Received 23 October 2023, accepted 6 November 2023, date of publication 8 November 2023,

date of current version 14 November 2023.

Digital Object Identifier 10.1109/ACCESS.2023.3331364

Fault-Tolerant Strategy for the MMC-Based PV

System With Faults Detection and Converter
Reconfiguration Using Permutation Algorithms
JUAN C. COLQUE 1 , ERNESTO RUPPERT1 , (Member, IEEE),
I. DELGADO-HUAYTA2 , (Member, IEEE), KARLOS A. C. CHIRAPO 2, (Member, IEEE),
AND JOSÉ LUIS AZCUE 3
1 Department of Systems and Energy, State University of Campinas, Campinas, São Paulo 13083-852, Brazil
2 ElectronicEngineering Professional School, National University of the Altiplano, Puno 21001, Peru
3 Engineering, Modeling, and Applied Social Sciences Center, Federal University of ABC, Santo André, São Paulo 09210-580, Brazil

Corresponding author: Juan C. Colque ([email protected])

This work was supported by the Brazilian National Council for Scientific and Technological Development (CNPq) under Grant
440138/2019-1.

ABSTRACT The modular multilevel converter has gained popularity in various applications, including
photovoltaic (PV) solar energy conversion. Its modular structure allows for the transformation of an
MMC into an MMC-based photovoltaic system, sharing key operational characteristics such as modularity,
flexibility, redundancy, increased efficiency, and fault tolerance. To ensure the reliability and uninterrupted
operation of the modified MMC, even in the event of potential failures in the photovoltaic submodules
(PVSMs), a fault-tolerant strategy is developed in this study. It assumes that the Maximum Power Point
Tracking (MPPT) of the PVSMs is already guaranteed. Redundant submodules (rSM) are utilized to maintain
power balance between the converter arms through voltage control, while reserve submodules (RSMs) are in
place to rescue the converter in case of a failure. The detection and localization of faults in the PVSMs/rSMs
are achieved through sliding mode observers (SMOs), and the converter reconfiguration is carried out using
the proposed permutation algorithms for switching signals and SMs voltages. For precise control of the
output current and electrical grid connection, the dq-reference frame control method is employed. To validate
these proposed algorithms, time-domain simulations are conducted using the Simulink/Matlab software.

INDEX TERMS Photovoltaic solar energy conversion, modular multilevel converter, fault-tolerant
operation, power mismatch, photovoltaic submodules, redundant submodules, permutation algorithm.

I. INTRODUCTION elimination of DC line cabling. Consequently, it presents an

The continually expanding energy market, coupled with appealing solution for grid-connected PV systems, despite
the rising trend in decentralized energy generation plants, its associated cost drawback, as noted in a previous
whether on a large or small scale, serves as a catalyst for study [1]. An even more sophisticated alternative involves
the advancement of cutting-edge electronic power systems in the utilization of permutation algorithms. These algorithms
photovoltaic energy generation. Traditionally, microinverters enable the extraction of maximum power from individual PV
have been employed for smaller PV arrays to enhance power panels, even under partial shading conditions, and facilitate
generation, particularly in scenarios involving partial shading the delivery of the greatest available power to the DC bus
of PV panels. This approach offers several advantages, connected to the inverter. This innovative approach addresses
including improved energy capture, seamless plug-and-play the challenges posed by PV panels operating under such
functionality, flexibility for expansion, redundancy, and the conditions [2].
One of the main electronic equipment to convert energy
The associate editor coordinating the review of this manuscript and is the voltage source converter (VSC) that use capacitors
approving it for publication was Jahangir Hossain . as energy storage elements. The two-level converter is
2023 The Authors. This work is licensed under a Creative Commons Attribution 4.0 License.
VOLUME 11, 2023 For more information, see https://creativecommons.org/licenses/by/4.0/ 126119
 J. C. Colque et al.: Fault-Tolerant Strategy for the MMC-Based PV System

the standard VSC (2L-VSC), which is typically used in based on the concept of virtual submodule (VSM) using
low voltage and low power applications [3], [4]. However, the selective virtual loop mapping control (developed in
to reach higher voltages and obtain advantages such as: these paper) that does not require voltage classification from
transformerless systems, lower voltage stress on devices, highest to lowest, it only identifies the capacitor voltage rating
smaller output filters, lower switching losses (due to low vc,min and vc,max , which makes it suitable for an MMC with
switching frequency), among others, multilevel converters a large number of SMs/arm. However, due to the use of
(MLC) are used, but are limited in terms of the number a centralized converter, the power generated under partial
of levels achievable, due to the large number of electronic shading is drastically reduced because of the general MPPT
devices (transistors, diodes, capacitors or DC sources) used in algorithm. In [24] a topology for a solar PV power generation
the different topologies of MLCs [4]. Therefore, to solve this system under partial shading conditions is proposed, in these
limitation, in [5] the Modular Multilevel Converter (MMC) article, the MMC employs PV panels in series that are
was introduced in the literature, which as a result of the directly connected to each half-bridge converter (also knew
advancement of technology made possible the development SM), this arrangement is called the power module (PM)
of large-scale applications. In the beginning, this converter which will operate as SM. The maximum power is extracted
was formed by half-bridge submodules (SMs), but due by regulating the average voltage between all capacitors
to its importance, several authors have proposed different (this value is close to the voltage at the maximum power
SM topologies (which depend on the type of application, point). The main drawback of this control is when the partial
resource optimization and to facilitate control). The MMC shading condition occurs, because the average voltage of
can normally operate with a different number of SMs/arm, all capacitors drops, generating energy generation losses.
however, for symmetrical operation this number must be Also, a redundant submodule (rSM) is added to each arm
identical on all arms. Modularity is its biggest advantage, to compensate for the voltage variation of energy generated
however, for this converter to work optimally, it is necessary by the PMs in partial shading condition. In [25] a new
to properly control the submodule voltage, the circulating topology is proposed, where the PV panels are connected to
current and the output current. an isolated DC/DC converter of dual active bridge (DAB)
To date, several studies have been reported in order to which is connected to each SM of the MMC. The energy
improve the reliability and performance of the MMC which balance between the converter arms is done through of the
mainly include: minimization of circulating current, capacitor power mismatch elimination strategy. The direct connection
voltage balancing, output current control, continuous-time between the isolated converters and the SMs allows the
and discrete-time modeling, modulation techniques, DC-bus independent action of the MPPT algorithm and the PV
fault tolerant operation (at the terminals of DC-bus) and modules ground connection.
submodule fault tolerant (in the internal devices of the MMC) In [28] a grid-connected PV system based on MMC
[3], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], is proposed, which realizes long-distance DC transmission
[17], [18]. The MMC based on the voltage source converter while feeding local power consumption. In addition, by using
(VSC), has several applications, such as: energy storage [19], the triple active bridge converter, there is no photovoltaic
drive of medium/high-power motors [3], [9], photovoltaic power mismatch between the upper and lower arms, since the
(PV) solar energy conversion [20], [21], [22], [23], [24], [25], isolation transformer couples two SMs (one in the upper arm
[26], [27], [28], electric vehicles [13], static synchronous and the other in the lower arm) on a three-ports transformer,
compensators (STATCOM) at medium voltage [29], among consequently, the differential mode components from the
others. However, in this article the fault tolerance of the MMC upper and lower arms cancel each other.
in photovoltaic system applications is specifically studied, To ensure that the MMC remains operational even with a
based on the information already existing in the literature faulty redundant PVSM or SM, some fault-tolerant strategy
and proposing a new utility of the MMC with photovoltaic must be included in the converter control. In this way, the
submodules, this information will be done in more detail in MMC can continue without affecting overall performance.
the next sections. This condition is insured by a minimum percentage number
The research of the MMC in solar PV applications is of failed SMs/arm, this is known as a redundancy factor
currently booming and is not yet an established technology which is around 10%. This means that the converter operation
in the literature compared to other applications or based can continue if less than 10% of the SMs fail, for a
on the MMC. In [20], the research studies are classified conventional MMC [16]. The detection methods mainly
in two types, projects that employ a common DC-bus and focus on semiconductor failures such as short-circuit and
others that use separate PV arrays to energize the submodules open-circuit, which can happen due to overcurrent, high
(SMs) with/without the isolated DC/DC stage. In [21] the temperature of the devices, or incompatibility of the thermal
MMC is studied as a centralized DC/AC inverter for the coefficients between silicon and aluminum [15]. Among
implementation of PV systems in distributed generation, the main detection and locations of faults methods we
in this application only the scalability capacity is used to have: sliding mode observers (SMO), Kalman filters, state
obtain large voltages at the converter output. In these article, observers, resilient structure, a clustering algorithm that
a new control for capacitor voltage balancing was introduced compares the calculated capacitances with the measured,

