Reduced Sensor Based Control of PV-DSTATCOM with Switch Current Limiting Scheme

Abstract

The work delineated in this paper deals with the reduced sensor-based operation of PV-DSTATCOM (Photovoltaic distribution static compensator) focusing on the voltage source inverter (VSI) switch current limiting control. In this study, the soft computing technique of PV-DSTATCOM control based on the variable leaky least mean square (VLLMS) algorithm is modified to incorporate both reduced sensor and switch current limiting schemes. DSTATCOM is predominantly introduced to improve current related power quality issues by providing non-real power (i.e., the reactive and harmonic component) of the load demand. However, during uncertain increase in the non-real power requirement at the load end due to sudden change in the load demand or any other transient conditions, the non-real current injected by the DSTATCOM may increase to a level which is beyond the current handling capacity of the DSTATCOM switches. As a consequence, the switches may become permanently damaged and, subsequently, the performance of the DSTATCOM will deteriorate. Hence, the current through the VSI switches can be limited by controlling the amount of non-real power flow from the DSTATCOM. During this transient or high current condition, the control transits from power quality improvement mode to protection mode. As a result, unity power factor operation at P.C.C (common point of coupling) is compromised for that period or few cycles of transient condition. Moreover, the efficient operation of any control demands proper information about the current and the voltage of the system, which are determined by sensors. However, the faults in these sensors may deteriorate the controller performance and degrade the system efficacy. Therefore, to mitigate this issue, the numbers of measurement units are reduced i.e., the measurement of current in only two phases for load as well as grid, and the measurement of line voltages using two voltage sensors instead of three sensors for each phase. The reduction of sensors is accomplished without compromising the controller efficiency. Further, the system is studied using MATLAB/Simulink under different condition of steady- state, varying solar irradiance, increasing load above rated capacity.

1. Introduction

The escalating interventions around power electronics devices have amplified the use of power electronic devices and loads in the distribution network, which subsequently deteriorated the quality of power in the system. Hence, power quality conditioners such as DSTATCOM (distribution static synchronous shunt compensator) and UPQC (unified power quality conditioner) are used in the system to compensate current related power quality issues [1]. In fact, DSTATCOM provides the reactive and harmonic part of the load current and assures purely sinusoidal load current and unity power factor operation at P.C.C (common point of coupling). Due to diminishing fossil fuel reserves and poor environmental condition, there is a significant drift towards the incorporation of photovoltaic (PV) array into the existing distribution system. In the case of DSTATCOM, the interconnection of PV is a progressive change as no extra inverter is required for PV [2]. Therefore, a combination of PV-DSTATCOM is gaining significant demand due to its multifunctional approach.

In literature, several control strategies for DSTATCOM and PV-DSTATCOM have been discussed. Conventional technique based on synchronous reference frame, has been discussed in [3] whereas this technique does not perform efficiently under dynamic condition due to the presence of low power filters. Similarly, in study [4], comparative analysis for different DSTATCOM control strategies is carried out. Moreover, an adaptive and advanced soft computing based control technique for DSTATCOM have delivered efficient performance in managing power quality.

The performance of any system depends on the efficiency of the controller and also on the efficacy and health of installed devices (e.g., IGBT switches, sensors) along with ancillary components (e.g., ICs used in the sensing unit). Subsequently, the accurate information of the power signals (voltage and current) is essential for the proper implementation of control schemes. Even though the sensors play a pivotal role, a fault in the sensor leads to malfunctioning of the overall system. Several literatures are available where emphasis is given on reduction in sensor. In [5], the authors have reduced sensors for dc-link voltage, VSI current and load current. The indirect current control technique is used where only grid current is considered and dc-link voltage is estimated. In literature [6], the authors have reduced source voltage and load current sensors, where phase lock loop (PLL) is used to generate unit templates. However, PLL could cause several issues, such as high cost, unstable operation under varying supply voltage, and also system complexity. Moreover, grid voltage sensors have been reduced and grid voltage is estimated in [7] using the virtual flux technique, which depends on the system parameters, such as inductance and resistance. However, the reliable generation of unit voltage templates is crucial for generating switching signals for VSI, irrespective of the parameter change of the system. In [8], the voltage sensors are eliminated and the enhanced second order generalized integrator-based voltage estimation technique, which is estimated for power quality improvement. Additionally, works related to reduced sensor discussed in the literature involves the extra burden of estimating voltages and current for the controller to operate. Hence, estimation technique is required for determining sensor-less quantities.

