Intelligent control of hybrid energy system for sustainability and enhanced power quality

Abstract

This thesis presents a comprehensive exploration into enhancing the efficiency, power quality, and overall performance of hybrid energy systems. Focusing on renewable energy sources such as wind, solar photovoltaic (PV), fuel cells, and batteries, the research integrates advanced control strategies and novel converter topologies to optimize energy utilization and reduce environmental impact. The first study presents the implementation of a Wind Energy Conversion System that utilizes artificial intelligence techniques for Maximum Power Point Tracking and the storage of surplus wind energy in a battery. The battery provided supplementary power when the demand for electricity exceeded the available wind power. The Multi-Level Inverter consists of ten switches and employs four direct current voltage sources to generate an output waveform with seventeen levels. Application of neural network (ANN) is noteworthy aspects to consider in this chapter. Artificial neural network is optimized by Levenberg Marquardt algorithm. These perspectives have been verified by simulation. The second study presents a simulation of a distributed generating system utilizing a Proton Exchange Membrane Fuel Cell and a lead-acid battery. The system has been subjected to analysis and simulation to improve power quality aspects. Artificial Neural Network has been employed as a controller for estimating reference current. This enables dynamic and real-time enhancements in power quality. The inverter obviates the necessity for a filter and supplies the nonlinear component of the load current. The total harmonic distortion of the grid current is decreased, and DC-link capacitor voltage regulation is also achieved. The performance of the artificial neural network was assessed and confirmed using MATLAB/Simulink. The proposed approach has been determined to effectively eliminate the need for a load filter, resulting in a reduction in system expenses. Consequently, the suggested approach is characterized by its cost-effectiveness and feasibility.