126120 VOLUME 11, 2023

J. C. Colque et al.: Fault-Tolerant Strategy for the MMC-Based PV System

supervision sensors and methods based on PWM modulation number of identical N −submodules, one arm at the top and
[14], [15], [17], [18], [30], [31]. All these fault diagnosis one arm at the bottom of each phase. Due to its modular
methods identify faults according to the internal dynamics of structure, the converter is scalable and flexible for any
the MMC. medium/high voltage application.
Based on all the information already mentioned, this paper Fig. 1 shows the structure of the MMC-based PV system
proposes a fault-tolerant strategy for the MMC-based PV proposed, the control algorithms are implemented in each arm
system and the redundant SM voltage distribution method, of the MMC, it is considering the photovoltaic SMs (PVSM),
the topology used is a version inspired by [16], [24], the redundant SM (rSM) and the reserve SMs (RSM). One
and [25]. The redundancy control and the fault-tolerant arm detailing is shown in Fig. 2. All SMs are connected in
operation is considered, which is necessary feature due to a series, the PVSMs generate energy in both arms of same
possible/unexpected failure in some of the SMs/PVSMs (this phase; due to partial shading conditions, there is a natural
failure could compromise the converter, generating cascade imbalance in the energy generated by the PV panels and,
failures in the arm or on the converter phase), to avoid this consequently, exist a energy discrepancy between the arms of
situation a backup system is required. Therefore, for the the MMC; the redundant SMs are used to correct this energy
system to be considered ‘‘fault-tolerant’’, three aspects are discrepancy and the possible voltage imbalance between the
necessary: arms. Also, the reserve SMs are used for the uninterrupted
(1) Fault diagnosis (detection and location of the SMs/PVSMs operation of the MMC in case of failure of any SM. For
at fault) and MMC reconfiguration. When a fault occurs in this, constant monitoring is carried out and when a failure is
any of the SMs, it is necessary to locate the fault in order detected, a contingency plan is activated [15], [17].
to reconfigure the converter to guarantee the continuous and
uninterrupted operation of the MMC. For this, the SMO is
used for diagnosis and, for the reconfiguration, a permutation
algorithm proposed, using the signals generated by the SMO.
(2) The redundant control and fault-tolerant MMC structure.
During normal operation (without faults) the redundant SMs
(rSM) are used to compensate for possible voltage imbalances
in the DC bus. When a rSM fault is detected a reserve SM
(RSM) is inserted in the circuit. However, when a PVSM
fault is detected, a new SM is inserted into the main circuit
and the rSM reference voltage is modified. For this, the
Redundant operation based on Spare SMs (RSS) strategy
based on submodule cold reserve method is used this way the
reserve SMs will be available to be inserted into the circuit
[32]. When a PVSM faults, the inserted RSM takes a period
of time to reach the reference voltage value generating a short
transient.
(3) The operation strategy. Basically consists of controlling
the converter together with the considerations above men-
tioned, for this purpose, it is necessary to guarantee the
adequate control of the following points: the energy balance
between the arms of each phase, the current control output,
the minimization of the circulating current and the control of
each sub-module (PVSM, rSM).
FIGURE 1. The proposed MMC-based PV system.
This paper is structured as follows. A system description
and its control is presented in Section II. The proposed fault-
tolerant strategy is detailed in Section III. Section IV presents B. MATHEMATICAL MODELING
the simulations results of the proposed fault-tolerant strategy The submodules are connected in series with an inductor in
for the MMC-based photovoltaic system. Finally, conclusions each converter arm, the upper and lower arm submodules are
are presented Section V. modeled as controlled AC voltage sources. The DC system
part is modeled as two virtual DC voltage sources.
II. MMC-BASED PV SYSTEM STRUCTURE AND ITS The upper and lower arm voltage are given by:
CONTROL

V di
A. SYSTEM DESCRIPTION
 vxu = dc − vxo − La xu − ra ixu

2 dt (1)
The conventional MMC is composed of two arms (single- V di
 vxl = dc + vxo − La xl − ra ixl

phase) or six arms (three-phase) and each arm contains a 2 dt

VOLUME 11, 2023 126121

 J. C. Colque et al.: Fault-Tolerant Strategy for the MMC-Based PV System

By Kirchhoff’s current law, the output current is given by: In this paper, the output current control approach is adopted
to manage active and reactive power. Then, the command
ixo = ixu − ixl (2) voltages for the submodules can be obtained through the
From (1), vxl is subtracted from vxu , and ixo is substituted following process: Substituting (4) in (5) (for each voltage),
for ixu − ixl , obtaining: these are the arm modulation voltage in steady state, defined
by:
La dixo ra vxl − vxu
vxo = · + ixo + (3) 
V
2 dt 2 2  vxu = dc − v∗xo − vxz

2 (6)
The effect of ixo can be neglected as long as the value of V
 vxl = dc + v∗xo − vxz
La and ra are relatively low and the new variable v∗xo is

2
introduced, this defines the reference voltage vxo , given by:
Then, the arm modulation index are given by:
vxl − vxu
v∗xo = (4) 
vxu 1 v∗ + vxz
2  mxu =
 = − xo
Vdc 2 Vdc (7)
v 1 v ∗ − vxz
 mxl = xl xo
= +

Vdc 2 Vdc
The equation (7) has two control variables v∗xo and vxz , v∗xo
is the reference AC output current control variable [34], and
vxz is the voltage drop across the arm inductor used to control
the circulating current.
mxu and mxl are processed by the modulation block to
generate switching signals Sxu k and S k for each submodule.
xl
However, in Section II-D the independent balancing factor
for each SM is added as shown Fig. 3.
In Fig. 3, the switching signal generation diagram is
presented, outlining the key components of the MMC control.
This diagram provides a comprehensive view of the control
strategy and its implementation within the MMC. The
switching signal generation diagram serves as a reference
for understanding the control mechanisms involved in the
MMC, particularly in this application. It encompasses various
elements, including signal generation, modulation technique,
capacitor voltage balancing control, and circulating and
output current control. This visual representation is essential
for gaining insight into the control process, as it showcases
the flow of signals and actions within the MMC, which will
later be added to fault tolerance algorithms.
FIGURE 2. One-arm detailing of Fig. 1.
For the proposed converter, in the control strategy, there
are multiple vectors to control, including: (1) Output current
From (1), add vxl and vxu and, replace ixz in place of control: This is essential to regulate the current delivered
ixu + ixl
. Then, the voltage across the inductor La and resistor to the grid or the load. The limiting factors for this control
2 include the power rating of the MMC and the grid/load
ra caused by current ixz is defined by:
requirements. A limitation here can result in not meeting
dixz Vdc − (vxu + vxl ) the desired output current, affecting the power output of the
La + ra ixz = vxz = (5)
dt 2 converter, but in this paper, the power generated is injected
In this particular application, it’s important to consider that into the electrical grid. (2) Capacitor voltage control: The
the converter exhibits symmetry in the number of submodules voltage across the arm capacitors needs to be controlled
within both the upper and lower arms, as highlighted in [33]. to maintain the desired output voltage. The limiting factor
As a result, half of the grid current must be provided by each for this control is the voltage rating of the capacitors.
arm. To achieve this, it’s crucial to maintain the voltage of If the voltage exceeds this limit, it can result in capacitor
the submodule capacitors at a specific level. Additionally, overvoltage and potential failure. (3) Circulating current
another parameter that requires control is the circulating control: To minimize circulating current between the arms of
current, which needs to be suppressed or mitigated. This the converter, control strategies are employed. The limiting
control is essential to ensure the energy balance between the factor here is the control bandwidth and the dynamic response
arms of the converter. of the control system. If the circulating current is not well-

