Sage Journals: Discover world-class research

Abstract

The incorporation of green energy sources into the network, along with the introduction of power electrical devices to regulate loads that are not linear, has significantly influenced power quality (PQ). Nevertheless, PQ is the primary challenge when integrating inconsistent power from renewable sources. Here, the reduced switch converter is selected for the Unified Power Quality Conditioner (UPQC) that is incorporated with the Solar PV Generation System (SPVGS), hydrogen Fuel cell (FC), and Battery Energy Storage Systems (BESS) to handle PQ problems effectively. This paper suggests a hybrid scheme that has the advantageous characteristics of a neural network controller (NNC), Proportional integral derivative controller (PIDC) and Fuzzy logic controller (FLC). NNC with Synchronous reference frame theory (dq) is adopted to generate the reference signals. Additionally, a meta-heuristic-based chimp algorithm (ChOA) is adapted to design the optimized FLC-PIDC hybrid controller for regulating the DC Link voltage (DLV), resistance, inductance, and capacitance variables of filters, proportional integral controller (PIC) gain values of FC and BESS. The primary goal of the suggested approach is to maintain a stable DLV despite fluctuations in both load and SPVGS. The system also seeks to reduce fluctuations in the load voltage and supply current, as well as to reduce sag, interruption, swell, and imbalances caused by variations in the source voltage. Testing the developed mode with four case studies each with a load like nonlinear, balanced, and unbalanced and grid voltage variations with variable irradiation was done to assess its performance. On the other hand, to validate the proposed technique the comparison analysis is conducted using conventional PIC. The proposed approach effectively decreases the total harmonic distortion (THD) to 2.30%, 2.63%, 2.79%, and 2.64%. These values are notably lower than those achieved by other methods in the literature.

Keywords

Unified power quality conditioner neural network controller PI controller fuzzy logic controller sliding mode controller

Introduction

There has been a growing push over the past decade for the association of clean sources such as tidal, wind, geothermal, and SPVGS into the distribution to lower the burden on the converters. The wind power is linked to DC bus of UPQC under several load circumstances and in the presence of faults. This investigation utilized both hysteresis and PWM approaches (DheyaaIed and Goksu, 2022). To reduce system THD, a distinctive modified UPQC has been created. This UPQC was designed to optimize MR variables by adapting the BHO and PSO methods to alter the PIC variables (Khosravi et al., 2023). Furthermore, a scheme was suggested to tie the AC/DC microgrid to the main power grid through the use of power-compensating techniques in order to reduce the harmonic magnitudes (Khosravi et al., 2021).

Next, the distribution network underwent a power flow analysis of the UPQC, taking into account multiple fault scenarios. The implementation of impedance-matching techniques was the primary focus (Xiaojun et al., 2021). Besides, to effectively handle PQ concerns, the solar-fed UPQC was proposed. To further ensure the effective extraction of the maximum power and the balancing of the DLV, a unique fuzzy PIC was created to optimize the MPPT process (Dongsheng et al., 2019). However, in order to simplify the transformations, the ANNC was chosen to generate the reference signal for the UPQC and to address PQ issues (Koganti et al., 2023). A solution utilizing a self-tuning filter was devised to handle PQ difficulties in the context of a UPQC connected with renewable sources (Mansoor et al., 2020). Besides, the extensively investigates was proven algorithms for controlling the functioning of SHAF, approaches for isolating harmonics, regulating DLV, and methods for controlling current (Yap et al., 2017). A sophisticated fuzzy-tuned PIC was created to optimize hybrid SHAF systems, with the primary objective of effectively minimizing THD in the current. An evaluation was carried out to measure its efficiency under various load circumstances, utilizing Clarke's transformation (Tzu-Chiao and Bawoke, 2022). Meanwhile, to suppress harmonics, such as sags and swells, the PV with NNC-based UPQC was introduced. In addition, this suggested technique underwent a comparison examination with the SRF theory, taking into account various load situations (Okech et al., 2022).

