Indoor Temperature Prediction of a Transpired Solar Collector Using Machine Learning

In this work, the application of artificial intelligence (AI) and machine learning techniques have been employed to predict the indoor temperature of a transpired solar collector based on a comprehensive set of input parameters which are the ambient temperature, inlet temperature, solar radiation, alongside with multiple channels of environmental data to develop a sophisticated predictive model. Leveraging the power of AI, the current study aims to discern complex patterns and correlations within the input variables that influence indoor thermal conditions. By employing machine learning algorithms, the goal is to create a robust and adaptive model capable of accurately forecasting indoor temperature, thereby providing a valuable tool to optimize building climate control systems. Following the experiment using Random Forest and Random Tree, two machine learning algorithms, the Random Tree algorithm produced better results with a mean absolute error of 0.2633 and a correlation coefficient of 0.9997.