The development of non-linear loads at consumers has significantly impacted power supply systems. Since, the poor power quality has been found in the three-phase distribution system due to unbalanced loads, harmonic current, undesired voltage regulation, and extreme reactive power demand. To overcome this issue, Distributed STATicCOMpensator (DSTATCOM) is implemented. DSTATCOM is a shunt-connected Voltage Source Converter (VSC) that has been utilized in distribution networks to balance the bus voltage in terms of enhancing reactive power control and power factor. DSTATCOM can provide both rapid and continuous capacitive and inductive mode compensation. A rectified resistive and inductive load eliminates current harmonics in a three-phase power supply. The synchronous fundamental DQ frame is a time-domain approach developed from three-phase system space vector transformations has been designed using MATLAB/Simulink. The DQ theory is used to produce the reference signal for the Pulse Width Modulation (PWM) generator. In addition, a traditional Proportional Integral Derivative (PID) controller is designed and compared with proposed soft computing approaches such as Fuzzy–PID and Artificial Neural Network (ANN-PID) and compared accurate reference current determination for Direct Current (DC) bus through DC link. An Analytical explores the proposed control strategies given to establish superior outcomes. Finally, total harmonic distortion analysis should be taken for performance analysis of the proposed system with IEEE standards.

A Distributed STATicCOMpensator (DSTATCOM) is used to compensate for Q and unbalanced loads in the distribution system. It is a shunt device that produces an Alternate Current (AC) voltage, injected through a shunt transformer into the system. In the control of DSTATCOM, traditional control schemes such as Proportional Integral (PI) controller and Proportional Integral Derivative (PID) controller are used. Still, now, intelligent controllers such as Fuzzy Logic Controllers (FLC) and Artificial Neural Network (ANN) are employed as alternative linear and non-linear control strategies.

The performance of a three-phase, two-level, and four-switch DSTATCOM is evaluated by modified synchronous reference frame theory based control scheme [

This control technique maintains the constant DC-link voltage, balances the split capacitor’s voltage, and generates switching pulses for the DSTATCOM’s four insulated-gate bipolar transistor switches to achieve the desired objective. The Perturbation and Observation algorithm is used to track the maximum power point and regulate the output voltage of the solar photo voltaic system. By integrating neural network control in DSTATCOM [

The Synchronous Reference Frame (SRF) based controlling of the DVR is responsible for successfully compensating balanced and unbalanced voltage sags [

Reactive power is created and provided to the AC system by increasing DSTATCOM’s output voltage [

One of the forms of power quality issues is the harmonics problem [

A new control mechanism for hybrid AC-DC microgrids, constructed on fuzzy control in D-STATCOM, is presented to minimize the unbalance voltage and solve the system’s power quality problems. These problems arise from exposing the intended power system to unanticipated fluctuations, such as developing a three-phase fault or abrupt dynamic load fluctuations. This paper presents the DSTATCOM-based SRF/DQ theory using PID, Fuzzy-PID and ANN-PID controllers. The DSTATCOM-based SRF with the controllers has been analyzed using MATLAB/Simulink software. Also, a comparative analysis of the controller’s Total harmonic distortion (THD) is illustrated.

A STATCOM is also called STATCON (STATicCONdenser). This is a FACTS device connected in parallel to the system (Shunt active filter) used to solve the power quality issues. A DSTATCOM is a VSC-based device with a V_{s} contained behind such a reactor. Because the V_{s} is derived from a DC link capacitor (C_{dc}), a DSTATCOM has relatively low P capabilities. However, connecting an appropriate energy storage device across the C_{dc} can enhance its P capability. The Q at the DSTATCOM terminals is proportional to the amplitude of the V_{s}. For instance, if the VSC’s terminal voltage is higher than the AC voltage at the Point of Common Coupling (PCC), the DSTATCOM generates reactive current; similarly, if the amplitude of the V_{s} is less than the AC voltage, it absorbs Q. A DSTATCOM’s response time is faster because of the quick switching times produced by the VSC’s MOSFET. Since the Q from a DSTATCOM decreases linear with the AC voltage, it provides greater Q support at low AC voltages. This involves current and voltage quality, load unbalanced problems such as unbalanced current at the PCC, reactive components and harmonic distortion. The DSTATCOM comprises a three-phase controlled VSC, DC link capacitor and an inductor, as shown in

The capacitor acts as an active filter for stored energy. Due to the switching algorithm, energy can be transmitted from the _{dc} to the inductor. The compensating principle of the DSTATCOM is constructed to compensate the current _{c} from supply to the load. Because of this _{c}, it eliminates the current harmonics and Q on the source side. Thus the source current (_{s}) can be purely sinusoidal and in phase with source voltage (_{s}).