In the case of PV-DSTATCOM, the correct information of power signals is crucial for power quality improvement otherwise the grid current will not track the reference accurately. Therefore, a fewer number of sensors will have less probability of encountering fault. Additionally, from an economic point of view, reduced sensor operation causes a reduction in expenses in terms of cost per sensor and other ancillary components utilized for sensing unit for individual sensor. Hence, reduced sensor-based control fosters the benefits of minimized system complexity, cost, and noise.

The efficient operation of VSI (DSTATCOM) demands healthily operating power electronic switches. Generally, when a system is designed, the current handling capacity of the switches is considered very high to protect the power electronic switches from damage during transient or high current condition. As the current handling capacity or the rating of the switches increases, the cost of VSI increases. However, there still exists the possibility of damage if the current goes beyond rating during any fault. In [9], the sizing of the DSTATCOM (shunt compensator) is discussed in detail. Moreover, the power handling scheme of UPQC is implemented by the SRF method. For the control of PV-DSTATCOM, there are various efficient and advance control techniques in the literature [10,11,12,13,14,15,16]. In [17], the authors have presented a multifunctional control approach of DSTATCOM based on the delayed µ-Law proportionate normalized least mean square (DMPNLMS) method, but discussion or approach has not been taken for sensor reduction and VSI switch current protection. Authors in [18] have presented dual staged PV fed DSTATCOM, where PV is connected to DSTATCOM through boost converter. Here, the control approach also does not consider sensor reduction and power electronic switch protection under high current condition. Nevertheless, the scheme for VSI switch current limiting for the protection of DSTATCOM is absent in all these studies. The safety factor for current rating is considered only during the sizing in the literature.

In order to protect the system, a current controlling factor or parameter must be in the control algorithm of VSI. This factor will give authority to the controller to limit the current through the VSI switches to the rated value whenever the current through the system goes above the rated value. In the presented work, this controlling factor is accounted in the control algorithm of PV-DSTATCOM. DSTATCOM alone compensates for the non-real (reactive and harmonics) component of load current to improve the current related power quality issues of the system. However, when PV is involved, it has to handle the maximum PV power. Therefore, in the case of PV-DSTATCOM, while designing DSTATCOM the maximum PV power and maximum non-real load power demand is considered. Thus, the amount of non-real power is the only factor which is to be controlled when it goes beyond the safe limit. During, transient or sudden increase in the load demand the limiting factor of the controller functions and shifts the control mode from power quality improvement to current limiting mode without switching off the DSTATCOM. In the protection mode of operation, the limiting factor decides the amount of non-real power to be provided by DSTATCOM and grid to keep the current through the switches within the limit. Here, the objective of power quality betterment is subsided for a certain period to safeguard the switches. Hence, there is no longer the requirement of considering a high rating of switches, rather a rating almost near to the system rating could be chosen, which will significantly reduce the cost of the VSI.

Table 1. Comparison of existing work with proposed work.

Reference	PLL-Less Grid Synchronization	Number of Low Pass Filters (LPF) Used for Filtering Oscillating Component	Grid Current Harmonic Mitigation	Reactive Power Compensation	Total Number of Sensor Used	VSI Switch Protection Scheme under High Current
[10]	Yes	4	Yes	Yes	9	No
[11]	No	2	Yes	Yes	12	No
[12]	Yes	1	Yes	Yes	10	No
[13]	Yes	0	Yes	Yes	12	No
[14]	Yes	4	Yes	Yes	9	No
[15]	Yes	4	Yes	Yes	9	No
[17]	Yes	2	Yes	Yes	10	No
Proposed work	Yes	0	Yes	Yes	7	Yes

shows a comparative analysis of the proposed work with the existing work in the literature.

The significant contributions of the proposed work are highlighted below:

I.

Reduced sensor operation by simple law of KCL and KVL in determining the sensor-less quantities. It reduces the extra burden of estimating the sensor-less signals using any complex techniques. Delivers the benefit of minimized probability of fault due to a greater number of sensor units, reduced complexity of the system, reduced cost due to reduction in sensors, and ancillary components for sensor unit.

