**6. Simulation Results**

 <sup>∗</sup> ൌ ሺ ሻ (13) where V<sup>୫</sup> is the maximum fixed reference rated voltage. The desired load voltage (V ∗ ) was equated with the measured load voltage (V) to produce the control signals for the series APF. **6. Simulation Results** This new sharing algorithm‐based UPC system was implemented in SIMULINK for an analysis considering the various operational conditions of the source and load. The three different types of disturbances in the source voltage were sag, swell and harmonics. Correspondingly, on the load side, the performance of the UPC was analyzed with three various load situations; i.e., with a non‐linear load only, a linear load only and a compo‐ This new sharing algorithm-based UPC system was implemented in SIMULINK for an analysis considering the various operational conditions of the source and load. The three different types of disturbances in the source voltage were sag, swell and harmonics. Correspondingly, on the load side, the performance of the UPC was analyzed with three various load situations; i.e., with a non-linear load only, a linear load only and a composite type (both linear and non-linear). The parameters considered for the simulation analysis are depicted in Table 1. With source voltage disturbances being common phenomena, the simulation results were analyzed under different categories of loading conditions. The simulation time was separated into various time divisions as per the source voltage disturbance. From 0 to 0.3 s, the source voltage was in a steady state rated condition; from 0.3 to 0.6 s, a voltage sag of 20% was introduced; from 0.6 to 0.9 s, a voltage swell of 20% was incorporated along with the rated source voltage; and from 0.9 to 1.2 s, the 3rd, 5th and 7th harmonic voltage component (10% each) was injected into the steady state voltage.


site type (both linear and non‐linear). The parameters considered for the simulation analysis are depicted in Table 1. With source voltage disturbances being common phe‐ nomena, the simulation results were analyzed under different categories of loading con‐ ditions. The simulation time was separated into various time divisions as per the source voltage disturbance. From 0 to 0.3 s, the source voltage was in a steady state rated condi‐ tion; from 0.3 to 0.6 s, a voltage sag of 20% was introduced; from 0.6 to 0.9 s, a voltage swell of 20% was incorporated along with the rated source voltage; and from 0.9 to 1.2 s, the 3rd, 5th and 7th harmonic voltage component (10% each) was injected into the steady

**Table 1.** Parameters for simulation and real-time analysis. Source impedance R = 0.06 Ω, L = 0.05 mH

**Table 1.** Parameters for simulation and real‐time analysis.

state voltage.

*Energies* **2022**, *15*, x FOR PEER REVIEW 6 of 19

#### *6.1. Simulation Results with a Composite Load* In the presence of both a linear and non‐linear load, the THD content of the source

In the presence of both a linear and non-linear load, the THD content of the source current without compensation was less than that obtained with the non-linear load alone, but remained on the higher side. With a constant maximum linear load of 5 kW and 5 kVAr, the series and shunt APFs took part equally in reactive power compensation, along with their respective voltage disturbance and current harmonic compensation. Figures 4 and 5 depict the different voltage and current waveforms under this load condition, respectively. Figure 4a illustrates the voltage waveforms for the complete duration of 1.2 s. Figure 4b–d depicts enlarged fragments of the voltage waveform observing the sag, swell and harmonic condition, respectively. It was observed that, under all scenarios, the load voltage was maintained as the same as the source voltage before any disturbance; i.e., before 0.3 s. Likewise, Figure 5b–d depicts enlarged fragments observing the sag, swell and harmonic condition, respectively, of the total current waveform shown in Figure 5a. The DC-link voltage profile was also maintained at its set reference value, as can be observed in Figure 6. Although a DC-link voltage ripple was observable, the UPC performance was not affected and was satisfactory with the considered operating conditions. current without compensation was less than that obtained with the non‐linear load alone, but remained on the higher side. With a constant maximum linear load of 5 kW and 5 kVAr, the series and shunt APFs took part equally in reactive power compensation, along with their respective voltage disturbance and current harmonic compensation. Figures 4 and 5 depict the different voltage and current waveforms under this load condition, re‐ spectively. Figure 4a illustrates the voltage waveforms for the complete duration of 1.2 s. Figure 4b–d depicts enlarged fragments of the voltage waveform observing the sag, swell and harmonic condition, respectively. It was observed that, under all scenarios, the load voltage was maintained as the same as the source voltage before any disturbance; i.e., before 0.3 s. Likewise, Figure 5b–d depicts enlarged fragments observing the sag, swell and harmonic condition, respectively, of the total current waveform shown in Figure 5a. The DC‐link voltage profile was also maintained at its set reference value, as can be ob‐ served in Figure 6. Although a DC‐link voltage ripple was observable, the UPC perfor‐ mance was not affected and was satisfactory with the considered operating conditions.

