Design and control of PV integrated qZS-CMI system with meta heuristic assisted artificial neural network. (November 2022)

• Record Type: Journal Article
• Authors: VC, Harish-Kumar; S, Amala-Shanthi
• Is Part Of: Advances in engineering software. Volume 173(2022)
• Publication Date: 2022-11
• Subjects: qZS-CMI -- SoSMC controller -- PR control -- NN -- FE-ROA algorithm; Computer-aided engineering -- Periodicals; Engineering -- Computer programs -- Periodicals; Engineering -- Software -- Periodicals; Periodicals; 620.0028553
• DOI: 10.1016/j.advengsoft.2022.103231
• Languages: English
• ISSNs: 0965-9978

Abstract

Highlights: The highlights of the article are given below for your kind perusal. Kindly, consider and forward my article for further processes. Exploits the Fitness Enabled-Rider Optimization Algorithm (FE-ROA) for modeling of controller for quasi-Z Source Cascaded Multilevel Inverter (qZS-CMI). Addresses the design and control of 3-phase qZS-CMI model based on two phases using FE-ROA approach. The performance of FE-ROA based qZS-CMI model is better than the standard conventional GA + (SoSMC & PR), PSO + (SoSMC & PR), ACS + (SoSMC & PR), ROA + (SoSMC & PR) and FE-ROA + (PI & PR) models. Abstract: "Quasi-Z source cascaded multilevel inverter (qZS-CMI) is a rising topology in photovoltaic (PV) power system. This resolves the drawbacks of traditional CMI-oriented PV systems as it has better balanced dc-link voltage". However, limited works are there for revealing controller design and modeling of qZS-CMI-oriented systems. This work introduces a controller model for a 3-phase qZS-CMI design. Two different phases are exploited in this work. In Phase I, an optimized "Second order Sliding Mode Controller (SoSMC)" is introduced to determine the PV voltage. For current regulation, Phase II uses an "Optimized Proportional Resonant (PR) controller with Artificial Neural Network (ANN)." A new "Fitness Enabled-Rider Optimization Algorithm (FE-ROA)" is used to optimize control gains in both phases. By using the FE-ROA convergence is improved. In Phase II, the suggested FE-ROA is used to train the ANN. The primary goal of phase I is to minimize the discrepancy between the reference and PV voltage, whereas phase II aims to minimize the error between computed and link grid current. As a result, the introduced model achieves "grid-connected current injection, " and system-level management improves "Maximum Power Point Tracking (MPPT)." Finally, the effectiveness of chosen methodology is demonstrated using a variety of measures.