Constant Power Generation of Photovoltaic Systems

Considering the Distributed Grid Capacity

Yongheng Yang, IEEE Student Member, Frede Blaabjerg, IEEE Fellow, Huai Wang, IEEE Member
Department of Energy Technology
Aalborg University
Pontoppidanstraede 101, Aalborg East DK-9220, Denmark
[email protected], [email protected], [email protected]

Abstract— With an imperative demand of clean and reliable set for RESs. For example, in Germany, electricity generation
electricity generation in some countries, the increasing adoption from RESs is expected to be over 80% by 2050 [7]. Thus, it
of new photovoltaic (PV) systems pushes the Distribution System calls for grid reinforcement measures, and it is of importance
Operators (DSOs) to expand the transmission/distributed lines.
However, the potential cost brought by such extensions and and necessity for those countries to realize high RES accep-
increased maintenances introduce new obstacles. In view of this tance plans. Alternatively, the Distributed System Operators
concern, the DSOs starts to reduce PV installations in order to (DSOs) have to reduce the RES installations. However, with
avoid an extension of the power infrastructure. Besides, another power infrastructure expansion, the existing systems have to
alternative solution is to limit the maximum feed-in power of be retrofitted and thus the total cost increases. Hence, the
the existing PV systems to a certain level. It can contribute to a
weakened requirement of grid expansion and at the same time line extension approach is not the optimal solution to increase
an increased penetration level. Therefore, to meet the need of renewable energy utilizations, including PV power, when it is
this emerging ancillary service provided by future PV systems, seen from a total cost perspective.
a Constant Power Generation (CPG) control concept of PV Among variable renewables, wind turbine power systems
inverters is proposed in this paper. Accordingly, it is worth have been witnessed as one of the most successful technologies
investigating into two main issues: a) analyzing the reduction of
the energy yield due to CPG control to study its feasibility from in some countries, e.g. Denmark [8]. As a result, many grid
an economic point of view and b) developing robust CPG control codes have been developed to flexibly regulate the connection
methods, otherwise, it may introduce instabilities. Thereby, the situations, such as output power control and voltage support
implementation possibilities for PV systems in CPG operation capability [9]–[12]. Regarding PV systems, they still account
mode are also discussed in this paper. Additionally, the loss of for a minor share of the overall electricity production in most
energy is calculated to reveal the viability of the proposed CPG
control method. Operation examples of a PV system are presented countries. Therefore, it is required PV systems to operate with
to show the effectiveness of the CPG control method to unload Maximum Power Point Tracking (MPPT) control, and they are
the distributed grid. not allowed to participate in the grid voltage regulation [6],
[13]–[15]. As aforementioned, the increasing adoption of PV
I. I NTRODUCTION systems poses more challenges for the distributed grid, which
Traditionally, the electricity generated from fossil-source- requires infrastructure upgrades. Thus, some of the grid codes
based power plants is transferred to the customers through long are undergoing a major revision to enable the grid support
transmission lines. Today, more Renewable Energy Sources capability for PV systems at a high penetration level [15]–
(RESs), e.g. PhotoVoltaic (PV) systems and wind turbine [22]. In that case, similar to wind power systems, the PV
systems, have been connected to the grid, which is mainly systems should be able to flexibly control the active power,
pushed by advanced power electronics technologies [1]–[4]. e.g. curtailment control and power ramp limitation, in order
The power of RESs is dependent on environmental conditions to maintain the grid stability.
(e.g. wind speed, solar irradiance and ambient temperature), Solutions to realize these advanced features include in-
and thus aging transmission lines and local substations have to tegrating energy storage systems and expanding the grid
cope with an increasing amount of fluctuated power. Moreover, capacity [7], [15], [22]. However, the increased cost, which
the imperative demand of clean, environmental-friendly, and may severely exceed the initial project outlay, hinders the
reliable electricity generation has been another principal driver development of both technologies. According to a study of
for a wide-scale adoption of RESs to reduce carbon-dioxide a grid-connected PV system based on a yearly mission profile
emissions from conventional power plants. This also makes (ambient temperature and solar irradiance level), limiting the
the power system highly decentralized and vulnerable. maximum feed-in power can contribute to the unloading of the
Consequently, the aging grids will experience severe over- distributed grid. Consequently, achievement of an even wide-
loading issues [5], [6], introducing further stability problems scale adoption of PV systems is quite feasible. Reducing the
and follow-up maintenance tasks, if the distributed grid re- maximum feed-in power will violate the MPPT function for
mains at its current capacity. This situation tends to be even PV systems with current control methodologies. Thus, it is
worse in those countries, where ambitious goals have been necessary to develop more robust control solutions.