To effectively eliminate voltage and current aberrations, a microgrid-connected multilevel DSTATCOM was designed (Krishna et al., 2020). Three control loops were optimized, including the system controller to reduce harmonics extend to a level that is within reasonable bounds (Nima et al., 2022). Additionally, a DLV feed forward NNC has been proposed to control and manage PV systems in conjunction with wind power, which is linked to the UPQC (Kumar et al., 2021). A novel method of compensating THD was developed for a hybrid micro-grid to lessen the harmonics intensity (Nima et al., 2021). In addition, a thorough investigation was conducted on phase synchronization techniques employed to manage the efficiency of SHPF (Yap et al., 2019). The energy industry is particularly interested in the environmentally sustainable attributes of Green Energy sources, as well as their ability to enhance energy security, minimize line losses, and minimize operational expenses. Consequently, multiple ongoing research has considered the incorporation of renewable energy in different sectors, together with electric vehicles and Microgrids (Choudhury, 2020). Conversely, electronic equipment generates a non-sinusoidal current which results in the presence of harmonics in the system. These harmonics can result in significant issues within the power system, such as power and economic losses (Wang and Blaabjerg, 2019). In addition, researchers are currently engaged in efforts to mitigate harmonics and boost PQ (Balasubramanian et al., 2019). A novel series of filters was created to decrease the expenses associated with traditional shunt filters used for low-capacity industrial loads (Labdai et al., 2019). The PIC was employed to ensure the stability of the DLV within the framework of a SHAF, effectively addressing PQ concerns by utilizing hysteresis current regulation for pulse generation. In addition, the system's performance was evaluated while being affected by both linear and nonlinear loads (Amir et al., 2020).

An innovative approach was suggested to reduce the harmonics in the distribution network. This involves designing a hybrid filter that incorporates both passive and active filters, as described in reference (Fabricio et al., 2018). In addition, the Soccer match optimization method was selected to successfully solve PQ difficulties and to determine the NNC optimized bias and weights for renewable sourced UPQC (Koganti et al., 2022). The FLC was created specifically for the SEAF in distribution networks to reduce current and voltage issues related to PQ (Pazhanimuthu and Ramesh, 2018). A grid-connected solar system was designed, consisting of a PV system, DC to DC converters, a battery, a three-phase inverter, power electronics devices, and loads (Bagi et al., 2020).

The hybridization of the meta-heuristic optimization methods was employed to address the PQ concerns by accurately determining the gain values of PIC (Rajesh et al., 2025). For the UPQC, a hybrid control strategy that combines the benefits of NNC and FLC was suggested. This technique aimed to maintain dynamic load balance in the DLV while concurrently minimizing defects and distortions in source current and grid voltage (Srilakshmi et al., 2022). To optimize the PIC's gain setting selection in the UPQC, an optimization technique known as Soccer-league was devised to effectively address PQ issues (Koganti et al., 2022). The 9-level VSC used in the UPQC coupled to solar systems was modified to achieve distortion-free voltage waveforms (Hassan et al., 2022). UPQC used the LMBP-trained NNC controller to successfully address the grid voltage and current concerns, yielding positive outcomes (Zhou et al., 2006). FOADRC was employed for UPQC with Innovative WOA to obtain optimal gain values to solve PQ scenarios (Ali et al. 2025). A TL-UPQC was presented to address PQ issues like dip, flickering in the network and a shunt compensator reduces THD. The PIC gain values are obtained using an EBES optimizer (Abdel et al., 2023). The UPQC was developed with WOA FOPIC tuning to decrease the THD and enhancing the PQ (Mahmoud et al., 2023).

However, the available research indicates that the majority of the works mainly focused only on the selected objectives but neglected consideration of reduced switch converters with the optimal design of hybrid controllers with metaheuritic algorithms. Here, the developed approach for multiobjectives, several load combinations, irradiation from the sun, and balanced as well as unbalanced grid voltage conditions. The primary contributions of the proposed work are:

Design of ChOA optimal selected hybrid FLC-PIDC in order to regulate DLV, gain values of PI controller of FC and BESS, resistance, inductance, and capacitance values of filters.

Implementation of reduced 10 switch converter for the UPQC to minimize the converter losses.

Development of NNC in association with SRF theory for appropriate reference signal generation.

Integration of SPVGS, fuel cell with energy storage to the UPQC to reduce the requirement of large ratings of converters in addition to satisfy the power demand.

The prime objectives are to reduce the THD of the source current, load voltage, DLV regulation during load variation, and compensate for voltage interruptions, sags, swells, and disturbances in addition to effective handling of unbalances in current and voltages.

Performance analysis of included four different test studies to demonstrate its superiority compared with PIC, NNC as well as other controllers mentioned in the literature survey.

The second section gives a developed configuration, the third section focuses on the FLC and NNC systems in the shunt controller, the fourth section introduces the series controller, and lastly, the fifth section gives an the results with discussions. Lastly, the sixth section concludes the study.

