Zhao 

et al. Prot Control Mod Power Syst (2021) 6:32 Protection and Control of
https://doi.org/10.1186/s41601-021-00210-1
Modern Power Systems

ORIGINAL RESEARCH Open Access

Model predictive control of grid‑connected

PV power generation system considering
optimal MPPT control of PV modules
Yingying Zhao1, Aimin An1,2,3*  , Yifan Xu1, Qianqian Wang1 and Minmin Wang1 

Abstract 
Because of system constraints caused by the external environment and grid faults, the conventional maximum power
point tracking (MPPT) and inverter control methods of a PV power generation system cannot achieve optimal power
output. They can also lead to misjudgments and poor dynamic performance. To address these issues, this paper
proposes a new MPPT method of PV modules based on model predictive control (MPC) and a finite control set model
predictive current control (FCS-MPCC) of an inverter. Using the identification model of PV arrays, the module-based
MPC controller is designed, and maximum output power is achieved by coordinating the optimal combination of
spectral wavelength and module temperature. An FCS-MPCC algorithm is then designed to predict the inverter
current under different voltage vectors, the optimal voltage vector is selected according to the optimal value func-
tion, and the corresponding optimal switching state is applied to power semiconductor devices of the inverter. The
MPPT performance of the MPC controller and the responses of the inverter under different constraints are verified,
and the steady-state and dynamic control effects of the inverter using FCS-MPCC are compared with the traditional
feedforward decoupling PI control in Matlab/Simulink. The results show that MPC has better tracking performance
under constraints, and the system has faster and more accurate dynamic response and flexibility than conventional PI
control.
Keywords:  Grid-connected PV power generation system, Model predictive control, Maximum power point tracking,
Inverter, Optimal value function

1 Introduction the voltage and current dynamically [5]. This improves

In addressing global climate change, the proposal of the energy conversion efficiency, power quality and sta-
reducing carbon dioxide emission and carbon neutrality bility of system operation without increasing the hard-
has accelerated the speed of energy low-carbon transfor- ware cost.
mation [1–3]. This has stimulated the rapid development The MPPT control method is the main technology
of solar energy, and the permeability of grid-connection to improve the efficiency of a PV system [6]. At pre-
photovoltaic (PV) has been increasing [4]. MPPT and sent, most MPPT control algorithms do not consider
inverter control strategy in a grid-connected PV power the impact of irradiance and temperature mutation on
generation system ensure that the system operates in a tracking speed and system control accuracy, e.g., the
stable and optimal state of maximum power by adjusting traditional incremental conductance method [7] finds it
difficult to meet the requirements of tracking efficiency
when there are interference and constraints from the
*Correspondence: [email protected] external environment. The MPPT algorithm in the case
1
College of Electrical and Information Engineering, Lanzhou University of external environment mutation has been studied in
of Technology, Lanzhou 730050, Gansu, China
Full list of author information is available at the end of the article
[8], while in [9], the mathematical model of PV modules

© The Author(s) 2021. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which
permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the
original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or
other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line
to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory
regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this
licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/.
Zhao et al. Prot Control Mod Power Syst (2021) 6:32 Page 2 of 12

under partial shadow interference is established, and the iPV idc

T1 T3 T5
relationship between output characteristics of PV arrays ic
ia R L ea
and outside interference factors is analyzed. The MPPT + +
ua
ib eb
udc ub uNO
algorithm combining current control and disturbance C
− uc ic ec
−
observation method is proposed in [10]. However, the PV
T2 T4 T6

