Energy Consumption Prediction of District Cooling Systems Using Bp-Ann Algorithm Based on Weather Data

Abstract

To achieve the low-carbon operation and maintenance of building HVAC system, the theoretical basis is accurate prediction of energy consumption. Existing studies either use only partial weather data resulting in low accuracy or need to build a three-dimensional building model which is hard to do for each building. Moreover, the energy consumption prediction model and its main influence parameters determined in the existing research for single building or residential building cannot be directly transferred to the district cooling system of building complex. To achieve the energy consumption prediction of DCS based on weather data, this investigation firstly analyzes the linear correlation between the outdoor environmental parameters with the measured energy consumption data. Then, the energy consumption prediction model based on the back propagation artificial neural network (BP-ANN) algorithm is constructed and the main parameters affecting its prediction performance are determined. Finally, the accuracy of the proposed BP-ANN model is tested. Although the proposed energy consumption prediction model driven by weather data was for a specific building complex, the proposed method is not only helpful to the energy consumption prediction and intelligent operation and maintenance of DCS, but also can be transferred to any building complex.

Keywords: District cooling system, Energy consumption prediction, HVAC system, Building complex, Weather data, BP-ANN model