Investigation on Stand-Alone Solar Energy Conversion System with Artificial Intelligence Techniques

Abstract

In the recent decades, the energy demand is increasing with the growth in infrastructure and population. To fulfill the requirement of energy demand, renewable energy sources are the best alternative. They also play an important role in the economic growth of any country. The REW energy sources are eco-friendly and non-polluting sources. Among the renewable energy sources, the solar photovoltaic (SPV) power plants are the most popular and economical. Nowadays, researches are focusing on renewable energy sources to bridge the gap between supply and demand. The working of SPV cells depends on the sunlight. Variations in the sunlight received from the sun affect the operation of SPV cells; therefore, this kind of renewable energy sources is highly unreliable and causes interruption in power supply to the load. A solar photovoltaic system with MPPT based on artificial neural network (ANN) is used for stable operation and for meeting the demand. The application of ANN-based MPPT techniques requires a detailed study and analysis of the solar photovoltaic (SPV) system. An ANN-based MPPT controller is used along with the SPV system as it can provide fast and accurate response to various environmental conditions. The Levenberg-Marquardt algorithm is used to train the neural network for MPPT in the photovoltaic system. This developed control strategy is able to control the devices and various power interface circuitry used therein. The main aim of this chapter is to ensure a maximum power output by coordinating appropriate control strategy with sources and to compare the performance of the ANN-based MPPT with conventional incremental conductance-based MPPT for the SPV system. The simulation studies are performed to find out the SPV system performance with different input conditions such as typical solar radiation, temperature, air, and battery charge or discharge. The simulation test results show variable power generation and verify that the performance of the integrated system with this control strategy is effective for real-time installation. The simulation results show that the performance of this developed control strategy for receiving maximum power output from the standalone SPV system is far-fetched.