above methods have high power consumption and find

it difficult to consider both the tracking speed and the Fig. 1  Topological structure of grid-connected PV power generation
control accuracy when the external constraints change system
rapidly.
The grid-connected PV inverter control technology has
become a research hotspot. Traditional control methods
tracking of the maximum power point of PV arrays and
include linear methods, such as feedforward decoupling
optimization of the inverter control performance.
control based on PI regulation [11], and nonlinear meth-
In order to achieve the optimal control of a grid-con-
ods, such as proportional resonance control [12]. The
nected PV power generation system, and maximize the
former needs the PWM modulation module, while the
utilization of solar energy, MPC strategies for PV mod-
static and dynamic performance of the system depend on
ules and the inverter are proposed, respectively. From the
the PI regulator. Although the latter has good dynamic
linear PV array model obtained by model identification, a
performance, the switching frequency is not fixed, which
model predictive controller is designed for modules. By
makes the harmonic filter design difficult. With the
predicting the incident light wavelength, the maximum
improvement of digital processors, new control methods
cut-off spectral wavelength is found and filtered to reduce
have been proposed, including fuzzy, adaptive and pre-
the module temperature and achieve maximum power
dictive control. The fuzzy logic control strategy of a grid- tracking. The FCS-MPCC algorithm is then applied to
connected PV inverter is used in [13], while fuzzy rules the inverter current control in the dq coordinate system.
and membership functions of fuzzy control depend on The voltage vectors are evaluated through the predictive
experience, which has some limitations, and the control model and the optimized value function is established in
stability needs to be improved. Reference [14] proposes each PWM sampling period. The voltage vector which
an adaptive control method for a 3-phase inverter in the minimizes the value function is used for the next sam-
distributed generation system. The design of adaptive
pling period to generate the switching control signals for
control is complex. The parameter adjustment is not only
the inverter. Finally, the system is simulated in Matlab/
time-consuming, but also needs to be tuned under vari-
Simulink to demonstrate the performance in steady-state
ous working conditions. Predictive control adopts strate- and dynamic operation. It shows that MPC can track the
gies of multi-step test, rolling optimization and feedback maximum power point of PV arrays and has an ability for
correction, and mainly includes deadbeat, repetitive [15] fast current regulation. It also verifies the effectiveness of
and model predictive control (MPC) [16–18]. Because FCS-MPCC for a 3-phase grid-connected PV inverter.
MPC can effectively overcome the uncertainty and non-
linearity of the process and deal with various constraints
2 Grid‑connected PV power generation system
of spatial state variables, optimal control can be achieved
model
by selecting the control mode that minimizes the evalu-
The grid-connected PV power generation system con-
ation function after establishing the system predictive
sists of PV modules, a DC capacitor, a 3-phase inverter
model. Reference [19] studies the performance verifica-
bridge, RL filters, step-up transformers, etc. [23], and
tion method of a finite control set model predictive con- its typical topology is shown in Fig. 1. Given the output
trol (FCS-MPC) strategy applied to the power electronic characteristics of the PV cell, the PV array converts the
converter, and evaluates the performance of the algo- absorbed solar energy into direct current under the con-
rithm using statistical model checking. Reference [20]
trol of MPPT, and the inverter converts the DC power to
studies the weighting factor design based on an artificial
AC power, which is then transmitted to the power grid
neural network of a converter under MPC, whereas in
after filtering and boosting [24].
[21], MPC is applied to modular multilevel converters.
The strategy of FCS-MPC is introduced into the voltage
2.1 Mathematical model of PV array
source converter and applied to an AC microgrid sys-
In practical application, PV modules connect multiple
tem in [22]. However, the above methods are focused on
single PV cells to transmit electrical energy. The relation
converters, whereas research on new energy power gen-
between the output current and voltage of the PV array is
eration systems has been limited. This paper combines given as:
a PV power generation system with MPC to realize fast
 Zhao et al. Prot Control Mod Power Syst (2021) 6:32 Page 3 of 12

coordinate system is transformed into the αβ coordinate

   
q(U + IRs ) U + IRs
I = IL − Isc exp −
AKT
−1 −
Rp (1) system, and the inverter output voltages are:

where IL is the photo-generated current and Isc is the

 � � �
 2 1
reverse saturation current. K is the Boltzmann constant,  uα = Udc Sa − (Sb + Sc )


3 2
q is the electron charge, T is the ambient temperature, √ (5)
 2
and A is the diode factor. Rs,Rp are the equivalent series


 uβ = Udc (Sb − Sc )
and parallel resistances, respectively. 2
When sunlight irradiates, only a small part of the inci- 
dent radiation is transformed into electrical energy by diα
 uα = L + Riα + eα


PV cells, while a large part of the energy increases the dt
(6)
module temperature through heat transfer. However, the di
 uβ = L β + Riβ + eβ


influence of irradiance and temperature on the output dt
power are diametrically opposite, and the relationship
between the PV module temperature and output power where uα , uβ and iα , iβ are the inverter output voltages
is given as: and currents in the αβ coordinate system, respectively.
eα , eβ are the grid voltages in the αβ coordinate system.
dTm By aligning the grid voltage synthesis vector E and the
Cmodule = qsw − qlw − qconv − Pout (2)
dt d-axis of the dq coordinate system [14], the inverter out-
put active power and reactive power in the dq coordinate
where Cmodule is the module heat capacity, Tm is the mod- system are obtained as:
ule temperature, qsw , qlw are the short and long wave
radiations of the incident light, respectively. qconv is the 3 3

