Building Energy Optimization Using an Improved Exponential Distribution Optimizer Based on Golden Sine Strategy Minimizing Energy Consumption Under Uncertainty

In this study, a new improved meta-heuristic algorithm is proposed for solving the energy building optimization (EBO) and also hybrid energy systems optimization considering uncertainty of conditioned surface area subjected to temperature control for BEO and renewable power and load uncertainties for hybrid system. The conventional Exponential distribution optimizer (EDO) is improved using the Golden sine strategy to boost search process efficiency and reduce the probability of getting trapped in local optima. The improved EDO (IEDO) effectiveness is evaluated in four cases including Case I: classical and CEC-2019 test functions, Case II: simplified and detailed building deterministic energy optimization (Without uncertainty), Case III: commercial, and residential complexes deterministic energy optimization, Case IV: simplified and detailed building stochastic energy optimization (With uncertainty), and Case V: commercial, and residential complexes stochastic energy optimization. The superiority of the IEDO is confirmed by comparing its performance with several well-known algorithms and previous studies in deterministic cases. Moreover, the competitive capability of the IEDO is enhanced in comparison with the previous studies to solve each case. The results demonstrate that IEDO outperforms other methods, reaching the optimal solution with the lowest annual initial energy consumption of 132.7519 kWh/m²a, compared to 132.9356 kWh/m²a for the grey wolf optimizer (GWO) and 133.0000 kWh/m²a for POSCO, as reported in previous studies for a single office building. Additionally, when applied to a detailed office building model, IEDO achieves the lowest annual initial energy consumption of 129.4297 kWh/m²a, surpassing the GWO, which recorded 129.6142 kWh/m²a. Moreover, the achieved outcomes of the stochastic cases incorporating uncertainty demonstrated that the annual energy consumption of the simplified, and detailed office buildings are increased by 2.16%, and 2.08%, and sizing cost of two commercial and residential buildings are increased by 4.30%, and 5.23%, respectively compared with the deterministic model without integrating the uncertainty. The uncertainty influences decision-making in the optimization process, causing trade-offs between minimizing costs and ensuring reliable energy supply, which can lead to both higher operational costs and larger investments in energy system capacity. Therefore, while the stochastic model offers a more realistic system design, it also highlights the challenge of managing uncertainty in building energy optimization.