Grid Connected PV System with Reactive Power
Compensation for the Grid
Asheesh Dhaneria
Research and Development Department
Electrical Research and Development Association
Vadodara, India
asheesh.dhaneria@erda.org
Abstract— Grid tied solar inverters are designed to generate
power at unity power factor which means they have the capability
to produce active power only. The reactive power requirement of
the load is catered by grid only. With the dramatic increase in the
deployment of renewable based Distributed Energy Resources,
reactive power drawn from the grid as compared to active power
has increased considerably. This affects the power quality of the
grid. If the grid tied solar inverter is made smart in terms of
supplying reactive power in addition to active power, the reactive
power requirement from the grid will reduce as the grid has to
supply lesser reactive power. In this paper, different methods for
operation of PV inverters in terms of absorbing and injecting
reactive power in addition to its normal functionality has been
discussed. Simulation results of various static reactive power
compensation control strategies has been presented and the
results are discussed in detail.
Index Terms—Reactive power compensation, Solar PV system,
VAR management, Volt VAR method.
I. INTRODUCTION
Grid tied solar inverter are traditionally designed to operate
at unity power factor which means that they have capability to
produce active power only. As electrical loads are
predominantly inductive loads, they tend to consume more
amount of inductive reactive power. Currently this reactive
power requirement is catered by grid only. With the increase in
penetration of large number of Distributed Energy Resources
(DERs) which pump only active power into the grid, the site
power factor becomes poor from utility point of view, affecting
the performance of the grid. Formerly, the reactive power
injection/absorption is done through the FACTS devices for
mitigating power quality issues. There are different types of
FACTS devices like series, shunt, series-series, series-shunt
which are used based on the requirement. But these devices
suffer from the drawback of large sizing, high cost, large area
of installation etc.
On account of these reasons, there is a need to regulate the
reactive power flow in the power system network as this may
affect the voltage regulation also. In this paper, control
strategies for grid tied solar inverter to supply reactive power in
addition to the active power are discussed. As the number of
978-1-7281-3103-0/20/$31.00 ©2020 IEEE
Authorized licensed use limited to: Auckland University of Technology. Downloaded on May 28,2020 at 10:46:39 UTC from IEEE Xplore. Restrictions apply.
grid tied-inverters increases, their usage as VAR compensators
will help in grid voltage regulation and reduce the need of
expensive capacitor banks.
In order to use the reactive power capability of the smart
inverters, there are two main options: oversizing the inverter or
active power curtailment. Oversizing the inverter has an extra
cost that needs to be taken into account. Active power
curtailment is the active power reduction and can be
implemented in different ways such as fixing the maximum
power point to say 70% of rated power or based on PCC voltage
as the reference voltage. This method leads to reduced
generation from the installed solar PV panels and hence
reduced earning to the PV system owners [1].
II. REACTIVE POWER COMPENSATION
The active power pumped by grid tied solar inverter into the
grid is a function of solar insolation. This means that the amount
of active power pumped into the grid will be lower than the
designed rated capacity of solar inverter if the solar irradiance
is less (which actually happens as the solar irradiance is not
uniformly maximum throughout the day). This leads to
underutilization of the inverter resource. If the inverter is
programmed to provide reactive power also in addition to active
power (based on solar irradiance availability) then the inverter
can be operated at its rated capacity even when the solar
resource is not fully available. Reactive power compensation
through solar inverter is an interesting method to manage
network voltages through reactive power injection and
absorption. Reactive power support as provided by solar PV
inverters are of two types: dynamic support and static support
[1].
Dynamic reactive power compensation is used to provide
stability to the grid when some events like short term voltage
sags or peaks occur. These type inverters use dynamic reactive
power provision for their riding-through capabilities.
Static reactive power support is used to maintain voltage
levels within an acceptable range. The idea of using reactive
power for maintaining voltage limitations derives from
traditional power system operation, where reactive power is
necessary to support the overall voltage stability in the grid.
 Control system can be developed and solar PV inverters can
be programmed for implementation of static reactive power
compensation by means of four different control strategies: (i)
fixed Q, (ii) fixed cos φ, (iii) cos φ (P) and (iv) Q (V). For all
the control strategies, the MPPT operation is given priority
which means that if abundant solar irradiance is available (say
from 1000 W/m2 to 800 W/m2) and the grid voltage is within
limits, then the complete solar PV generation is fed to grid in
the form of active power. However, if the cloud blocks the sun
and causes the PV array’s maximum power to drop below a set
value, say 80%, then any of the above stated control mode can
be switched ON. This will enable the system to utilize the
unused capacity of the solar inverter which otherwise would not
be used when the solar inverter is operated at unity power
Figure 2. Operational window for Fixed power factor control
factor.
A. Fixed Q
When the inverter is designed to operate at fixed reactive
power (Q) method, then irrespective of the irradiance and the
active power generated by the PV panel, the PV inverter will
deliver or absorb fixed amount of set reactive power with only
constraint of designed VA rating of the inverter. The reactive
power value will be prescribed by the plant operator depending
on the load forecasting and load scheduling on an hourly basis.
Figure 1 shows the operational window for fixed Q method.