126122 VOLUME 11, 2023

J. C. Colque et al.: Fault-Tolerant Strategy for the MMC-Based PV System

FIGURE 3. Switching signal generation block diagram.

controlled, it can lead to losses and a power mismatch in the where j indicates the j−th element of the variable x(j), for this
converter. case j ∈ {1, . . . , M }, M represents the number of rSM/arm
The interplay between these controls is complex. Changes and vPVSM
c,xy is the PVSMs voltage vector. Initially the converter
in one control loop can affect the others. For instance, if the operates only with one rSM/arm.
output current control is too aggressive, it may lead to In this paper, the following clarification is made, the
increased circulating currents. If the voltage control is not reference voltage v∗c,xy
rSM
can be distributed among several rSMs
precise, it can affect the output current regulation. Therefore, of the same arm, if this value exceeds the rated value of
careful tuning and coordination of these control strategies are 0.75 vPVSM
c,xy (normally in case of failure of any PVSM). Then,
crucial to ensure the proper and efficient operation of the the control of rSMs is carried out through:
MMC. These controls are detailed to follow.
(
C. REDUNDANT SUBMODULE (RSM) CONTROL
1u∗xurSM (j) = K1u (v∗c,xu
rSM
c,xu (j)) sgn(ixu )
(j) − vrSM
(9)
This control is for specific applications, in this case for 1uxl (j) = K1u (vc,xl (j) − vc,xl (j)) sgn(ixl )
∗rSM ∗rSM rSM

the photovoltaic system based on the MMC. In [24] the

redundant control of the upper and lower arm voltage is where the sgn(x) function denotes the direction of the current,
proposed in order to obtain Vdc,xu = Vdc,xl . This is just to taking values of {−1, 0, 1}; u∗xurSM (j) represents the balancing
compensate the DC-bus voltage to a desired voltage, on each factor of rSM(j) and K1u is the proportional gain.
arm of the converter, and consequently improve its reliability.
In practical application, it is expected that the sum of the
D. PHOTOVOLTAIC SUBMODULE (PVSM) BALANCE
voltage of all submodules (photovoltaic and redundant) of
CONTROL
each arm will be constant and its value equal to the DC
∗ of the MMC. For this, the reference In [34] this method (for PWM signal) is studied in more
reference voltage Vdc
detail. In summary, balancing is possible because the MMC
voltage of the rSM is expressed by:
X submodules are modulated independently using iPSC-PWM
( ∗rSM
vc,xu (j) = Vdc∗
− (vPVSM ) technique. The voltage balance of the capacitors can be
X c,xu (8) achieved by adjusting the reference signal for each factor of
v∗c,xl
rSM ∗
(j) = Vdc − (vPVSM
c,xl ) (7). The reference signals u∗xuPVSM (j) and u∗xlPVSM (j) together with

VOLUME 11, 2023 126123

 J. C. Colque et al.: Fault-Tolerant Strategy for the MMC-Based PV System

the voltage balancing variable are given by: In (15), ido and iqo are coupled due to the presence of Ls ωo .
( To decouple the dynamic behavior, the control variables udo
u∗xuPVSM (j) = vxu + 1u∗xuPVSM (j) and uqo are introduced.
(10)
u∗xlPVSM (j) = vxl + 1u∗xlPVSM (j)   
ki4
 ∗
vdo = kp4 + (i∗do − ido ) −Ls ωo iqo + vds
1u∗xuPVSM (j) and 1u∗xlPVSM (j) represent the reference adjust-

s




ment of each submodule in the upper and lower arms, 
 | {z
u
}
do
respectively. For this case j ∈ {1, . . . , N }. These reference 
ki4 (16)
adjustments are intended to control the PVSMs voltage

 v∗
qo = kp4 + (i ∗
qo − i qo ) +Ls ω o ido + vqs
s



balance and can be calculated as [35] and [34]: 

 | {z }
( uqo
1u∗xuPVSM (j) = K1u (v∗c,xu
PVSM
c,xu (j)) sgn(ixu )
(j) − vPVSM
(11) v∗do and v∗qo are reference variables of vdo and vqo , respectively.
1uxl (j) = K1u (vc,xl (j) − vPVSM
∗PVSM ∗PVSM
c,xl (j)) sgn(ixl ) Therefore, v∗do ≈ vdo and v∗qo ≈ vqo .
Therefore, the normalized reference signals for each SM
with the voltage balancing factor, are given by: F. CIRCULATING CURRENT CONTROL
The AC component flows only between the legs of the MMC
1 v∗ + vxz 1u∗xuXSM (k)

 Mxu

 ∗XSM
(k) = − xo + and is derived from the upper and lower arm instantaneous
2 Vdc Vdc (12) currents, given by:
∗XSM 1 v∗xo − vxz 1u∗xlXSM (k)
 Mxl (k) = + +

 ixu + ixl
2 Vdc Vdc ixz = (17)
2
where the superscript XSM denotes all types of SMs in
The control of the circulating current consists basically in
operation (PVSM and rSM/RSM) and k denotes the k−th
the adequate distribution of the energy between the arms, that
element of variable x(k) that takes values {1, . . . , (N +M )}.
is, in the leg of the MMC. This ensures power balance and
symmetrical operation.
E. OUTPUT CURRENT CONTROL
The average voltage (V c,x ) measured from each leg is given
The output current of the MMC is controlled using the by:
synchronous dq−reference frame [36], and the block diagram
for the three-phase MMC is shown in Fig. 3. 1 h X OP X i
V c,x = (vc,xu ) + (vOP
c,xl ) (18)
 2
vas (t) = v̂s cos(ω
 o t) OP
Note that vOP c,xu and vc,xl are the the upper and lower

 
2π


vs (t) = vbs (t) = v̂s cos ωo t − 3
voltages vectors (SM in operation) that will be observed

−
→ (13)
  by the fault-tolerant algorithm. These vectors are composed
2π


 vcs (t) = v̂s cos ωo t + of the voltage of the arm capacitors as elements, i.e.


3
c,xu = [vc,xu , vc,xu , . . . , vc,xu , vc,xu ], with the SMs in
PVSM1 PVSM2 PVSMk rSM1
vOP
where v̂s represents the peak value of the grid voltage, ωo is initial operating status.
the angular frequency of the AC output voltage. Note that for Fig. 3 shows the control diagram, which has two loops.
closed loop AC voltage and AC current control is defined, The external one, regulates the average voltage of the leg
∗ , this loop minimizes
where, θ ∈ {0, −2π/3, 2π/3} is the phase angle for single- (2N −SMs) to a constant value of Vdc
phase or three-phase system. The dynamics of the AC three- the error voltage and provides the reference circulating
phase side of the VSC system is described by: current. The internal one, takes care of the circulating current
−
→ minimization.
d ixo −
→ → −
Ls = −rs ixo + −vxo − v→xs (14)
dt III. PROPOSED FAULT-TOLERANT STRATEGY BASED ON
where Ls represents the inductance of the output filter and rs SMO FOR THE MMC-BASED PV SYSTEM
is the resistance of the filter material. Then, the equation that The operating strategy mainly consists of:
expresses (14) in the frame of reference dq, is defined by [36]: (1) Output current control.

di (2) SM capacitor voltage control (SM: rSM/RSM and
 Ls do = −rs ido + vdo − vds + Ls ωo iqo

SMFV): Ensuring uniform voltage levels across SMs.
dt (15)
di • SM voltage equalization.
 Ls qo = −rs iqo + vqo − vqs − Ls ωo ido

dt • Voltage distribution of redundant SMs: Managing

where ido and iqo represent the current components in dq voltage distribution among redundant SMs when
frame, injected into the electrical grid; vds and vqs represents PVSMs fail.
the components in dq frame of the electrical grid three-phase (3) Command signals generation.
voltage; vdo and vdo are defined as the components in dq frame • Circulating current minimization.
of the converter output three-phase voltage. • Energy balance of the between the arms.