 P = (ed id + eq iq ) = Eid

heat exchange and Pout is the output power. 2 2
(7)
 Q = 3 (e i + e i ) = − 3 Ei

q d d q q
2 2
2.2 Mathematical model of 3‑phase PV grid‑connected
inverter From (7), it can be seen that the active current id and
Considering all possible switching states of the 3-phase reactive current iq have a linear relationship with P, Q ,
inverter, there are eight combinations. The inverter respectively. By independently adjusting the active and
switching states are defined as: reactive currents, the inverter output active power
 and reactive power can be decoupled and separately
Sk =
1 (The upper arm is on ,the lower arm is off) controlled.
0 (The upper arm off on, the lower arm is on)
(3) 3 MPPT control of PV modules based on model
where k = a,b, c is the three-phase sequence. predictive control
According to Kirchhoff ’s voltage law and switching MPPT has become the key technology to increase out-
function, the mathematical model of an inverter in the put power, improve photoelectric conversion efficiency
3-phase static coordinate system is shown as: and decrease power generation cost [25]. The influence of
   temperature on power generation efficiency is negative,

 dia 1 � so when the surface temperature of the PV module rises,
 Ria + L = udc Sa − Sk  − ea
the photoelectric conversion efficiency reduces [26].

dt 3



k=a,b,c
Therefore, the MPC controller is designed to predict the



  
influence of incident light wavelength on module tem-


 dib 1 �
Rib + L = udc Sb − Sk  − eb (4) perature and power, control the active filter to filter the
dt 3
maximum cut-off spectral wavelength, and dynamically



 k=a,b,c
adjust the relationship between module temperature and

  

di 1

c
optimal output power, so as to minimize module temper-
 �
 Ri + L

= udc Sc − Sk  − ec
 c

 dt 3 ature while maximizing output power at the same time.
k=a,b,c
This not only achieves the purpose of maximum power
where ia , ib , ic are the inverter output currents, udc is the tracking, but also improves power generation efficiency
DC bus voltage, and ea , eb , ec are the 3-phase grid volt- and durability. Figure  2 shows the structure block dia-
ages. Through Clark transformation, the 3-phase static gram of a model predictive controller for PV modules.
Zhao et al. Prot Control Mod Power Syst (2021) 6:32 Page 4 of 12

Incident
irradiance
Table 1 Coefficients of state space equation of the linearized
Wind speed
model
Ambient
temperature
A B C D
PV module Module power  T 0
   
Reference Tmodule ∆Tmodule
MPC
∆λcutoff λcutoff 0 1 0 −0.0007632 0
 
generator setpoint  setpoint  Module 0 0 1   −0.0007444  0
Controllor 
temperature
Tmodule nominal
0.9681 −2.936 2.968 −0.000726 0
Tmodule ∆Tmodule



Tmodule nominal λcutoff nominal

Fig. 2  The structure block diagram of MPC controller of PV arrays

Fig. 4  Fitting curves of actual plant data and linear model

r (k ) + xm (k + 1) xm (k ) y (k )
+
Ky B q −1 C
- +
A

Kx
Fig. 3  Input parameters and output power of model identification
Fig. 5  Structure block diagram of MPC system

3.1 Mechanism model identification of PV modules

To design a model predictive controller it is necessary 3.2 Design of the model predictive controller for PV arrays
to obtain a linear model of PV modules. The two-step 3.2.1 Theory of model predictive control
change of ultraviolet cut-off wavelength UV is set as MPC is a model-based closed-loop optimal control
the excitation signal of system input, and the infrared strategy [27], and the mathematical model describing
cut-off wavelength IR is set as 1100 nm. The ambient the research object in discrete form is given as:
temperature and wind speed are obtained from a data 
x(k + 1) = Ax(k) + Bu(k)
table. The reference module temperature Tm_f  , the (8)
y(k) = Cx(k) + Du(k)
change of ultraviolet cut-off wavelength UV and the
actual output power Pout are shown in Fig. 3. where k is the sampling time, u is the input vector, y is the
In order to ensure that the PV modules fully reach a output vector, and x is the state variable. When the input
stable state, the total simulation time is set to 2.22  h. and output variables are defined as U and Y  , the output
Using the system identification toolbox in Matlab, the prediction sequence is defined as:
initial linearization model is generated by a subspace
identification method, which is then further optimized Y = Fx(ki ) + �U (9)
by the prediction error method. The coefficients of whereF = [CA CA2 · · · CANp ]T , and
the state space equation obtained by the identifica-
tion toolbox are given in Table 1, and the fitting curve
 
CB 0 0 ··· 0
between the actual output of the system and the lin-  CAB CB 0 ··· 0 
= .. .. .. ..
 
ear model is obtained, as shown in Fig.  4. The fitting .. 
 . . . . . 
degree reaches 92.36%. Final prediction error (FPE) is CANp −1 B CANp −2 B CANp −3 B CANp −Nc B
0.01307, and mean squared error (MSE) is 0.01287.
 Zhao et al. Prot Control Mod Power Syst (2021) 6:32 Page 5 of 12

Fig. 6  Simulation diagram of MPC controller of PV modules in Simulink

where Kx is the first row of matrix (�T � + R)−1 �T F  ,

Ky is the first element of matrix (�T � + R)−1 �T Rs , and
r(ki ) is the expected value of r dimensional output at a
certain sampling time.
The output prediction value of the system is:
y(ki + 1) = C(A − BKx )x(ki ) + CBKy r(ki ) (13)
From the above inference, the structure of the MPC
system is shown in Fig. 5.

