comparison with Three-Phase Models, Product
Id: 3002010928, Date Published: Dec 06, 2017
https://www.epri.com/#/pages/product/3002010928/
[25] Proposed REGC_B Model, March 2017, WECC REMTF & MVWG Meeting
https://www.wecc.biz/Administrative/REMTF_REGC_A_and_REGC_B_0317.pdf
[26] D. Ramasubramanian, Z. Yu, R. Ayyanar, V. Vittal and J. M. Undrill, “Converter Model for
Representing Converter Interfaced Generation in Large Scale Grid Simulations”, IEEE
Trans. PWRS, April 2016.
[27] Ö. Göksu, P. Sørensen, J. Fortmann, A. Morales, S. Weigel, P. Pourbeik, “Compatibility of
IEC 61400-27-1 Ed 1 and WECC 2nd Generation Wind Turbine Models”, Conference: 15th
International Workshop on Large-Scale Integration of Wind Power into Power Systems as
well as on Transmission Networks for Offshore Wind Power Plants, November 2016.
[28] D. Ramasubramanian, E. Farantatos, A. Gaikwad and W. Wang, “Proposal for Improved
Voltage Source Converter Model – An Update”, Presented at WECC MVWG Meeting May
10, 2018.
https://www.wecc.biz/Administrative/08%20Proposed%20Improved%20Voltaged%20Conve
rter%20Model-%20Ramasubramanian.pdf
[29] P. Pourbeik, “Discussion on Improvements and Additions to the 2nd Generation Generic
RES Models”, Presented at WECC MVWG Meeting May 10, 2018.
https://www.wecc.biz/Administrative/08%202nd%20Generation%20Generic%20RES%20M
odels-%20Pourbeik.pdf

14496639 6-3
14496639
A
MODEL NAMES IN THE MAJOR SOFTWARE
PLATFORMS USED IN NORTH AMERICA
The models presented in this report have slightly different names in the various commercial
software platforms simply because of the inherent naming conventions used by the software
vendors. The table below gives a cross-reference for the model names in the four most
commonly used tools in North America. The table below is current as of February 2018.
Table A-1
Model names across four main commercial software tools used in North America

Model Model Name in

Model Name in the Model Name in PowerWorld PowerTech Labs
TM TM
Specification Document GE PSLF Model Name in Siemens PTI PSS®E Simulator TSAT
New Models (developed 2011 - present)
REGC_A regc_a REGCAU1 (V33); REGCA1 (V34) REGC_A REGC_A
REEC_A reec_a REECAU1 (V33); REECA1 (V34) REEC_A REEC_A
REEC_B reec_b REECBU1 (V33); REECB1 (V34) REEC_B supported*
REEC_C reec_c REECCU1 (V33 & V34) REEC_C supported*
REPC_A repc_a REPCTAU1 & REPCAU1 (V33); REPCTA1 & REPCA1 (V34) REPC_A REPC_A
WTGT_A wtgt_a WTDTAU1 (V33); WTDTA1 (V34) WTGT_A WTGT_A
WTGAR_A wtga_a WTARAU1 (V33); WTARA1 (V34) WTGA_A WTGA_A
WTGPT_A wtgp_a WTPTAU1 (V33); WTPTA1 (V34) WTGP_A WTGP_A
WTGTRQ_A wtgq_a WTTQAU1 (V33); WTTQA1 (V34) WTGTRQ_A WTGQ_A
WT1P_B wt1p_b not yet part of the standard model library WT1P_B not yet supported
REPC_B repc_b PLNTBU1 (V33 & V34) REPC_B supported**
Existing Models (developed prior to 2009)
WT1G wt1g WT1G1 WT1G & WT1G1 WGNA
WT2G wt2g WT2G1 WT2G & WT2G1 WGNB
WT2E wt2e WT2E1 WT2E & WT2E1 WGNAE
LHVRT lhvrt VTGTPAT (trips machine), VTGDCAT (disconnects bus) LHVRT supported*
LHFRT lhfrt FRQTPAT (trips machine), FRQDCAT (disconnects bus) LHFRT supported*
*supports PSS®E and PSLF formats
** supports PSLF format

Note: In the TSATTM column for some of the models it is said supported and no model name is given. As
explained by the software vendors there are no equivalent models in DSAToolsTM (or TSATTM) format.
However, the mentioned models are supported. That is, if user’s dynamic data are in GE PSLF or
Siemens PTI PSS®E format, TSATTM recognizes the said models and uses them in the simulation.
Contact the specific software vendor for more details.

14496639 A-1
14496639
B
CURRENT LIMIT LOGIC
VDL1 is a piecewise linear curve define by four pairs of numbers:
{(vq1, Iq1), (vq2, Iq2), (vq2, Iq3), (vq4, Iq4),}

VDL2 is a piecewise linear curve define by four pairs of numbers:

{(vp1, Ip1), (vp2, Ip2), (vp2, Ip3), (vp4, Ip4),}

If (Pqflag = 0) Q – priority
Iqmax = min {VDL1, Imax}
Iqmin = -1×Iqmax
Ipmax = min{VDL2, Imax 2 − Iqcmd 2 )
Ipmin = 0
Else P – priority
Iqmax = min {VDL1, Imax 2 − Ipcmd 2 }
Iqmin = -1×Iqmax
Ipmax = min{VDL2, Imax)
Ipmin = 0
End

14496639 B-1
14496639
C
BLOCK DIAGRAMS FOR THE WIND TURBINE
RELATED MODELS

Figure C-1
Wind turbine generator drive-train model (wtgt_a)

Figure C-2
Wind turbine generator aero-dynamic model (wtgar_a)

14496639 C-1
Figure C-3
Wind turbine generator pitch-controller model (wtgpt_a)

14496639 C-2
 Freeze State Upon
Voltage Dip

ωg
(from drive-train model)

Prefo Temax

+ Tflag Kip Temax

_ 1 s
1
Pe +
1 + s Tp s2
+ Pref (to reec_a model)
s1 0 Temin
+
ωref _ Kpp
Temin
1
f(Pe)
1 + s Tωref
s0

(p4,spd4)

(p3,spd3)
f(Pe)
(p2,spd2)

(p1,spd1)

speed

Figure C-4
Wind turbine generation torque model (wtgtrq_a)

14496639 C-3
14496639
D
BLOCK DIAGRAMS FOR RE*** MODELS
State Transition – switch position:
State 0 - If Voltage_dip = 0; normal operation (Iqinj = 0)
State 1 - If Voltage_dip = 1; Iqinj goes to position 1
State 2 - If Thld > 0, then after voltage_dip goes back to
Warning!! zero, set value to Iqfrz for t= Thld, after which go
Extreme care should be taken in coordinating back to state 0.
- If Thld < 0, then after voltage_dip returns to zero
the parameters dbd1, dbd2 and Vdip, Vup so
stay in State 1 for t = Thld, after which go back to
as not to have an unintentional response state 0.
from the reactive power injection control loop.
dbd1, dbd2
_ Verr
If (Vt < Vdip) or (Vt > Vup) then Vt 1 Iqv
Kqv
Voltage_dip = 1 1 + s Trv
Vt_filt