126124 VOLUME 11, 2023

J. C. Colque et al.: Fault-Tolerant Strategy for the MMC-Based PV System

• SM capacitor voltage balancing. The Disturbance compensation is the quality that an

(4) Fault-tolerant strategy. estimator must have to recover the estimated variable, even
• SMO for fault detection and location.
if there is a failure or disturbances caused by measurement
• Command signal permutation algorithm for the MMC
error. In [35] and [31] the estimated value is used to
reconfiguration. compensate the observation controller, and thus robustly
• SM voltage permutation algorithm: Manage voltage
detect and locate faults. Due to the function sgn(x), L is
permutations, especially in cases involving PVSMs. obtained from L > |ax̃|. Therefore, (21) keeps x̃ → 0 and
x̃˙ → 0, so the system is on the sliding surface.
Due to the nature of the study, which primarily focuses
To compensate for disturbances in the estimated variable in
on fault tolerance within the PV system based on MMC,
(19), it is consider perturbations in A, B and u and it is added
an analysis of harmonic levels (i.e., THD) has not been
a general perturbation variable D, this results in:
included. The central objective of this research is to ensure
x̂˙ = (A + 1A)x̂ + (B + 1B)(u + 1u) + L sgn x − x̂ + d


the uninterrupted operation of the converter under various
conditions, with a particular emphasis on fault tolerance (23)
mechanisms, fault detection, and system reconfiguration.
While THD analysis holds significance in the context of where 1A and 1B are the perturbations of A and B,
photovoltaic applications, this study prioritizes the successful respectively. 1u is the measurement error and/or the scaling
implementation of fault tolerance strategies. This includes error of the observed variable. It is assumed that the values
the use of sliding mode observers and redundant structures of these perturbations are limited and smaller than the
to ensure the MMC can adapt and operate effectively even error value when a failure occurs. Subtracting (23) from
in the event of a fault. The research aims to maintain the measured variable in (19), the error between the measured
converter’s continuous functionality and reliable operation by and estimated states is obtained by:
minimizing downtime and disruptions.
x̃˙ = ẋ − x̂˙
A. SLIDING MODE OBSERVER = Ax̃ − [1Ax̂ + (u + 1u)1B + B1u + d] −L sgn(x̃)
| {z }
In [37], the SMO is studied using the equivalent control D
method for a first order system. However, for clarity in [35] (24)
an introduction to the derivation of SMO equations from
[37] using the equivalent control method for a second-order The updated L is taken from (24):
system is presented. However, this method is complex to L > |Ax̃| + |D| (25)
implement (when the number of observed variables is high).
( Once (24) enters the sliding surface (sliding mode), this is
ẋ = Ax + Bu when x̃˙ → 0 and x̃ → 0. Then, (24) is expressed as:
(19)
x̂˙ = Ax̂ + Bu + L sgn(v)
D = −L sgn(x̃) (26)
Note that x is a measured variable, therefore, x̂ is the
estimated/observed variable of x; v is the sliding mode Due to the high frequency switching generated by the
variable, L is the observer gain and sgn(x) is the signal term ‘‘−L sgn(x̃)’’, a low-pass filter is applied to obtain the
function defined by: estimated value of D̂:
 Kd L sgn(x̃)
 1, x>0 D̂ = − (27)
1 + τs

sgn(x) = 0, x=0 (20)
x<0 where D̂ represents the estimated value of the perturbations,

−1,

Kd is the filter gain and τ is the low-pass filter time constant.
The sliding mode variable and the observer gain need to be Therefore, the observed state in (19) is rewritten as:
set to ensure that x̂ → x in a short finite period of time. In [37]
x̂˙ = (Ax̂ + Bu) + L sgn x − x̂ − D̂


and [35] the sliding mode variable is adopted as v = x − x̂. (28)
Replacing the sliding mode variable ‘‘v’’ by x − x̂, and
doing ẋ − x̂, ˙ the dynamic error between observed and The value of D̂ varies aggressively when there is a failure,
measured variables is obtained. thus forcing the correction of the estimated variable.

x̃˙ = ẋ − x̂˙ = Ax̃ − L sgn(x̃), x̃ ≜ x − x̂ (21) 1) SMO FOR MMC APPLICATIONS

The average value of ‘‘L sgn(x̃)’’ performs as a control In [30] the SMO is implemented using the arm currents
function that counterbalances ‘‘Ax̃’’, for this way, the control and in [38] the circulating current as the main parameter.
keeps the observer variable on the sliding surface, and thus In this paper, the variable defined by x is implemented by
guarantees x − x̂ = 0. the arm current (to avoid false positives that could change
the performance of the algorithm on the opposite arm of the
Ax̃ − L sgn(x̃) = 0 (22) same phase or on another phase of the converter) and the

VOLUME 11, 2023 126125

 J. C. Colque et al.: Fault-Tolerant Strategy for the MMC-Based PV System

SM capacitor voltage. The aim of this variable estimator is One of the critical parameters to consider is the gain ‘‘L1 ’’,
to locate the open-circuit fault after its occurrence. For this, This value should be adjusted based on the dynamic behavior
two modes of operation are necessary, suppose that initially, of the estimated variable. If the gain is too high, the estimation
it is in normal operation. Fig. 4 shows the flowchart used for of the measured variable will consistently resemble the
the detection and location of defective SMs. estimated one. In the event of a possible fault, the estimation
will exhibit the same behavior as the measured variable,
a: FAULT DETECTION MODE making it challenging to detect discrepancies between them.
This mode determines whether the MMC is working normally Conversely, if the gain value is low, the estimation will not
or not. By observing the arm current. adequately track the measured variable and won’t be able to
   keep up with its changes. A choice of observer gain is:
d 1 Vdc  
 îxu = SM
− vxo − (ϑxy ) + L1 sgn ĩxu + D̂


dt La  2 ixo
d 1 Vdc
   L1 ≥ (34)
 îxl =

 SM
+ vxo − (ϑxy ) + L1 sgn ĩxl + D̂ La
dt La 2
(29) b: FAULT LOCATION MODE
N +M To locate the faulty SM, the voltage of all SMs is observed
X+R
ϑxy =
SM XSM XSM
(Sxy (j) · vc,xy (j)) (30) too. For that, if 1ixy is greater than a current threshold value
j=1
ixy,th (for more than 1 ms), and the estimated voltage error is
greater than the voltage threshold value vc,th , a faulty SM is
ĩxy = ixy − îxy (31) detected. Next, all parameters are represented in vector form
Kds  
and, the estimated voltage error used to locate the fault is
D̂ = L1 sgn ĩxy (32)
1 + 0, 1s given by:

1vc,xy = |vc,xy − v̂c,xy | (35)


d ixy  
 vPVSMc,xy = SPVSM + L2a sgn vPVSM
c,xy − v̂xy
PVSM
;


xy


 dt C PVSM
d rSM ixy
  
vc,xy = SrSMxy + L2b sgn vrSM rSM
c,xy − v̂c,xy ;

 dt C SM
d


 RSM xy
i  
 vRSM
 = S + L2b sgn v RSM
− v̂RSM

dt c,xy xy
CSM c,xy c,xy

(36)

L2a and L2b represent the observer gains.

Similar to how the current estimation gains are adjusted.
To ensure effective fault detection in the SMs, the tuning of
these parameters, L2a and L2b , can be performed by:

vPVSM
c,xy
 L2a <


2La (37)
vSM
 L2b < c,xy


2La

B. FAULT-TOLERANT STRATEGY
The block diagram shown in Fig. 5 provides an explanation
of the proposed fault-tolerant procedure.

