Hybrid control for capacitor-assisted Z-source inverter in grid-connected photovoltaic system

Abstract

Renewable energy integration into the power grid is essential for sustainable energy systems, but maintaining efficiency and reliability in such systems remains a key challenge. This study introduces a novel hybrid control topology for a Capacitor-Assisted Extended Boost Z-Source Multilevel Inverter in a grid associated solar photovoltaic (PV) structure. The suggested hybrid technique implies a united implementation of a Gannet Optimization Algorithm and Spiking Deep Residual Network and usually referred as GOA-SDRN technique. Initially, the modelling of the inverter is collected to get the best signal by the proposed controller. The proposed inverter configuration consists of a minimal quantity of diodes, switches, and sources. This configuration also offers advantages like less total harmonic distortion (THD) and reduced electromagnetic interference, making it a favourable choice for PV integration. The GOA is employed to determine the most favourable gain limiting factor based on a variation of power from their regular values. This control method ideally satisfies the load demand while reducing changes in the system limiting factor and external disturbances. The proposed control topology is executed in MATLAB and the concept is contrasted to different techniques. The THD values of existing methods such as Particle Swarm Optimization, Genetic Algorithm, and Grasshopper Optimization Algorithm are 1.36 %,0.89 %, and 1.99 % respectively, while the THD value of the proposedmethod is 0.63 %, demonstrating its optimal performance over existing methods. The proposed GOA-SDRN technique provides an effective solution for enhancing inverter efficiency and reducing distortion in grid-associated PV systems.