Prediction of Thermal Energy Demand Using Fuzzy-Based Models Synthesized with Metaheuristic Algorithms

Increasing consumption of energy calls for proper approximation demand towards a sustainable and cost-effective development. In this work, novel hybrid methodologies aim to predict the annual thermal (ATED) by analyzing characteristics building, such as transmission coefficients elements, glazing, air-change conditions. For objective, an adaptive neuro-fuzzy-inference system (ANFIS) was optimized with equilibrium optimization (EO) Harris hawks (HHO) provide globally optimum training. Moreover, these algorithms were compared two benchmark techniques, namely grey wolf optimizer (GWO) slap swarm algorithm (SSA). The performance designed hybrids evaluated using different accuracy indicators, based on results, ANFIS-EO ANFIS-HHO (with respective RMSEs equal 6.43 6.90 kWh·m?2·year?1 versus 9.01 ANFIS-GWO 11.80 ANFIS-SSA) presented most accurate analysis ATED. Hence, models are recommended practical usages, i.e., early estimations ATED, leading more efficient design buildings.