1) SWITCHING SIGNAL PERMUTATION

In order to avoid the addition of comparators that could
generate the signal of the reserve SMs (once these are
inserted into the main circuit) a switching signal permutation
FIGURE 4. Fault detection and location flowchart.
algorithm is implemented which basically consists of two
steps, (a) blocking signal generation and (b) switching signal
where: L1 is the observer gain, Kds is the disturbance permutation.
correction gain and, the estimated current error is given by: (a) The SMO provides the vector stsxy that contains the
instantaneous detection states of the all SMs, these states are
1ixy = |ixy − îxy | (33) used to generate the blocking vector Bxy and limit the number

126126 VOLUME 11, 2023

J. C. Colque et al.: Fault-Tolerant Strategy for the MMC-Based PV System

FIGURE 5. Diagram of fault-tolerant strategy for MMC-based PV system.

of substitutions of each arm of MMC (the reserve SMs instead (b) The modulation technique provides NOP −switching
of the defective SMs), for a finite RSMs. signals (corresponding to SMs in operation), however, there

VOLUME 11, 2023 126127

 J. C. Colque et al.: Fault-Tolerant Strategy for the MMC-Based PV System

are NT −SMs to control, in each arm of the MMC. For this, TABLE 1. Simulation parameters.
the proposed algorithm allows the adjustment from NOP −SMs
signals (initially, includes the PVSMs and the rSMs of the
arm in operational state) to NT −SMs signals (includes all the
SMs of the arm), when ‘‘Bxy ([p]) = 1’’ the switching signal
of q−th element of Sxy is transported to the 1–st position of
RSM (that is, (NOP + 1)–th place of Sxy ), thus the defective
SMs is replaced by the reserve SM and the SSP algorithm
generates NT −switching signals for NT −SMs. Permutation
takes place only between the NOP −SMs and RSMs.

2) SUBMODULE VOLTAGE PERMUTATION

As previously discussed, the modulation technique requires
NOP −modulating signals. However, considering the presence
of NT −SMs that need control, an adjustment in the number
of capacitor voltage signals becomes necessary.
In this approach, the strategy involves excluding the signals
originating from SMs that are in bypassed-state due to faults
or placed in reserve-state. Instead, the strategy exclusively
utilizes the signals originating from the SMs in operation, that
1) INITIAL CONSIDERATIONS
actively participate in the primary circuit.
• The initial value of the SM capacitors voltage is of great
importance for the fast recovery of the converter (for
3) CONTROL AND VOLTAGE DISTRIBUTION OF THE steady-state), after confronting a failure in some SM.
REDUNDANT SMs Therefore, this value must be previously defined by a
This control ensures that the average voltage on the upper method of capacitors pre-charging.
arm (vdc,xu ) will be equal to the lower arm (vdc,xl ). However, • The transient effect (period of time that the algorithm
if a failure occurs in a PVSM, the imbalance of power is takes in the MMC reconfiguration) must be short as
relatively high, then, it is necessary to distribute power to the possible, since it is always tried to avoid the failures in
redundant SMs, in this case the reference voltage is readjusted cascade.
as needed. The initial reference voltage for the redundant • If the voltage of the submodule inserted in the main
SMs is determined as a function of the reference voltage circuit undergoes a change in the voltage level there may
∗n ) and the summation of all PVSMs voltages (per arm),
(Vdc be a transient, in this case it is necessary to compensate
as depicted in Figure 5. This is expressed by the following the total voltage of the arm until reaching the steady
equation: state, this effect will be detailed in case 2 and 3.
In the following, Table 2 presents a comprehensive summary
c,xy ([1]) = Vdc − vc,xy ([1 + 2 + . . . + N ])
∗n
v*rSM *PVSM
(38)
of fault detection and recovery times in Figure 6.

TABLE 2. Summary of fault detection and recovery times.

IV. SIMULATION RESULTS
To validate the effectiveness of the fault-tolerant strategy
for the MMC-based photovoltaic system of Fig. 2 that
is simulated in Simulink/Matlab software environment.
To decrease the number of interactions and the computational
burn, allowing the simulation to be faster, each isolated dc-
dc converter is represented by a controllable DC source.
The MMC is constituted by twelve-PVSMs, one-rMS and
two-RSMs in each arm (N = 12, M = 1 and R = 2).
The PV array is constituted by 13 series-connected panels
and 3 parallel-connected strings (39 panels in total) of the
Kyocera Solar KD180GX-L type, each panel generates 180W
under standard test condition (STC) obtaining 7020W of
nominal power in each PVSM. In each PVSM, the isolated
dc-dc converter is used to control the voltage at the maximum In Figure 6, for a conventional MMC, the normalized
power point using the perturb and observe MPPT algorithm. voltage of the SMs from both arms, denoted as ‘‘vnc,y ’’
1
The system parameters are shown in Table 1. (normalized to ‘‘ vSM k ’’), is shown. In these tests, the initial
c,xy

126128 VOLUME 11, 2023

J. C. Colque et al.: Fault-Tolerant Strategy for the MMC-Based PV System

capacitor voltage is considered to be at 50%, 75%, 100%,

and 125% of the nominal value, respectively, before being
introduced into the main circuit. It can be observed that:
• When the voltage of the SM to be introduced is
less than or equal to 75% of the nominal value, the
reconfiguration of the MMC converter is slower due
to the difference between the voltages of the SMs in
operation and the voltage of the new SM. Consequently,
voltage balancing is slow.
• Similarly, when the voltage exceeds 125% of the
nominal value, voltage balancing is slow.

2) PV SYSTEM BASED ON MMC

For the proposed MMC converter. Initially, the photovoltaic
generators are all subject to STC (irradiance 1000W /m2 and
temperature 25o C). Then, at t = 0.25 s, the solar irradiation
of the photovoltaic generators of each PVSM goes from
1000W /m2 to 750W /m2 and at t = 0.55 s, it goes to
500W /m2 . Consequently, the power generated by the PV
generators in the interval 0 → 0.25 s is reduced from 505 kW
to 384 kW and finally, the power in the interval 0.25 s →
0.55 s is reduced to 257 kW while the reactive power output
remains unchanged at zero, as Fig. 7(a) shows. The current
in dq−synchronous frame are shown in Fig. 7(b). Hence,
as Fig. 7(c) shows, the sinusoidal three-phase output current
injected into the electrical grid is balanced at all times despite
the existence of PVSMs and rSMs faults.

FIGURE 7. (a) Active and reactive power, (b) dq−currents and

(c) three-phase output currents.

Three case studies are carried out in which the behavior

of the converter against SM faults is studied, in Fig. 8
the sequence of Faults ‘‘Fi’’ and the moment in which the
converter is reconfigured ‘‘Ri’’ are presented. This analysis
is supported by a graphical representation illustrating the
temporal aspects of fault detection and the MMC recovery.
FIGURE 6. Normalized voltage of SM capacitors. By examining these key time frames, it is gain valuable

VOLUME 11, 2023 126129

 J. C. Colque et al.: Fault-Tolerant Strategy for the MMC-Based PV System

insights into the efficiency and reliability of the fault-tolerant

algorithms employed in the system.