3.2.2 Design process of PV modules MPC controller

The aim of the MPC controller is to maximize the out-
put power by filtering the spectral wavelength to con-
trol the module temperature to track the reference
value. Because the PV power generation is limited by
Fig. 7  Simulation results of the MPC controller of PV modules the natural environment, the input, output and errors
are constrained as follows:
300 nm ≤ UV ≤ 430 nm, 0 nm ≤ UV ≤ 130 nm
The objective function is defined as:
262 K ≤ Tm ≤ 482 K, −100 K ≤ Tm ≤ 100 K
J = (Rs − Y )T (Rs − Y ) + U T RU (10)
To obtain the optimum output power at a given ambi-
where R is the weight of the input influence on the objec- ent temperature, the quantitative relationship between
tive function and Rs is the expected output. the ambient temperature and the module temperature is
When the predictive control achieves the optimal simulated according to the actual measurement data. The
control objective, the objective function takes the mini- reference value of the module temperature is calculated
mum value, as: as:

∂J Tmf = 0.9155Tamb + 80.01 (14)

= 0 ⇒ U = (�T � + R)−1 �T (Rs − Fx(ki )) (11)
∂U where Tamb is the ambient temperature.
The predicted value of the state variable at the next The simulation of an MPC controller of PV modules is
time is: established in Simulink/Matalb, as shown in Fig. 6, with
sampling time Ts =10 , prediction step NP =10 and control
x(ki + 1) = Ax(ki ) + Bx(ki ) = (A − BKx )x(ki ) + BKy r(ki ) step Nm =2.
(12)
Zhao et al. Prot Control Mod Power Syst (2021) 6:32 Page 6 of 12

rotating flux position angle θ; (2)id , iq are obtained by

coordinate transformation; (3) According to the feedback
currents idref , iqref  , the PI controllers are used to con-
trol the current inner loop to obtain ud , uq ; (4)ud , uq are
transformed into uα , uβ through the dq to αβ coordinate
transformation; (5)uα , uβ are input to the SVPWM mod-
ule to generate PWM signals to control the inverter.

4.2 Model predictive current control (MPCC)

Fig. 8  Feedforward decoupling control structure of the
of the 3‑phase inverter
grid-connected PV inverter based on PI regulation The MPCC technology uses current and past states to
predict inverter current values in the next switching cycle
derived from an accurate mathematical model of the sys-
tem. The optimal value function is then used to select
It can be seen from the results in Fig. 7 that the module
the optimal switch combination, so as to precisely track
temperature is slightly lower than the reference temper-
the current information [28]. The prediction model and
ature in the first hour of the simulation. The minimum
system optimization value function are determined as
value of UV is 300 nm and the maximum value is 430 nm.
follows:
This does not exceed the constraint range. In the second
The predictive model is the basic element of the MPCC
hour of the simulation, although there is a small fluctua-
algorithm. The Forward Euler is used to discretize (6) to
tion in the module temperature, the temperature tracking
obtain the current prediction value at k + 1 time, as:
effect has been achieved. At about 2 h, the power gradu-
ally reaches the maximum value. Overall, MPC tracking  RTs Ts
performance is fast and good, and the purpose of MPPT  iα (k + 1) = (1 −
 )iα (k) + [uα (k) − eα (k)]
L L
is realized.  i (k + 1) = (1 − RTs )i (k) + Ts �u (k) − e (k)�

β β β β
L L
4 Control of the 3‑phase PV grid‑connected (16)
inverter where Ts is the sampling period. Transforming (16) into
4.1 Feedforward decoupling control of inverter based dq coordinates yields the discrete space state prediction
on PI regulation model of the grid-connected inverter, as:
The feedforward decoupling control strategy based on
id (k + 1) = iα (k + 1)cosθ + iβ (k + 1) sin θ

double-loop PI regulation is a common PWM inverter
control strategy. Its core is the design of a DC voltage iq (k + 1) = −iα (k + 1) sin θ + iβ (k + 1) cos θ
outer loop, a current inner loop and PI regulators. The (17)
structure of inverter control strategy is shown in Fig. 8. where θ is the angle between the a-axis and d-axis, and
In order to prevent the mutual influence between d, q id (k + 1), iq (k + 1) are the dq axis components of the
axes in the current inner loop and eliminate the influence predicted current at k + 1 time.
between active and reactive power, decoupling control is The optimal value function uses the quadratic perfor-
used, as: mance index of prediction error to select the optimal
 �
Kil
� combination of switching variables from different pre-
 Ud = − Kip + s (idref − Id ) + ωLIq + ed

