Harnessing Solar Energy: A Comprehensive Review of Solar Thermal Collector Applications

Abstract

This study examines the importance of efficient solar collector design for optimal performance in low to medium-temperature applications, emphasizing the role of Artificial Neural Networks (ANN) in enhancing system efficiency. ANN, known for its speed and accuracy in solving complex problems, is widely used across various fields such as science, engineering, and business. The primary aim is to review ANN applications in predicting solar collector performance and to identify future research gaps. The systematic literature review (SLR) and meta-analysis focus on ANN applications in solar thermal collectors, covering research from 2000 to 2021. Out of 374 initial papers, 86 utilized ANN methods. The review analyzes data collection methods, ANN model setups, evaluation metrics, and meta-analysis results. Findings indicate that most studies focus on solar water heaters (SWH) using water as the heat exchanger, with a significant concentration of research conducted in Asia. The Multilayer Perceptron (MLP) model is predominantly used, appearing in 76 of the 86 reviewed papers. A key research gap is identified in the limited use of deep learning and other traditional machine learning models, as well as the underrepresentation of African data in ANN modeling for solar thermal devices. Despite covering two decades of studies, there is an observed exponential growth in research interest, highlighting the need for future studies to incorporate a broader range of machine learning algorithms for comprehensive conclusions.