Proposed configuration

Figure 1 shows the proposed hybrid UPQC framework and Figure 2(a) represents the block diagram of the developed system. While Figure 2(b) shows the structure of the planned UPQC. Boost and Buck-Boost converters are used to create the DC Link, which connects the SPVGS and BESS. Table 1 shows the performance comparison between the proposed method and literature survey. By supplying compensating voltage (CV), SEAF aims to reduce voltage swings. Additionally, an isolating transformer is used to create isolation between the power line and the VSC. The main role of SHAF is to suppress the defects in the current signal by providing the compensatory current (CC) and quickly managing the DLV constant.

Figure 1.

Proposed hybrid UPQC framework.

Figure 2.

Configuration of reduced switch UPQC with CHOA control circuits.

Table 1.

Performance comparison of the proposed method versus literature survey.

Ref [no]	Controller adapted	Grid voltage issues mitigation	Current harmonics reduction	DC Bus voltage control	Power factor enhancement	Renewable source integration	Reduced switch based voltage source converters	Meta-heuristic algorithm based optimized controller
Khosravi et al. 2023	PI-MR	✓	✓					✓
Dongsheng et al., 2019	Fuzzy-PIC	✓	✓			✓
Koganti al., 2023	NNC	✓	✓		✓	✓
Mansoor et al., 2020	PI	✓	✓	✓		✓
Okech et al., 2022	NNC	✓	✓	✓		✓
Krishna et al., 2020	PI	✓	✓			✓
Kumar et al., 2021	NNC		✓		✓
Labdai et al., 2019	PI		✓			✓		✓
Amir et al., 2020	PI		✓	✓
Proposed	ChOA based Fuzzy-PIDC	✓	✓	✓	✓	✓	✓	✓

This article proposes a reduced switching configuration for non-linear systems in order to improve performance and, in turn, reduce the overall number of switches. It is recognized for using only ten switches and maintains all of the 12-switch's execution advantages while reducing its underutilization without raising the switch's VA rating. As shown in Table 2, the suggested configuration is accepted by combining the switches of leg C of the filters VSI separately into a single leg that uses a common setup of two switches. To achieve the steady action of the suggested structure in different operational situations, an appropriate control calculation is generated. The lower switches work in tandem with the series compensator section in Figure 2, while the upper portion of the switches is a part of the shunt filter. By combining the phase C switches of the shunt and series VSI [SC1, SC2] and [SC’1 SC’2] into a single leg with a common arrangement of C1 and C2 switches, the recommended Topology is recognized.

Table 2.

Switching for C phase VSI in reduced switches.

Voltage (If)	Switching table (then)
Voltage across shunt and voltage across series converters is equal to V_dc	S_c1 gets ON and S_c2 will OFF
Voltage across shunt and voltage across series converters is equal to Zero	S_c2 gets ON and S_c1 gets OFF
Voltage across shunt converters is equal to V_dc and voltage across series is equal to zero	Not possible
Voltage across shunt converters is equal to is equal to zero and voltage across series is equal to V_dc	Not possible

External support for DCL

For the UPQC, a solar/battery-powered DC link is recommended. The structure is made up of a hybrid energy source system that adjusts DLV in accordance with variations in power demand by combining solar and battery technologies. By lessening the burden on the utility, external help can be used to lower converter ratings and stress.

SPVGS

PV cells are joined in series to form a string, which is then connected in parallel with other strings to generate the desired power. Here, incremental conductance method is adopted for MPPT to extract maximum output and PV system with control circuit is illustrated in Figure 2(b).

P_{P V} = V_{P V} \times i_{P V}

(1)

BESS

The BESS assists in stabilizing the DLV network. To achieve the required power, cells in a battery can be connected in parallel or series. The Li-ion type battery was selected because of its benefits, including cheap maintenance costs and a gradual discharge rate. The SOCOB is analyzed by equation (2).

S O C O B = 80 (1 + \int i_{B S S} d t Q)

(2)

Table 3 displays the ratings for the battery, fuel cell, and solar systems used in this work. In Figure 2(b), the BESS control mechanism is emphasized.

Table 3.

Specifications of PV and BES.

Equipment	Factor	Considered value
PV panel Type: SPR-305E -WHT-D)	Short circuit current	5.96 A
	Max output power	305.226
	Voltage and current during max power	54.7 V/5.58 A
	Open circuit voltage	64.2 V
Li-ion battery	Capacity	35 Ah
Li-ion battery	Normal voltage	350
FC	Cells	65
	Resistance	0.70833 ohm
	Efficiency of stack	55%

Fuel cell

The energy conversion device that transforms chemical energy into electrical energy is called a fuel cell. Electricity is produced by the electrochemical process of hydrogen and oxygen. To generate high voltage, several FC's are joined to create a FC stack. With the aid of a DC-DC boost converter, the FC's voltage level is raised. The FC controller and boost converter are shown in Figure 2(b).

