Suggesting a Stochastic Fractal Search Paradigm in Combination with Artificial Neural Network for Early Prediction of Cooling Load in Residential Buildings

Early prediction of thermal loads plays an essential role in analyzing energy-efficient buildings’ energy performance. On the other hand, stochastic algorithms have recently shown high proficiency dealing with this issue. These are reasons that study is dedicated to evaluating innovative hybrid method for predicting cooling load (CL) buildings residential usage. The proposed model a combination artificial neural networks and fractal search (SFS–ANNs). Two benchmark algorithms, namely grasshopper optimization algorithm (GOA) firefly (FA) also considered be compared SFS. non-linear effect eight independent factors on CL analyzed using each model’s optimal structure. Evaluation results outlined all three metaheuristic (with more than 90% correlation) can adequately optimize ANN. In regard, tool’s error declined by nearly 23%, 18%, 36% applying GOA, FA, SFS techniques. Moreover, used accuracy criteria indicated superiority over schemes. Therefore, it inferred utilizing along ANN provides reliable early CL.