Figure 3. Calculations for fixed power factor mode

C. cos φ (P)
In this mode of operation, the inverter varies its power factor
according to the characteristic curve (cos φ versus P). This
curve may be given by the plant operator. The idea is to vary
the power factor when the active power output reaches a pre-
defined value say 50% of the rated output of inverter. Figure 4
depicts the characteristic curve and operational window for cos
φ (P) method.
Figure 1. Operational window for Fixed Q control

B. Fixed cos φ
The grid tied inverter can be programmed to operate at a
fixed power factor (figure 2). In this mode, the inverter varies
the reactive power output to maintain a constant user defined
power factor. The generated reactive power is proportional to
the active power. The power factor may be commanded to any
value between 0.8 and 1.0, leading or lagging. For pf < 1, the
inverters real output power is limited by the power factor
command. The power factor value may be changed in the
control systems setting parameter of the software. In this mode,
the power limit curve has the effect of curtailing the real output
power during periods of high irradiance for pf < 1. Hence, in
order to fully utilize the active power generation capability of
solar PV, the reactive power is allowed to change. Hence during
the period of high irradiance say 800 to 1000 W/m2, the reactive
power compensation block is suppressed. The calculation block
diagram for fixed power factor mode is as shown in figure 3.

Authorized licensed use limited to: Auckland University of Technology. Downloaded on May 28,2020 at 10:46:39 UTC from IEEE Xplore. Restrictions apply.

Figure 4. Characteristic curve and operational window for cos φ (P) control
D. Q (V)
This method is also referred as Volt-VAR function, the
inverter will provide an amount of reactive power as a function
of voltage at the point of interconnection according to the
setting points programmed in the inverter [2]. Figure 5 depicts
the characteristic curve that may be set to an inverter for
performing reactive power compensation utilizing Volt-VAR
method. Following this curve, the inverter will deliver or absorb
the reactive power when the operating range of voltage is
outside the specified voltage range. If the voltage is between U1
and U2 (say 0.95 and 1.05 respectively), the inverter will
operate at unity power factor with no reactive power
compensation. If the voltage is less than Ulow or greater than
Uhigh, then the inverter will deliver or absorb 100% reactive
power respectively [3].
Figure 5. Characteristic curve and operational window for Q (V) control
Authorized licensed use limited to: Auckland University of Technology. Downloaded on May 28,2020 at 10:46:39 UTC from IEEE Xplore. Restrictions apply.
III. SIMULATION RESULT
To verify the feasibility of all the four control strategies, the
simulation of three phase, 50 kW solar PV inverter is presented
in this section. MATLAB Simulink platform is used for this
simulation. The configuration of three phase PV system
connected to the utility grid is shown in figure 6.
Figure 6. A grid connected three phase inverter PV system
Test 1: Constant irradiance commanding constant active
power pump to grid, various reactive power [4].
Figure 7 shows the waveform of the reactive power
command (Q*) and injected or absorbed reactive power (Q), u-
phase inverter voltage and current, active power and DC
voltage. Irradiance is set at 500 W/m2 during the simulation
time, which means that 50 kVA rating inverter will continue to
feed active power of 25 kW owing to MPPT operation as can
be seen from active power waveform of figure 7. In the
beginning, the Q* is set to 0 VAR. At time t = 0.7, Q* is
abruptly changed from 0 to 30 kVAR (reactive power
absorbed). Thus the amplitude of the current fed by the inverter
increases due to the increment of reactive power command. The
phase angle of the inverter current also becomes leading and
can be seen from the inverter phase voltage and current
waveform. Likewise, at t = 0.9, the Q* is changed from 0 to 30
kVAR (reactive power supplied). In this case, the phase angle
of the inverter current becomes lagging. During the time period,
0.8 < t < 0.9, the inverter operates at unity power factor with 25
kW injected active power into the grid. It can be found that the
VAR compensation scheme using fixed Q method can control
the PV inverter reactive output power without causing
noticeable drop in the active power generation.
45
Reference Q (kVAR)
30
Measured Q (kVAR)
15
0
-15
-30
-45
0.6 0.7 0.8 0.9 1 1.1
 400
Inverter phase voltage (V) The inverter current settles to the required waveform in less
Inverter phase current (A)
than a cycle time period showing smooth and quick transition.
200 Time stamp Active power Reactive power
(kW) (kVAR)
0 0 to 1 second 49.5 0
1 to 2 second 39.5 23.9
-200
2 to 3 second 29.3 17.8

-400 3 to 4 second 19.2 11.6

0.6 0.7 0.8 0.9 1 1.1
4 to 5 second 9.2 5.5
50
Active Power (kW) TABLE I. ACTIVE AND REACTIVE POWER
40
1200