Shunt active power filter

The main objectives of SHAF are to bring stability to the DLV and reduce imperfections in the current signal by injecting CC. It carries out (i) SRF transformations; (ii) FLC-PIDC is designed to accomplish goals in association with NNC-based pulse generation. Since SRF theory conversions are already available in the research review, this work highlights the control system used for the recommended approaches, such as chimp optimization algorithm, ChOA optimized FLC, PIDC.

Chimp optimization algorithm

In 2020, Khishe and Mosavi presented the ChOA algorithm, one among of the latest nature-inspired algorithms. The mobility of chimps during collective hunting and its sexual motives provide as inspiration for CHOA. Prey hunting is used in the ChOA to solve the optimization problem in the best possible way. Driving, blocking, chasing, and attacking are the four primary stages of hunting, according to ChOA. In the first, a random chimp population is generated to initialize ChOA. After that, chimps are separated into four groups: driver, barrier, attacker, and chaser. Equation (1) has been proposed for demonstrate driving and chasing the prey.

\begin{aligned} d = | c . X_{p r e y} (t) - m . X_{c h i m p} (t) | \\ w h e r e, \\ X_{c h i m p} (t + 1) = X_{p r e y} (t) - a . d \\ a = 2. f . r_{1} - f \\ c = 2. r_{2} \\ m = C h a o t i c V e c t o r \end{aligned}

(3)

where t is the current iteration, a, c, and m are the coefficient vectors, f is the dynamic vector ɛ[0, 2.5], r1 and r2 are the random vectors ɛ[0, 1], m resembles the chaotic vector, X_prey represents vector of prey's position and X_chimp gives the vector of chimp's position. In the hunting process, the chimps use driver, blocker, and chaser chimps to first determine the location of the prey. Attackers carry out the hunting phase of the exploitation process. The attacker, barrier, chaser, and driver chimps assess the prey's position for this reason, and other chimps use the prey to update their positions. Equations (4)–(8) are used to formulate this process.

d_{a t t a c \ker} = | c_{1} . X_{a t t a c \ker} - m_{1} . X |, d_{B a r r i e r} = | c_{2} . X_{B a r r i e r} - m_{2} . X |

(4)

d_{c h a s e r} = | c_{3} . X_{c h a s e r} - m_{3} . X |, d_{D r i v e r} = | c_{4} . X_{D r i v e r} - m_{4} . X |

(5)

X_{1} = X_{a t t a c \ker} - a_{1} . (d_{a t t a c \ker}), X_{2} = X_{B a r r i e r} - a_{2} . (d_{B a r r i e r})

(6)

X_{3} = X_{c h a s e r} - a_{3} . (d_{c h a s e r}), X_{4} = X_{D r i v e r} - a_{4} . (d_{D r i v e r})

(7)

X (t + 1) = \frac{X 1 + X 2 + X 3 + X 4}{4}

(8)

where the best search agent is indicated by X_attacher, the second-best by X_barrier, the third-best by X_chaser, the fourth-best by X_driver, and the latest position of each chimp is shown by X(t + 1). Additionally, a parameter is applied to build exploration procedure so that chimps and prey diverge when a > 1 and a < −1 are present. Additionally, a variable with value in between +1 and −1 aids in the convergence of chimps and prey, which will boost exploitation. Furthermore, the c parameter facilitates the algorithm's exploration phase. Lastly, regardless of their responsibilities, all chimps assault their victim to obtain social privileges following prey hunting the social behavior is formulated by equation (9). The generalized CHOA working process is given in detail in the flow chart is illustrated in Figure 3.

X_{c h i m p} (t + 1) = {\frac{X p r e y (t) - a . d i f μ \leq 0.5}{C h a o t i c_v a l u e i f μ \geq 0.5}

(9)

CHOA optimized FLC-PIDC

The FLC produces a duty cycle as output that indicates the necessary change in current after receiving the error (E) and change in error (CE). The created fuzzy parameters are represented by triangular type of membership functions (MESF) represented by µ. A triangular MF is specified by three parameters {a, b, c} as follows:

t r i a n g l e (y; a, b, c) = {\begin{matrix} 0, y \leq a . \\ \frac{y - a}{b - a}, a \leq y \leq b . \\ \frac{c - y}{c - b}, b \leq y \leq c . \\ 0, c \leq y . \end{matrix}

(10)

Figure 3.

ChOA flow chart.

The variables {a, b, c} (with a < b < c) give the y coordinates of the corners of the triangular MF as given in Figure 4. This involves PB, medium positive (PM), PS, ZE, NB, medium negative (NM), and NS. The MESF are as shown in Figure 5.

