Load forecasting and operation optimization of residential fresh air system based on artificial neural network

Radiant cooling and heating fresh air system is more widely used in residential buildings as a high-comfort, energy-saving efficient air-conditioning system. The handles all the moisture load part of building. In actual operation, there are some problems, such high proportion energy consumption mismatch between operation characteristics. this paper, zone-level artificial neural network (ANN) model established to predict building Compared with measured data, zonelevel ANN verified. total data for training testing 13260 864 respectively. This paper also introduces control optimization model, optimizes combined forecasting results model. Under scenario potential storage time use price, strategy formulated improve flexibility show that has prediction accuracy. root mean square error variation coefficients corresponding 8.72%. can reduce cost by 27.2% 29.2% respectively whole conditioning season.