978-1-4799-2325-0/14/$31.00 ©2014 IEEE 379

 LCL Active Power MPPT Control
PV Strings CPV Boost Cdc Inverter Po
Filter
PPV DC DC Q Grid Possible active power
C MPPT Control
Zg
DC AC Absolut (constant)
iPV vPV PWMb PWMinv production constraint
PCC
Boost Control vdc vg
(MPPT, Plimit) Inverter
Gradient production Power ramp
v*dc Control ig Delta production
S, Ta constraint constraint
Monitoring and Control constraint

Time
Communication Supervisory command

Fig. 1. A two-stage single-phase grid-connected PV system with MPPT Fig. 2. Active power control functions for wind turbine power systems
control: S - Solar irradiance and Ta - Ambient temperature. defined in the Danish grid code [10].

In view of the above issues, a Constant Power Generation [12], [15], [25]–[27]. It can be achieved by the droop function
(CPG) control concept of PV inverters is proposed in this of the grid frequency and the active power.
paper. Since the total energy loss in CPG operation mode is Besides, when a high penetration degree of PV systems is
limited, its feasibility is discussed in §II. The detailed imple- achieved, imbalances in or overloading of the grid will appear
mentation of this control method is presented in §III, followed due to intermittency and/or reconnection from fault situations.
by operation examples to show the potential contributions to Hence, similar to the requirements for wind power systems
the reduction of grid expansion requirements, and thus pave in Denmark [10], [28], it is better for PV systems to be
the way for the next generation PV systems to take a major equipped with active power control functions, such as active
part in electricity generation. Finally, §V concludes this work. power production constraint, delta production constraint, and
power ramp constraint. Those active power control functions
II. V IABILITY OF L IMITING F EED - IN P OWER OF are indicated in Fig. 2. In that case, the PV systems have to
P HOTOVOLTAIC S YSTEMS provide a wide range of active power controllability, including
A typical single-phase grid-connected PV power generation constant power generation, the feasibility of which will be
system is shown in Fig. 1. Maximizing the feed-in power detailed in the following.
from such systems, also known as MPPT control, is required
in current grid integration policies. This objective can be B. Constant Power Generation Viability
achieved with a dedicated MPPT system as it is shown in Fig. Integrating energy storage systems into PV systems seems
1, and normally a DC-DC boost stage is adopted. The resultant to be a good solution to desirably control the active power, as it
two-stage system offers a flexible controllability of the input has been presented in [7], [16], [18] and [19]. However, it will
PV power, and an extension of operational hours when the introduce additional cost and additional lifetime-limited energy
solar insolation reduces. Moreover, the PV systems have to storage components, making it not an optimal approach [27].
disconnect from the grid in the case of disturbances, e.g. volt- Considering the expenses, energy storage systems are mostly
age variations and frequency deviations. Those requirements installed at the substation side instead of integrated in the PV
were introduced initially based on a low penetration level of inverters in order to achieve constant power production.
PV systems, where adverse impacts were negligible. However, The constant (absolute) power production control can also
catering for a vast amount of PV hosting capacity magnifies be attained at the PV inverter level. A representative study
the influences from PV systems on the grid, and thus those of a 3 kW single-phase PV system in a certain place with a
policies are going to change [1], [9], [15] yearly mission profile has revealed its feasibility. The yearly
mission profile shown in Fig. 3 includes recorded ambient
A. Active Power Control Functions temperature and solar irradiance level, and hence the annual
The rapid installation of PV systems can cause voltage energy production of the same place has little variation in
rise, which could damage household appliances in worst case different years. As it is shown in the results (Fig. 4), a 20%
scenarios. In order to solve the problem of voltage rise on the reduction of the maximum feed-in power from PV systems
distributed feeders, the PV inverters have to either enable reac- only leads to a 6.23% reduction of yearly total energy yield.
tive power control or to limit the active power generation [12], Thus, it is viable and reasonable to increase PV penetration
[15], [23]–[25]. Due to the high R/X ratio of low-voltage degree without violating the grid capacity by limiting the
grids, reduction of active power shows more effectiveness maximum feed-in power from current existing PV systems. At
in voltage regulation than reactive power control. Curtailing the same time, frequency stability can potentially be improved
active power output is similar to constant power production by enabling absolute power control [27]. Such issues are
when the voltage amplitude reaches the upper level. Moreover, already being discussed in some countries [1], [6], [7], where
frequency control through active power has been introduced the PV systems share a considerable proportion of electricity
for PV systems as well to stabilize the grid frequency [8], generation, e.g. Germany.

