Advanced Power Quality Control through Artificial Neural Networks Controller and Fuzzy Logic Controller Based Dynamic Voltage Restorer Systems

Abstract

Power quality is a major concern in modern power systems because it directly affects both consumers and utility operations. Issues like voltage harmonics, sags, and swells can cause significant damage to sensitive equipment, which is increasingly prevalent in modern electrical systems. As more sensitive devices are used, ensuring high power quality becomes critical for the reliable and secure operation of these systems. To manage these power quality issues, a device known as a Dynamic Voltage Restorer (DVR)is widely used in distribution systems. The DVR is a type of Distribution Flexible AC Transmission System device is used to correct non- standard voltage conditions by injecting voltages that help maintain the desired voltage profile. This ensures a continuous and stable voltage supply to loads, protecting them from voltage fluctuations. Artificial neural networks control based DVR do not give sufficient Total harmonic distortion (THD), voltage balance, efficiency. This paper focuses on reducing voltage sags, voltage swells, and total harmonic distortion (THD) by employing a Dynamic Voltage Restorer (DVR). A hybrid control scheme that combines an Artificial Neural Network (ANN) with a Fuzzy Logic Controller is adopted to improve the effectiveness of voltage compensation. The integrated control strategy addresses the drawbacks of conventional ANN-controlled DVR systems and enhances overall power quality performance. The performance and the control strategy of the DVR by the combine of Artificial neural network (ANN) & Fuzzy logic controller is simulated using MATLAB SIMULINK software.

Introduction

Modern electrical and electronic advancements have increased power systems’ vulnerability to disturbances such as voltage sags, swells, harmonics, and flicker. These issues arise from distribution faults, sudden load changes, large inductive equipment, and nonlinear devices. Poor power quality can damage sensitive equipment and reduce system reliability, making voltage compensation essential in modern networks.

Among custom power devices, the Dynamic Voltage Restorer (DVR) is widely recognized as an effective solution for mitigating voltage-related disturbances. A DVR injects a compensating voltage in series with the supply to maintain the load voltage within acceptable limits during abnormal conditions. However, the performance of a DVR largely depends on its control strategy, especially under nonlinear and rapidly changing system conditions.

Limitations of Conventional Controllers

Traditional Proportional–Integral (PI) and Proportional–Integral–Derivative (PID) controllers are commonly used in DVR systems due to simplicity. However, they:

• Exhibit slower transient response during disturbances

• Perform poorly under nonlinear conditions

• Increase harmonic distortion

To overcome these limitations, intelligent control techniques such as Fuzzy Logic Control (FLC) and Artificial Neural Networks (ANN) have been introduced.

ANN-Based Controller

ANN controllers:

• Handle nonlinear and time-varying system behavior effectively

• Learn from training data using backpropagation

• Provide fast voltage compensation during sag and swell

• Exhibit strong generalization ability

However, standalone ANN systems have drawbacks:

• Depend heavily on high-quality training data

• Operate as “black-box” models with low interpretability

• Require high computational effort

• May perform poorly under rare or severe disturbances

Proposed Hybrid ANN–Fuzzy Logic Controller

To overcome these limitations, a hybrid ANN–Fuzzy Logic Controller is proposed for DVR systems.

Working Principle

• The Fuzzy Logic Controller (FLC) processes voltage error and its rate of change using membership functions and rule-based reasoning.

• It generates an initial control signal.

• The ANN module is trained using both system data and fuzzy outputs to refine the control signal.

• The optimized output is used to generate PWM pulses for the voltage source inverter (VSI).

• The inverter injects the required compensating voltage through a series transformer.

Advantages of the Hybrid Approach

• Combines interpretability of fuzzy logic with adaptability of ANN

• Reduces dependence on large datasets

• Improves transient response

• Minimizes steady-state error

• Significantly reduces Total Harmonic Distortion (THD)

• Enhances robustness under uncertainties

Comparative results indicate that the hybrid ANN–Fuzzy controller outperforms conventional PI, standalone ANN, and standalone FLC approaches.

Mathematical Modelling of DVR

The DVR operates by injecting a voltage:

• Vinj=Vref−VPCCV_{inj} = V_{ref} - V_{PCC}Vinj?=Vref?−VPCC?

• Final load voltage: VL=Vs+VinjV_L = V_s + V_{inj}VL?=Vs?+Vinj?

Voltage sag/swell is modeled considering source impedance and load current.

Three compensation techniques are discussed:

1. Pre-sag compensation

2. In-phase compensation

3. Phase-advanced compensation (reactive power only, suitable for DVRs without energy storage)

DVR systems are classified as:

• With energy storage

• Without energy storage

Unlike UPS systems, DVR compensates only the missing voltage portion rather than supplying full load power.

Control Strategy

The control strategy includes:

1. Voltage Error Calculation

• Error: e(t)=Vref−VLe(t) = V_{ref} - V_Le(t)=Vref?−VL?

