Heat supply prediction method of a heat pump system based on timing analysis and a neural network

• DOI: 10.1016/j.enbenv.2024.02.005
• Publication Date: 2024-02-29T07:11:56Z
• Authors: Xin Liu; Xiuhui Wu; Jingmeng Sang; Kailiang Huang; Guohui Feng; Mengmeng Song; Xiangdong Wang
• Keywords: Adaptability

The prediction of heat pump system has more complicated characteristics, and the accuracy existing single model is not ideal. From perspective energy efficiency consumption, it necessary to improve prediction. A sewage source in Shenyang, China, was used as research object this paper. ARIMA model, BP neural network ARIMA-BP integrated were built. predicted values supply obtained by models verified. verified extreme weather. completeness validation improved. Three had been applied water soil system. adaptability generalization number training sets for divided. at beginning heating season analyzed. results showed that mean absolute percentage errors 5.37 %, 5.97 % 3.21 respectively. root square 177.31, 186.98, 139.44, a improved 2.16 compared model. 2.76 In weather, error 7.83 296.42. overall also within reasonable range. high good applicability generalization. At season, can be better when 4 days.