Iql1

Iqh1
else
s0 +
Voltage_dip =0 1 0
Vref0 (user defined) 2
Iqfrz 0
Freeze State if Voltage_dip = 1

pfaref tan Qmax Vmax Iqmax

Vmax Iqinj Iqmax
VFlag QFlag +
1 1 + + Kqp + Kqi 1 + Kvp + Kvi 1 +
Pe s
1 + s Tp _ s Iqcmd
_
s1 Qext s2 s3 0
0 0 Vmin Iqmin
Qmin Vmin Iqmin
(Qext is initialized to a PfFlag
constant, or can be Qgen Vt_filt
connected to an external (s0)
model, e.g. wppc) Vref1 (user defined)
+ + VDL1
1
1 + s Tiq
Current
0.01

s4
Limit Pqflag
VDL2 Logic 0 – Q priority
1 – P priority
Vt_filt
(s0)

Freeze State if Voltage_dip = 1

Vt_filt
ωg (s0)

0.01
Pmax Thld2
dPmax Ipmax
PFlag
(Pref is initialized to a
constant, or can be s6 1 1 Pord
Pref
connected to an external 1 + s Tpord Ipcmd
model)
dPmin s5
0 Ipmin = 0
Pmin

Figure D-1
Renewable energy electrical controls model a (reec_a)

14496639 D-1
Warning!!
Extreme care should be taken in coordinating
the parameters dbd1, dbd2 and Vdip, Vup so
as not to have an unintentional response
from the reactive power injection control loop.
dbd1, dbd2
_ Verr
If (Vt < Vdip) or (Vt > Vup) then Vt 1
Kqv
Voltage_dip = 1 1 + s Trv
Vt_filt
else
s0 +
Voltage_dip =0
Vref0 (user defined)
Freeze State if Voltage_dip = 1

pfaref tan Qmax Vmax Iqmax

Vmax Iqinj Iqmax
VFlag QFlag
1 1 + + Kqp + Kqi 1 + Kvp + Kvi 1 +
Pe s
1 + s Tp _ s Iqcmd
_
s1 Qext s2 s3 0
0 0 Vmin Iqmin
Qmin Vmin Iqmin
(Qext is initialized to a PfFlag
constant, or can be Qgen Vt_filt
connected to an external (s0)
model, e.g. wppc) Vref1 (user defined)
+ +
1
1 + s Tiq
Current

0.01
s4
Limit Pqflag
Logic 0 – Q priority
1 – P priority
Vt_filt
(s0)

Freeze State if Voltage_dip = 1

Vt_filt
(s0)

0.01
Pmax
dPmax Ipmax
(Pref is initialized to a
constant, or can be s6 1 Pord
Pref
connected to an external 1 + s Tpord Ipcmd
model)
dPmin s5
Ipmin = 0
Pmin

Figure D-2
Renewable energy electrical controls model B (reec_b)

14496639 D-2
Warning!!
Extreme care should be taken in coordinating
the parameters dbd1, dbd2 and Vdip, Vup so
as not to have an unintentional response
from the reactive power injection control loop.
dbd1, dbd2
_ Verr
If (Vt < Vdip) or (Vt > Vup) then Vt 1 Iqv
Kqv
Voltage_dip = 1 1 + s Trv
Vt_filt

Iql1

Iqh1
else
s0 +
Voltage_dip =0
Vref0 (user defined)
Freeze State if Voltage_dip = 1

pfaref tan Qmax Vmax Iqmax

Vmax Iqinj Iqmax
VFlag QFlag +
1 1 + + Kqp + Kqi 1 + Kvp + Kvi 1 +
Pe s
1 + s Tp _ s Iqcmd
_
s1 Qext s2 s3 0
0 0 Vmin Iqmin
Qmin Vmin Iqmin
(Qext is initialized to a PfFlag
constant, or can be Qgen Vt_filt
connected to an external (s0)
model, e.g. wppc) VDL1
1
1 + s Tiq
Current

0.01
s4
Limit Pqflag
VDL2 Logic 0 – Q priority
1 – P priority
Vt_filt
(s0)

Freeze State if Voltage_dip = 1

Vt_filt
(s0)

0.01
Pmax
dPmax Ipmax
(Pref is initialized to a
constant, or can be s6 1 Pord + Ipcmd
Pref
connected to an external 1 + s Tpord
model)
s5 +
dPmin Ipmin
Pmin Paux

SOCmax
_ If SOC >= SOCmax
1 SOC Ipmin = 0
Pgen
T.s elseif SOC <= SOCmin
+ Ipmax = 0
s7
SOCmin
SOCini

Figure D-3
Renewable energy electrical controls model C (reec_c)

14496639 D-3
 Rate limits on reactive current for recovery after fault.
Upward limit is active when Qgeno > 0
Downward limit is active Qgeno < 0
Qgeno

Iqrmax

Iqcmd
-1 Iq
1 + s Tg
High Voltage
s0 Vt
Iqrmin Reactive Current
Management

Interface to Network Model

LVPL & rrpwr

Ipcmd
1 Ip
1 + s Tg
s1

LVPL Low Voltage

Lvplsw Active Current
0 Lvpl1
1
Management
1 + s Tfltr
1
s2
Zerox Brkpt V

Figure D-4
Renewable energy generator/converter model A (regc_a)