II.

A scheme to limit the VSI switch current to protect the DSTATCOM from damage during transient and high current due to uncertain load demand by introducing a limiting factor into the controller. It prevents the switches from uncertain risk of damage due to the current of value higher than the safety factor considered while designing. This scheme enables to consider the size of DSTATCOM or VSI almost near to the rating or the requirement of the system. Hence, economically it reduces the unnecessary burden of considering comparatively higher rating of switches for the safety of the switches.

III.

Power management between grid, load, PV and DSTATCOM, for both active as well as reactive power of the system, is carried out smoothly even during the transition of the operating modes of DSTATCOM from power quality mode to protection mode.

Thus, the focus of the proposed work is the operation of the grid integrated PV-DSTATCOM with a reduction in the number of sensors and the protection of the VSI switches, along with power quality improvements and power management. Further, the arrangement of the work in the paper is as follows: Section 1 is the introduction followed by system configuration in Section 2, a proposed scheme is elaborated in Section 3 with result and discussion in Section 4. Real time simulation results and discussion are presented in Section 5 and finally, the conclusion is depicted in Section 6.

2. System Configuration

Figure 1 displays the system considered for the presented work. A 3ph system is taken with grid, single staged PV system, DSTATCOM and non-linear load. PV is directly connected to DSTATCOM through a DC-link capacitor C_dc with DC-link voltage of V_dc. DSTATCOM is connected at P.C.C through coupling inductors (L_ca, L_cb and L_cc). R-C filters are used at P.C.C to eliminate ripples generated due to the switching of VSI (DSTATCOM). Perturb and Observe technique is implemented to track maximum power point [2].

3. Proposed Scheme

The proposed work based on reduced sensor-based shunt VSI current limiting-based control algorithm involves the estimation of fundamental active component and reactive component of load current, the calculation of harmonic component, the determination of required current and voltages with a reduced number of sensors, and the computation of limiting factor to control the current.

3.1. Sensor Reduction

A fault in the sensor causes DSTATCOM to deteriorate power quality rather than improving it due to inadequate information provided to the controller. In this work, a total of seven sensors are used instead of 10 sensors. Two sensors for the sensing line voltage of grid V_sab and V_sbc; two sensors for the load phase current I_La and I_Lb; two sensors for the grid phase current I_sa and I_sb; and one sensor for the DC-link voltage V_dc. The phase voltage of all three phases are required to determine the unit vector template. Hence, the following equation is used to get V_sa, V_sb and V_sc from line voltage:

$$V_{sa}=\frac{2V_{sab}+V_{sbc}}{3}, V_{sb}=\frac{-V_{sab}+V_{sbc}}{3},V_{sc}=\frac{-V_{sab}-2V_{sbc}}{3}$$

(1)

Here, the current sensors are used to sense any two phases of the load and grid current, but the current of all three phases are important to estimate the active and reactive component and to generate switching signals. Therefore, the basic Kirchhoff’s law of current (KCL) is applied at the source side and load side to get the rest of the phase current, as in (2):

$$I_{sc}=-(I_{sa}+I_{sb}),I_{Lc}=-(I_{La}+I_{Lb})$$

(2)

Therefore, there is a reduction of three number of sensors. In the physical system, the Hall Effect current and voltage transducers are used to sense currents and voltages. The cost of one Hall Effect voltage transducer (LEM LV 25-P) is USD 76.46 and cost of one Hall Effect current transducer (LEM LA 55-P) is USD 29.44 [10]. As the current and voltage rating increases, the cost will also increase. Hence, in the present work, the total cost reduced is given as:

$$\begin{array}{l} Total cost reduced\\ =2\times (cost of one current sensor)+1\times (cost of one voltage sensor)\\ =2\times \$29.44+\$76.46\\ =\$135.44 \end{array}$$

(3)

So, a fewer number of sensors provide benefits w.r.t cost, maintenance, simplified calibration, and reduces probability of fault, a fewer number of ancillary components for the sensing unit, fewer complex circuit and noise reduction.

When the number of sensors used is more, then the probability of encountering faulty data will also increase. Additionally, a greater number of sensors will require more supporting components (ICs, regulated DC power supply system of ±15 V, resistors etc.) to acquire data from the sensor, thereby increasing the installation as well as maintenance cost.