Figure 4.

Triangular MSF.

Figure 5.

Fuzzy MESF for DLV balancing.

Through the use of pharos, the load current is transformed into a dq0, and a PLL uses the supply voltage to calculate the frequency. The generation of the proper reference signal and DLV control are necessary for SHAF to function. Furthermore, a variation in DLV may arise from an imbalance in power flow brought on by load fluctuations. For the DLV to be stabilized, the power in the SHAF needs to match the loss experienced during switch operations. The ChOA optimized FLC-PIDC introduces a direct current error signal, which is computed based on the discrepancy between the reference and actual DLV as determined by equation (11).

Δ i_{d c} = e_{1} (t) = V_{d c}^{r e f} - V_{d c} (t)

(11)

It is constructed for reducing THD of the problem selected with filter parameters, PI controller gain values of BESS and FC, in addition to PIDC gain variables while satisfying the upper and lower bonds.

M a x i m i z e F = \frac{1}{1 + {T H D}}

(12)

Here, THD is calculated by equation (13).

T H D = \frac{\sqrt{(I_{2}^{2} +} I_{3}^{2} + \dots . . I_{n}^{2})}{I_{1}}

(13)

The dth component of current at the load terminals is combined with the error acquired from the ChOA optimized FLC and PIDC. The dq0 is transformed into abc and given to as inputs of NNC with the compensating currents as target. Finally, using a hysterics controller is adapted for the pulse generation. The SHAF with suggested controller is illustrated in Figure 6.

Figure 6.

Fuzzy-PIDC and NNC-based control system for SAPF.

NNC with PWM for reference signal generation

In this study, the suggested NNC is adapted to generate signals for the shunt VSC. NNC is built by input, hidden, output layers (IL, HL, and OL) respectively. The IL sends the input data to the HL. The corresponding weights on the links are then multiplied when it is connected between the IL and the HL. In this context, computations are executed with a selected preference towards HL, and the results are collected in OL.

The NNC controller with LMBP training is chosen. During the training phase, the link weights are adjusted by examining the error to obtain the desired outcome. MSE is the performance function that is used in training. The LMBP method changes the weight by using the calculated derivatives, which leads to quick convergence and effective learning. Each neuron in a multilayer perceptron network consists of activation and summation functions. However, the neurons are connected by means of numerical weights. The MSE is computed using equation (14). The received output is denoted as O, while the intended output is denoted as $\bar{O}$ . The number of instances represented by the variables n, p, and m is neurons.

M S E = \frac{1}{n} \sum_{p = 1}^{m} {(O_{p} - {\bar{O}}_{p})}^{2}

(14)

Figure 7 depicts the reference abc current signals (i_{refsh_abc}) as the goal data, while the injected abc currents (i_{sh_abc}) are regarded as the input. The NNC, in conjunction with the hysteresis controller, is utilized to generate pulses in order to accomplish the desired goals.

Figure 7.

Structure of NNC for reference current and voltage generation.

Series controller

The controller generates signals by comparing the load voltage, which has been transformed into a dq frame. It is later converted to an abc frame, as seen in Figure 8. A Neural Network NNC connected to PWM generates the gating pulses. The main task of SAPF is to mitigate voltage imperfections originating in the grid side by injecting an appropriate CV to ensure a steady load voltage. Figure 8 illustrates the proposed technique for generating the VSC reference signal; it depicts the structure of the NNC with an HL of 200 neurons. The reference voltages are generated using the compensating voltages as input data and target data for the NNC.

Figure 8.

Control system for series filter.

Results analysis with discussions

The working of the UPQC with the designed control system was evaluated on a three-phase network. This method was created using Matlab version 2022b. The UPQC and load specifications are presented in Table 4. Four test studies were conducted to evaluate the operation of the proposed control system. These studies involved various combinations of nonloads, grid voltages, and conditions such as harmonics, swell, and sag. The tests were carried out under constant/ variable irradiation. The results of these studies demonstrated the superior performance of the suggested system. The grid voltage is considered to be balanced in cases 1, 2, and 3, but unbalanced for test study 4. On the other hand, the THD was assessed using equation (23) for all the scenarios and compared to the PIC, NNC, and SMC approaches, employing the traditional SRF theory. Additionally, the controllers mentioned in the literature were also included for THD comparison, as presented in Table 5.

Table 4.

UPQC ratings.

