Energy-saving diagnosis model of central air-conditioning refrigeration system in large shopping mall

Abstract

Aiming at the problems of extensive management and energy waste in the operation of shopping malls, this paper proposed a shopping mall building energy-saving diagnosis model based on an improved PSO-SVM neural network. Firstly, the operation data with good energy-saving characteristics were selected from the collected historical data of energy consumption by clustering algorithm as the data basis for the establishment of the model. Secondly, the Adaboost algorithm was used to optimize the PSO-SVM neural network to complete the model construction, and the energy-saving data was used to complete the training of the model. Finally, the judgment criteria for abnormal energy consumption were established, and the energy consumption situation at the current moment was diagnosed by using the trained energy-saving diagnosis model. The abnormal judgment criteria of energy consumption were used to analyze the diagnostic results, and the energy-saving diagnosis process was completed, which was verified by case analysis. Through case analysis and verification, the energy-saving rate of the energy-saving diagnosis model established in this paper has been increased from -1.1% to 11.9%, and the problem moments generated during the operation of air conditioning in shopping malls have been successfully diagnosed. The model established in this paper can not only find out the unreasonable conditions in the operation of the air-conditioning system, but also provide a reference for building energy-saving management and operation.