14496639 D-4
E
CONVERSION BETWEEN THE 1ST AND 2ND
GENERATION WIND TURBINE GENERATOR MODELS
E.1 Type 1 and Type 2WTG
For the type 1 WTG, the wt1g and wt1t models are still valid. The only change is to remove and
replace the wt1p model with the wt1p_b model. The wt1p and wtp1_b models are incompatible.
The reason is that the wt1p model was found, in early 2010 or so, to be deficient and not able to
properly represent certain aspects of type 1 WTG controls. Thus, wt1p_b was developed as a
simple and generic emulation of the general behavior of the pitch-control functionality used in
type 1 WTGs with active-stall control (see [6], [17] and [14]).
If a type 1 or 2 WTG model is to be converted to the newer generation models and no data exists
for using the wt1p_b model, it is suggested that at minimum the wt1p model be removed and no
turbine controls be modeled. This may give conservative results, but it is certainly less likely to
cause erroneous or optimistic results that may otherwise be yielded using the wt1p model.
Furthermore, if it is known that the type 1 WTG being modeled is a stall regulated unit (i.e. with
fix blades) then most certainly no turbine controls should be modeled at all since none exist on
the actual WTG.
Although this has been explained elsewhere (e.g. [6]) for the sake of completeness, we will
present a brief explanation here of the above statements. Figure E-1 shows the block diagram of
the old wt1p so-called pseudo-governor model. As can be seen the model changes mechanical
power based on changes in speed from the system reference frequency and the machines
electrical power from the initial power reference. The issues with this model are therefore
twofold:
• For cases where system frequency events are simulated the change in the slip-speed of the
unit due to system frequency variations can results in a governor type action from the model.
This was observed in some simulations in WECC and noted as an issue, since the type 1 and
2 turbines do not generally provide primary frequency response.
• Many type 1 OEMs provide functionality that quickly ramps down mechanical power when a
nearby severe voltage dip (e.g. electrical fault) is detected in order to help with the low-
voltage ride-through of the unit (see [14], Figure 2-12). This is not represented by this older
model.
In contrast the simpler and newer model shown in Figure E-2, overcomes these concerns. First,
it does not exhibit the unexpected behavior for frequency event simulations. Secondly, it
provides for a rather simple emulation of the mechanical power ramp-down for nearby faults.
See [6] and [17] for a more detailed explanation. In brief terms, this model emulates the fact that
for many type 1WTGs with active-pitch control, the turbine controls will ramp down mechanical
power when a nearby fault is observed, as described above. The model emulates this by looking
at filtered terminal voltage (filter time constant Tr). If the voltage falls below specific set-points
(see [1] of the applicable software user’s manual for a full listing of the model parameters for the
look-up table F(Vt)), then Flag1 is changed by the model from 0 to 1. This then initiates the

14496639 E-1
ramping down of the mechanical power at a rate of rmin to the Pmin. This occurs for the amount
of time determined from the look-up table F(Vt) 10. Then once the timer times-out, Flag1
switches back to position 0, and the mechanical power ramps back up to the initial value Po, at a
rate of rmax.
ωref
_

ω + Kw

pimax
_
Kp + Ki 1pimin 1
Pmech
s 1 + sT1 1 + sT2
+
pimin

1 _
Pe
Kdroop
1 + sTpe
+

Pref

Figure E-1
Model wt1p which was part of the 1st generation generic WTG models and is no longer
recommended for use.

Figure E-2
Model wt1p_b, which is the new pitch-controller for the type 1 and 2 WTGs and the recommended
model for use with these turbines.

10
Briefly, the look-up table consists of four (4) voltage levels (v1, v2, v3, v4) and four-time settings (t1, t2, t3
and t4). If Vt falls below v1, then the time is set to t1. If it is below v2 and above v1, it is set to t2, and so on.

14496639 E-2
Summary:
• Continue to use wt1g, wt1t for type 1 WTGs, and wt2g, wt2e, and wt2t for type 2 WTGs.
• Do not use the wt1p and wt2p models – removed them from your data sets.
• Use the wt1p_b model, where data is available, for representing the behavior of the pitch-
controller.
• For “stall” type 1 WTGs do not use any pitch-controller/pseudo-governor model at all.
Disclaimer: This report, and the organization/author that produced this report, make no
guarantees or warranties, expressed or implied, with respect to the accuracy or applicability of
the proposed conversion from the old to the new models for the type 1 and 2 WTGs. In all
cases the best approach is to consult the equipment vendor to come up with appropriate
parameters for the 2nd generation models to represent existing wind power plants.

E.2 Type 3 WTG

The following tables show how to convert the old (1st generation) generic stability models for
type 3 WTGs to the new (2nd generation) models. The 1st generation models are a subset of the
more general 2nd generation models, with a few exceptions:
1. The older model did not have a current limit on the output; this is estimated based on the
flux limit.
2. The older model had the ability to bypass the second integrator in the reactive control
path (i.e. set vltflg = 0 in the old model). This is not available in the new reec_a model,
since at the time of developing the reec_a model the consensus was that this option is
rarely if ever used.
3. The new RES suite of models were developed through a collaborative effort and so a
single central model specification was developed [1]; this meant that the new models are
essentially identical among the commercial software platforms that have adopted them.
Much effort was made to ensure a one to one correspondence in the parameter lists and to
compare simulations across the platforms. This is not necessarily true of the older (1st
generation) models. As such, the user will quickly notice significant differences between
the implementation of the 1st generation models across some of the software platforms.
We clearly cannot address these issues now, since we neither had an influence on the
initial development of the older (1st generation) models, nor is it wise to try to address the
differences now when those models are to be in time replaced by the newer models.
Therefore, with the above in mind, the conversion tables developed here are based on the GE
PSLF implementation of the 1st generation (older) generic models. The response of the
converted model will not be exactly the same due to the differences explained above. In
particular, if a two-mass shaft model is used Dshaft may have to be slightly modified in the 2nd
generation model to yield the same level of damping.
The reader should also remember that the 2nd generation models were developed and validated
against multiple measured field response of individual turbines from various OEMs (see [1], [2],
[3] and [4]).

14496639 E-3
Disclaimer: This report, and the organization/author that produced this report, make no
guarantees or warranties, expressed or implied, with respect to the accuracy or applicability of
the conversion tables below. It should also be noted that the new 2nd generation models were
developed with the expressed intention of making them more flexible to allow modeling of a
larger variety of equipment. With that in mind where older (1st generation) models exist in
dynamic databases to represent existing equipment, they may not necessarily have been very
representative of the equipment performance (particularly for none GE units); thus,
converting them through the conversion tables presented below will not yield a better
representation of the units. In all cases the best approach is to consult the equipment vendor
to come up with appropriate parameters for the 2nd generation models to represent the actual
wind power plants (even in the case of GE units).

New Model Older Model

regc_a wt3g
reec_a wt3e (part of)
repc_a wt3e (part of)
wtgp_a wt3p
wtgt_a wt3t (part of)
wtga_a wt3t (part of)
wtgq_a wt3e (part of)

14496639 E-4
New Model Old Model Explanatory Comments
regc_a wt3g Comment
Lvplsw Lvplsw
Rrpwr Rrpwr
Brkpt Brkpt
Zerox Zerox
Lvpl1 Lvpl1
Vtmax Volim
Lvpnt1 Lvpnt1
Lvpnt0 Lvpnt0
qmin Iolim
accel Khv
Tg Td
Tfltr Tfltr
iqrmax 999 disable limit; no equivalent in old model
iqrmin -999 disable limit; no equivalent in old model
Xe Lpp disable limit; no equivalent in old model

wtgp_a wt3p Comments

Kiw Kip
Kpw Kpp
Kic Kic
Kpc Kpc
Kcc 0 disable; no equivalent in old model
Tpi Tpi
Pimax Pimax
Pimin Pimin
Piratmx Pirat
Piratmn -Pirat

wtgt_a wt3t Comments

Ht Hfrac * H
Hg H - Ht
Dshaft Dshaft In the wtgt_a model D does not exist; Dshaft may also need to be adjusted to get the same result as in the wt3t model
Kshaft 2*(2*pi*Freq1)^2*Ht*(Hg/H)
wo 1 this value will be set to ωref upon initialization by the wtgq_a model

wtga_a wt3t Comments

Ka Kaero
Theta0 0 set to default; no equivalent in old model

wtgq_a wt3e Comments

Kip Kitrq
Kpp Kptrq
Tp 0 not available in old model so disable
Twref Tsp
Temax Pmax/Wp100
Temin Pmin/Wpmin
p1 Pmin
spd1 Wpmin
p2 0.2
spd2 Wp20
p3 0.6
spd3 Wp60
p4 Pwp100
spd4 Wp100
Tflag 0 not available in old model so disable

