Risk Control of Energy Performance Fluctuation in Multi-Unit Housing for Weather Uncertainty

With the acceleration of urban development, population density cities has increased. As spatial characteristics multi-unit housing (MUH) perfectly fit this developmental trend and, simultaneously, have high energy efficiency, number MUHs increased rapidly in recent decades. Although many studies proposed efficiency strategies, weather uncertainty leads to errors between operational performance building and simulated values. This study introduces a robust optimization framework that incorporates considerations into process suppress consumption fluctuations improve average performance. Neural networks are used model multiple elements as normal distributions for each hour, accuracy is validated by calculating mean absolute percentage error (MAPE) values distribution measurement values, which ranges from 3% 13%. The clustering algorithm replace sampling method complete work space serve subsequent process. Compared with traditional method, results show better representativeness sample space. optimal scheme decreases 13.4%, standard deviation approximately 17.2%, means scheme, generated study, lower more stable face uncertainty.