Heat load prediction in district heating and cooling systems through recurrent neural networks

Abstract

Heat load is the amount of cold water, hot water and steam used for air conditioning in a district heating and cooling system. Heat load prediction in district heating and cooling (DHC) systems is one of the key technologies for economical and safe operations of DHC systems. The heat load prediction method through a simplified robust filter and a three-layered neural network has been used in an actual DHC plant on a trial basis. Unfortunately, however, there exists a drawback that its prediction becomes less accurate in periods when the heat load is non-stationary. In this paper, for adapting the dynamical variation of heat load together with a new kind of input data in consideration of the characteristics of heat load data, a novel prediction method through a recurrent neural network is presented. Several numerical experiments with actual heat load data demonstrate the feasibility and efficiency of the proposed method.