 diction values to minimize the error between the out-
� � (15) put current and the reference current at k + 1 time. The
K value function is constituted by the quadratic difference
 Uq = − Kip + il (iqref − Iq ) − ωLId + eq


s between the predicted current value and the reference
current value at k + 1 time, as:
where Kil , Kip are the proportional and integral regula-
tion coefficients of the current inner loop PI regulators, 2  2
g = id∗ (k + 1) − id (k + 1) + iq∗ (k + 1) − iq (k + 1)

respectively.
The feedforward decoupling control of an inverter (18)
based on a PI regulator combines SVPWM and feedfor- where id∗ (k + 1), iq∗ (k + 1) are the dq axis current refer-
ward decoupling control strategy. The control process ences at k + 1 time.
is as follows: (1) Phase-locked loop is used to obtain the
 Zhao et al. Prot Control Mod Power Syst (2021) 6:32 Page 7 of 12

1. The measurement module. Through real-time meas-

urements of grid voltages, inverter currents and DC
bus voltages, the actual operating data is transmitted
to the prediction model.
2. PLL. PLL provides the phase angle for coordinate
transformation and calculation of the prediction
model.
3. Prediction model. The prediction model uses the
recursive relationship in (17) to calculate the current
Table 2  Parameters of the simulated system
Parameters of the PV value Parameters of the value
module system

S/W·m−2 1000 P/W 0.35 × ­104

T/℃ 25 f /Hz 60
Uoc/V 85.40 Cdc/F 0.06
Isc/A 6.09 L/H 0.01 × ­10–6
Um/V 72.90 R/Ω 0.03 × ­10–6
Im/A 5.69 Udc/V 480
Ns 7 PT1/VA 0.25 × ­106
Np 88 PT2/VA 0.50 × ­108

prediction value at k + 1 time through the measured

value of the system at k time.
4. Optimize the value function. The quadratic error of the
reference current and the predicted current at k + 1
time is taken as the optimization value function in (18).
5. Current reference. The active current reference
comes from the outer DC voltage loop, i.e., the ref-
erence voltage Udcref of the PV module is generated
by MPPT control, and is then compared to the meas-
Fig. 9  The algorithm flow of FCS-MPCC ured DC bus voltage Udc , before passing to a PI regu-
lator. In order to be connected to the power grid with
unit power factor, the reactive current reference is set
to 0.
6. Signal driver. From the optimized value function, the
switch state corresponding to the minimum value
function is taken as the optimal switch state of the
following sampling period, and the power semicon-
ductor devices are switched accordingly.

Fig. 10  FCS-MPCC control structure of the PV inverter 5 Simulation verification and analysis

A simulation model is built in MATLAB to verify the fea-
sibility of the control strategy. In order to implement the
The FCS-MPCC algorithm flow is presented in Fig.  9, control effect, the FCS-MPCC algorithm is programmed
and the structure of FCS-MPCC for the PV inverter through the S-function. The system model is shown in
is shown in Fig.  10. The details of the control strategy Fig.  11, the sampling period is set to 2 × ­10–5  s, and the
mainly include the following 6 parts: parameters of the simulated system are shown in Table 2.
Zhao et al. Prot Control Mod Power Syst (2021) 6:32 Page 8 of 12

Fig. 11  Simulation model of a grid-connected PV power generation system

Fig. 12  Response curves of PV arrays output during external conditions changes a PV arrays output during sudden change of irradiance, b PV
arrays output during sudden change of temperature

Figure 12a, b show the response curves of the PV array output power. This is in line with the theory of PV power
output DC voltage and current, and the active and reac- generation.
tive power into the grid during sudden changes of irra- Figure 13 presents the active powers into the grid and
diance and temperature, respectively. As can be seen, the voltages at the point of common coupling (PCC)
when irradiance and temperature change suddenly under different grid faults. Four different types of faults
at 1.0  s and 2.0  s respectively, the voltage is adjusted are applied at 1.5 s, and cleared after 0.5 s. It is seen from
quickly to follow the maximum power point, and the Fig.  13 that the active power and voltage can reach sta-
MPPT tracking effect is effectively realized by the MPC ble states within 0.2 s after fault clearance. This indicates
controller. It is also seen that the FCS-MPCC algorithm that the grid-connected PV inverter with FCS-MPCC
quickly stabilizes the inverter output power at the maxi- strategy can operate during faults with fast response.
mum power point after disturbances at 1.0 s and 2.0 s. In It can be seen from the partial enlarged diagram that
addition, as shown in Fig. 12, the increase of irradiance the power oscillation and voltage fluctuation caused by
leads to a power increase indicating a positive relation- the three-phase fault are the most severe, and have the
ship between the light intensity and PV output power, greatest impact on the stability of the system, while the
while temperature change is inversely related to PV
 Zhao et al. Prot Control Mod Power Syst (2021) 6:32 Page 9 of 12