14496639 E-5
New Model Old Model Explanatory Comments
reec_a wt3e Comments
vdip 0 not available in old model so disable
vup 2 not available in old model so disable
Trv Tr
dbd1 -1 not available in old model so disable
dbd2 1 not available in old model so disable
kqv 0 not available in old model so disable
iqh1 0.001 not available in old model so disable
iql1 -0.001 not available in old model so disable
vref0 1 not available in old model so disable
iqfrz 0 not available in old model so disable
thld 0 not available in old model so disable
thld2 0 not available in old model so disable
Tp Tp
Qmax Qmax
Qmin Qmin
Vmax Vmax
Vmin Vmin
kqp 0 not available in old model so disable
kqi Kqi
kvp 0 not available in old model so disable
kvi Kqv
vref1 0 not available in old model so disable
tiq 0.02 set to default value
dpmax Pwrat
dpmin -Pwrat
Pmax Pmax
Pmin Pmin
imax Xiqmax/Lpp wt3e had no current limit so set to estimated current limit based on flux limit; Lpp comes from wt3g model
Tpord Tpc
pfflag 1 if varflg = -1 or 0 if varflg = 0 or 1
vflag 1 set to 1 since no equivalent in old model
qflag 1 if vltflg = 1 for the case that vltflg = 0 in the old model, this option is not available in the new models; this was rarely used
pflag 0 must be set to 0 for the implementation in GE PSLF®
pqflag 1 set to 1 since no equivalent in old model
vq1 0
iq1 Xiqmax
vq2 0.1
iq2 Xiqmax
vq3 0.9
iq3 Xiqmax
vq4 1
iq4 Xiqmax
not available in old model so set to constant value across all voltages
vp1 0
ip1 ipmax
vp2 0.1
ip2 ipmax
vp3 0.9
ip3 ipmax
vp4 1
ip4 ipmax

repc_a wt3e Comments

Tfltr Tr
Kp Kpv
Ki Kiv
Tft 0 not available in old model so disable
Tfv Tc
refflg 1 not available in old model so disable
vfrz 0 not available in old model so disable
rc 0 not available in old model so disable
xc Xc not available in old model so disable
Kc 0 not available in old model so disable
vcmpflg 1 the branch for monitoring the voltage is set differently in various software platforms; check the software user's manual
emax 99 not available in old model so disable
emin -99 not available in old model so disable
dbd 0 not available in old model so disable
Qmax Qmax
Qmin Qmin
kpg 0 not available in old model so disable
kig 0 not available in old model so disable
Tp Tp
fdbd1 0 not available in old model so disable
fdbd2 0 not available in old model so disable
femax 0 not available in old model so disable
femin 0 not available in old model so disable
pmax Pmax not available in old model so disable
pmin Pmin not available in old model so disable
tlag 0 not available in old model so disable
ddn 0 not available in old model so disable
dup 0 not available in old model so disable
frqflg 0 not available in old model so disable
outflag 0 not available in old model so disable

14496639 E-6
E.3 Type 4 WTG
The following tables show how to convert the old (1st generation) generic stability models for
type 4 WTGs to the new (2nd generation) models. It should be noted that the 1st generation
models are essentially a subset of the more general 2nd generation models, with a few exceptions:
1. The older model had a rather more complex current limit, which was specific to one
vendor. This can now be emulated using the current limit logic together with the VDL1
and VDL2 tables. This is shown below.
2. The wt4t model in the 1st generation (older) generic models has no real counterpart in the
new 2nd generation generic RES models.
3. The new RES suite of models were developed through a collaborative effort and so a
single central model specification was developed [1]; this meant that the new models are
essentially identical among the commercial software platforms. Much effort was made to
ensure a one to one correspondence in the parameter lists and to compare simulations
across the platforms. This is not necessarily true of the older (1st generation) models. As
such, the user will quickly notice significant differences between the implementation of
the 1st generation models across some of the software platforms. We clearly cannot
address these issues now, since we neither had an influence on the initial development of
the older (1st generation) models, nor is it wise to try to address the differences now when
those models are to be in time replaced by the newer ones.
Therefore, with the above in mind, the conversion tables developed here are based on the GE
PSLF implementation of the 1st generation (older) generic models. The response of the converted
model will not be exactly the same due to the differences explained above.
Disclaimer: This report, and the organization/author that produced this report, make no
guarantees or warranties, expressed or implied, with respect to the accuracy or applicability of
the conversion tables below. It should also be noted that the new 2nd generation models were
developed with the expressed intention of making them more flexible to allow modeling of a
larger variety of models. With that in mind where older (1st generation) models exist in dynamic
databases to represent existing equipment, they may not necessarily have been very
representative of the equipment performance (particularly for none GE units); thus, converting
them through the conversion tables presented below will not yield a better representation of the
units. In all cases the best approach is to consult the equipment vendor to come up with
appropriate parameters for the 2nd generation models to represent the actual wind power plants
(even in the case of GE units).
New Model Older Model
regc_a wt4g
reec_a wt4e (part of)
repc_a wt4e (part of) and wt4t (part of)

14496639 E-7
New Model Old Model Explanatory Comments
regc_a wt4g Comments
Lvplsw Lvplsw
Rrpwr Rrpwr
Brkpt Brkpt
Zerox Zerox
Lvpl1 Lvpl1
Vtmax Volim
Lvpnt1 Lvpnt1
Lvpnt0 Lvpnt0
qmin Iolim
accel Khv
Tg Td
Tfltr Tfltr
iqrmax 99 disable limit; no equivalent in old model
iqrmin -99 disable limit; no equivalent in old model
Xe 0 disable; no equivalent in old model

