Modelling of Adaptive Neuro-fuzzy Inference System (ANFIS) - Based Maximum Power Point Tracking (MPPT) Controller for a Solar Photovoltaic System

Abstract

Aim: The aim of this research is to model and simulate Adaptive Neuro-Fuzzy Inference System (ANFIS) - based MPPT Controller and also compare its performance with the Perturb and Observe MPPT controller for Photovoltaic systems. Study of the Design: The PV system consists of a PV module, a PWM inverter, an MPPT controller and a DC-DC converter, all of which are connected using Matlab-Simulink environment. Methodology: The ANFIS reference model is constructed based on two input parameters: solar irradiance and temperature. Its output parameter is the reference maximum power output. The temperature and irradiance data employed for the particular site examined in this study have been acquired from an online global database. The P&O MPPT method was used as a benchmark against the proposed ANFIS-based MPPT technique using Matlab-Simuink Enviroment. Results: In the absence of the controller, the PV voltage registered at 50V. However, through the ANFIS-based MPPT, the incoming voltage from the PV was able to roughly double. Conversely, in the absence of the controller, the PV current stabilized at 4.5A, but the ANFIS-based MPPT managed to reduce this incoming current from the PV by approximately half. The generated PV power follows a similar trajectory to the theoretical PV power, reaching a peak of around 420W, which closely aligns with the theoretical peak power of 440W. The power spans the range from nearly 0 to 420W. As a result, the overall efficiency of the ANFIS-based MPPT charge controller is estimated to be around 60%. Conclusion: The evaluation of the ANFIS-based MPPT and Perturb and Observe-based MPPT Controllers reveals that the Perturb and Observe-based controller demonstrated superior efficiency. Based on this investigation, it can be inferred that both MPPT controllers effectively address uncertain weather scenarios and can readily accommodate challenges such as partial shading and other irregularities commonly associated with varying weather conditions.