3.2. Estimation of Real, Reactive and Harmonic Component of Load Current Using VLLMS Algorithm

The VLLMS (variable leaky least mean square algorithm) is a soft computing technique based on neural network [16]. The VLLMS based algorithm is used to determine the fundamental real and reactive component of load current. The steps involved in the estimation of these components using VLLMS is discussed in detail in [1]. Figure 2 depicts the control architecture of the proposed work. This figure provides information regarding the implementation of VLLMS in the estimation process. Then, by using the magnitude of estimated real (W_pa, W_pb and W_pc) and reactive part (W_qa, W_qb and W_qc), the magnitude of the harmonic component (W_ha, W_hb and W_hc) is calculated.

The in-phase and quadrature unit templates are used in the estimation process and are given in:

$$j_{pa}=\frac{V_{sa}}{V_{sp}}, j_{pb}=\frac{V_{sb}}{V_{sp}}, j_{pc}=\frac{V_{sc}}{V_{sp}} (where, V_{sp}=\sqrt{\frac{2}{3}(V_{sa}^2+V_{sb}^2+V_{sc}^2)})$$

(4)

$$j_{qa}=\frac{-j_{pb}+j_{pc}}{\sqrt{3}}, j_{qb}=\frac{\sqrt{3}{j}_{pa}}{2}+\frac{(j_{pb}-j_{pc})}{2\sqrt{3}}, j_{qc}=\frac{-\sqrt{3}{j}_{pa}}{2}+\frac{(j_{pb}-j_{pc})}{2\sqrt{3}}.$$

(5)

These unit templates and the magnitude of fundamental real and reactive component are used to obtain the fundamental real current, and reactive current is obtained as in (6) and (7):

$$I_{pa}=W_{pa}\times j_{pa}, I_{pb}=W_{pb}\times j_{pb}, I_{pc}=W_{pc}\times j_{pc}$$

(6)

$$I_{qa}=W_{qa}\times j_{qa}, I_{qb}=W_{qb}\times j_{qb}, I_{qc}=W_{qc}\times j_{qc}$$

(7)

Then, the harmonic current coponent of load current is calculated as:

$$I_{ha}=\sqrt{I_{La}^2-(I_{pa}^2+I_{qa}^2)}, I_{hb}=\sqrt{I_{Lb}^2-(I_{pb}^2+I_{qb}^2)}, I_{hc}=\sqrt{I_{Lc}^2-(I_{pc}^2+I_{qc}^2)}$$

(8)

3.3. Determination of N_limit to Limit VSI Switch Current

In Figure 3, the algorithm for determination of the current limiting factor N_limit is shown. In order to limit the VSC switch current based on the proposed work, power limitation strategies are introduced to the VLLMS based control of PV-DSTATCOM controller, to limit the flow of power through the converter, so that the switches can be protected even if the load demand increases beyond the rated value. Here, the DSTATCOM handles the non-real power (i.e., harmonic and reactive power) of the load demand and maximum possible power produced by PV. By kee** this in mind, the design and sizing of the VSC is performed, as given in (9). The grid provides the active power demand of the load. However, during an uncertain increase in the non-real power demand of load, the current through the switches will be more than the rated and will damage the switch. To resolve this unwanted scenario, a controlling parameter (N_limit) is introduced to control the flow of non-real power from the VSC:

$$\begin{array}{c}{S}_{sh}=\sqrt{P_{PV}^{MPP}{}^2+Q_{load\_max}^2+H_{har\_max}^2}\hfill \\ =\sqrt{{\left(1490\right)}^2+{\left(0.4\ast 1100\right)}^2+{\left(0.2\ast 0.9\ast 1100\right)}^2}\hfill \\ =1566.17 \mathrm{VA}=1.56 \mathrm{kVA}\hfill \end{array}$$

(9)

where $Q_{Load\_max}$ is the maximum reactive power requirement of the load (as maximum load of 1.1 kVA 0.9 P.F is considered) and $H_{har\_max}$ is the maximum harmonic power corresponding to maximum load i.e., here, 20% of the real power is considered as maximum harmonic power.