reec_a wt4e Comments

vdip 0 disable; no equivalent in old model
vup 2 disable; no equivalent in old model
Trv Tr
dbd1 -1 disable; no equivalent in old model
dbd2 1 disable; no equivalent in old model
kqv 0 disable; no equivalent in old model
iqh1 0.001 disable; no equivalent in old model
iql1 -0.001 disable; no equivalent in old model
vref0 1 disable; no equivalent in old model
iqfrz 0 disable; no equivalent in old model
thld 0 disable; no equivalent in old model
thld2 0 disable; no equivalent in old model
Tp Tpwr
Qmax Qmax
Qmin Qmin
Vmax Vmax
Vmin Vmin
kqp 0 disable; no equivalent in old model
kqi Kqi
kvp 0 disable; no equivalent in old model
kvi Kvi
vref1 0 disable; no equivalent in old model
tiq 0.02 disable; no equivalent in old model
dpmax 99
disable; no equivalent in old model
dpmin -99
Pmax 1/fn set so as to respect number of units online for representing the aggregate unit
Pmin 0 assume zero; no equivalent variable in old model
imax ImaxTD
Tpord 0 disable; no equivalent in old model; (could set Tpord to reasonable small value = 0.02 s)
pfflag if (varflg =1 and pfaflg =1) then pfflag = 1; else pfflag = 0 set appropriate value
vflag 1 set appropriate value
qflag 1 set appropriate value
pflag 0 set appropriate value
Pqflag Pqflag
vq1 0
iq1 Iqh1 if Iqh1 > Viqlim; then set this to Viqlim and if Iqh1 < qmax there is an error with the model
vq2 (Viqlim - Iqh1)/(Viqlim - qmax) if Iqh1 > Viqlim; then set this to 0.5
iq2 Iqh1 if Iqh1 > Viqlim; then set this to (Viqlim -qmax)*0.5
vq3 1
iq3 qmax
vq4 1.05
iq4 qmax
vp1 0
ip1 Iph1
vp2 0.1
ip2 Iph1
vp3 0.9
ip3 Iph1
vp4 1
ip4 Iph1

14496639 E-8
New Model Old Model Explanatory Comments
repc_a wt4e Comments
Tfltr Tr
Kp Kpv
Ki Kiv
Tft 0 disable since no equivalent in old model
Tfv Tc
refflg 1 disable since no equivalent in old model
vfrz 0 disable since no equivalent in old model
rc 0 disable since no equivalent in old model
xc 0 disable since no equivalent in old model
Kc 0 disable since no equivalent in old model
vcmpflg 1 disable since no equivalent in old model
emax 99 disable since no equivalent in old model
emin -99 disable since no equivalent in old model
dbd 0 disable since no equivalent in old model
Qmax Qmax
Qmin Qmin
kpg 0 disable since no equivalent in old model
kig 0 disable since no equivalent in old model
Tp Tpwr
fdbd1 0 disable since no equivalent in old model
fdbd2 0 disable since no equivalent in old model
femax 0 disable since no equivalent in old model
femin 0 disable since no equivalent in old model
pmax 0 disable since no equivalent in old model
pmin 0 disable since no equivalent in old model
tlag 0 disable since no equivalent in old model
ddn 0 disable since no equivalent in old model
dup 0 disable since no equivalent in old model
frqflg 0 disable since no equivalent in old model
outflag 0 disable since no equivalent in old model

14496639 E-9
14496639
F
PARAMETER LIST FOR THE 2ND GENERATION
GENERIC MODELS
The following tables were initially developed as a spreadsheet through collaboration with the
WECC Renewable Energy Modeling Task Force (REMTF) and receiving comments and input
from various equipment vendors and other stake holders. We are truly grateful to them all.
These tables are provided simply as a guide. They are not to be taken as definitive and final. In
the end the values of the parameters of the models must be based on thorough consultation with
the equipment vendor, and evidence from testing and/or model validation exercises.
REGC_A:
regc_a Range Comment Units
Xe 0-1 For type 4 WTGs typically it is set to 0; for type 3 WTGs some vendors will supply a number, typically around 0.5 to 0.8 pu pu
Lvplsw 1 or 0 Either value appropriate; set as instructed by OEM N/A
Rrpwr 1 to 20 Rate at which active current (power) recovers after a fault; typical values are between 2 to 10 pu/sec pu/s
Brkpt 0.05 - 0.9 Used only when Lvplsw = 1; voltage point below which active current is linearly reduced as a function of voltage until it reaches zero at Zerox pu
Zerox 0.02 - 0.5 Used only when Lvplsw = 1; voltage point below which active current become zero pu
Lvpl1 1.1 - 1.5 Active current limit at Brkpt voltage and above; typically set to a value such as 1.22 pu
Vtmax 1.1 - 1.3 pu
Lvpnt1 0.1 - 0.8 pu
Lvpnt0 0.05 - 0.4 These parameters are all associated with the numerical interface between the "current source" and the network equations. They are essential for pu
qmin -1.3 maintaining numerical stability and convergence of the network solution. Vtmax, Lvpnt1, Lvpnt0, qmin and accel should not be set to "zero". pu
accel 0.7 - 1.0 numerical acceleration factor N/a
Tg 0.02 - 0.05 Emulated delay in converter controls s
Tfltr 0.02 - 0.05 Filtering time constant for voltage measurement s
iqrmax 1 - 999 Rate at which reactive current (power) recovers after a fault when the initial reactive output of the unit is less than zero; typically set to -999 pu/s
iqrmin -999 to -1 Rate at which reactive current (power) recovers after a fault when the initial reactive output of the unit is greater than zero; typically set to -999 pu/s

Xe >= 0
rrpwr > 0
Brkpt > Zerox
Lvpnt1 > Lvpnt0
Relational
qmin < 0
Requirements
accel > 0
iqrmax > 0
iqrmin < 0
time constants check software user's manual; typically can be zero, but if greater than zero must be at least three to four times integration time-step