380
 1500 Voltage Voltage Limitation (Vlimit)

Solar Irrandiance (W/m²) 1250

Power
1000
P1 P2 P3 Pm Pn
750
C
500

250 P*1=Pmax1 P*2=Pmax2 P3*=0 Pm* =0 Pn*=Pmaxn

0 Vlimit Central Control Unit
Oct. Nov. Dec. Jan. Feb. Mar. Apr. May Jun. Jul. Aug. Sep.
(a) Cutoff PV systems Communications
50
Ambient Temperature (ºC)

Fig. 5. Voltage reduction by cutting off PV systems during peak

40 production period in a highly PV-penetrated system.
30
20

10 This control strategy can be implemented by means of: a)

0 enabling the energy storage systems, b) managing the total
-10 power of an aggregated PV system, and c) modifying the
-20 MPPT control of the PV inverters.
Oct. Nov. Dec. Jan. Feb. Mar. Apr. May Jun. Jul. Aug. Sep.
(b)
A. Integrating Energy Storage Systems
Fig. 3. A yearly mission profile from October 2011 to September 2012 for Although integration of energy storage systems into highly
a 3 kW single-phase grid-connected PV system in a certain place: (a) solar PV-penetrated grids will increase the total expenses, signifi-
irradiance and (b) ambient temperature.
cant improvements of active power control functions can be
120
achieved [7], [16], [19]. Mostly, energy storage systems are
installed at the medium-voltage side (substations), and thus for
Reduction Percentage of Annual

100
large PV power plant applications. This technology has several
80 advantages and gives flexibility to the active power control by
Energy Production

60 20 % Reduction providing an energy ”reservoir”, and thus smoothing the active

of Feed-in Power power [16], [19]. One important feature with energy storage
40
system is that the PV output power can be free of fluctuations
6.23 % Reduction
20
of Energy Yield
and higher than its nominal value by continuously discharging
0 the energy storage systems. The penalty is additional cost and
0 20 40 60 80 100 120 the introduction of lifetime-limited energy storage components
Feed-in Power Percentage of PV Peak Power
into the system. Nevertheless, due to the significance in the
Fig. 4. Energy reduction due to the limitation of maximum feed-in power increase of self-consumption abilities and the reduction of
from a 3 kW grid-connected single-phase PV system. energy bills, many efforts have been made into the integration
of PV systems with batteries or mobility energy storage
systems (e.g. electric vehicles) [21], [29], [30].
As it has been illustrated, reducing the maximum feed-in
B. Power Management Control
power from the PV systems can contribute to decreased burden
of the distributed grid with a high penetration level. This Inspired by the concept to alleviate voltage rise through
means that a certain percentage of the energy from the current active power curtailment, appropriately managing the power
PV systems is cut off arbitrarily in a long-term operation, at the secondary control level in an aggregated PV system can
which is shown in Fig. 4, and thus in a short-term operation, contribute to a constant power production. As it is shown in
the PV systems should be able to operate in CPG mode when Fig. 5, when the distributed grid voltage exceeds the limitation
the output power reaches a certain level (e.g. 80% of the due to strong solar insolation in a highly PV-penetrated system,
nominal maximum power). The following section presents the several PV systems have to be cut-off in order to bring the
implementation of the CPG approach. voltage back [25]. Similarly, the total output power can be
maintained constant by properly managing the power of each
III. I MPLEMENTATION OF C ONSTANT P OWER PV unit, as it is shown in Fig. 6, where the entire system is
G ENERATION C ONTROL controlled centrally. The power reference command for each
The objectives of active power control functions include: PV system is sent out from a central control system. The total
1) further increasing the penetration degree of PV systems output power can be expressed as,
⎧ n
without the necessity of grid expansion and 2) participating in ⎪  n
⎪
⎨ Pi , when Pmaxi < Plimit
frequency regulations to ensure grid stability. Hence, the PV
Ptotal = i=1 i=1
n (1)
system should be able to enter into a CPG mode when the ⎪
⎪ ≥
⎩ limit
P , when P maxi Plimit
output reaches a certain level (e.g. 80% of the peak power). i=1