is provided by the voltage outer loop, with the reactive

current reference iqref = 0A . The parameters of the tra-
ditional feedforward decoupling PI controller are given in
Table 3.
It can be seen from Figs.  14 and 15 that after initial
transients, both methods reach stable operational states.
The voltage and current are in phase, the system realizes
the operation of unity-power factor, and the total har-
monic distortion (THD) rate of grid current is less than
5%. This achieves the grid connection requirements.
Comparing the THD of the grid current shown
in Figs.  14b and 15b shows that the current THD is
reduced from 4.82% with the traditional feedforward
decoupling PI control to 2.54% with the proposed FCS-
MPCC. In addition, as seen from Figs. 14c, d and 15c,
d, the system under FCS-MPCC strategy only takes
0.08  s to reach stable operation, compared to 0.15  s
with the traditional feedforward decoupling PI control.
Overall, the quality of the output waveforms is greatly
Fig. 13  Response curves of system under short circuit faults a improved, and the control accuracy is significantly
comparison of active powers into power grid under different faults, b increased with the proposed FCS-MPCC method.
comparison of voltages at PCC under different faults
To test the stability of the system under abnormal
operating conditions, the initial conditions of the sys-
tem are changed. The references of the active and reac-
single-phase fault has the least impact. This is in line with tive currents are suddenly changed to idref = 0.78A and
practical working conditions. iqref = −0.45A at 0.2 s. The simulation results with the
System performance with the traditional feedforward FCS-MPCC method and the traditional feedforward
decoupling PI control and the proposed FCS-MPCC decoupling PI control are compared in Fig. 16.
control method is shown in Figs. 14 and 15, respectively. As can be seen from Fig.  16a, b, compared with the
During the simulation, the active current reference idref traditional feedforward decoupling PI control, id , iq

Fig. 14  A-phase waveforms of inverter with traditional feedforward decoupling PI control under initial condition a A-phase voltage and current at
PCC, b harmonic analysis of A-phase grid current, c active and reactive power into power grid, d power factor
Zhao et al. Prot Control Mod Power Syst (2021) 6:32 Page 10 of 12

Fig. 15  A-phase waveforms of inverter with FCS-MPCC under initial condition a A-phase voltage and current at PCC, b harmonic analysis of
A-phase grid current, c active and reactive power into power grid, d power factor

Table 3  Parameters of PI controller can better track idref , iqref with FCS-MPCC strategy.
Comparing the response speed of the active power and
Parameters Value reactive power shown in Fig.  16c, d, FCS-MPCC strat-
Kil 369 egy takes 0.05  s, while the feedforward decoupling PI
Kip 0.852 controller requires 0.08  s, indicating better dynamic
Kvl 600 response of FCS-MPCC through decoupling control
Kvp 5 of active and reactive power. In addition, the reactive

Fig. 16  Comparisons of waveforms with FCS-MPCC and traditional feedforward decoupling PI control after changing initial conditions a active
current, b reactive current, c inverter output active power, d inverter output reactive power
 Zhao et al. Prot Control Mod Power Syst (2021) 6:32 Page 11 of 12