14496639 F-1
REEC_A:
reec_a Range Comment Units
vdip 0 - 0.9 Note: to disable vdip it is typical to set it to -1; The voltage below which voltage-dip logic is initiated pu
vup 1.1 - 1.3 Note: to disable vup it is typical to set it to 2; The voltage above which voltage-dip/up logic is initiated pu
trv 0 - 0.1 voltage measurement transducer time constant s
dbd1 -0.1 - 0 lower deadband in voltage error pu
dbd2 0 - 0.1 upper deadband in voltage error pu
kqv 0 - 20 reactive current injection proportional gain; active only during a voltage-dip/rise; 0 to disable pu/pu
iqh1 1 to Imax maximum limit on reactive current injection during a voltage-dip/rise pu
iql1 -Imax to -1 minimum limit on reactive current injection during a voltage-dip/rise pu
vref0 N/A if vref0 is left as zero most commercial software tools will initialize it appropriately; otherwise typically set to 1 pu (nominal voltage) pu
iqfrz 0 - Iqh1 Value at which reactive current injection (during a voltage-dip) is held for thld seconds following a voltage dip if thld > 0, p.u.; 0 to disablepu
Time for which reactive current injection is held at some value following termination of the voltage-dip; if value is +ve, then current is
thld -1 to 1 held at iqfrz , if -ve then held at the value just prior to ending of the voltage-dip; 0 to disable s
thld2 0-1 Time for which active current is held at its value during a voltage-dip, following the termination of the voltage-dip; 0 to disable s
tp 0-1 Electrical power transducer time constant s
qmax 0-1 Maximum reactive output limit (to disable set to 9999) pu
qmin -1 - 0 Minimum reactive output limit (to disable set to -9999) pu
vmax 1.05 - 1.1 Voltage control maximum limit; typically 110% (higher values than shown may be used in some cases) pu
vmin 0.8 - 0.95 Voltage control minimum limit; typically 90% (lower values than shown may be used in some cases) pu
kqp 0 - 999 pu/pu
These are tunable PI gains of the local q-control
kqi 0 - 999 pu/pu/s
kvp 0 - 999 pu/pu
These are tunable PI gains of the local v-control
kvi 0 - 999 pu/pu/s
vref1 0 Typically set to zero; only change if instructed by OEM data, in some cases may be set to 1 or other appropriate value pu
tiq 0 - 0.02 Controller time-constant s
dpmax 0.1 - 99 Up ramp-rate on power reference; typically set to 99 to disable pu/s
dpmin -99 to -0.1 Down ramp-rate on power reference; typically set to -99 to disable pu/s
pmax 1.0 - 1.15 maximum power reference (high values than shown may be used in some cases) pu
pmin 0 - 0.05 minimum power reference pu
imax 1 to 2 Maximum current limit of the device pu
tpord 0.02 - 1.0 time constant for power order s
pfflag 0 or 1 N/A
vflag 0 or 1 N/A
qflag 0 or 1 See spreadsheet tab on flag settings N/A
pflag 0 or 1 N/A
pqflag 0 or 1 N/A
vq1 pu
iq1 pu
vq2 pu
iq2 pu
vq3 pu
iq3 These are the voltage-dependent limits on active and reactive current. There is no typical set of values. Presently, looking at extending pu
vq4 these to ten (10) points and to extend their features. However, as they stand today, the following guidelines should be followed: (i) do pu
iq4 not set any iq/ip values to a negative number, (ii) the values should generally be monotonically increasing, and (iii) to disable the block pu
N/A
vp1 simply set all iq/ip values to Imax for a set of monotonically increases voltage values. Recommendation: use values from OEM data - pu
ip1 be aware that the number of points and some functionally needs to be improved in future model updates before these tables can pu
vp2 emulate certain OEM data. pu
ip2 pu
vp3 pu
ip3 pu
vp4 pu
ip4 pu

vup > vdip

iqh1 > iql1
dbd2 >= dbd1
iqfrz >= 0
qmax > qmin; qmax > 0
Relational pmax > pmin > 0
Requirements vmax > vmin
Kqv >= 0
dpmax > dpmin
imax > 0
kqp, kqi, kvp, kvi >= 0 These are tunable gains so any value within reason, that is greater than zero and tuned to give a good response is acceptable.
time constants check software user's manual; typically can be zero, but if greater than zero must be at least three to four times integration time-step

14496639 F-2
REEC_B: (model not recommended for use any more – table provided for completeness)
reec_b Range Comment Units
vdip 0 - 0.9 Note: to disable vdip it is typical to set it to -1; The voltage below which voltage-dip logic is initiated pu
vup 1.1 - 1.3 Note: to disable vup it is typical to set it to 2; The voltage above which voltage-dip/up logic is initiated pu
trv 0 - 0.1 voltage measurement transducer time constant s
dbd1 -0.1 - 0 lower deadband in voltage error pu
dbd2 0 - 0.1 upper deadband in voltage error pu
kqv 0 - 20 reactive current injection proportional gain; active only during a voltage-dip/rise; 0 to disable pu/pu
iqh1 1 to Imax maximum limit on reactive current injection during a voltage-dip/rise pu
iql1 -Imax to -1 minimum limit on reactive current injection during a voltage-dip/rise pu
vref0 N/A if vref0 is left as zero most commercial software tools will initialize it appropriately; otherwise typically set to 1 pu (nominal voltage) pu
tp 0-1 Electrical power transducer time constant s
qmax 0-1 Maximum reactive output limit (to disable set to 9999) pu
qmin -1 - 0 Minimum reactive output limit (to disable set to -9999) pu
vmax 1.05 - 1.1 Voltage control maximum limit; typically 110% (higher values than shown may be used in some cases) pu
vmin 0.8 - 0.95 Voltage control minimum limit; typically 90% (lower values than shown may be used in some cases) pu
kqp 0 - 999 pu/pu
These are tunable PI gains of the local q-control
kqi 0 - 999 pu/pu/s
kvp 0 - 999 pu/pu
These are tunable PI gains of the local v-control
kvi 0 - 999 pu/pu/s
tiq 0 - 0.02 Controller time-constant s
dpmax 0.1 - 99 Up ramp-rate on power reference; typically set to 99 to disable pu/s
dpmin -99 to -0.1 Down ramp-rate on power reference; typically set to -99 to disable pu/s
pmax 1 - 1.15 maximum power reference pu
pmin 0 - 0.05 minimum power reference pu
imax 1 to 2 Maximum current limit of the device pu
tpord 0-1 power order time constant s
pfflag 0 or 1 N/A
vflag 0 or 1 N/A
qflag 0 or 1 See spreadsheet tab on flag settings N/A
pflag 0 or 1 N/A
pqflag 0 or 1 N/A

vup > vdip

iqh1 > iql1
dbd2 >= dbd1
iqfrz >= 0
qmax > qmin; qmax > 0
Relational vmax > vmin
Requirements pmax > pmin > 0
Kqv >= 0
dpmax > dpmin
imax > 0
kqp, kqi, kvp, kvi >= 0 These are tunable gains so any value within reason, that is greater than zero and tuned to give a good response is acceptable.
time constants check software user's manual; typically can be zero, but if greater than zero must be at least three to four times integration time-step