381
 Total Power Power Limitation (Plimit) Operation Mode
ipv
MPPT kmpp Ppv < Plimit
Power vpv
PWMb
P1 P2 P3 Pm Pn Ppv
kcpg Ppv ≥ Plimit
C Plimit (a)

P1* P2* P3* Pm* Pn* vdc PI Controller

ki ig* Current
Plimit Central Control Unit *
vdc kp+ s PWMinv
Controller
Communications
Phase ig
vg
Fig. 6. Power management control to achieve a total constant power Locked Loop
(b)
production in an aggregated PV system.

Fig. 7. Control diagram of a two-stage single-phase PV system with

constant power generation ability by modifying MPPT control: (a) boost
where Pmaxi is the maximum power of each PV system control diagram and (b) PV inverter control system.
achieved by MPPT control and Plimit is the total power
limitation of the distributed grid. I, III, V- MPPT II, IV- CPG
As it is illustrated in Fig. 6, in this control strategy, some III IV
PV units operate in MPPT mode, while some in CPG mode. I II V
Pmax
This means that an individual PV inverter should be able to
adjust its output power (e.g. constant power) according to Plimit

Power
the power set-points from the central control unit. Moreover,
this system can also operate in delta power production mode PPV

or in a controlled power ramp-rate mode, only when the

Energy yield
power reference for each PV unit is allocated appropriately.
Notably, the implementation of this control strategy requires
t0 t1 t2 t3 t4 t
the knowledge of the forecasted mission profiles (a better
Time
monitoring system and forecast algorithms) and an enhanced
communication system. Fig. 8. Operation regions for a single-phase PV system during a day.

C. Modifying MPPT Control

For an individual PV inverter, the controllability of active PV system during a day. Then, the energy production in
power can be achieved by modifying the MPPT algorithm [6]. different operation modes can be calculated as ,
In contrast to constant power production with energy storage ⎧
systems, with modified MPPT control in the CPG operation ⎪ t
⎪
⎨Empp = PP V (t)dt
mode, there is no need to install extra devices and thus there t0
t2 t4 (3)
are no additional expenses in order to increase the penetration ⎪
⎪
degree of PV systems. ⎩Ecpg = Plimit dt + Plimit dt
t1 t3
In regard to Fig. 1, the control structure of a two-stage
single-phase PV system with CPG control by modifying the where Empp is the available energy during a day with MPPT
MPPT control algorithm can be represented as shown in Fig. control, and Ecpg is the energy production in CPG operation
7. As it is shown in Fig. 7(a), the CPG control is implemented mode in regions of II and IV in Fig. 8. Subsequently, the cut-
in the control of the DC-DC stage, and hence the whole system off energy can be expressed as Eess = Empp − Ecpg , which
can be divided into two operation modes based on the output could be used to size, e.g. an energy storage system.
power of the PV panels - the MPPT control mode and the According to Fig. 7, Fig. 8, and (3), the operation principle
constant power production mode. Hence, the actual power of CPG control for a PV inverter can be described as follows.
production can be expressed as, When Ppv ≥ Plimit , the system enters into CPG operation
 mode and the MPPT control is deactivated. The PV output
PP V (t), when PP V (t) < Plimit
Po = (2) power is regulated by a proportional controller (kcpg ). The
Plimit , when PP V (t) ≥ Plimit
proportional controller is able to maintain the output power
in which Po is the output power, PP V (t) is the instantaneous constant during the operation. When Ppv < Plimit , the system
power of the PV system, and Plimit is the power limitation. should deliver as much energy as possible to the grid with an
During a day, the output power of PV strings is changing advanced MPPT control algorithm, and thus the CPG control
with the environmental conditions, which means that the is disabled. In both operation modes, the DC-link voltage (vdc )
operation modes will also alter with solar insolation and/or is regulated via a Proportional Integrator (PI) controller to
ambient temperature, especially in a cloudy day. Fig. 8 shows be constant to ensure the power injection. Current controllers
an example of different operation regions for a single-phase should be selected in terms of good harmonics rejection.

