Conference Year
July 2018
Keywords
central cooling plants, association rules, operational parameter optimization, apriori algorithm
Abstract
More than 50% of the total energy consumption of central air-conditioning system is consumed by the central cooling plants. It is crucial to optimize central cooling plants operation parameter settings which is also significant for improving its operating efficiency, reducing the energy consumption, and promoting the overall energy saving of air-conditioning systems. The regular methods of central cooling plants optimization can be divided into three categories: engineering method, mechanism modeling and artificial intelligence modeling. In recent years, with the development of the internet of things, the monitoring and control platform for air-conditioning system provides data mining with mass ground truth data for central cooling plants optimization. Compared with the other methods, the data mining method for optimizing the key operation parameters of central cooling plants takes the advantages of simple, wide applicable and practical. In this paper, the association rule data mining method is proposed to optimize the operation parameters of the whole central cooling plants from the ground truth data. The central cooling plants in a shopping mall in Guangzhou is taken as the case study. Through historical data processing, like data cleaning, selection of optimization parameters, discrete transform of data and so on, the association rules are mined between the optimal energy efficiency Ratio and the running parameters of the central cooling plants under different operating conditions by Apriori algorithm. Finally, from the simulation results, it’s shown that by the association rules, the total energy consumption of the whole central cooling plants under two different working conditions are reduced 13.33% and 11.6% less than by the original operational parameters in the transition season and summer respectively. The simulation results verify the validity of the mining rules. This method excavates the energy saving potential of central cooling plants from the point of view of engineering practice, which is suitable for the central cooling plants which has accumulated a large amount of operation data and provides a reference for the energy-saving optimization operation of central cooling plants.