The algorithm required to calculate the limiting factor N_limit is presented in Figure 3. Let us calculate N_limit per phase:

$$N_{limit}=\frac{I_{non-real\_max}}{I_{non-real}}-1=\frac{I_{non-real\_max}}{\sqrt{I_{loadrms}^2-I_{realrms}^2}}-1$$

(10)

where $I_{non-real\_max}$ is the maximum nominal non-active current required by the load, $I_{non-real}$ is the actual non-active current required by the load and $I_{load}$ is the total load current. Then, $N_{limit}$ should be introduced in the evaluation of the reactive component of the reference grid current and in the reference active component of grid current for considering harmonic current as shown in the control architecture in Figure 2. If $N_{limit}$ > 0, then the total non-real part of load demand is fulfilled by the shunt VSC. If $N_{limit}$ < 0, then the extra amount of non-real part of the load demand is provided by grid. So, in case of transient or high demand of non-real power, both DSTATCOM and grid fulfills the requirement. Hence, by controlling $N_{limit}$ the VSC current can be limited.

3.4. Generation of Switching Pulses for PV-DSTATCOM

The overall control algorithm for the generation of switching the signal for VSI is shown in Figure 2. The generated fundamental real, reactive, harmonic component and the switch current limiting factor are then further utilized to generate the reference grid current signal to get the required pulses for switching of VSI.

Firstly, the average of fundamental real component (W_pavg) and reactive component (W_qavg) are calculated by using the magnitude real and reactive component of each phase:

$$W_{pavg}=\frac{W_{pa}+W_{pb}+W_{pc}}{3} W_{qavg}=\frac{W_{qa}+W_{qb}+W_{qc}}{3}$$

(11)

Then, the DC loss component is considered, which is provided by the grid to regulate the DC-link voltage and is given as:

$$W_{loss}(\kappa +1)=W_{loss}(\kappa )+k_p\{e_{dc}(\kappa +1)-e_{dc}(\kappa )\}+k_i{e}_{dc}(\kappa +1)$$

(12)

where κ is the discrete time instant, k_p as the proportional constant and k_i as the integral gain constant of DC link PI controller. The amount of PV power contributed to the system is depicted as:

$$W_{PV}=\frac{2P_{PV}}{3V_{sp}}$$

(13)

Finally, the magnitude of total in-phase and quadrature component of reference grid is obtained as in (14) and (15):

$$W_{sp}=W_{pavg}+W_{loss}-W_{PV}$$

(14)

$$W_{sq}=-W_{qavg}\times N_{limit}$$

(15)

Here, in the case of a reactive component in (15) switch current limiting factor is introduced to control the amount of reactive power contribution by grid during steady state and high current or transient condition.

Additionally, during fewer cycles of transient or high current condition the higher value of harmonic current component of load current is needed to be fulfilled by the grid in order to prevent DSTATCOM from damage. Therefore, the harmonic component obtained in (8) is incorporated with the reference grid current as in (16):

$$\begin{array}{l}{i}_{sp\_a}=W_{sp}{j}_{pa}+N_{limit}\times I_{ha} \\ i_{sp\_b}=W_{sp}{j}_{pb}+N_{limit}\times I_{hb}\\ i_{sp\_c}=W_{sp}{j}_{pc}+N_{limit}\times I_{hc}\end{array}$$

(16)

$$i_{sq\_a}=W_{sq}{j}_{qa}, i_{sq\_b}=W_{sq}{j}_{qb},i_{sq\_c}=W_{sq}{j}_{qc}$$

(17)

$$i_{sa}^{*}=i_{sp\_a}+i_{sq\_a},i_{sb}^{*}=i_{sp\_b}+i_{sq\_b},i_{sc}^{*}=i_{sp\_c}+i_{sq\_c}$$

(18)

Then, the generated reference grid currents ($i_{sa}^{*},i_{sb}^{*} \mathrm{and} i_{sc}^{*}$) are compared with sensed grid current and the error signal is the forwarded to hysteresis current controller (HCC) to get the switching pulses for DSTATCOM.

4. Simulation Results and Discussion

The proposed controller involving scheme for VSC switch current limiting of PV-DSTATCOM is implemented and analyzed with the help of the MATLAB/Simulink. The parameters involved in the following studies are given in

Table 2. Parameters for simulation.