14496639 F-3
REEC_C:
reec_c Range Comment Units
vdip 0 - 0.9 Note: to disable vdip it is typical to set it to -1; The voltage below which voltage-dip logic is initiated pu
vup 1.1 - 1.3 Note: to disable vup it is typical to set it to 2; The voltage above which voltage-dip/up logic is initiated pu
trv 0 - 0.1 voltage measurement transducer time constant s
dbd1 -0.1 - 0 lower deadband in voltage error pu
dbd2 0 - 0.1 upper deadband in voltage error pu
kqv 0 - 20 reactive current injection proportional gain; 0 to disable pu/pu
iqh1 1 to Imax maximum limit on reactive current injection pu
iql1 -Imax to -1 minimum limit on reactive current injection pu
vref0 N/A if vref0 is left as zero most commercial software tools will initialize it appropriately; otherwise typically set to 1 pu (nominal voltage) pu
SOCini SOCmin to SOCmax Initial state of charge; must be between SOCmin and SOCmax pu
SOCmax 0.8 - 1 Maximum allowable state of charge in pu (these are typical ranges, they could be different for various projects) pu
SOCmin 0 - 0.2 Minimum allowable state of charge in pu (these are typical ranges, they could be different for various projects) pu
Time it takes for the battery to discharge when putting out 1 pu power; typically set to 9999 since most batteries are large as compared to the typical
T N/A simulation time in a transient stability study s
tp 0-1 Electrical power transducer time constant s
qmax 0-1 Maximum reactive output limit (to disable set to 9999) pu
qmin -1 - 0 Minimum reactive output limit (to disable set to -9999) pu
vmax 1.05 - 1.1 Voltage control maximum limit; typically 110% (higher values than shown may be used in some cases) pu
vmin 0.8 - 0.95 Voltage control minimum limit; typically 90% (lower values than shown may be used in some cases) pu
kqp 0 - 999 pu/pu
These are tunable PI gains of the local q-control
kqi 0 - 999 pu/pu/s
kvp 0 - 999 pu/pu
These are tunable PI gains of the local v-control
kvi 0 - 999 pu/pu/s
vref1 0 Typically set to zero; only change if instructed by OEM data pu
tiq 0 - 0.02 Controller time-constant s
dpmax 0.1 - 99 Up ramp-rate on power reference; typically set to 99 to disable pu/s
dpmin -99 to -0.1 Down ramp-rate on power reference; typically set to -99 to disable pu/s
pmax 1 - 1.15 maximum power reference pu
pmin -1 to 0 minimum power reference pu
imax 1 to 2 Maximum current limit of the device pu
tpord 0-1 power order time constant s
pfflag 0 or 1 N/A
vflag 0 or 1 N/A
qflag 0 or 1 See spreadsheet tab on flag settings N/A
pflag 0 or 1 N/A
pqflag 0 or 1 N/A
vq1 pu
iq1 pu
vq2 pu
iq2 pu
vq3 pu
iq3 pu
These are the voltage-dependent limits on active and reactive current. There is no typical set of values. Presently, looking at extending these to ten (10)
vq4 pu
points and to extend their features. However, as they stand today, the following guidelines should be followed: (i) do not set any iq/ip values to a
iq4 pu
N/A negative number, (ii) the values should generally be monotonically increasing, and (iii) to disable the block simply set all iq/ip values to Imax for a set of
vp1 pu
monotonically increases voltage values. Recommendation: use values from OEM data - be aware that the number of points and some functionally
ip1 pu
needs to be improved in future model updates before these tables can emulate certain OEM data.
vp2 pu
ip2 pu
vp3 pu
ip3 pu
vp4 pu
ip4 pu

vup > vdip

iqh1 > iql1
dbd2 >= dbd1
iqfrz >= 0
qmax > qmin; qmax > 0
pmax > pmin Note Pmax should be greater than zero; but since this model is typically used for Energy Storage Pmin will be negative
Relational
vmax > vmin
Requirements
Kqv >= 0
dpmax > dpmin
imax > 0
kqp, kqi, kvp, kvi >= 0 These are tunable gains so any value within reason, that is greater than zero and tuned to give a good response is acceptable.
SOCmax >= SOCini >= SOCmin
time constants check software user's manual; typically can be zero, but if greater than zero must be at least three to four times integration time-step

14496639 F-4
REPC_A (and REPC_B):
repc_a Range Comments Units
Filtering time constant for transducers; Note: to represent delays in communication or other aspects
tfltr 0 - 0.05 this time constant can sometimes be set as high as 2 seconds s
kp 0 - 20 Proportional control gain pu/pu
ki 0 - 10 Integral control gain pu/pu/s
tft 0 Lead time constant; typically set to zero unless otherwise specified by OEM s
tfv 0.02 - 0.2 Lag time constant s
Note with repc_b a third option (=2)
0 - constant Q control; 1 - voltage control
refflg 0 or 1 N/A for power factor control here
vfrz 0 - 0.9 Voltage below which the integrator state (of the volt/var controller) is frozen pu
rc 0 Resistive part of current compensation (typically set to zero) pu
xc 0 to 0.05 Reactive part of current compensation pu
Kc 0 to 0.15 Reactive droop pu
vcmpflg 0 or 1 0 - reactive droop; 1 - current compensation
emax 0.05 to 99 Maximum voltage/var error
emin -99 to -0.05 Minimum voltage/var error
dbd 0 - 0.1 deadband; typically zero or a small number, e.g. 0.01
IMPORTANT NOTE: qmax/qmin is the maximum/minimum reactive limits at the plant level if the down-
stream reec_* model is in local Q/V or Q control, however, if the reec_* model is in local V-control (i.e.
qmax N/A pu
Vflag = 0 and Qflag = 1 in the reec_* model) then qmax/qmin become voltage limits. So when they are
reactive limits, typical values might be 0.4/-0.4, while if they are voltage limits typical values might be
qmin N/A 1.1/0.9 pu
kpg 0 - 10 Proportional control gain pu/pu
kig 0 - 10 Integral control gain pu/pu/s
tp 0-1 Power measurement transducer time constant s
fdbd1 -0.01 to 0 Frequency deadband, downside; a typical value might be -0.0003 pu pu
fdbd2 0 to 0.01 Frequency deadband, downside; a typical value might be 0.0003 pu pu
femax 0.01 to 99 Maximum frequency error pu
femin -99 to -0.01 Minimum frequency error pu
Maximum power; IMPORTANT NOTE: when modeling a RES power plant operating at it maximum
available resource limit (e.g. producing maximum power from current available wind or solar energy)
then Pmax may be set to a value below the "nameplate" maximum capability of the plant to avoid the
plant responding to a simulate under-frequency event. Thus, a value less than 1 may be reasonable in
many cases. This, however, is a decision to be made by the planner using the model, rather than the
GO providing the model. The model, as provided by a GO, should indicate here the maximum
pmax 1 to 1.15 "nameplate" capability of the plant. pu
pmin 0 - 0.1 Minimum power pu
tlag 0-1 Frequency response lag time constant s
Downside droop gain; typically, wind and PV plants will have a droop of 4 to 5% thus ddn = 1/0.04 to
ddn 0 - 30 1/0.05, i.e. 25 to 20. pu/pu
Upside droop gain; typically, wind and PV plants will have a droop of 4 to 5% thus ddn = 1/0.04 to
dup 0 - 30 1/0.05, i.e. 25 to 20. pu/pu
frqflg 0 or 1 0 - no primary frequency response; 1 - with primary frequency response N/A
0 - if output of the model is a Qref; 1 - if output of the model is Vref (see important note under Note these two parameters do not
outflag 0 or 1 qmax/qmin above) N/A apply to repc_b since it is always on
0 - Pbranch and Qbranch feedback to model are on system MVA base; 1 - Pbranch/Qbranch are on the system MVA base; it has a set of
puflag 0 or 1 model MVA base N/A parameters that follow which link it to