The SRF control, commonly called DQ control, is primarily based on a reference frame transformation unit that converts the utility grid’s current and voltage characteristics to a synchronously rotating reference frame. The Clarke and Park transformation procedures, which convert abc to and DQ, are used to accomplish this transformation. The SRF control method is based on the current transformations in the d-q frame. Shunt Active Power Filter (APF) sensed the source voltage (V_{a}, V_{b}, V_{c}). The PCC current (Ia, Ib, Ic) and the injected Va, Vb, Vc are in phase. These voltages and source current can be converted into D-Q, shown in

where, the DQ axes are represented revolving with an angular velocity equal to, the same as the phase voltages and currents. V_{ED} and V_{EQ} are the representation of rotating reference frame to a three-phase (abc) signal. The shunt APF signals are given to the PID controller to generate control signals, converted to the ABC domain for switching signals in PWM. It again provides the pulse to the shunt APF converter by synchronized decision.

The current components are generated in

SRF control extracts the DC component of id and _{q}; further, the dc components are extraction of _{ddc} and _{qdc} are transformed into an

The current transformation is obtained as _{abc}. The compensation of ‘Q’ can be provided as it is zero to calculate the reference source current _{ref},_{abc}.

The function of DSTATCOM needs an AC supply to require real power (P) to the load and losses. So, the switching DSTATCOM is used to decide _{ref},_{abc} by the load current component, which is extracted from SRF and the losses can be estimated using PID, Fuzzy-PID and ANN-PID control the DC link voltage (_{dc}).

A controller in a control system is a technique that attempts to reduce the difference between a system’s actual value (i.e., the process variable) and its desired value (i.e., the setpoint). All sophisticated control systems require controllers; these are critical points of control technique. The second-order active reference current is compared with reference _{dc}^{*} and sensed _{dc} is a voltage error in nth sample _{dc(error,n)}, which is expressed below,

_{dc(error,n),}which is progressed in PID control, the output _{out(n)} in the n^{th} sample,

where _{pdc}, _{ids,} and _{ddc} are 1, 0.5 and 0.001 are the gains of the PID controller. The PID controller’s output is considered the current loss component of the DSTATCOM. This can be added to the d-axis of the current.

A Fuzzy-PID is used to regulate _{dc} in DSTATCOM. The fuzzy rules are consequences of the knowledge of the system’s behavior, whereas mathematical modeling is needed to incorporate into the system. Capacitor voltage (_{c}) and reference voltage (_{dref}) are compared. Fuzzy control has the input variables like error (_{c}. the n^{th} sampling of the system time is expressed below,

where

Then the _{adj} is the output of the change in adjusting the current, _{adj} is the actual adjusting current and _{adj} is supplied to the converter loss, and it will include the peak value of the _{ref},_{abc}. The fuzzy control adds appropriate rules that narrate the input variables to the output model. Here, the inputs and output were selected to the triangle type Membership Functions (MF). This arranged MF can be combined with (

The input MFs can ensure that the converter output voltage with the 49 rules is derived, as shown in

e | NH | NM | NL | Z | PL | PM | PH |
---|---|---|---|---|---|---|---|

NL | P | P | P | A | Z | L | Z |

NM | P | P | A | Z | Z | Z | L |

NS | P | A | L | Z | Z | L | L |

Z | A | L | L | Z | L | L | A |

PS | L | L | Z | Z | L | A | P |

PM | L | Z | Z | Z | A | P | P |

PL | Z | L | Z | A | P | P | P |

Then the attained values from fuzzy can be given to the PID control. It can be further tuned the required value to the DSTATCOM.

In this ANN control, the data can be trained with initial inputs and targets from the presented system, as shown in

Then select the samples in matrix rows. After that, validation and test data should be checked with 15% of 1810 samples of 2 inputs and 200 hidden layers of the neurons with two output layers. Finally, train the system.

The load can sense the reference current. Then the waveform can be expressed by Fourier analysis,

where _{n} and _{n} are the nt^{h} harmonics component of cos and sin amplitude,

The sample singles are arranged with a uniform rate (^{T} is the weighting vector. _{esti}^{T} are given by

Where,

Here the error

The attained weight is used as the next iteration _{(k+1)} to minimize the _{dc} of the DSTATCOM.

The simulation of the DSTATCOM system is designed using MATLAB/Simulink in the Simscape platform, which is shown in _{f} and L_{f} are connected with the range of 1 Ω and 120 mH, respectively.

The simulated SRF control is shown in _{dc}, which is kept at 250 V. At t = 0.4–0.6 s, the impact of a load change is observed.

