Optimizing Photovoltaic Systems: A Decision Matrix Approach for Selecting the Optimal Maximum Power Point Tracking Technique

Abstract

Numerous strategies exist for improving maximum power point tracking (MPPT) techniques, which vary in terms of tracking speed, accuracy, sensor usage, complexity, and cost. Due to the importance of sustainable energy use and related challenges, it is essential to choose an appropriate algorithm that can reliably provide energy without exhausting resources. While much research has examined the benefits and drawbacks of different MPPT algorithms, guidance on selecting the best suited algorithm for a particular solar system application is lacking. For a fair and thorough comparison, this study analyzes four MPPT algorithms, including two artificial intelligence-based techniques, an adaptive neuro-fuzzy inference system, and a conventional technique. A decision matrix model and requirement analysis are used to determine their suitability for a standalone PV application, using particularly ranked evaluation criteria, which include tracking efficiency, implementation costs, rise time, settling time, tracking error and variance. Simulation results show that the perturb and observe technique has the fastest settling and rise times and high tracking efficiency for varying irradiance and temperature levels, with little implementation effort required for a stand-alone photovoltaic application.