382
 TABLE I TABLE III
PARAMETERS OF THE BP 365 S OLAR PV PANEL . C ONTROL S YSTEM PARAMETERS .

Rated power Pmpp = 65 W MPPT control gain kmpp = 23.8

Voltage at Pmpp Vmpp = 17.6 V CPG control gain kcpg = 0.1
Current at Pmpp Impp = 3.69 A DC-link (PI) controller kp = 0.1, ki = 1.25
Open circuit voltage VOC = 21.7 V PR controller for grid current krp = 20, kri = 2000
Short circuit current ISC = 3.99 A
1200
TABLE II
PARAMETERS OF THE T WO -S TAGE S INGLE -P HASE PV S YSTEM . 2925 W

Solar Irradiance (W/m2)

1000
Boost converter inductor L1 = 5 mH
DC-link capacitor Cdc = 2200 μF 800
PV-side capacitor CP V = 220 μF 2030 W
Inverter-side inductor:
Linv = 2 mH, Grid-side 600
LCL-filter 1450 W
inductor: Lgf = 3 mH,
Capacitor: Cf = 4.7 μF Ta = 25°C
Damping resistor of LCL-filter Rd = 10 Ω 400
0.5 1 1.5 2 2.5 3
Switching frequencies fb = finv = 10 kHz Time (s)
MPPT sampling frequency fmppt = 300 Hz
Grid nominal voltage (RMS) Vg = 230 V Fig. 9. Solar irradiance level profile under a constant ambient temperature.

Grid nominal frequency ω0 = 2π × 50 rad/s

objective is to ensure the quality of grid current, and thus a

Proportional Resonant (PR) controller shown in (4) is adopted
The constant power production control discussed in this
in the current control loop, since it can track a sinusoidal
paper is achieved by changing the operation modes. An
reference quickly and accurately [14], [34]. In some cases
alternative way to attain constant power production is to
of high power quality demands, harmonic compensations are
directly modify the MPPT algorithm, as it is presented in [6].
necessary. Regarding the grid synchronization, a second-order
Both solutions require a better knowledge of MPPT control
generalized integrator [34] based phase locked loop is adopted
algorithms. Typically, the Perturb-and-Observe (P&O) and
in this system. Parameters of the control system are shown in
incremental-conductance MPPT methods can be adopted in
Table III.
a PV system. However, a sudden change of the operation kri s
condition (e.g. from MPPT operation to CPG operation mode) GP R (s) = krp + 2 (4)
s + ω02
may introduce instability [31], being a challenge for those
MPPT control systems, which needs in-depth investigations. where ω0 is the grid nominal frequency, and krp , kri are the
control parameters.
IV. O PERATION E XAMPLES OF C ONSTANT P OWER Firstly, the single-phase PV system with CPG control by
G ENERATION C ONTROL modifying MPPT algorithm is tested under a constant ambient
In order to verify the effectiveness of the proposed CPG temperature, but the solar irradiance level has experienced
control concept, simulations of a two-stage single-phase PV ramp-changes, as it is shown in Fig. 9. The results of the
system are carried out in MATLAB/Simulink referring to Figs. PV system under this mission profile are presented in Figs.
1 and 7. The model of BP 365 PV panel [32] is adopted 10 and 11. It has been shown in those test results that, when
in the verifications. Parameters of the PV panel are given in the maximum power exceeds the limitation (80% of nominal
Table I. Three PV strings are connected in parallel to the DC- power set by the central control system), the CPG control has
DC converter, and each consists of 15 PV panels in series. been activated and thus constant output power of PV strings
Thus, the rated maximum power is Pmax = 2925 W. Other is achieved effectively. Once the output power goes below
specifications of the system are listed in Table II. the limitation due to low solar insolation or high ambient
Control systems are implemented according to Fig. 7. The temperature, the system returns to the MPPT operation with
P&O MPPT control method is adopted in the MPPT operation ensured stability, as it is shown in Figs. 10 and 11. Then, the
mode [33]. In the CPG operation mode, the control unit maximum power from the PV strings are fed into the grid.
receives the power set-point command, and the PV output Fig. 12 demonstrates the performance of the PV system
power is regulated using a simple proportional controller. with CPG control under two daily profiles according to Fig.
As for the control of PV inverter, the first objective is to 3. The energy production is also calculated based on equation
∗
maintain the DC-link voltage, i.e. vdc = 400 V. The second (3). The effectiveness of the proposed method by modifying