**Figure 4.** Simulation results with composite load (linear and non‐linear): source voltage, load voltage and series injected voltage (top to bottom) at (**a**) 0–1.2 s, (**b**) 0.2–0.4 s, (**c**) 0.5–0.7 s and (**d**) **Figure 4.** Simulation results with composite load (linear and non-linear): source voltage, load voltage and series injected voltage (top to bottom) at (**a**) 0–1.2 s, (**b**) 0.2–0.4 s, (**c**) 0.5–0.7 s and (**d**) 0.8–1.0 s.

(**a**)

0.8–1.0 s.

0.8–1.0 s.

*Energies* **2022**, *15*, x FOR PEER REVIEW 7 of 19

(**b**)

(**c**)

(**d**)

**Figure 4.** Simulation results with composite load (linear and non‐linear): source voltage, load voltage and series injected voltage (top to bottom) at (**a**) 0–1.2 s, (**b**) 0.2–0.4 s, (**c**) 0.5–0.7 s and (**d**)

**Figure 5.** Simulation results with composite load (linear and non‐linear): source current before compensation, source current after compensation and injection current from shunt APF (top to bottom) at (**a**) 0–1.2 s, (**b**) 0.2–0.4 s, (**c**) 0.5–0.7 s and (**d**) 0.8–1.0 s. **Figure 5.** Simulation results with composite load (linear and non-linear): source current before compensation, source current after compensation and injection current from shunt APF (top to bottom) at (**a**) 0–1.2 s, (**b**) 0.2–0.4 s, (**c**) 0.5–0.7 s and (**d**) 0.8–1.0 s.

**Figure 6.** Simulation result with composite load (linear and non‐linear): DC‐link voltage.

(**d**)

bottom) at (**a**) 0–1.2 s, (**b**) 0.2–0.4 s, (**c**) 0.5–0.7 s and (**d**) 0.8–1.0 s.

**Figure 5.** Simulation results with composite load (linear and non‐linear): source current before compensation, source current after compensation and injection current from shunt APF (top to

*Energies* **2022**, *15*, x FOR PEER REVIEW 8 of 19

(**b**)

(**c**)

**Figure 6.** Simulation result with composite load (linear and non‐linear): DC‐link voltage. **Figure 6.** Simulation result with composite load (linear and non-linear): DC-link voltage.

### *6.2. Simulation Results with Linear Load 6.2. Simulation Results with Linear Load*

In the absence of any non-linear load, the source current without compensation was free from harmonics and did not require any harmonic compensation from the shunt APF. Figure 7 illustrates the response of the UPC system with the proposed controller for a constant linear load of 5 kW and 5 kVAR. Under varying source voltage disturbances, the series injected voltage made adjustments to maintain a steady state and rated load voltage profile. The compensating current from the shunt APF was only responsible for partly compensating the VAR demand whereas an equal amount of reactive power was compensated for by the series APF. In the absence of any non‐linear load, the source current without compensation was free from harmonics and did not require any harmonic compensation from the shunt APF. Figure 7 illustrates the response of the UPC system with the proposed controller for a constant linear load of 5 kW and 5 kVAR. Under varying source voltage disturbances, the series injected voltage made adjustments to maintain a steady state and rated load voltage profile. The compensating current from the shunt APF was only responsible for partly compensating the VAR demand whereas an equal amount of reactive power was compensated for by the series APF.

**Figure 7.** Simulation results with linear load only: source current before compensation, source current after compensation and injection current from shunt APF (top to bottom). **Figure 7.** Simulation results with linear load only: source current before compensation, source current after compensation and injection current from shunt APF (top to bottom).

