Energy optimization of PV systems under partial shading conditions using various technique-based MPPT methods.

Area of Science:

• Renewable Energy Systems
• Artificial Intelligence in Engineering
• Power Electronics

Background:

• Photovoltaic (PV) systems face challenges with nonlinear power-voltage characteristics, particularly under partial shading conditions (PSC).
• These issues lead to reduced energy efficiency and tracking accuracy in conventional maximum power point tracking (MPPT) systems.
• Existing methods like perturb-and-observe (P&O) suffer from steady-state oscillations and slow dynamic responses.

Purpose of the Study:

• To develop and evaluate novel MPPT controllers for PV systems to overcome limitations caused by PSC.
• To improve energy efficiency, tracking accuracy, and dynamic response compared to traditional methods.
• To ensure reliable and efficient power extraction under rapidly changing environmental conditions.

Main Methods:

• Proposed two improved MPPT controllers utilizing Artificial Neural Network (ANN) and Adaptive Neuro-Fuzzy Inference System (ANFIS) techniques.
• Implemented a predictive, non-iterative control architecture using power-voltage and voltage time derivatives as input features.
• Simulated and compared the performance of the proposed controllers against the conventional P&O algorithm and other AI-based MPPT approaches.

Main Results:

• The ANN and ANFIS controllers achieved high average tracking efficiencies of 99.4% and 99.75%, respectively.
• Demonstrated a 55% reduction in response time and over 70% suppression of steady-state oscillations compared to P&O.
• The ANFIS controller showed superior stability, reducing duty-cycle fluctuations by 20% compared to ANN, with execution times under 0.2 ms.

Conclusions:

• The proposed intelligent MPPT framework, particularly the ANFIS controller, offers a fast, accurate, and computationally efficient solution for PV systems.
• These advanced controllers significantly improve reliability and energy yield, especially under dynamic and partial shading conditions.
• The low computational complexity makes the controllers suitable for real-time deployment on low-cost digital signal processors (DSPs).