Solar photovoltaic converter controller using opposition-based reinforcement learning with butterfly optimization...

Area of Science:

• Renewable Energy Systems
• Electrical Engineering
• Artificial Intelligence in Energy

Background:

• Photovoltaic (PV) systems aim to maximize power generation.
• Partial shading causes power fluctuations, leading to energy loss.
• Existing maximum power point tracking (MPPT) methods struggle with these fluctuations.

Purpose of the Study:

• To propose a hybridized MPPT technique to address power fluctuations under partial shading.
• To enhance the energy yield and stability of photovoltaic systems.
• To improve the adaptability and convergence of MPPT algorithms.

Main Methods:

• A novel hybridized MPPT technique combining opposition-based reinforcement learning and a butterfly optimization algorithm was developed.
• The methodology was tested on various photovoltaic configurations (6S, 3S2P, 2S3P) under diverse shading conditions.
• Performance was compared against established MPPT techniques including butterfly optimization, grey wolf optimization, whale optimization, and particle swarm optimization.

Main Results:

• The proposed hybridized MPPT technique demonstrated superior adaptation compared to conventional methods.
• The method effectively mitigated issues related to load variation convergence.
• It also addressed frequent exploration and exploitation patterns, leading to more stable power output.

Conclusions:

• The developed opposition-based reinforcement learning with butterfly optimization algorithm offers a robust solution for MPPT in PV systems under partial shading.
• This hybridized approach significantly improves energy generation efficiency and system reliability.
• The findings suggest a promising direction for advanced control strategies in renewable energy applications.