Research On Energy-saving Transformation And Operation Energy Efficiency Optimization Control Of A Existing Data Center’s Chiller Plant

With the rapid development of digital economy in China,the scale and number of data centers have exploded,and data center energy efficiency optimization has become an important part of "dual carbon" goal.In order to meet the strict requirements of data center’s IT equipment on the environment temperature and humidity of the computer room,the air conditioning system has become the second most energy-consuming system in the data center.However,because of extensive management,the parameter adjustment of the central air conditioning water system are unreasonably adjusted and have low energy efficiency,and the phenomenon of small flow and temperature difference has caused the generally high PUE of the data center.Therefore,a further discussion on the energy-saving modification scheme and energy efficiency optimization control strategy of the existing data center air-conditioning system is of great significance for guiding the energy-saving and optimized operation of the data center cooling system and reducing the data center PUE.In this paper,a data center in hot summer and warm winter zone is studied,The load variation characteristics of data center and the operation status of cold source system equipment are analyzed,and the energy efficiency optimization control scheme of cold source is designed;studies the cooling load forecasting modeling method and energy efficiency optimization control of the central air conditioning system Strategy;Develop a software module for energy efficiency optimization of cold source system operation to provide a reference for the engineering application of energy-saving optimization of cold source system.The main research work of this dissertation is as follows:(1)According to the overall energy consumption status of data center and the operation control level of cold source system,combined with the energy consumption evaluation index of data center,the problems existing in the operation process of air conditioning system are analyzed,and the necessity of cold source system transformation is expounded.By analyzing the energy-saving transformation requirements of cold source system,the energy-saving transformation scheme of cold source system and the energy efficiency optimization control system of cold source are designed,which provides an effective technical means to reduce the PUE of data center.(2)The load of data center has the characteristics of strong periodicity and randomness,and the single load forecasting model is difficult to ensure its load forecasting accuracy.In this paper,a Prophet-SA-SVR combined forecasting model based on load characteristics is proposed.Compared with the single forecasting model,the forecasting accuracy is increased by 28.11 % and 9 %,respectively,which verifies the effectiveness of the combined forecasting model.(3)An energy efficiency optimization control strategy of cold source system based on load forecasting is proposed to optimize the operating parameters of cold source system.For unsteady data and atypical working conditions,quasi-steady state discriminant method based on moving time window and similar working condition matching method based on Mahalanobis distance are proposed respectively to solve the problems of quasi-steady state data screening and atypical working condition matching optimization rules.(4)The “software module of cold source system operation energy efficiency optimization”is developed,which encapsulates data preprocessing,load combination forecasting method and cold source system operation energy efficiency optimization control strategy,and provides convenient load forecasting and operation energy efficiency optimization analysis platform for engineering operators.