power transmission to the grid can be controlled by the Author’s information

Yingying Zhao, (1997-), female, postgraduate, the main research direction is
q-axis current. renewable energy access to Smart Grid.
Aimin An, (1972-), male, Ph.D. and Professor, Major in the artificial intelligence
6 Conclusion control technology, advanced industrial process control, renewable energy
access to Smart Grid.
This paper addresses the optimal control problem of a Yifan Xu, (1996-), male, postgraduate, the main research direction is renewable
grid-connected PV inverter system and optimizes the energy access to Smart Grid.
tracking performance of MPPT. To better deal with the Qianqian Wang, (1995-), female, Ph.D. candidate, Major in the artificial intel-
ligence control technology and advanced industrial process control.
small external constraints and system interference, an Minmin Wang, (1995-), female, postgraduate, the main research direction is
MPC-based MPPT method for PV arrays and an FCS- renewable energy access to Smart Grid.
MPCC strategy for the inverter in dq coordinate are pro-
Funding
posed. The steady-state and dynamic performance of the This work was supported by National Science Foundation of China (61563032,
MPC-based system are verified by simulation under dif- 61963025), Project supported by Gansu Basic Research Innovation Group
ferent external constraints, faults, and change of initial (18JR3RA133), Industrial Support and Guidance Project for Higher Education
Institutions of Gansu Province (2019C-05), Open Fund Project of Key Labora-
conditions. The performance is also compared with the tory of Industrial Process Advanced Control of Gansu Province (2019KFJJ02).
traditional feedforward decoupling PI controller. The fol-
lowing conclusions are obtained: Availability of data and materials
All data used or analyzed during this study are included in the published
article.
1. During sudden change of the external factors (ambi-
ent temperature, irradiance) or power grid faults, the Declarations
grid-connected PV power generation system with
MPC has rapid response speed, good anti-interfer- Competing interests
The authors declare that they have no known competing financial interests
ence ability and satisfactory dynamic response. The or personal relationships that could have appeared to influence the work
MPC controller can quickly track the maximum reported in this paper.
power point of the PV array and output stable active
Author details
power. This effectively increases the tracking sensitiv- 1
 College of Electrical and Information Engineering, Lanzhou University
ity and accuracy of a PV power generation system, of Technology, Lanzhou 730050, Gansu, China. 2 Key Laboratory of Gansu
and improves the MPPT control performance. Advanced Control for Industrial Processes, Lanzhou 730050, Gansu, China.
3
 National Demonstration Center for Experimental Electrical and Control
2. Under FCS-MPCC, the grid current and voltage can Engineering Education, Lanzhou University of Technology, Lanzhou 730050,
be precisely controlled to operate in unity-power Gansu, China.
factor. Compared to the feedforward decoupling PI
Received: 16 April 2021 Accepted: 16 September 2021
controller, the proposed FCS-MPCC results in lower
current THD, better power quality, faster current
tracking and improved dynamic performance.
References
1. Ostad-Ali-Askari, K., Ghorbanizadeh Kharazi, H., Shayannejad, M., &
Zareian, M. J. (2019). Effect of management strategies on reducing nega-
tive impacts of climate change on water resources of the Isfahan-Borkhar
Abbreviations aquifer using MODFLOW. River Research and Applications, 35(6), 611–631.
S: Irradiance; T : Ambient temperature; Uoc: Open-circuit voltage; Isc: Short- 2. Ostad-Ali-Askari, K., Ghorbanizadeh Kharazi, H., Shayannejad, M., &
circuit current; Um: Peak voltage; Im: Peak current; Ns: Number of PV modules in Zareian, M. J. (2020). Effect of Climate change on precipitation patterns
series; Np: Number of PV modules in parallel; P: Rated power of grid-connected in an arid region using GCM models: Case study of Isfahan-Borkhar plain.
PV power generation system; f : Power frequency; Cdc: Voltage stabilizing Natural Hazards Review, 21(2), 04020006.
capacitor; L: Filter inductor; R: Filter resistor; Udc: The DC bus voltage; PT1: Rated 3. Ostad-Ali-Askari, K., Su, R. D., & Liu, L. M. (2018). Water resources and
power of 1#transformer; PT2: Rated power of 2#transformer. climate change. Journal of Water and Climate Change, 9(2), 239.
4. Magdy, G., Mohamed, E. A., Shabib, G., et al. (2018). Microgrid dynamic
Acknowledgements security considering high penetration of renewable energy. Protection
Not applicable. and Control of Modern Power Systems, 3(1), 1–11.
5. Shan, Y. H., Hu, J. F., & Guerrero, J. M. (2019). A model predictive power
Authors’ contributions control method for pv and energy storage systems with voltage support
All authors contributed to the study conception and commented on previous capability. IEEE Transactions on Smart Grid, 11(2), 1018–1029.
versions of the manuscript. YZ modeled the system, designed the algorithms 6. Chaibi, Y., Allouhi, A., Salhi, M., & EI-jouni, A. (2019). Annual performance
of control strategy and wrote the paper. AA brought up the original concep- analysis of different maximum power point tracking techniques used in
tion and provided the technical guidance as a corresponding author. YX photovoltaic systems. Protection and Control of Modern Power Systems,
checked the data and tested simulation model in the paper. QW and MW con- 4(1), 1–10.
tributed to revise and typeset the manuscript. All authors read and approved 7. Shang, L., Guo, H., & Zhu, W. (2020). An improved MPPT control strategy
the final manuscript. based on incremental conductance algorithm. Protection and Control of
Modern Power Systems, 5(1), 1–8.
Zhao et al. Prot Control Mod Power Syst (2021) 6:32 Page 12 of 12

