Artificial Neural Networks in MPPT Algorithms for Optimization of Photovoltaic Power Systems: A Review

Area of Science:

• Renewable Energy Systems
• Electrical Engineering
• Artificial Intelligence

Background:

• Growing demand for clean electrical energy necessitates improved photovoltaic (PV) system efficiency.
• Existing Maximum Power Point Tracking (MPPT) inverters require advanced algorithms for optimal energy extraction.
• Low efficiency of conventional PV systems drives research into more effective MPPT techniques.

Purpose of the Study:

• To review and present Artificial Neural Networks (ANN) applications for Maximum Power Point Tracking (MPPT) control in PV systems.
• To evaluate the performance of ANN-based MPPT algorithms compared to other intelligent methods.
• To highlight the advantages of using ANNs for fast and accurate MPPT.

Main Methods:

• A comprehensive literature review of research over the last six years focusing on ANN for MPPT control.
• Analysis of algorithms based on ANN, or hybrid combinations with Fuzzy Logic (FL) or metaheuristic algorithms.
• Evaluation of controller effectiveness based on ANN training and hidden layer algorithms.

Main Results:

• ANN MPPT algorithms demonstrate an average performance of 98% under uniform conditions.
• ANN-based MPPT exhibits significantly faster convergence speeds compared to traditional methods.
• ANN MPPT controllers show fewer oscillations around the Maximum Power Point (MPP).

Conclusions:

• Artificial Neural Networks are highly effective for MPPT control in PV systems.
• ANNs provide a robust solution for enhancing PV energy yield and system performance.
• Future research should continue exploring ANN and hybrid approaches for advanced MPPT strategies.