The Intelligent RT Forecaster of Ice-Storage Air-Conditioner by Neural Network

In this paper, an intelligent refrigeration ton (RT) forecaster of the ice-storage air-conditioning system by using neural network is proposed. It is expected to help the operator could precisely set the RT values needed for the air-conditioner and then reach the goal of energy saving. To demonstrate the superiority of the forecaster we developed, the same simulations by using traditional methods, including expert and CLTD-CLF models, are also performed as a comparison. From the simulation results, it could be clearly found that the forecaster we developed indeed greatly improved the accuracy of the RT forecasting.