383
 3000
MPPT CPG MPPT 80% Feed-in of
Power to Grid
350 Maximum Power
(a)
PV Voltage (V)
2500 (2400 W)
300

Power (W)
250 2000

200
PV Power
0.5 1 1.5 2 2.5 3
1500
Time (s)
12 MPPT
MPPT CPG
PV Current (A)

(b)
1000
9
0.5 1 1.5 2 2.5
6 Time (s)
(a)
3 3000
0.5 1 1.5 2 2.5 3 CPG
Time (s) S=1000 W/m2
20 2500
(c) S = 700 W/m2
Grid Current (A)

Power (W)
10
0 2000

-10 MPPT
-20 1500
0.5 1 1.5 2 2.5 3
Time (s) S = 500 W/m2
1000
200 250 300 350
Fig. 10. Performance of the single-phase PV system with constant power
Voltage (V)
generation control: (a) PV voltage, (b) PV current and (c) grid current.
(b)

Fig. 11. Simulation results of a single-phase PV system with constant

MPPT control to limit maximum feed-in power has been power generation control by limiting the maximum feed-in power: (a)
demonstrated in the results. However, the injected power to output power and (b) PV power v.s. PV voltage.
the grid presents large overshoot as shown in Fig. 12(c) due
to rapid solar insolation changes of a cloudy day. This means
that the robustness of the control system has to be enhanced by the system reliability is possible to be enhanced. However,
optimizing the control parameters. Nonetheless, the proposed installations of energy storage systems increase the total cost.
concept is able to control the PV systems with a constant Based on the discussions in the paper, further research can
power output and limited loss of energy, and thus more PV be put on: stability analysis of MPPT methods and the entire
systems can be installed without upgrading the grid capacity. control system, detailed calculation of energy losses, controller
parameter design, experimental verifications, and application
V. C ONCLUSION
of the CPG concept in advanced power control strategies.
In this paper, a Constant Power Generation (CPG) control
concept of PV systems considering the distributed grid capac- R EFERENCES
ity has been explored. Since a certain limitation (e.g. 80%) [1] D. Rosenwirth and K. Strubbe, “Integrating variable renewables
of PV maximum feed-in power has an insignificant impact as Germany expands its grid,” [Online]. Available: http://www.
on the annual total energy production (e.g. 6.23% energy renwableenergyworld.com/, Mar. 2013.
[2] K.O. Kovanen, “Photovoltaics and power distribution,” Renewable En-
reduction), the CPG control concept has been proposed, which ergy Focus, vol. 14, no. 3, pp. 20–21, May/Jun. 2013.
can help the DSOs to avoid grid expansion and thus save [3] REN 21, “Renewables 2013: Global Status Report (GSR),” [Online].
money, but at the same time increase the PV hosting capacity. Available: http://www.ren21.net/, Jun. 2013.
Moreover, this concept can alleviate overloading issues for PV [4] C. Winneker, “Worlds solar photovoltaic capacity passes 100-Gigawatt
landmark after strong year,” [Online]. Available: http://www.epia.org/
systems operating during peak production. The implementa- news/, Feb. 2013.
tions of CPG control could achieved by power management [5] D. Maxwell, “Parts of Northern Ireland’s electricity grid overloaded,”
control, integration of energy storage systems and modification BBC News, [Online]. Available: http://www.bbc.co.uk/, 13 Nov. 2013.
[6] A. Ahmed, L. Ran, S. Moon, and J.-H. Park, “A fast PV power tracking
of the MPPT control. The results verified the effectiveness control algorithm with reduced power mode,” IEEE Trans. Energy
of the CPG control by modifying MPPT algorithm. While Conversion, vol. 28, no. 3, pp. 565–575, 2013.
the PV systems with integrated energy storage systems can [7] T. Thien, R. Alvarez, Z. Cai, P. Awater, M. Leuthold, A. Moser, and D.U.
Souer, “Storage- and grid expansion needs in a European electricity-
contribute to not only the avoidance of grid expansion but supply-system with a high share of renwable energy,” in Proc. of
also smoothing of PV feed-in power fluctuations, and thus IRES’12, pp. 656-663, Nov. 2012.

