Predication of Solar Still Production Rate using Artificial Neural Network and Step Regression Model

Solar desalination presents a promising solution for freshwater production, and this study evaluates its effectiveness by analyzing experimental data to estimate the performance of a solar still. The study shows that the performance of a solar still is affected by several key factors, such as solar radiation, feed flow rate, ambient temperature, relative humidity, and wind speed. To better understand and predict these influences, researchers used a back-propagation artificial neural network (ANN) and compared its results to those of a traditional stepwise regression model. The results revealed that the ANN offered significantly higher accuracy in predicting the freshwater output, outperforming the regression-based approach. This underscores the potential of machine learning in optimizing solar desalination systems under dynamic environmental conditions.