Grid supply	Voltage: 415 V; resistance: 0.1 ohm; inductance: 0.15 mH
SEAF	$R_{s e}$ : 1 ohm; $L_{s e}$ : 3.6 mH; $C_{s e}$ : 60 microfarad
SHAF	$R_{s h}$ : 0.0010 ohm; $L_{s h}$ : 2.15 mH; $C_{s h}$ : 1 µF
DC Link	$C_{d c}$ : 9400 microfarad; $V_{d c}^{r e f}$ = 700 V
Rectifier based nonlinear balanced Load (L-1)	R = 30 ohm and 20 mH
Rectifier based nonlinear balanced Load (L-2)	R = 50 ohm and 1 mH
Un-balanced RL branch Load (L-3)	R: 10, 20 and15 ohm; L: 11.50, 14.50 and 18.50 mH.
Active reactive power load	P = 10e3, Q = 1000 Vars

Table 5.

THD comparison.

Controller	Case 1	Case 2	Case 3	Case 4
Without UPQC	19.69	22.28	26.03	11.92
PIC	3.68	4.22	3.18	3.77
SMC	3.15	4.09	3.56	3.38
NNC	2.87	3.66	3.42	3.91
PIC (Pazhanimuthu et al., 2018)	3.880	–	–	–
PIC (Balasubramanian et al., 2019)	3.480	–	–	–
PIC (Amir et al., 2020)	14.740	–	–	–
ANFIS (Amir et al., 2020)	6.130	–	–	–
PIC (Koganti et al., 2022)	2.430	–	–	–
FLC (Koganti et al., 2022)	3.650	–	–	–
ChOA optimized FLC-PIDC	2.30	2.63	2.79	2.64

Case 1: Performance during source voltage sag and swell for the balanced supply with L-1 and L-2

In case 1, the balanced grid voltage with 30% of sag, and swell in the time intervals of 0.2–0.3 s, and 0.35–0.45 s was considered as shown in Figure 9(a). The devised approach detects voltage irregularities and applies the necessary corrective measures by using a coupling transformer to keep a stable voltage across load terminals. To assess the efficiency of SHAF, a combined three-phase balanced non-linear loads (L₁ + L₂) were selected for study with fixed solar irradiation. The load current waveform was determined to be non-sinusoidal yet balanced, as depicted in Figure 9(a). The suggested approach mitigates the disturbances in the waveform, which in turn reduces THD. Nevertheless, the THD of the current was reduced from 19.69% to 2.30%, which is lower than the THD values reported in the literature for other controllers such as PIC (between 3.8% and 14.7%), ANFIS (about 6.30%), and FL-C (about 3.65%). These comparisons are presented in Table 5. In addition, the DLV rapidly achieves a stable voltage of 700 V within a smaller time period of 0.056 s as shown in Figure 9(b).

Figure 9.

Case 1.

Case 2: Performance during short interruption and harmonic content for balanced supply voltage with sudden addition of load L-2 to L-1

In case 2, short interruption is considered for 0.5 to 0.6 s with voltage distortion from 0.65sec to 0.7sec respectively, as shown in Figure 10(a). Here, the developed control approach ensures a constant V_l. The load current of the non-linear load with balanced phases exhibited a significant increase in amplitude at 0.55 s due to addition of load 2 to load 1, resulting in a non-sinusoidal waveform. However, the load remained balanced, as shown in Figure 10(a). The suggested UPQC minimizes the THD from 22.28% to 2.63%. The effectiveness of the suggested method in addressing both voltage and current-related PQ problems is evident. Figure 10(b) illustrates the effective performance of the recommended controller in swiftly maintaining a constant DLV of 700 V within a short period of time of 0.1 s.

Figure 10.

Case 2.

Case 3: Performance during unbalanced sag /swell in supply with dynamic variation of solar irradiation and loads of L-4, L-3, and L-2

In example 3, the V_S was assumed to be in balanced till 0.2 due to L-4 and at 0.2 load 3 is added and L-1 is added to existing loads at 0.3 s due to which the measured load current exhibited non-sinusoidal and imbalanced characteristics, with significant levels of harmonic distortions, as depicted in Figure 11(a). The developed optimized hybrid model effectively reduces the THD from 26.03% to 2.79% and successfully maintains a balanced load voltage with steady DLV within short period of time after disturbance, even in the presence of load and irradiation variations as shown in Figure 11(b).

Figure 11.

U-SEBES for case 3.