384
 1200 35 3500
12 samples/h 2 samples/h
Solar Irradiance Level (W/m2)

MPPT
1000 12 samples/h 30 3000 20

Ambient Temperature (°C)

80% Feed-in
2 samples/h

Output Power (W)

25 2500 CPG (2400 W)
800
20 2000
600
15 1500
400
10 1000
200 5 500

0 0 0 15
1200 35 3500
MPPT
Solar Irradiance Level (W/m2)

2 samples/h
1000 30 3000 20
80% Feed-in

Ambient Temperature (°C)

Output Power (W)

12 samples/h 25 2500 CPG (2400 W)
800
2 samples/h
20 12 samples/h 2000
600
15 1500
400
10 1000
200 5 500
0 0 0 15
0 50 100 150 200 250 0 50 100 150 200 250 0 2 4 6 8 10 12 14
Number of Samples Number of Samples Simulation Time (s) Energy (kWh)
(a) solar irradiance level (b) ambient temperature (c) output power and enery production

Fig. 12. Results of a 3 kW single-phase PV system with constant power generation control using recorded daily data (up: a clear day, down: a cloudy day).

[8] F. Blaabjerg and K. Ma, “Future on power electronics for wind turbine [21] G. Hoffmann, “Intelligent PV- Technologies, costs and potential,” in
systems,” IEEE Journal of Emerging and Selected Topics in Power Proc. of IRES’12, pp. 976-1009, Nov. 2012.
Electronics, vol. 1, no. 3, pp. 139–152, 2013. [22] M. Vandenbergh, D. Craciun, V. Helmbrecht, R. Hermes, R. Lama,
[9] Y. Bae, T.-K. Vu, and R.-Y. Kim, “Implemental control strategy for grid P.M. Sonvilla, M. Rekinger, G. Concas, H. Loew, C. Mateo, P. Frias,
stabilization of grid-connected PV system based on German grid code E. Collado, Z. Musilova, and D. Busek, “Prioritization of technical
in symmetrical low-to-medium voltage network,” IEEE Trans. Energy solutions available for the integration of PV into the distribution grid
Conversion, vol. 28, no. 3, pp. 619–631, Sept. 2013. published,” PV GRID Project, Tech. Rep.
[10] ENERGINET.DK, “Wind power plants with a power output greater than [23] G. Mokhtari, A. Ghosh, G. Nourbakhsh, and G. Ledwich, “Smart robust
11 kW,” Technical Regulation 3.2.5, Sept. 2010. resources control in LV network to deal with voltage rise issue,” IEEE
Trans. Sustain. Energy, vol. 4, no. 4, pp. 1043–1050, 2013.
[11] E.ON GmbH, “Grid Code-high and extra high voltage,” 2006.
[24] R. Caldon, M. Coppo, and R. Turri, “Distributed voltage control strategy
[12] VDE Verband der Elektrotechnik Elektronik Informationstechnik e.V., for LV networks with inverter-interfaced generators,” Electric Power
“Power generation systems connected to the low-voltage distribution Systems Research, vol. 107, pp. 85–92, 2014.
network - Technical minimum requirements for the connection to and [25] Y. Miyamoto, “Technology for high penetration residential PV systems
parallel operation with low-voltage distribution networks,” VDE-AR-N on a distribution line in Japan,” in Proc. of the fifth Int’l Conf. on
4105:2011-08, Aug. 2011. Integration of Renewable and Distibuted Energy Resources, 4-6 Dec.
[13] S.B. Kjaer, J.K. Pedersen, and F. Blaabjerg, “A review of single-phase 2012.
grid-connected inverters for photovoltaic modules,” IEEE Trans. Ind. [26] G. Kaestle and T.K. Vrana, “Improved requirements for the connection
Appl., vol. 41, no. 5, pp. 1292–1306, Sept./Oct. 2005. to the low voltage grid,” in Proc. of CIRED, pp. 1-5, 6-9 Jun. 2011.