FIGURE 9. Capacitor voltages in phase a. (a) upper arm and (b) lower arm.

state. Then, ‘‘tre = t2 − t1 ’’ is the converter reconfiguration

time and, ‘‘tss = t3 − t1 ’’ is the steady-state reach time.
For F1: t1 = 0.15 s, t2 = 0, 1543 s and t3 = 0, 162 s, then,
FIGURE 8. Fault occurrence and reconfiguration time of phase a, b and c. is tre = 4.3 ms and tss = 12 ms. For F2: t1 = 0.3 s, t2 =
0.3021 s and t3 = 0.309 s, then, tre = 2.1 ms and tss = 9 ms.
To comprehensively investigate the converter’s behavior
under SM failures, three distinct case studies are conducted. B. CASE 2: REDUNDANT AND PHOTOVOLTAIC
To carry out a comprehensive study of the converter SUBMODULE FAULTS IN OPPOSITE ARMS
performance under SMs failures, three distinct case studies This case presents the behavior of the capacitors voltage in
are carried out. In these scenarios, F1, F2, F3, F4, and F7 the presence of two faults, a redundant SM fault in the upper
represent failures in the rSMs, while F5, F6, and F8 represent arm and a photovoltaic SM fault in the lower arm of phase c.
failures in the PVSMs. These failures are also distributed Fig. 10 shows the voltage of the PVSMs, rSM and RSMs,
across the converter’s phases. The purpose of these case when the fault F7 occurs, the new SM (RSM1cu ) replaces
studies is to analyze and understand how the converter the defective SM (rSM1cu ); and when the fault F8 occurs,
responds to different fault scenarios, contributing to the the new SM (RSM1cl ) replaces the defective PVSM5cl , at
the same time vrSM 1 RSM1
assessment of its robustness and fault tolerance capabilities. c,cl and vc,cl increase its average voltage
level from 443 V to 803 V, due to the fact that the RSM that
replaces the PVSM generates a power imbalance between the
A. CASE 1: REDUNDANT SUBMODULES FAULTS IN
two arms, this voltage adjustment is made through the voltage
OPPOSITE ARMS
distribution algorithm.
This case demonstrates the performance of the proposed
fault-tolerant technique in the presence of two redundant
SM faults, one in the upper arm and the other in the lower
arm of phase a. This type of fault is normally studied in
the conventional MMC. When a fault occurs, the MMC
must locate the fault as quickly as possible for uninterrupted
operation of the MMC.
Figure 9 shows the capacitor voltages in phase a. Note
that there are only five voltage signals per arm, this is to
facilitate the visualization of the capacitor voltages. Due to
the proposed fault-tolerant strategy, the new SM (RSM1ay )
replaces the defective SM (rSM1ay ), in both arms, which
goes RSM1au and RSM1al from a reserve to an operational
state, through the proposed permutation algorithms; the
voltage signal sent to the modulation stage is also updated FIGURE 10. Capacitor voltages in phase c. (a) upper arm and (b) lower
considering the new SM. arm.

F1 and F2 represent rSM failures caused in the upper

and lower arms, respectively. As noted, the converter is For F7: t1 = 0.75 s, t2 = 0, 751 s and t3 = 0, 759 s, then,
reconfigured accordingly. From now on, in t1 the fault occurs, tre = 1 ms and tss = 9 ms. For F8: t1 = 0.85 s, t2 = 0.8736 s
in t2 the new SM is inserted, in t3 the converter is in steady- and t3 = 0.925 s, then, tre = 23.6 ms and tss = 75 ms.

126130 VOLUME 11, 2023

J. C. Colque et al.: Fault-Tolerant Strategy for the MMC-Based PV System

FIGURE 11. Procedure performed by the permutation algorithms.

C. CASE 3: REDUNDANT AND PHOTOVOLTAIC

SUBMODULE FAULTS IN THE SAME ARM
For this case, the rSM1bu , rSM1bl , PVSM4bu and PVSM8bl
fail in that order (a rSM and a PVSM in each arm), this is
to demonstrate the robustness of the fault-tolerant strategy,
for that, in this case the fault detection and location process
is detailed. The arm current is used as a detection parameter
(through equation (29)).
Fig. 11 shows the procedure performed by the algorithms
(i) Switching signal permutation and (ii) SM voltage
permutation, concerning to fault F3 and F5 shown in Fig. 12.
Initially (MMC arm without faults), all elements of the vector
stsbu are zeros (0: healthy SM and 1: faulty SM); the vector
Bbu has NOP −elements with value zero and R−elements with
value one (0: normal operation and 1: bypass); the vector Sbu
contains the switching signals of the operational NOP −SMs
and the signals of the SMs in bypass state are represented by
zeros; the voltage of all the capacitors of the submodules are
in the vector vc,bu , in sequence, and the vector vOP
c,bu contains
the first NOP −elements of vc,bu that will be processed by the
modulation method. When F3 occurs, in (i), if stsbu ([13]) =
1, Bbu ([13]) turns 1 and in Sbu the 13th element g13 goes to
the 14th place and the 13th place becomes zero. Similarly in
(ii), vc,bu the 13th and 14th element are swapped. The same
FIGURE 12. Waveforms of upper arm of phase b. (a) Arm current
logic is applied to the fault F5 in 4th column. (measured and estimated), (b) Capacitor voltages (measured and
Fig. 12(a) shows as the estimated current îbu closely PVSMj
estimated), (c) 1vc,bu observed variable error with details and (d)
follows the measured current ibu , and when a failure occurs, xSMj
1vc,bu observed variable error with details.
the estimated signal quickly returns to follow the measured
value, avoiding cascading failures. Fig. 12(b) shows the
capacitor voltages with details, demonstrating the rapid
replacement of a faulty SM allowing uninterrupted operation immediately, vRSM 1 RSM2
c,bl and vc,bl increase its average voltage level
of the converter. Fig. 12(c)-(d) shows the observed voltage from 443 V to 803 V. F3 and F4 represent rSM failures and,
PVSMj xSMj
error of the PVSMs and rSM/RSMs, 1vc,bu and 1vc,bu , F5 and F6 represent PVSM failures. As noted, all faults are
respectively, these signals are compared with a threshold reconfigured accordingly.
value. If the two conditions are met (1ibu > ibu,th and For F3: t1 = 0.4 s, t2 = 0.4102 s and t3 = 0.417 s, then,
XSMj XSMj
1vc,bu > vc,bu ), the defective SM is detected and replaced tre = 10.2 ms and tss = 12 ms. For F4: t1 = 0.5 s, t2 =
efficiently, in both case of PVSM and rSM. 0.5016 s and t3 = 0.509 s, then, tre = 1.6 ms and tss = 9 ms.
Fig. 13 shows the capacitor voltages in the lower arm of For F5: t1 = 0.6 s, t2 = 0.6158 s and t3 = 0.679 s, then, tre =
phase b, when the fault F4 occurs at t = 0.5 s in the rSM1bl . 15.8 ms and tss = 79 ms. For F6: t1 = 0.7 s, t2 = 0.7083 s
Then, the reserve SM (RSM1bl ) replaces the defective SM and t3 = 0.752 s, then, tre = 8.3 ms and tss = 52 ms.
(rSM1bl ). Similarly, when fault F6 occurs in PVSM8bl , the Note that F5 represents an eventual fault in an upper-arm
reserve SM (RSM2bu ) replaces the defective SM (PVSM8bl ); PVSM, and F6 represents a deliberate fault in a lower-arm

VOLUME 11, 2023 126131

 J. C. Colque et al.: Fault-Tolerant Strategy for the MMC-Based PV System

array through isolated DC-DC converters, which allows inde-

pendent monitoring of the maximum power point (MPPT)
and provides grounding for each submodule. The proposed
system encompasses various types of SMs, including photo-
voltaic and half-bridge SMs, each exhibiting distinct dynamic
behavior. Consequently, it becomes imperative to conduct
fault observation and diagnosis separately for each SM type.
The performance of the SMO and the proposed switching
and voltage permutation algorithms are verified, observing
FIGURE 13. Capacitor voltages in lower arm of phase b.
results consistent with those presented in the scientific
literature in relation to the SMO, with this, the possibility
of using this fault-tolerant strategy in the modified MMC
PVSM. This is done to minimize the power mismatch.
is made possible. By employing the arm current as an
While the distribution algorithm manages to maintain the
observation variable, the strategy effectively eliminates false
bus voltage Vdc , the generated energy is unbalanced. For
positives or potential failures that might impede diagnosis in
this reason, a PVSM in the opposite arm is disconnected,
the opposite arm. This approach demonstrates robustness in
as indicated by F6.
identifying defective internal semiconductor devices within
Finally, the total DC voltage of the upper arm (sum of
the submodule. Furthermore, the permutation and signal
the voltages of the SMs in operation) and lower arm of all
blocking algorithms were subject to specific testing. These
phases are shown in Fig. 14(a)-(c). When an rSM fault occurs,
tests resulted in a reduction in the number of comparators
a small transient is observed, while when a PVSM fault
required for the PWM signal. To stabilize the DC-bus
occurs, the transient is larger, this is due to the unbalance
voltages of each leg, redundant submodules are deployed.
generated in the voltage of the arms and also because the
This ensures that one-third of the total photovoltaic energy
inserted SM voltage must still reach the reference voltage
generated flows through each phase of the MMC and
(guaranteed by the distribution control), so that vdc,xu is
subsequently feeds into the electrical grid. Consequently,
equal to vdc,xl , showing that the voltages of the arms in
balanced currents are injected into the electrical grid,
each phase are balanced. While Fig. 14(d) shows the DC-bus
mitigating the impact of potentially uneven photovoltaic
voltage of MMC. The voltages of the PVSMs and rSMs are
energy generation due to shading or submodule failures.
regulated to approximately 1163 V and 443 V, respectively,
The voltage distribution algorithm comes into play
and the DC link voltage of the MMC has been set to
exclusively in the presence of photovoltaic submodule
approximately 14400 V (initially it corresponds to 12 PVSMs
failures. This algorithm prevents voltage overstress in certain
and 1 rSM per arm).
redundant submodules or Reserve Submodules (ex-float) and
guarantees that the total average voltage on one arm equals
that of the opposing arm. In summary, this article offers a
comprehensive exploration of the mathematical foundations
and flowcharts underpinning the proposed fault-tolerant
strategy. These concepts have been rigorously validated
through simulation results.
In this study, the analysis of harmonic levels (THD) has
not been included due to the primary focus on fault tolerance
within the PV system based on MMC. The central objective
of this research is to ensure the uninterrupted operation of
the converter under various conditions, with a particular
emphasis on fault tolerance mechanisms, fault detection, and
system reconfiguration.