#### *6.3. Simulation Results with a Non‐Linear Load 6.3. Simulation Results with a Non-Linear Load*

With the inclusion of only a non‐linear load, the UPC system operation was confined to the compensation of source voltage disturbances and current harmonic compensation. As is clear from Figure 8, the source current profile was free from harmonics, with the THD content being reduced to 3.4% from 23.6%. Figure 8a depicts the time duration of 1.2 s whereas Figure 8b is an enlarged fragment for the time duration of 0.3–0.6 s for a better illustration of the effect of the non‐linear load on the current waveform. It was evident that the source current before compensation was highly contaminated with harmonics whereas after compensation it was closer to sinusoidal. The THD analysis of the source current is discussed in detail further on in this work. With a proper compen‐ sating series injected voltage by the series APF and a compensating current by the shunt APF, the load voltage profile was maintained at its rated condition, as illustrated in Fig‐ ure 9. With the inclusion of only a non-linear load, the UPC system operation was confined to the compensation of source voltage disturbances and current harmonic compensation. As is clear from Figure 8, the source current profile was free from harmonics, with the THD content being reduced to 3.4% from 23.6%. Figure 8a depicts the time duration of 1.2 s whereas Figure 8b is an enlarged fragment for the time duration of 0.3–0.6 s for a better illustration of the effect of the non-linear load on the current waveform. It was evident that the source current before compensation was highly contaminated with harmonics whereas after compensation it was closer to sinusoidal. The THD analysis of the source current is discussed in detail further on in this work. With a proper compensating series injected voltage by the series APF and a compensating current by the shunt APF, the load voltage profile was maintained at its rated condition, as illustrated in Figure 9.

(**a**)

*Energies* **2022**, *15*, x FOR PEER REVIEW 9 of 19

In the absence of any non‐linear load, the source current without compensation was free from harmonics and did not require any harmonic compensation from the shunt APF. Figure 7 illustrates the response of the UPC system with the proposed controller for a constant linear load of 5 kW and 5 kVAR. Under varying source voltage disturbances, the series injected voltage made adjustments to maintain a steady state and rated load voltage profile. The compensating current from the shunt APF was only responsible for partly compensating the VAR demand whereas an equal amount of reactive power was

**Figure 7.** Simulation results with linear load only: source current before compensation, source

With the inclusion of only a non‐linear load, the UPC system operation was confined to the compensation of source voltage disturbances and current harmonic compensation. As is clear from Figure 8, the source current profile was free from harmonics, with the THD content being reduced to 3.4% from 23.6%. Figure 8a depicts the time duration of 1.2 s whereas Figure 8b is an enlarged fragment for the time duration of 0.3–0.6 s for a better illustration of the effect of the non‐linear load on the current waveform. It was evident that the source current before compensation was highly contaminated with harmonics whereas after compensation it was closer to sinusoidal. The THD analysis of the source current is discussed in detail further on in this work. With a proper compen‐ sating series injected voltage by the series APF and a compensating current by the shunt APF, the load voltage profile was maintained at its rated condition, as illustrated in Fig‐

current after compensation and injection current from shunt APF (top to bottom).

*6.2. Simulation Results with Linear Load*

compensated for by the series APF.

*6.3. Simulation Results with a Non‐Linear Load*

ure 9.

**Figure 8.** Simulation results with non‐linear load only: source current before compensation, source current after compensation and injection current from shunt APF (top to bottom) at (**a**) 0–1.2 s and (**b**) 0.3–0.6 s. **Figure 8.** Simulation results with non-linear load only: source current before compensation, source current after compensation and injection current from shunt APF (top to bottom) at (**a**) 0–1.2 s and (**b**) 0.3–0.6 s. **Figure 8.** Simulation results with non‐linear load only: source current before compensation, source current after compensation and injection current from shunt APF (top to bottom) at (**a**) 0–1.2 s and (**b**) 0.3–0.6 s.

**Figure 9.** Simulation results with non‐linear load only: source voltage, load voltage and series in‐ jected voltage (top to bottom). **Figure 9.** Simulation results with non‐linear load only: source voltage, load voltage and series in‐ jected voltage (top to bottom). **Figure 9.** Simulation results with non-linear load only: source voltage, load voltage and series injected voltage (top to bottom).

