인공신경망을 이용한 건물의 에너지 모델 최적화알고리즘 개발 및 검증에 관한 연구

Abstract

This paper discusses the modeling methodologies and optimization methods for building energy systemusing time series auto regression artificial neural networks. The model can be integrated into energy solution tools forbuilding energy assessment, optimization, and many other applications. The model predicts whole building energy consumptionsas function of four input variables, dry bulb and wet bulb outdoor air temperatures, hour of day and typeof day. To train and test the models, data from simulations are used for the analysis. Advanced computational methodsare used for data analysis and preprocessing. Different neural network structures are tested along with various inputand feedback delays to determine the one yielding the best results. The optimization method was also developed toautomate the process of finding the best model structure that can produce the best accurate prediction against the actualdata. The results show that the developed model can provide results sufficiently accurate for its use in various energyefficiency and saving estimation applications.