ACKNOWLEDGMENT
The authors would like to thank the State University of
Campinas and the Federal University of ABC, and also would
like to thank the authorities of National University of the
FIGURE 14. DC-bus voltage. (a) phase a, (b) phase b, (c) phase c and
(d) MMC-based PV system. Altiplano for their support and encouragement in research
endeavors.

REFERENCES
V. CONCLUSION
[1] J. Feng, H. Wang, J. Xu, M. Su, W. Gui, and X. Li, ‘‘A three-phase grid-
The photovoltaic system is based on a modular multilevel connected microinverter for AC photovoltaic module applications,’’ IEEE
converter, each PVSMs is interconnected with photovoltaic Trans. Power Electron., vol. 33, no. 9, pp. 7721–7732, Sep. 2018.

126132 VOLUME 11, 2023

J. C. Colque et al.: Fault-Tolerant Strategy for the MMC-Based PV System

[2] I. Abdalla, J. Corda, and L. Zhang, ‘‘Multilevel DC-link inverter and [22] H. Nademi, A. Das, R. Burgos, and L. E. Norum, ‘‘A new circuit
control algorithm to overcome the PV partial shading,’’ IEEE Trans. Power performance of modular multilevel inverter suitable for photovoltaic
Electron., vol. 28, no. 1, pp. 14–18, Jan. 2013. conversion plants,’’ IEEE J. Emerg. Sel. Topics Power Electron., vol. 4,
[3] S. Du, A. Dekka, B. Wu, and N. Zargari, Modular Multilevel Converters: no. 2, pp. 393–404, Jun. 2016.
Analysis, Control, and Applications. Hoboken, NJ, USA: Wiley, 2018. [23] A. Rashwan, ‘‘A new topology for the large-scale photovoltaic systems
[4] K. K. Gupta, A. Ranjan, P. Bhatnagar, L. K. Sahu, and S. Jain, ‘‘Multilevel grid connection based on modular multilevel converter,’’ in Proc. 20th Int.
inverter topologies with reduced device count: A review,’’ IEEE Trans. Middle East Power Syst. Conf. (MEPCON), Dec. 2018, pp. 286–291.
Power Electron., vol. 31, no. 1, pp. 135–151, Jan. 2016. [24] F. Rong, X. Gong, and S. Huang, ‘‘A novel grid-connected PV system
[5] A. Lesnicar and R. Marquardt, ‘‘An innovative modular multilevel based on MMC to get the maximum power under partial shading
converter topology suitable for a wide power range,’’ in Proc. IEEE conditions,’’ IEEE Trans. Power Electron., vol. 32, no. 6, pp. 4320–4333,
Bologna Power Tech. Conf., Jun. 2003, p. 6. Jun. 2017.
[25] H. Bayat and A. Yazdani, ‘‘A power mismatch elimination strategy for an
[6] S. Debnath, J. Qin, B. Bahrani, M. Saeedifard, and P. Barbosa,
MMC-based photovoltaic system,’’ IEEE Trans. Energy Convers., vol. 33,
‘‘Operation, control, and applications of the modular multilevel converter:
no. 3, pp. 1519–1528, Sep. 2018.
A review,’’ IEEE Trans. Power Electron., vol. 30, no. 1, pp. 37–53,
[26] H. Bayat and A. Yazdani, ‘‘A hybrid MMC-based photovoltaic and battery
Jan. 2015.
energy storage system,’’ IEEE Power Energy Technol. Syst. J., vol. 6, no. 1,
[7] Y. Luo, Z. Li, L. Xu, X. Xiong, Y. Li, and C. Zhao, ‘‘An adaptive voltage-
pp. 32–40, Mar. 2019.
balancing method for high-power modular multilevel converters,’’ IEEE
[27] C. A. Rojas, S. Kouro, M. A. Perez, and J. Echeverria, ‘‘DC–DC MMC
Trans. Power Electron., vol. 33, no. 4, pp. 2901–2912, Apr. 2018.
for HVdc grid interface of utility-scale photovoltaic conversion systems,’’
[8] A. Dekka, B. Wu, R. L. Fuentes, M. Perez, and N. R. Zargari, ‘‘Evolution IEEE Trans. Ind. Electron., vol. 65, no. 1, pp. 352–362, Jan. 2018.
of topologies, modeling, control schemes, and applications of modular [28] X. Pan, S. Pan, P. Jin, L. Yao, J. Gong, F. Liu, and X. Zha, ‘‘Decoupling
multilevel converters,’’ IEEE J. Emerg. Sel. Topics Power Electron., vol. 5, capacitor minimization of an MMC-based photovoltaic system with three-
no. 4, pp. 1631–1656, Dec. 2017. winding power channel,’’ IEEE Trans. Power Electron., vol. 37, no. 1,
[9] M. A. Perez, S. Bernet, J. Rodriguez, S. Kouro, and R. Lizana, ‘‘Circuit pp. 1012–1026, Jan. 2022.
topologies, modeling, control schemes, and applications of modular [29] T. H. Nguyen, K. A. Hosani, M. S. E. Moursi, and F. Blaabjerg,
multilevel converters,’’ IEEE Trans. Power Electron., vol. 30, no. 1, ‘‘An overview of modular multilevel converters in HVDC transmission
pp. 4–17, Jan. 2015. systems with STATCOM operation during pole-to-pole DC short circuits,’’
[10] B. Chang, O. Cwikowski, M. Barnes, R. Shuttleworth, A. Beddard, and IEEE Trans. Power Electron., vol. 34, no. 5, pp. 4137–4160, May 2019.
P. Coventry, ‘‘Review of different fault detection methods and their impact [30] S. Shao, P. W. Wheeler, J. C. Clare, and A. J. Watson, ‘‘Fault detection
on pre-emptive VSC-HVDC DC protection performance,’’ High Voltage, for modular multilevel converters based on sliding mode observer,’’ IEEE
vol. 2, no. 4, pp. 211–219, Dec. 2017. Trans. Power Electron., vol. 28, no. 11, pp. 4867–4872, Nov. 2013.
[11] A. Dekka, B. Wu, V. Yaramasu, and N. R. Zargari, ‘‘Model predictive [31] S. Shao, A. J. Watson, J. C. Clare, and P. W. Wheeler, ‘‘Robustness analysis
control with common-mode voltage injection for modular multilevel and experimental validation of a fault detection and isolation method for
converter,’’ IEEE Trans. Power Electron., vol. 32, no. 3, pp. 1767–1778, the modular multilevel converter,’’ IEEE Trans. Power Electron., vol. 31,
Mar. 2017. no. 5, pp. 3794–3805, May 2016.
[12] A. António-Ferreira, C. Collados-Rodríguez, and O. Gomis-Bellmunt, [32] B. Li, Y. Zhang, R. Yang, R. Xu, D. Xu, and W. Wang, ‘‘Seamless transition
‘‘Modulation techniques applied to medium voltage modular multilevel control for modular multilevel converters when inserting a cold-reserve
converters for renewable energy integration: A review,’’ Electric Power redundant submodule,’’ IEEE Trans. Power Electron., vol. 30, no. 8,
Syst. Res., vol. 155, pp. 21–39, Feb. 2018. pp. 4052–4057, Aug. 2015.
[13] D. Ronanki and S. S. Williamson, ‘‘Modular multilevel converters for [33] R. Hasan, S. Mekhilef, M. Seyedmahmoudian, and B. Horan, ‘‘Grid-
transportation electrification: Challenges and opportunities,’’ IEEE Trans. connected isolated PV microinverters: A review,’’ Renew. Sustain. Energy
Transport. Electrific., vol. 4, no. 2, pp. 399–407, Jun. 2018. Rev., vol. 67, pp. 1065–1080, Jan. 2017.
[14] F. Deng, Y. Tian, R. Zhu, and Z. Chen, ‘‘Fault-tolerant approach for [34] B. Li, R. Yang, D. Xu, G. Wang, W. Wang, and D. Xu, ‘‘Analysis of the
modular multilevel converters under submodule faults,’’ IEEE Trans. Ind. phase-shifted carrier modulation for modular multilevel converters,’’ IEEE
Electron., vol. 63, no. 11, pp. 7253–7263, Nov. 2016. Trans. Power Electron., vol. 30, no. 1, pp. 297–310, Jan. 2015.
[15] F. Deng, R. Zhu, D. Liu, Y. Wang, H. Wang, Z. Chen, and M. Cheng, [35] S. Shao, ‘‘The application of sliding mode observers to fault detection
‘‘Protection scheme for modular multilevel converters under diode open- and isolation for multilevel converters,’’ Ph.D. dissertation, School Elect.
circuit faults,’’ IEEE Trans. Power Electron., vol. 33, no. 4, pp. 2866–2877, Electron. Eng., Power Electron., Mach. Control Group, Univ. Nottingham,
Apr. 2018. U.K., 2015.
[16] J. V. M. Farias, A. F. Cupertino, H. A. Pereira, S. I. S. Junior, and [36] A. Yazdani and R. Iravani, Voltage-Sourced Converters in Power Systems,
R. Teodorescu, ‘‘On the redundancy strategies of modular multilevel vol. 39. Hoboken, NJ, USA: Wiley, 2010.
converters,’’ IEEE Trans. Power Del., vol. 33, no. 2, pp. 851–860, [37] Y. Shtessel, C. Edwards, L. Fridman, and A. Levant, Sliding Mode Control
Apr. 2018. and Observation, vol. 10. Berlin, Germany: Springer, 2014.
[38] S. Shao, J. C. Clare, A. J. Watson, and P. W. Wheeler, ‘‘Detection and
[17] B. Li, Z. Xu, J. Ding, and D. Xu, ‘‘Fault-tolerant control of medium-voltage
isolation of multiple faults in a modular multilevel converter based on
modular multilevel converters with minimum performance degradation
a sliding mode observer,’’ in Proc. IEEE Energy Convers. Congr. Expo.
under submodule failures,’’ IEEE Access, vol. 6, pp. 11772–11781,
(ECCE), Sep. 2014, pp. 3491–3495.
2018.
[18] S. Yang, Y. Tang, and P. Wang, ‘‘Seamless fault-tolerant operation of a
modular multilevel converter with switch open-circuit fault diagnosis in
a distributed control architecture,’’ IEEE Trans. Power Electron., vol. 33,
no. 8, pp. 7058–7070, Aug. 2018.
JUAN C. COLQUE received the B.S. degree in
electronic engineering from the National Uni-
[19] N. Li, F. Gao, T. Hao, Z. Ma, and C. Zhang, ‘‘SOH balancing control
method for the MMC battery energy storage system,’’ IEEE Trans. Ind.
versity of the Altiplano (UNA), Peru, in 2013,
Electron., vol. 65, no. 8, pp. 6581–6591, Aug. 2018. and the M.Sc. degree in electrical engineering
[20] A. I. Elsanabary, G. Konstantinou, S. Mekhilef, C. D. Townsend,
from the Federal University of ABC (UFABC),
M. Seyedmahmoudian, and A. Stojcevski, ‘‘Medium voltage large-scale in 2017. He is currently pursuing the Ph.D. degree
grid-connected photovoltaic systems using cascaded H-bridge and modular in electrical engineering with the University of
multilevel converters: A review,’’ IEEE Access, vol. 8, pp. 223686–223699, Campinas, funded by the CNPq Scholarship
2020. Program. His research interests include the various
[21] J. Mei, B. Xiao, K. Shen, L. M. Tolbert, and J. Y. Zheng, ‘‘Modular aspects of power electronics, such as applications
multilevel inverter with new modulation method and its application to in renewable energy sources, energy storage systems, predictive control,
photovoltaic grid-connected generator,’’ IEEE Trans. Power Electron., fault-tolerant strategies, and estimation parameters methods for power
vol. 28, no. 11, pp. 5063–5073, Nov. 2013. electronics converters.

