Adaptive Neuro-Fuzzy Inference System-Based Bass Gura Controller for Solar-Powered SEPIC Converter

Abstract

A systematic design and detailed circuit analysis of a solar-powered ROSEPIC (reduced order single ended primary inductance converter) converter for maximum power point tracking (MPPT) and load voltage regulation by cascading an adaptive neuro-fuzzy inference system (ANFIS)-based Bass Gura controller with particle swarm optimization for standalone DC nanogrid applications has been proposed. An accurate mathematical model is derived to evaluate the gains of the Bass Gura controller. The order of the transfer function is reduced by using the moment matching method. The performance of the proposed controller is compared with the conventional Bass Gura controller to show the effectiveness of the system. Also, to substantiate the simulation results, an experimental prototype model controlled by C2000 Piccolo TMS320F28035MCU digital controller is set up. The inferences arrived from the results are (a) the proposed ANFIS-based Bass Gura controller has the ability in tracking 81.48–99.42% of MPPT and load voltage simultaneously against various irradiation levels, temperatures, and loads and (b) the moment matching method eases the controller design by retaining only the dominant modes. Despite of its potential benefits, the ROSEPIC system is also subjected to partial shading conditions, and the results guaranteed that the proposed ANFIS-based Bass Gura controller is effective enough in tracking the MPPT as well as the load voltage.