Stochastic optimization of energy systems configuration for nearly-zero energy buildings considering load uncertainties

Abstract

The energy systems configuration in nearly-zero energy buildings (NZEBs) has traditionally been optimized under deterministic conditions. However, building energy load often exhibiteds uncertainties in practice, influenced by factors such as occupant behavior and weather conditions. These uncertainties may lead to suboptimal solutions of capacity configuration or failure in achieving building design targets. This study introduces an stochastic optimization method for energy systems configuration that accounts for load uncertainties. The process begins with the characterization of uncertain parameters, followed by the construction of a scenario set, and concludes with the multi-objective optimization within a 70%–90% load guarantee. The NSGA-II, coupled with entropy weight-TOPSIS method, was utilized to formulate and solve the multi-objective optimization problem. This approach was then compared with results obtained under deterministic and robust conditions based on load guarantee rate, cost, and carbon emissions. The results show that the most optimal solution was obtained by the stochastic optimization with a load guarantee rate of 90%, which decreases equipment investment by 58.61% and carbon emissions by 15.8 %, and increases load guarantee rate by 133.69% compared to the initial design. These results underscore the significant effectiveness of incorporating load uncertainties in designing robust and flexible energy systems in NZEBs.