VOLUME 11, 2023 126133

 J. C. Colque et al.: Fault-Tolerant Strategy for the MMC-Based PV System

ERNESTO RUPPERT (Member, IEEE) received KARLOS A. C. CHIRAPO (Member, IEEE)

the B.S. degree in electrical engineering and received the B.S. degree in electronic engineer-
the M.S. and Ph.D. degrees from the Uni- ing from the National University of the Alti-
versity of Campinas (UNICAMP), Campinas, plano, Puno, Peru, in 2004, the M.Sc. degree
Brazil, in 1971, 1974, and 1983, respectively. in telecommunications from Tarapacá-Arica Uni-
From 1972 to 1978, he was with the Electrical and versity, Chile, in 2010, and the M.Sc. degree in
Computer Engineering School, UNICAMP, as an electrical engineering from the Federal University
Assistant Professor in electromechanical energy of ABC, São Paulo, Brazil, in 2018. He is
conversion. From 1979 to 1983, he was with currently pursuing the Ph.D. degree with the
General Electric, Brazil, designing large induction Polytechnic School, University of São Paulo,
and synchronous motors and working as an Application Engineer dedicated Brazil. From 2007 to 2015, he was a Professor with the National University
to large motors and generators. He is currently a Full Professor with the of the Altiplano. From 2016 to 2018, he was a Researcher with the Federal
Electrical and Computer Engineering School, UNICAMP, researching and University of ABC. From 2019 to 2022, he was a Researcher with the
teaching in the areas of electrical machines, power electronics, drives, and Polytechnic School, University of São Paulo. His research interests include
electrical power systems. power electronics, digital signal processing, microcontrollers, and digital
systems.

JOSÉ LUIS AZCUE received the B.S. degree in

I. DELGADO-HUAYTA (Member, IEEE) received electronic engineering from the National Univer-
the B.S. degree in electronic engineering and sity of the Altiplano (UNA), Peru, in 2004, and the
the M.Sc. and D.Sc. degrees in mechatronic M.Sc. and Ph.D. degrees in electrical engineering
engineering from the National University of San from the University of Campinas (UNICAMP),
Agustín, Arequipa, Peru, in 2001, 2015, and 2018, in 2010 and 2013, respectively. He is currently
respectively. Since 2001, he has been a full-time a Professor with the Federal University of ABC
Professor with the National University of the Alti- (UFABC), Brazil. His research interests include
plano, Puno, Peru. His research interests include electric drives, power electronics, fuzzy and neu-
power electronics, the IoT industrial technology, ral controllers, wind, and photovoltaic energy
digital signal processing, microcontrollers, and systems.
embedded digital systems.

126134 VOLUME 11, 2023