• Change in error: Δe(t)\Delta e(t)Δe(t)

2. Fuzzy Logic Controller

• Fuzzification of error signals

• Rule-based inference

• Defuzzification using centroid method

3. ANN Model

• Feedforward network with hidden layer

• Uses tansig activation in hidden layer

• Trained using gradient descent to minimize error

• Produces refined control signal

4. PWM Generation

• Modulation index calculation

• Switching pulse generation

Conclusion

Enhancing power quality with a The Dynamic Voltage Restorer, controlled using a combination of Artificial Neural Networks and Fuzzy Logic, presents an effective method for mitigating voltage sags, swells, and harmonics, in modern distribution systems. By integrating these intelligent control strategies, the DVR can respond dynamically and accurately to voltage variations, achieving performance levels far superior to conventional controllers. The ANN controller adjusts to changing operating conditions by learning the nonlinear behavior of the system, while the Fuzzy Logic controller strengthens decision- making under uncertain or rapidly changing load conditions. Simulation results indicate that this hybrid control approach maintains the load voltage close to its nominal value, minimizes interruptions, and enhances overall power quality. Consequently, the intelligent DVR design delivers a reliable, efficient, and adaptive means of voltage regulation, ensuring stable operation and uninterrupted power supply in smart distribution networks subject to frequent disturbances and load variations.

References

[1] M. H. Bollen, R. Das, S. Djokic, P. Ciufo, J. Meyer, and S. K. Rönnberg, “Power quality concerns in implementing smart distribution-grid applications,” IEEE Transactions on Smart Grid, vol. 8, no. 1, pp. 391–399, 2017. [2] V. Khadkikar, D. Xu, and C. Cecati, “Emerging power quality problems and state-of-the-art solutions,” IEEE Transactions on Industrial Electronics, vol. 64, no. 2, pp. 761–763, 2017. [3] X. Liang, “Emerging power quality challenges due to integration of renewable energy sources,” IEEE Transactions on Industry Applications, vol. 53, no. 1, pp. 855–866, 2017. [4] T. Sutradhar, J. R. Pal, and C. Nandi, “Voltage sag mitigation by using SVC,” International Journal of Computer Applications, vol. 71, no. 3, pp. 1–5, 2013. [5] F. M. Mahdianpoor, R. A. Hooshmand, and M. Ataei, “A new approach to multifunctional dynamic voltage restorer implementation for emergency control in distribution systems,” IEEE Transactions on Power Delivery, vol. 26, no. 2, pp. 882–890, 2011. [6] E. Babaei, M. F. Kangarlu, and M. Sabahi, “Mitigation of voltage disturbances using dynamic voltage restorer based on direct converters,” IEEE Transactions on Power Delivery, vol. 25, no. 4, pp. 2676–2683, 2010. [7] G. N. Kumar and D. D. Chowdary, “DVR with sliding mode control to alleviate voltage sags on a distribution system for three-phase short-circuit fault,” in Proc. IEEE Region 10 Conf. Industrial and Information Systems (ICIIS), 2008, pp. 1–4. [8] M. Messiha, C. Baraket, A. Massoud, A. Iqbal, and R. Soliman, “Dynamic voltage restorer for voltage sag mitigation in oil & gas industry,” in Proc. 1st Workshop on Smart Grid and Renewable Energy (SGRE), 2015, pp. 1–6. [9] M. Önci, M. Büyük, A. Tan, K. Bayindir, and M. Tümay, “Survey of inverter topologies implemented in dynamic voltage restorers,” 2017. [10] A. M. Gee, F. Robinson, and W. Yuan, “A superconducting magnetic energy storage– emulator/battery supported dynamic voltage restorer,” IEEE Transactions on Energy Conversion, vol. 32, no. 1, pp. 55–64, 2017. [11] A. H. Abed, J. Rahebi, and A. Farzamnia, “Improvement for power quality by using dynamic voltage restorer in electrical distribution networks,” in Proc. IEEE 2nd Int. Conf. Automatic Control and Intelligent Systems (I2CACIS), Kota Kinabalu, 2017, pp. 122–127. [12] D. Francis and T. Thomas, “Mitigation of voltage sag and swell using dynamic voltage restorer,” in Proc. Annual Int. Conf. Emerging Research Areas: Magnetics, Machines and Drives (AICERA/iCMMD), 2014, pp. 1–6. [13] T. Jimichi, H. Fujita, and H. Akagi, “A dynamic voltage restorer equipped with a high-frequency isolated DC–DC converter,” IEEE Transactions on Industry Applications, vol. 47, no. 4, pp. 169–175, 2011. [14] M. Shahabadini and H. Iman-Eini, “Improving the performance of a cascaded H-bridge-based interline dynamic voltage restorer,” IEEE Transactions on Power Delivery, vol. 31, no. 3, pp. 1160–1167, 2016. [15] M. S. H. Sunny, A. N. R. Ahmed, and M. K. Hasan, “Design and simulation of maximum power point tracking of photovoltaic system using ANN,” in Proc. 3rd Int. Conf. Electrical Engineering and Information Communication Technology (ICEEICT), Dhaka, 2016, pp. 1–5. doi: 10.1109/CEEICT.2016.7873105.

Copyright

Copyright © 2026 B. Chandrashekar, S. Tara Kalyani. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.