#### *6.4. Cumulative UPC Performance Parameters under Different Voltages and Loading Conditions 6.4. Cumulative UPC Performance Parameters under Different Voltages and Loading Conditions 6.4. Cumulative UPC Performance Parameters under Different Voltages and Loading Conditions*

Table 2 illustrates the various performance parameters used to obtain a better un‐ derstanding of the UPC performance analysis under different voltage and loading con‐ ditions. The performance parameters listed in the table are: I: the RMS value of the load voltage in volts (without a UPC); II: the RMS value of the load voltage (with a UPC); III: the THD% of the load voltage (without a UPC); IV: the THD% of the load voltage (with a UPC); V: the THD% of the source current (without a UPC); VI: the THD% of the source current (with a UPC); and VII: the reactive power share between the shunt APF (QSH) and the series APF (QSR) in VAR (volt‐ampere reactive). As observed from the table, the load voltage RMS value was maintained close to the reference RMS value of 230 V under dif‐ ferent voltage and loading conditions. The THD% of the load voltage with the harmonics under consideration was within the limits of 5% with an appropriate compensation from the UPC in all cases. As can be seen from the table, the THD% of the source current Table 2 illustrates the various performance parameters used to obtain a better un‐ derstanding of the UPC performance analysis under different voltage and loading con‐ ditions. The performance parameters listed in the table are: I: the RMS value of the load voltage in volts (without a UPC); II: the RMS value of the load voltage (with a UPC); III: the THD% of the load voltage (without a UPC); IV: the THD% of the load voltage (with a UPC); V: the THD% of the source current (without a UPC); VI: the THD% of the source current (with a UPC); and VII: the reactive power share between the shunt APF (QSH) and the series APF (QSR) in VAR (volt‐ampere reactive). As observed from the table, the load voltage RMS value was maintained close to the reference RMS value of 230 V under dif‐ ferent voltage and loading conditions. The THD% of the load voltage with the harmonics under consideration was within the limits of 5% with an appropriate compensation from Table 2 illustrates the various performance parameters used to obtain a better understanding of the UPC performance analysis under different voltage and loading conditions. The performance parameters listed in the table are: I: the RMS value of the load voltage in volts (without a UPC); II: the RMS value of the load voltage (with a UPC); III: the THD% of the load voltage (without a UPC); IV: the THD% of the load voltage (with a UPC); V: the THD% of the source current (without a UPC); VI: the THD% of the source current (with a UPC); and VII: the reactive power share between the shunt APF (QSH) and the series APF (QSR) in VAR (volt-ampere reactive). As observed from the table, the load voltage RMS value was maintained close to the reference RMS value of 230 V under different voltage and loading conditions. The THD% of the load voltage with the harmonics under consideration was within the limits of 5% with an appropriate compensation from the UPC in all cases. As

without the UPC, considering both a linear and non‐linear load (composite load), was in the range of 9–10% whereas for a non‐linear load, it was in the range of 16–20%. How‐

the UPC in all cases. As can be seen from the table, the THD% of the source current without the UPC, considering both a linear and non‐linear load (composite load), was in the range of 9–10% whereas for a non‐linear load, it was in the range of 16–20%. How‐

for the composite load and linear load cases. With the composite load, the total load re‐ active power demand was around 6 kVAR and equal sharing was observed by the shunt and series APF. With only a linear load, the load reactive power demand was 5 kVAR

limit. Thereafter, reactive power compensation sharing by the UPC APFs were observed for the composite load and linear load cases. With the composite load, the total load re‐ active power demand was around 6 kVAR and equal sharing was observed by the shunt and series APF. With only a linear load, the load reactive power demand was 5 kVAR

and, consequently, there was equal sharing between the shunt and series APF.

and, consequently, there was equal sharing between the shunt and series APF.

can be seen from the table, the THD% of the source current without the UPC, considering both a linear and non-linear load (composite load), was in the range of 9–10% whereas for a non-linear load, it was in the range of 16–20%. However, with the UPC, the THD% of the source current was brought back to within a 5% limit. Thereafter, reactive power compensation sharing by the UPC APFs were observed for the composite load and linear load cases. With the composite load, the total load reactive power demand was around 6 kVAR and equal sharing was observed by the shunt and series APF. With only a linear load, the load reactive power demand was 5 kVAR and, consequently, there was equal sharing between the shunt and series APF.