8. Peter, P. K., & Agarwal, V. (2016). Photovoltaic module-integrated stand- 19. Novak, M., Dragicevic, T., Blaabjerg, F., & Nyman, U. M. (2017). Analyti-
alone single-stage switched capacitor inverter with maximum power cal performance verification of FCS-MPC applied to power electronic
point tracking. IEEE Transactions on Power Electronics, 32(5), 3571–3584. converters: A model checking approach. In 2017 IEEE 18th workshop on
9. Nishant, K., Ikhlaq, H., Bhim, S., et al. (2017). Rapid MPPT for uniformly and control and modeling for power electronics (pp. 1–6). IEEE: California.
partial shaded PV system by using JayaDE algorithm in highly fluctuating 20. Dragicevic, T., & Novak, M. (2018). Weighting factor design in model
atmospheric conditions. IEEE Transactions on Industrial Informatics, 13(5), predictive control of power electronic converters: An artificial neural
2406–2416. network approach. IEEE Transactions on Industrial Electronics, 66(11),
10. Venkatramanan, D., & John, V. (2019). Dynamic modeling and analysis 8870–8880.
of buck converter based solar PV charge controller for improved MPPT 21. Wang, J., Liu, X., Xiao, Q., Zhou, D., Qiu, H., & Tang, Y. (2020). Modulated
performance. IEEE Transactions on Industry Applications, 55(6), 6234–6246. model predictive control for modular multilevel converters with easy
11. Gao, F. Y., Du, Q., Qiao, Y., & Qiang, G. D. (2018). Decoupling control strat- implementation and enhanced steady-state performance. IEEE Transac-
egy with inverter-side current feedback for LCL-type three-phase PV grid- tions on Power Electronics, 35(9), 9107–9118.
connected inverter. Power System Protection and Control, 46(09), 122–128. 22. Dragicevic, T. (2017). Model predictive control of power converters for
12. Dhimish, M., & Chen, Z. (2019). Optimized design of multiresonant AC robust and fast operation of AC microgrids. IEEE Transactions on Power
current regulators for single-phase grid-connected photovoltaic invert- Electronics, 33(7), 6304–6317.
ers. IEEE Journal of Photovoltaics, 9(6), 1815–1818. 23. Liao, C. Y., Lin, W. S., Chen, Y. M., & Chou, C. Y. (2016). A PV micro-inverter
13. Hannan, M. A., Ghani, Z. A., Mohamed, A., & Uddin, M. N. (2015). Real-time with PV current decoupling strategy. IEEE Transactions on Power Electron-
testing of a fuzzy-logic-controller-based grid-connected photovol- ics, 32(8), 6544–6557.
taic inverter system. IEEE Transactions on Industry Applications, 51(6), 24. Salman, S., Xin, A. I., & Zhou, W. U. (2018). Design of a P-&-O algorithm
4775–4784. based MPPT charge controller for a stand-alone 200W PV system. Protec-
14. Wai, R. J., Lin, Y. F., & Liu, Y. K. (2015). Design of adaptive control and fuzzy tion and Control of Modern Power Systems, 3(1), 1–8.
neural network control for single-stage boost inverter. IEEE Transactions 25. Dabra, V., Paliwal, K. K., Sharma, P., & Kumar, N. (2017). Optimization of
on Industrial Electronics, 62(12), 5434–5445. photovoltaic power system: A comparative study. Protection and Control
15. Qu, A. W., & Chen, D. L. (2017). Three-phase voltage-fed quasi-z-source of Modern Power Systems, 2(1), 1–11.
photovoltaic grid-connected inverter system with repetitive control. 26. Abdel-Rahim, O., & Wang, H. (2020). A new high gain DC-DC converter
Power System Technology, 41(06), 1997–2006. with model-predictive-control based MPPT technique for photovoltaic
16. Jin, N., Hu, S. Y., Cui, G. Z., & Jiang, S. X. (2015). Finite state model predictive systems. CPSS Transactions on Power Electronics and Applications, 5(2),
current control of grid-connected inverters for PV systems. Proceedings of 191–200.
the CSEE, 35(1), 190–196. 27. Scoltock, J., Geyer, T., & Madawala, U. K. (2015). Model predictive direct
17. Yang, J., Gu, D. D., Sun, M. H., et al. (2016). Multi-objective optimization power control for grid-connected NPC converters. IEEE Transactions on
model-predictive control of PV grid-connected inverters. Power System Industrial Electronics, 62(9), 5319–5328.
Protection and Control, 44(15), 112–119. 28. Bayhan, S., Trabelsi, M., Abu-Rub, H., & Malinowski, M. (2017). Finite-
18. Morfin-Magaa, R., Rico-Melgoza, J. J., Ornelas-Tellez, F., & Vasca, F. (2019). control-set model-predictive control for a quasi-Z-source four-leg inverter
Complementarity model of a photovoltaic power electronic system under unbalanced load condition. IEEE Transactions on Industrial Electron-
with model predictive control. IEEE Transactions on Circuits and Systems I: ics, 64(4), 2560–2569.
Regular Papers, 66(11), 4402–4414.