Parameters	Value
Grid	110 V (line-line) RMS, 50 Hz
Interfacing inductor (shunt VSC)	4 mH
DC link capacitor	4.4 mF
DC link voltage	200 V
Non-linear load	(i) 850 W, 100 VAR (ii) 2400 W, 510 VAR (iii) 1000 W, 200 VAR
SPV voltage and current at MPP	203 V, 7.35 A
SPV power at MPP	1.49 kW

. The analysis is carried out under different cases of steady state, varying load, and varying irradiance.

4.1. Operation of the Proposed Work under Steady-State Condition

The performance of grid integrated PV-DSTATCOM with the proposed control scheme is studied under steady state condition with constant demand of 850 W, 100 VAR of three phase nonlinear load. The PV system is working at irradiance of 500 W/m² as in Figure 4a. Here, the load demand is within the maximum switch current limit of DSTAT-COM. So, it can be seen in the Figure 4b that the DSTATCOM performs the multiple operation of reactive power compensation, regulation of DC-link voltage and eliminating harmonics from the grid current. The real power demand of load is fulfilled both by grid and PV as in Figure 4c. Additionally, DSTATCOM manages the non-real power demand of load, as in Figure 4d. Therefore, in Figure 4e it can be seen that the grid voltage and current are in phase and unity power factor operation is obtained at P.C.C.

4.2. Operation of the Proposed Work under Varying Solar Irradiance Condition

The performance of grid integrated PV-DSTATCOM with the proposed control scheme is studied under changing irradiance condition from 500 W/m² to 1000 W/m². In Figure 5a, it is observed that with the change in irradiance there is negligible change in PV voltage, whereas PV current is increased and the power contributed by PV array is also increased. When the PV current increases then the current through the DSTATCOM also increases as PV is interfaced to load and grid through the DSTATCOM (shunt compensator). When the power generated by PV is greater than the active power of the load demand. The extra amount of active power from PV is fed to the grid and the direction of grid current is reversed. It can be seen in Figure 5b,c.

During the sizing of DSTATCOM, as in (9), the active current handling capacity of the switches includes the maximum value of PV current. The control of DSTATCOM therefore focuses on the power quality improvement task. The switch current limit of DSTATCOM is decided by considering the maximum PV current and maximum non-real power demand of the load. So, at t = 0.25 s, when there is increase in PV current due to an increment in irradiance, the system operation is not effected and DSTATCOM performs its function, as shown in Figure 5d,e.

4.3. Operation of the Proposed Work by Varying the Load Demand from Rated Value to Very High Value Such That the Switch Current Is beyond the Limit at Solar Irradiance of 500 W/m²

In Figure 6, from period t = 0 s to t = 0.25 s the load demand is 1019 VA (1000 W, 200 VAR) where the reactive power demand is less than designed reactive power handling capacity of DSTATCOM i.e., 300 VAR. Therefore, under this condition the controller is operating for power quality improvement and power management of the system.

However, at t = 0.25 s, there is sudden increase in load demand of value 2453 VA (2400 W, 510 VAR) where the reactive power requirement of the load is very high as compared to the rated capacity of the DSTATCOM, which will eventually lead high current to flow through the switches, hence damaging it. So, in order to prevent the switches from being damaged, the switch current limiting part of the controller will come into action. Now, the extra amount of non–real power, which is above the rated value of the DSTATCOM, is provided by the grid. Though the real power demand of the load is also high, this will not affect the VSI switch. This is because the share of the real power demand that is contributed by PV is within the real power handling capacity of the switch, and the maximum power produced by PV is taken into consideration, while designing DSTATCOM. The rest of the real power requirement of the load is provided by the grid. Therefore, under this transient condition, the total reactive power demand of the load is managed by the grid and DSTATCOM and the total real power demand is fulfilled by PV and the grid, as shown in Figure 7 and Figure 8, respectively.

In Figure 9, it can be observed that before t = 0.25 s, the current through the DSTATCOM is within the safe limit, and as a result unity power factor operation is obtained. However, from t = 0.25 s, when the current is beyond the safety limit, then the priority shifts towards preventing DSTATCOM from damage instead of power quality improvement. Therefore, in this high current period, the grid supports DSTATCOM by feeding an extra amount of non–linear current to load. Eventually, the grid current and grid voltage deviates from their in–phase condition.

