Machine Learning Method Based on Symbiotic Organism Search Algorithm for Thermal Load Prediction in Buildings

This research investigates the efficacy of a proposed novel machine learning tool for optimal simulation building thermal load. By applying symbiotic organism search (SOS) metaheuristic algorithm to well-known model, namely an artificial neural network (ANN), sophisticated optimizable methodology is developed estimating heating load (HL) in residential buildings. Moreover, SOS comparatively assessed with several identical optimizers, political optimizer, heap-based Henry gas solubility optimization, atom stochastic fractal search, and cuttlefish optimization algorithm. The dataset used this study lists HL versus corresponding conditions model tries disclose nonlinear relationship between them. For each mode, extensive trial error effort revealed most suitable configuration. Examining accuracy prediction showed that SOS–ANN hybrid strong predictor as its results are great harmony expectations. verify SOS–ANN, it was compared benchmark models employed study, well earlier literature. comparison superior suggested model. Hence, utilizing highly recommended energy-building experts attaining early estimation from designed building’s characteristics.