A Feed Forward Back Propagation Neural Network Approach For Reliability And Sensitivity Analysis Of A Standalone Integrated Renewable Energy System

Abstract

Today electricity is one of the basic requirements for the development of any country. Presently, there is a rapid growth in the industry and technology sector; still, there are many remote areas where electricity is a dream. It may require a huge amount of money and manpower to establish a strong infrastructure and seamless power supply. The problem could be solved if the electricity is generated at the same places by the utilization of naturally available resources with the help of integrated renewable energy systems. In this piece of work, the authors have analyzed reliability and various other parameters for an integrated renewable energy system (IRES) with distinct substitutes of power generation using renewable energy resources, their availability, and load. The resources, used in the current study, are Photovoltaic modules, Wind turbines, power storage units as battery banks on a hybrid basis. The failure and repair rates are distributed exponentially. Due to the complexity of integrated structure, the degradation of system functioning may increase, therefore authors have incorporated an intelligent computational approach. Neural Networks (NN) are employed to develop state probabilities and various parameters in proposed IRES. To improve the accuracy and consistency of reliability parameters, a Feed Forward Back Propagation Neural Network (FFBPNN) is applied. FFBPNN’s learning mechanism can optimize the values of parameters by modifying neural weights. The MATLAB codes are used by authors to demonstrate the numerical examples and iterations are repeated until the precision in error tends up to 0.0001. The sensitivity and cost analysis can help in managing the real-time operations of the system.