Machine Learning-driven Enhancement of Concentrated Solar Power Systems: a Hybrid Ann and Ga Approach
Year: 2025
Type: article
Abstract: The imperative for efficient and sustainable energy systems has spotlighted the need to enhance the thermal performance of Concentrated Solar Power (CSP) technologies. This paper introduces a novel AI-powered control architecture designed to optimize the thermal dynamics of parabolic trough CSP collectors under fluctuating environmental conditions. The framework incorporates a hybrid model combining Artificial Neural Networks (ANN) with Genetic Algorithms (GA) to predict, regulate, and adapt system parameters such as fluid flow rate and receiver temperature. Real-time data-driven optimization enables superior heat exchange performance, reduced thermal losses, and enhanced responsiveness. Simulation outcomes demonstrate a substantial improvement in thermal efficiency, energy yield, and system adaptability compared to traditional fixed-control schemes. The proposed methodology exemplifies the transformative role of AI in fostering resilient, intelligent, and high-performance solar thermal infrastructures suitable for deployment in hot climate zones.
Author: Shaymaa W. Al-Shammari
Language: English
Field: Energy
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W4414475048