Design optimization and uncertainty analysis of multi-energy complementary system for residential building in isolated area

Abstract

Due to extreme climatic and geographical factors, conventional power sources, which typically use large power-grids, have difficulties to supply energy to isolated areas. In addition, many isolated areas rely mainly on fossil fuel to ensure adequate energy-supply, which involves transport across long distances and high costs. Thus, finding a multi-energy complementary system (MECS) for residential buildings can help ensure the survival and improve the well-being of about 1.5 billion people that are currently living in isolated areas worldwide. In this study, a rough interval-Copula stochastic planning (RI-CSP) programming model is proposed to address this problem. Its purpose is to optimize the MECS in isolated areas, with the aim to reduce cost and provide high-quality energy. RI-CSP can deal with multiple uncertain parameters that are characterized by rough intervals and probability distributions in the MECS. It can also reveal the combined effect of different random variables. To prove the feasibility of RI-CSP, it was applied to a residential building on Yongxing island in the South China Sea. 24 different scenarios were selected that consider four renewable energy scarcity levels and 2 constraint-violation levels (p). Then, the optimal scenarios with minimum total system costs were selected. This study can help decision makers find better configurations for many alternative power-generation scenarios. The results show that the uncertainties that exist in the MECS have a significant effect on the power-generation configuration and total system cost. The results indicate that the total cost of MECS decreases with the increase in scarcity levels. Our results also revealed that, as the scarcity levels for renewable energy decrease, the fraction of PV power-generation gradually increases.