Case 4: Performance during unbalanced supply with unbalanced load under variable irradiation condition

According to Figure 12(a), both the V_S and load in case 4 are determined to be unbalanced. The proposed UPQC rectifies voltage imbalances and ensures a consistent load voltage supply. The load current is seen to have a sinusoidal waveform, although it is unbalanced, as highlighted in Figure 12(b). The developed control method effectively reduces the defects in the current waveform and decreases the THD from 8.92% to 3.29%. Figure 12(a) demonstrates its ability to regulate DLV during load variation, achieving a response time of approximately 0.18 s when subjected to simultaneous changes in load, temperature, and irradiation. Figure 13 displays the frequency spectrum of the proposed system for all case studies.

Figure 12.

Case-4.

Figure 13.

THD spectrum.

Conclusion

The SPVGS and BESS connected UPQC was developed in with the CHOA optimized Fuzzy-PIDC along with PI controller gain values of FC and BESS, filter parameters of NNC hybrid control system for SHAF and SEAF for UPQC. The main goal of this control system is to successfully regulate the DLV during load variations, while also removing harmonics, interruption, swell, and sag in the source voltage. Additionally, the controllers aim to remove imperfections in current waveforms, thereby reducing THD, as well as improving the PF by lowering the THD. A comparative study with conventional PIC, SMC, NNC, and other controllers available in literature shows that the suggested controller performs better than the others. It is clear from the analysis of the performance of the proposed technique on four test cases that the system produced lesser THD values of 2.30% for case 1, 2.63% for case 2, 2.79% for case 3, and 2.64% for case 4. These findings clearly show that the suggested method was successful in bringing the THD down to manageable levels. Future study could perhaps focus on further examining the suggested system for Multi-level UPQC.

Footnotes

Abbreviations

ORCID iD

Praveen Kumar Balachandran

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

References

Abdel

Ibrahim

Elbarbary

, et al. (2023) Unified power quality conditioner using recent optimization technique: A case study in Cairo Airport Egypt. Sustainability 15(4): 3710.

Ali

Akram

Abdelhady

, et al. (2025) Power quality enhancement based disturbance rejection controller in microgrid system using walrus optimization algorithm and multi-functional recurrent fuzzy neural network. Ain Shams Engineering Journal 16(9): 103540.

Amir

Kumar

Yusuf

(2020) Modeling and simulation of a PI controlled shunt active power filter for power quality enhancement based on P-Q theory. Electronics 9(4): 637.

Bagi

Kudchi

Bagewadi

(2020) Power quality improvement using a shunt active power filter for grid connected photovoltaic generation system. IEEE Bangalore Humanitarian Technology Conference (B-HTC), Vijiyapur, India: 1–4.

Balasubramanian

Parkavikathirvelu

Sankaran

, et al. (2019) Design, simulation and hardware implementation of shunt hybrid compensator using synchronous rotating reference frame (SRRF)-based control technique. Electronics 8(1): 44.

Choudhury

(2020) A comprehensive review on issues, investigations, control and protection trends, technical challenges and future directions for microgrid technology. International Transaction Electrical Energy System 30: e12446.

DheyaaIed

Goksu

(2022) Design and control of three-phase power system with wind power using unified power quality conditioner. Energies 15(19): 7074.

Dongsheng

Zhanchao

Xiaoting

, et al. (2019) Control strategy of intergrated photovoltaic-UPQC system for DC-bus voltage stability and voltage sags compensation. Energies 12(20): 4009.

Fabricio

Junior

Jacobina

, et al. (2018) Analysis of main topologies of shunt active power filters applied to four-wire systems. IEEE Transactions on Power Electronics 33(3): 2100–2112.

10.

Hassan

Carlo

Issam

, et al. (2022) New modulation technique in smart grid interfaced multilevel UPQC-PV controlled via fuzzy logic controller. Electronics 11(6): 919.

11.

Khosravi

Abdolmohammadi

Bagheri

, et al. (2021) Improvement of harmonic conditions in the AC/DC microgrids with the presence of filter compensation modules. Renewable and Sustainable Energy Reviews 143: 110898.

12.

Khosravi

Echalih

Hekss

, et al. (2023) A new approach to enhance the operation of M-UPQC proportional-integral multiresonant controller based on the optimization methods for a stand-alone AC microgrid. IEEE Transactions on Power Electronics 38(3): 3765–3774.

13.

Koganti

Canavoy

Praveen

, et al. (2022b) Optimal design of an artificial intelligence controller for solar-battery integrated UPQC in three phase distribution networks. Sustainability 14(21): 13992.

14.

Koganti

Krishna

Kalyani

, et al. (2023) Design of UPQC with solar PV and battery storage systems for power quality improvement. Cybernetics and Systems: An International Journal 56(5): 599–628.

15.

Koganti

Nakka

Praveen

, et al. (2022a) Design of soccer league optimization based hybrid controller for solar-battery integrated UPQC. IEEE Access 10: 107116–107136.

