Research on energy-saving optimization of commercial central air-conditioning based on data mining algorithm

Central air conditioning accounts for more than 40 % of the energy consumption of social buildings. Ensuring the efficient operation of central air conditioning at full working conditions has a positive impact on reducing the total energy consumption of society. In this paper, the historical operation data of central air conditioning in a commercial building in Shenzhen is used as the source and the energy -saving strategy of this central air conditioning system is studied using 2019 data as the main research data. Firstly, the measured parameters are extracted as the Cooling factor, Delivery factor, and Load factor based on factor analysis, and the K-Means algorithm is used as input parameters for pattern recognition of the central air conditioning system. Then, according to the results of pattern recognition, C5.0 and the CHAID decision tree algorithm are used to obtain the energy-saving strategy model. Finally, the 2020 operational data of the building's central air conditioning system was used as the validation data set for energy efficiency verification. The energy-saving strategy based on the C5.0 decision tree algorithm is calculated to save 78,183 kW?h, which is 32.4 % of the total energy consumption, and the energy -saving strategy based on the CHDIA decision tree algorithm is calculated to save 73,182 kWh, which is 30.3 % of the total energy consumption. (c) 2022 Elsevier B.V. All rights reserved.

Automated summary

This paper studies an energy-saving strategy for a central air conditioning system and validates it using historical operational data. Based on pattern recognition and decision tree algorithms, two energy-saving strategies are proposed, which can save 32.4% and 30.3% of the total energy consumption respectively.