These are tunable gains so any value within reason, that is greater than zero and tuned to give a good
kp, ki, kpg, kig response is acceptable.
emax > emin
qmax > qmin; qmax > 0
pmax > pmin Note Pmax should be greater than zero, but Pmin might be negative for an Energy Storage Plant
femax > femin
Relational
femax > 0
Requirements
femin < 0
fdbd1 >=0
fdbd2 <= 0
ddn and dup >= 0
check software user's manual; typically can be zero, but if greater than zero must be at least three to
time constants four times integration time-step

14496639 F-5
Flag Settings in REEC_* and REPC_* models:
reec_* flags repc_a flag
Functionality Models Needed PfFlag Vflag Qflag RefFlag Comments
Constant pf control at unit level reec_* 1 0 or 1 0 N/A
Constant Q control at unit level reec_* 0 0 or 1 0 N/A
V-control only at unit level reec_* 0 0 1 N/A
Q/V control only at unit level reec_* 0 1 1 N/A
Plant level Q control + unit level Q control reec_* + repc_a 0 0 or 1 0 0
Plant level V control + unit level V control reec_* + repc_a 0 0 or 1 0 1 In this case see note on qmax/qmin and setting outflag, in repc_a
Plant level Vcontrol + unit level Q control reec_* + repc_a 0 0 1 1
Plant level V Control + coordinated unit level Q/V control reec_* + repc_a 0 1 1 1
Plant level Q Control + coordinated unit level Q/V control reec_* + repc_a 0 1 1 0

WTGP_A:
wtgp_a Range Comments Units
kiw 1 - 50 pitch controller integral gain deg/pu/s
kpw 1 - 200 pitch controller proportional gain deg/pu
kic 1 - 50 pitch controller integral gain deg/pu/s
kpc 1 - 10 pitch controller proportional gain deg/pu
kcc 0-1 cross term gain pu/pu
tpi 0.1 - 1 pitch time constant s
pimax 25 - 35 maximum pitch angle deg
pimin -5 - 0 minimum pitch angle deg
piratmx 5 - 15 maximum rate of increase of pitch angle deg/s
piratmn -15 to -5 maximum rate of decrease of pitch angle deg/s

These are tunable gains so any value within reason, that is

greater than zero and tuned to give a good response is
kiw, kpw, kic, kpc acceptable.
kcc >= 0
Relational pimax > pimin
Requirements piratmx > 0
piratmn < 0
check software user's manual; typically can be zero, but if greater
than zero must be at least three to four times integration time-
time constants step

WTGT_A:
wtgt_a Range Comments Units
Ht 4 to 6 The inertia of the turbine. MUST be obtained form OEM; typically quite high and around 5 or so. MWs/MVA
Hg 0.5 to 1 The inertia of the generator. MUST be obtained form OEM; typically low and around 0.7 or so. MWs/MVA
The estimated damping factor; in reality much of this is achieved by the active-drive train damping controls in
Dshaft 1 to 2 type 3 & 4 machines that is not explicitly modeled. pu/pu
Shaft spring constant. This should be either obtained from the OEM or calculated based on the following
information from the OEM: Kshaft = 2*(2*pi*Freq1)^2*Ht*(Hg/H), where Freq1 = frequency in Hz of the first
Kshaft 90 - 150 torsional mode, H = Ht + Hg = total shaft inertia in MWs/MVA pu/pu
Note: this value will be set to Wref upon initialization by the wtgq_a model; if the model is being used with a
wo 0.7 - 1.3 type 4A model, then typically this is set to 1 pu pu

Ht > Hg
Relational Dshaft > 0
Requirements Kshaft > 0
1.4 > wo > 0.6 suggested check; typically neither type 3 nor type 4 WTGs will be able to operate outside this range of speeds

14496639 F-6
WTGA_A:
wtga_a Range Comments Units
Ka 0.005 - 0.008 Aerodynamics gain factor (provided by OEM); a typical value is 0.007 pu/deg2
The initial pitch angel; it is typically set to 0 to indicate WTG is producing maximum available power.
When modeling primary frequency response, and using this model, set Theta0 to a non-zero value
Theta0 0 - 10 such as 10. deg

Relational Ka > 0
Requirements Theta0 >= 0

WTGQ_A:
wtgq_a Range Comments Units
kip 0.1 - 1 Integral gain of torque controller pu/pu/s
kpp 1 to 10 Proportional gain of torque controller pu/pu
tp 0.05 - 1 Power measurement time constant s
twref 10 to 60 Speed reference time constant s
temax 0.8 - 1.2 Maximum torque pu
temin 0.05 - 0.08 Minimum torque pu
p1 N/A pu
spd1 N/A pu
p2 N/A pu
spd2 N/A OEM provided speed versus power curve. A typical curve might pu
p3 N/A be: ps = [0.2 0.4 0.6 0.74] and spds = [0.69 0.78 0.98 1.2] pu
spd3 N/A pu
p4 N/A pu
spd4 N/A pu
Tflag 0 or 1 From OEM; type of torque control N/A

These are tunable gains so any value within reason, that is greater
kip, kpp >= 0 than zero and tuned to give a good response is acceptable.
twref > 0
Relational
temax > temin
Requirements
suggested check; typically neither type 3 nor type 4 WTGs will be
1.4 > spd4 > spd3 > spd2 > spd1 > 0.6 able to operate at speeds below 0.6 pu or above 1.4 pu
1.0 > p4 > p3 > p3 > p1 > 0.1 suggested check; typical range the values of the p's should lie

14496639 F-7
14496639
14496639
 The Electric Power Research Institute, Inc.
(EPRI, www.epri.com) conducts research and
development relating to the generation, delivery and
use of electricity for the benefit of the public. An
independent, nonprofit organization, EPRI brings
together its scientists and engineers as well as
experts from academia and industry to help address
challenges in electricity, including reliability,
efficiency, affordability, health, safety and the
environment. EPRI members represent 90% of the
electric utility revenue in the United States with
international participation in 35 countries. EPRI’s
principal offices and laboratories are located in Palo
Alto, Calif.; Charlotte, N.C.; Knoxville, Tenn.; and
Lenox, Mass.

Together…Shaping the Future of Electricity

© 2018 Electric Power Research Institute (EPRI), Inc. All rights reserved.
Electric Power Research Institute, EPRI, and TOGETHER…SHAPING THE
FUTURE OF ELECTRICITY are registered service marks of the Electric
Power Research Institute, Inc.
3002014083

Electric Power Research Institute

3420 Hillview Avenue, Palo Alto, California 94304-1338 • PO Box 10412, Palo Alto, California 94303-0813 • USA
14496639800.313.3774 • 650.855.2121 • [email protected] • www.epri.com