In the sub-system of the voltage stabilizer, Controllers are designed to control the _{dc}, _{ref} and capacitive DC are the source. The estimated values are given to the _{d,loss}. Whereas the proposed implemented controllers are shown in

The DSTATCOM-based SRF control method is designed with soft computing techniques of Fuzzy-PID and ANN-PID. The results follow the observations. An Implementation of a short-circuit fault either between phase and ground. Whenever the exterior switching time mode is nominated, the fault action is controlled by a Simulink logical signal. The fault of switching time is at 0.4 to 0.6 s. At the same time, the Vs and I_{s} are attained from the presented DSTATCOM-based SRF control method shown in

_{s} and I_{s}, which means without compensation. The V_{s} and I_{s} are 300 V and 11 A, respectively. At the fault period, the V_{s} and I_{s} can be unbalanced. So the voltage can be decreased from 300 to 250 V and the current is unbalanced from 11 to 60 A.

The voltage is decreased and the current is increased due to the inductive nature of the load, which receives a significant amount of power from the supply side. Also, the system’s P is increased from 5.7 to 12 kW and Q is decreased from 3.2 to −10 kVAR at 0.4 to 0.6 s because of the unbalances.

The dynamics are displayed in reverse order after 0.6 s. I_{L} waveforms show a delay in compensating. The Low Pass Filter (LPF) causes this delay used to filter power signals. An LPF has a cut-off frequency that determines which frequencies are allowed to pass and which are filtered. A signal component would pass if its frequency is less than the cut-off frequency, or it would be filtered. These filters are employed to reduce circuit noise. When a sound is sent via LPF, most of the noise is eliminated, leaving a clean sound. Furthermore, SRF theory-based PID computes instantaneous P and Q using voltage signals. Any voltage distortion or unbalance will consequence in an incorrect computation of i_{ref,abc} should only include the true fundamental frequency component of the current of the load.

_{L}, I_{L}, P and Q. At 0.4 s, the load was raised from 300 to 350 V, and the unbalance was introduced at 0.4 s. This waveform shows the compensation of the voltage and current due to the injection of Q by the DSTATCOM. In compensation, P is 5.5 to 7 kW with a slight peak at 0.45 s; Q is 2.4 to 1.6 kVAR at 0.4 to 0.6 s.

After that, a Fuzzy-PID controller is proposed to control the unbalances. Simultaneous load adjustments and imbalanced circumstances are simulated as in the preceding scenario. The extracted reference current waveform shows the influence of delay caused by the LPF utilized in the DQ frame signal filtering. The development of voltage is essential in calculating i_{ref,abc}.

The ANN-PID extractor has the benefit of requiring less computing effort, making implementation of this approach considerably simpler. Furthermore, ANN-PID has intrinsic linearity which is the models having non-linear parameters, but they can be converted into linear regression models with the appropriate transformation that enables it to a quick solution.

It can be seen that DSTATCOM with SRF-based ANN-PID is sufficient to meet load fluctuations within a one sine wave. Using this technique, the unbalance load voltage is compensated from 250 to 310 V at 0.4–0.6 s. Likewise, the load current also compensated from 60 to 11 A at 0.4–0.6 s. Here the injection of Q is 2.4 to 2.2 kVAR. In this technique, the Q injection is low than in other techniques, which shows that the voltage and current are balanced and maintain the power quality, which leads to a unity power factor. If the power factor is lesser than one, then the load current will be higher and so the unity power factor is maintained. It also prevents transformers, motors, and other electrical components from overheating. The unity power factor develops even though there is no phase shift across voltage and current or when they are both in the same phase. When the power factor is one, the primary power source is doing beneficial work. In this situation, no electricity is being wasted. Shallow losses occur at the unity power factor. Whereas the system can be stable at the unity power factor, due to power quality issues, the system may unbalance, which is explained above, which leads to 0.8 PF.

Total harmonic distortion (THD) is a measurement that indicates the voltage or current distortion caused by harmonics in the signal. It is defined as the ratio of the total of all harmonic components’ powers to the fundamental frequency’s power.

Improving power quality is essential because of the increasing use of electronic devices more vulnerable to electrical interference. This article recommended implementing DSTATCOM to enhance power quality in the distribution system. The SRF control, which is time domain-based, is presented here. Furthermore, soft computing approaches such as Fuzzy-PID and ANN-PID are proposed to improve power quality concerns. The proposed controllers are compared to classic PID, Fuzzy, and ANN controllers. The proposed system is simulated in MATLAB/Simulink, demonstrating that the DSTATCOM-based SRF has a rapid reaction and precise load adjustment utilizing Fuzzy-PID and ANN-PID controllers. It is observed that DSTATCOM-based SRF with ANN-PID control offers THD of 1.48%. To overcome the complexity of the presented system, the future scope of this work is to mitigate the power quality issue using Genetic Algorithm (GA)-PID control and metaheuristic optimization techniques for more effectiveness.