30 1000
Irradiance (W/m sq)
800
20
600
10
400

0 200
0.6 0.7 0.8 0.9 1 1.1
0
600 0 1 2 3 4 5
Reference DC voltage (V) 80
Measured DC voltage (V) Active Power (kW)
550 60 Reactive Power (kVAR)

40
500 20

0
450
-20

-40
400
0.6 0.7 0.8 0.9 1 1.1 -60
0 1 2 3 4 5
Figure 7. Key waveforms of solar PV inverter for fixed Q mode with
constant active power but various reactive power
400
Inverter phase voltage (V)
Test 2: Varying irradiance commanding active power pump 300 Inverter phase current (A)
to grid in accordance with MPPT operation, power factor set to 200
fixed value.
100
Figure 8 shows the waveform of set irradiance, active and
0
reactive power fed by inverter, u-phase inverter voltage and
current. The irradiance is varied from 1000 to 200 W/m2 in steps -100
of 200 units. The power factor for this mode of operation is set -200
as 0.85. The active power fed to grid follows the MPPT
-300
operation to extract and pump maximum power to grid so as not
to lose revenue. The reactive power fed by inverter varies in -400
proportion to the active power so as to maintain fixed set value 1.85 1.9 1.95 2 2.05 2.1 2.15
of power factor of 0.85 value. Values of active power and
reactive power for different time stamps as taken from the Figure 8. Key waveforms of solar PV inverter for fixed power factor mode
active, reactive power plot of figure 8, are tabulated in Table 1.
The inverter current shows a constant time lagging from the Test 3: Fixed irradiance commanding constant active power
inverter voltage showing a constant power factor operation. pump to grid, Volt-VAR operation
This waveform is shown for the transition period in which the Figure 9 shows the waveforms for grid voltage, inverter’s
irradiance is changing from 800 to 600 W/m2 at t = 2 second. reactive power and the active power. The irradiance is kept

Authorized licensed use limited to: Auckland University of Technology. Downloaded on May 28,2020 at 10:46:39 UTC from IEEE Xplore. Restrictions apply.
fixed at 500 W/m2 during the simulation. During the time IV. CONCLUSION
period, 0 < t < 1 and 3 < t < 5 when the grid voltage (in per unit) This paper presents a grid connected photovoltaic inverter
is between U1 and U2 values of characteristic curve of figure 5, system having reactive power compensation feature. The need
the inverter operates at unity power factor with zero reactive of reactive power generation in the PV inverter has been
power. During the time period, 1 < t < 2, when the gird voltage discussed. Simulink model of three phase dual stage grid
is below Ulow (refer characteristic curve of figure 5), reactive connected solar PV inverter is developed in MATLAB. PV
capacitive power injected into the grid (positive value set). inverter operation in active and reactive power generation mode
When the grid voltage is above Uhigh (6 < t < 7), reactive has been simulated. The various active and reactive power
inductive power injected into the grid (negative value set). For measurements were captured for analysis. Simulation results
other values of grid voltages, the magnitude of reactive presented for various control modes showed that the VAR
inductive or capacitive power to be injected into the grid compensation schemes can be implemented in solar inverter
depends upon the equation of characteristic curve of figure 5. control system to control the PV inverter reactive output power
1.5 without causing noticeable drop in the active power generation.
Grid voltage (p.u.) The reactive power compensation should be made available in
1
solar a solar plant to improve the effective utilization of the
system when the solar irradiance is not abundant.
0.5 REFERENCES
[1] J. F. Gomez-Gonzalez, D. Canadillas-Ramallo, B. Gonzalez-Diaz, J. A.
0 Mandez-Perez, J. Rodriguez, J. Sanchez and Guerrero-Lemus, "Reactive
0 1 2 3 4 5 6 7 power management in photovoltaic installations connected to low-
voltage grids to avoid active power curtailment," Renewable Energy and
Power Quality Journal, ISSN 2172-038 X, No. 16 April 2018.
100 [2] Huijuan Li, Yan Xu, Sarina Adhikari, D. Tom Rizy, Fangxing Li and
Reactive power (kVAR) Philip Irminger, "Real and reactive power control of a three-phase single-
50 stage pv system and pv voltage stability," IEEE Power and Energy
Society General Meeting, 2012.
0 [3] "Active and reactive power control and LVRT" Schneider Electric,
Application note, 976-0381-01-01/B, March 2017
-50 [4] A. Cagnano, E. De Tuglie, M. Liserre, and R. A. Mastromauro, "Online
optimal reactive power control strategy of PV inverters," IEEE Trans.
-100 Ind. Electron., vol. 58, no. 10, pp. 4549–4558, Oct. 2011.
0 1 2 3 4 5 6 7

Figure 9. Key waveforms of solar PV inverter for volt-VAR control scheme

Authorized licensed use limited to: Auckland University of Technology. Downloaded on May 28,2020 at 10:46:39 UTC from IEEE Xplore. Restrictions apply.