5. Real Time Simulation Results and Discussion

In addition to the simulation study, the proposed work is also implemented in the Opal–RT platform for real time simulation validation, as shown in Figure 10. The real time simulation is conducted for steady state, load change, and irradiance change condition, to substantiate the simulation result obtained in the MATLAB/Simulink.

5.1. Under Steady-State Condition

Under constant load demand of 850 W, 100 VAR of three phase nonlinear load and solar irradiance of 500 W/m², the system is studied for real time simulation. The load considered is less than the rated value (1000 W and 440 VAR). Since the current injected by the shunt compensator is within the reactive current handling capacity of the switches, the normal operation of power quality improvement is performed. The load current, grid current, and shunt for all three phases is shown in Figure 11a, Figure 11b, and Figure 11c, respectively, for the steady state condition. In Figure 11d, the phase relation between the grid current and the grid voltage under the unity power factor operation of the system is depicted. Additionally, the regulated DC-link voltage at 200 V is presented in Figure 11d.

5.2. Under Solar Irradiance Change from 500 W/m² to 1000 W/m² and Rated Load Demand

The irradiance of the system is increased from 500 W/m² to 1000 W/m² at a constant load such that at irradiance of 500 W/m², PV generated power is less than the active power demand of the load, and at 1000 W/m², the PV generated power is greater than the load. It can be seen in Figure 12a, that the current injected by the shunt compensator is increased as PV current flows to load and the grid through DSTATCOM. So, DSTATCOM carries both real and non-real current. Here, the current through the DSTATCOM is within its rated capacity as the maximum PV power is taken into account while the sizing of DSTATCOM. Therefore, the overall objective of the system under this condition is power quality improvement (the elimination of grid current harmonics, reactive power compensation and DC-link voltage regulation) as the current is within its rated limit, as can be seen in Figure 12a. When PV power more than the load demands then the extra amount of real power is given to the grid, as can be seen in Figure 12b; the direction of the grid current is reversed as it is consuming power.

5.3. Under Load Change from Rated Value (1000 W, 200 VAR) to Higher Value (2400 W, 510 VAR)

When the proposed system is subjected to load change to a value higher than the rated value, then the VSI switch current limiting scheme is activated automatically by the controller. This happens in order to protect the DSTATCOM switches from encountering permanent damage. This operation of the proposed controller is validated by real time simulation, as shown in Figure 13a. When applied load (1000 W, 200 VAR) is less than or equal to the rated load (1000 W, 440 VAR), the power is within the power handling capacity of DSTATCOM (440 VAR), so the controller performs the operation of power quality improvement. Whereas, during applied load greater (2400 W, 510 VAR) than the rated load, which is more than the current handling capacity of DSTATCOM. Therefore, to protect the switches, the control mode shifts from power quality improvement to switch current limiting mode. Here, the amount of reactive power provided by DSTATCOM is restricted and the extra amount of reactive power is provided by the grid compromising the unity power factor at P.C.C, as validated in Figure 13b.

6. Conclusions

An efficient strategy for PV–DSTATCOM control with a reduced number of sensors along with a switch current limiting scheme is analyzed in the proposed study. Here, three sensors have been reduced with a reduction of USD 135.44 from the sensor expenses without vitiating with the effectiveness of the controller. The introduction of the switch current limiting scheme has also protected the system from irreversible damage during transient and high current condition, even under dynamic condition of varying load (load more than the rated load of 1000 W and 440 VAR) and solar irradiance (500 W/m² to 1000 W/m²). The current limiting factor is efficiently synchronized with the controller operation such that it successfully switches from power quality enhancement mode to switch protection mode untill the transient persists. Whenever the reactive power demand of the load goes beyond the value of 440 VAR, the current limiting scheme operates to protect the DSTATCOM. Henceforth, there is no need to take unnecessarily higher rating switches, so, subsequently, the rating of the DSTATCOM is reduced leading to the reduced cost of the overall system. The system is able to economically maintain the power quality of the system along with power management with reduced sensors and a switch current limiting strategy. Furthermore, the real time simulation of the proposed work is carried out using the Opal–RT platform. The results obtained under steady state, as well as dynamic condition, validate the simulation results. The proposed work can also be studied and investigated by implementing it to an experimental prototype. Additionally, this work can be implemented to UPQC to obtain the multiple benefits of power quality improvement of voltage, as well as current, along with the UPQC switch protection scheme.