[14] F. Blaabjerg, R. Teodorescu, M. Liserre, and A.V. Timbus, “Overview [27] A. Hoke and D. Maksimovic, “Active power control of photovoltaic
of control and grid synchronization for distributed power generation power systems,” in Proc. of SusTech’13, pp. 70-77, 2013.
systems,” IEEE Trans. Ind. Electron., vol. 53, no. 5, pp. 1398–1409, [28] J. Cochran, L. Bird, J. Heeter, and D.J. Arent, “Integrating variable
Oct. 2006. renewable energy in electric power markets: Best practices from inter-
[15] Y. Yang, P. Enjeti, F. Blaabjerg, and H. Wang, “Suggested grid code national experience,” National Renewable Energy Laboratory (NREL),
modifications to ensure wide-scale adoption of photovoltaic energy in Golden, CO, USA, Tech. Rep., Apr. 2012.
distributed power generation systems,” in Proc. of IEEE-IAS Annual [29] R. Doelling, “Greater economic viability with smaller storage capaci-
Meeting, pp. 1-8, 6-11 Oct. 2013. ties?” [Online]. Available: http://www.solarenergystorage.org/en/, 2013.
[16] H. Gaztanaga, J. Landaluze, I. Etxeberria-Otadui, A. Padros, I. Beraza- [30] J. Traube, F. Lu, D. Maksimovic, J. Mossoba, M. Kromer, P. Faill,
luce, and D. Cuesta, “Enhanced experimental PV plant grid-integration S. Katz, B. Borowy, S. Nichols, and L. Casey, “Mitigation of solar
with a MW Lithium-Ion energy storage system,” in Proc. of ECCE, pp. irradiance intermittency in photovoltaic power systems with integrated
1324-1329, 15-19 Sept. 2013. electric-vehicle charging functionality,” IEEE Trans. Power Electron.,
[17] T. Stetz, F. Marten, and M. Braun, “Improved low voltage grid- vol. 28, no. 6, pp. 3058–3067, 2013.
integration of photovoltaic systems in Germany,” IEEE Trans. Sustain. [31] L. Nousiainen, J. Puukko, A. Maki, T. Messo, J. Huusari, J. Jokipii,
Energy, vol. 4, no. 2, pp. 534–542, 2013. J. Viinamaki, D.T. Lobera, S. Valkealahti, and T. Suntio, “Photovoltaic
generator as an input source for power electronic converters,” IEEE
[18] H. Beltran, E. Perez, N. Aparicio, and P. Rodriguez, “Daily solar energy Trans. Power Electron., vol. 28, no. 6, pp. 3028–3038, 2013.
estimation for minimizing energy storage requirements in PV power [32] Plexim GmbH, “PLECS User Manual Version 3.4,” [Online]. Available:
plants,” IEEE Trans. Sustain. Energy,, vol. 4, no. 2, pp. 474–481, 2013. http://www.plexim.com/, 14 Jun. 2013.
[19] H. Beltran, E. Bilbao, E. Belenguer, I. Etxeberria-Otadui, and P. Ro- [33] D. Sera, L. Mathe, T. Kerekes, S.V. Spataru, and R. Teodorescu, “On
driguez, “Evaluation of storage energy requirements for constant pro- the perturb-and-observe and incremental conductance MPPT methods
duction in PV power plants,” IEEE Trans. Ind. Electron., vol. 60, no. 3, for PV systems,” IEEE Journal of Photovoltaics, vol. 3, no. 3, pp. 1070–
pp. 1225–1234, 2013. 1078, 2013.
[20] W. Deis and C. Kolligs, “Yet another storage project? Going beyond [34] M. Ciobotaru, R. Teodorescu, and F. Blaabjerg, “Control of single-stage
just storing photovoltaic energy,” in Proc. of IRES’12, pp. 1029-1047, single-phase PV inverter,” in Proc. of EPE’05, pp. P.1-P.10, 2005.
Nov. 2012.

385