Fuzzy cognitive network-based maximum power point tracking using a self-tuned adaptive gain scheduled fuzzy proportional integral derivative controller and improved artificial neural network-based particle swarm optimization

摘要

The increased demand for electrical energy has driven the development of renewable energy sources. In particular, the conversion of solar energy into electrical energy using photovoltaic (PV) systems has become popular because of its simplicity and low cost. However, the nonlinear characteristics and power fluctuations due to changes in the temperature and irradiation hinder the maximum utilization of the power with a PV system. Thus, the maximum power point tracking (MPPT) control technique is used to extract the maximum available power from PV arrays. Due to insolation and variations in temperature in a PV system, the conventional MPPT techniques are readily trapped by local maxima to significantly reduce the conversion efficiency. In order to overcome this issue, we developed a novel perturb and observe algorithm based on an adaptive fuzzy PID controller with an improved artificial neural network-based particle swarm optimization method for tracking the maximum power point with high tracking speed as well as maintaining the system's stability. In addition, we used a fuzzy cognitive network to maintain the equilibrium state, which is essential for improving the conversion efficiency. Simulation results and performance evaluations using our proposed method demonstrated its suitability for applications in PV systems. (C) 2019 Elsevier B.V. All rights reserved.