16.

Krishna

Debashis

Goswami

(2020) A modified PV-wind-PEMFCS-based hybrid UPQC system with combined DVR/STATCOM operation by harmonic compensation. International Journal of Modeling and Simulation 41(4): 243–255.

17.

Kumar

Jaisiva

Clement

, et al. (2021) Hybrid renewable energy based smart grid system for reactive power management and voltage profile enhancement using artificial neural network. Energy Sources, Part A: Recovery, Utilization, and Environmental Effects 43(19): 2419–2442.

18.

Labdai

Hemici

Bounara

, et al. (2019) Control of a DFIG based WECS with optimized PI controllers via a duplicate PSO algorithm. International Conference on Control, Automation and Diagnosis, Grenoble, France: pp.1–6.

19.

Mahmoud

Atia

Esmail

, et al. (2023) Application of whale optimization algorithm based FOPI controllers for STATCOM and UPQC to mitigate harmonics and voltage instability in modern distribution power grids. Axioms 12: 420.

20.

Mansoor

Hasan

Othman

, et al. (2020) Construction and performance investigation of three phase solar PV and battery energy storage system integrated UPQC. IEEE Accesses 8: 103511–103538.

21.

Nima

Hamid

Sajad

, et al. (2021) A novel control approach for harmonic compensation using switched power filter compensators in micro-grids. IET Renewable Power Generation 15(16): 3989–4005.

22.

Nima

Salwa

Rasoul

, et al. (2022) Enhancement of power quality issues for a hybrid AC/DC microgrid based on optimization methods. IET Renewable Power Generation 6(8): 1773–1791.

23.

Okech

Zhang

Mali

, et al. (2022) Neural network controlled solar PV battery powered unified power quality conditioner for grid connected operation. Energies 15(18): 6825.

24.

Pazhanimuthu

Ramesh

(2018) Grid integration of renewable energy sources (RES) for power quality improvement using adaptive fuzzy logic controller based series hybrid active power filter (SHAPF). Journal of Intelligent & Fuzzy Systems 35(1): 749–766.

25.

Rajesh

Shajin

Uma sankar

(2025) Novel control scheme for PV/WT/FC/battery to power quality enhancement in micro grid system: A hybrid technique. Energy Sources, Part A: Recovery, Utilization, and Environmental Effects 47(1): 9126–9142.

26.

Srilakshmi

Krishna Jyothi

Surender Reddy

(2022) Design of multi-objective-based artificial intelligence controller for wind/battery-connected shunt active power filter. Algorithms 15(8): 256.

27.

Tzu-Chiao

Bawoke

(2022) Intelligent tuned hybrid power filter with fuzzy-PI control. Energies 15(12): 4371.

28.

Wang

Blaabjerg

(2019) Harmonic stability in power electronic-based power systems: concept, modeling, and analysis. IEEE Transactions on Smart Grid 10(3): 2858–2870.

29.

Xiaojun

Xiuhui

Xiaoqiang

, et al. (2021) Impedance matching-based power flow analysis for UPQC in three-phase four-wire system. Energies 14(9): 2702.

30.

Yap

Mohd

, et al. (2017) Control algorithms of shunt active power filter for harmonics mitigation: a review. Energies 10(12): 2038.

31.

Yap

Mohd

Muhammad

, et al. (2019) Shunt active power filter: A review on phase synchronization control techniques. Electronics 8(7): 791.

32.

Zhou

Wan

Wei

, et al. (2006) Control method for power quality compensation based on levenberg-marquardt optimized BP neural networks. CES/IEEE 5th International Power Electronics and Motion Control Conference: 1–4.

Chimp algorithm based Solar-Hydrogen energy powered UPQC for distribution system

Abstract

Keywords

Introduction

Proposed configuration

External support for DCL

SPVGS

BESS

Fuel cell

Shunt active power filter

Chimp optimization algorithm

CHOA optimized FLC-PIDC

NNC with PWM for reference signal generation

Series controller

Results analysis with discussions

Case 1: Performance during source voltage sag and swell for the balanced supply with L-1 and L-2

Case 2: Performance during short interruption and harmonic content for balanced supply voltage with sudden addition of load L-2 to L-1

Case 3: Performance during unbalanced sag /swell in supply with dynamic variation of solar irradiation and loads of L-4, L-3, and L-2

Case 4: Performance during unbalanced supply with unbalanced load under variable irradiation condition

Conclusion

Footnotes

Abbreviations

ORCID iD

Funding

Declaration of conflicting interests

References