Optimization of solar PV–UPQC performance using enhanced IPQ theory with DNN technique

Dispersed power systems powered by renewable energy sources such as wind and solar often face significant power quality challenges. To mitigate these issues in microgrid systems, specialized power equipment like D-STATCOM, UPQC, UPS, TVSS, and DVR are commonly employed. Due to its many benefits, UPQC may be a viable option for renewable energy systems. The Unified Power Quality Conditioner (UPQC), which is a product of emerging trends and technological advancements in power electronics, performs better than various approaches. The proposed work focuses on the comparative performance analysis of a distributed generation system that combines a Unified Power Quality Conditioner using Adaptive Neuro-Fuzzy Interference System (ANFIS) and Deep Neural Network (DNN) controllers. A self-tuning filter (STF) that is integrated with the unit vector generator (UVG) technology guides the UPQC controller’s phase synchronisation function. By implementing STF, the UPQC will be certain to function properly even in the presence of distorted and imbalanced grid voltage. As a result, the STF-UVG is used to generate the synchronisation phases for the series and shunt active power filter (APF) compensator in the UPQC controller, eliminating the need for a phase-locked loop (PLL). The main objective of the suggested UPQC-DNN is to outperform the UPQC-ANFIC and conventional PI controller.in both steady and dynamic states. The DNN controller achieves the lowest THD, with 0.76% for voltage and 0.81% for current, demonstrating superior performance over the other controllers. The suggested system has been developed and tested in various operating scenarios using the MATLAB Simulink.