Evaluation of Hybrid AI-based Techniques for MPPT Optimization

Solar energy's intrinsic recurring and time-varying nature leads to limitations of the stability and trustworthiness of solar power grid systems. To have a stable energy supply and optimal output, Artificial Intelligence (AI) based Maximum Power Point Tracking (MPPT) techniques are now being exploited. Literature has shown that most AI-based MPPT algorithms have a faster convergence time, reduced steady-state oscillation, and higher efficiency but need a lot of processing and are expensive to implement. Overall, hybrid AI-based MPPT has a good performance-to-complexity ratio. It incorporates the benefits of traditional and AI-based MPPT methodologies but choosing the appropriate hybrid MPPT techniques is difficult since each has advantages and disadvantages. In this work, we evaluated and proposed suitable hybrid AI-based MPPT techniques that exhibit the right balance between performance and complexity when utilizing AI in MPPT for solar power system optimization. The selected MPPT techniques evaluated here includes Particle Swarm Optimization (PSO) trained Adaptive Neural-Fuzzy Interface System (ANFIS) and PSO trained Neural Network (NN) MPPT. Lastly, we compared these methods with our proposed AI-based technique, Genetic Algorithm (GA)-trained ANFIS method from previous work.