Learning Automata and Soft Computing Techniques Based Maximum Power Point Tracking for Solar PV Systems

Abstract

Solar PV (SPV)-based power generation is one of the promising solutions among the alternative energy solutions. The characteristic of SPV module is nonlinear, and maximum power point (MPP) is dependent on the input conditions. To extract maximum power from SPV module, MPP tracking controller is incorporated in SPV system. The main aim of this chapter is to introduce learning automata concept and its adaptability for the development of MPPT algorithm. As an overview, learning the optimal actions in a stochastic environment using LA is explained with n-arm bandit problem and the learning process using pursuit algorithm (PA) is described in detail. Then, the development of perturb & observe (P&O) and LA-based hybrid MPPT technique is explained in detail. Performance of the proposed hybrid MPPT technique is analyzed by conducting extensive simulation studies for different input conditions. A comprehensive comparison and performance analysis of the MPPT techniques like P&O, variable step size (VSS) P&O, the hybrid MPPT technique developed using P&O and LA is carried out and presented. The soft computing technique-based MPPT techniques such as fuzzy MPPT, GA-optimized fuzzy MPPT and ANFIS-optimized fuzzy MPPT are also reviewed with the help of simulation studies in this chapter.