Volume 51

SPXAI: Solar Power Generation with Explainable AI Technology Rifat Al Mamun Rudro, Afroza Nahar, Md. Faruk Abdullah Al Sohan

Abstract

The integration of machine learning and deep learning technologies has revolutionized solar power production by addressing challenges such as variability and unpredictability. This paper explores the application of Explainable AI (XAI) through the proposed SPXAI model to enhance the efficiency and reliability of solar energy systems. SPXAI collects extensive power production data from solar farms and employs machine learning and deep learning models to analyze this data on an hourly basis. This analysis provides clear insights into predictions, identifies influential factors, and offers rule-based explanations for complex model decisions. Additionally, SPXAI makes real-time, data-driven decisions to optimize solar panel performance, such as adjusting panel orientations, scheduling predictive maintenance, and refining energy storage and distribution strategies. This approach enhances transparency and reliance on AI-driven recommendations, reducing operational costs and increasing solar power production reliability.

Keywords deep learning, explainable AI, solar power, prediction, optimized performance