Monitoring,Evaluation And Analysis On Property Of Chilled Water Storage

As an effective means of relieving the pressure of the power system,chilled water storage plays an important role in peak shaving and valley filling.At present,although the technology of chilled water storage has been developed and popularized in our country,the research on the monitoring data of chilled water storage has not yet started.At run time,a large amount of data will be accumulated in a short time.How to use these monitoring data to evaluate the chilled water storage system,and to forecast the cold load efficiently and accurately has become a new research topic.On the basis of time series analysis,the paper makes pretreatment,abnormal detection and trend prediction analysis in turn,and the three processes are connected with each other to form a monitoring model of the chilled water storage.In this paper,the principle of the chilled water storage is introduced.Secondly,the monitoring model of chilled water storage is deeply studied: In the pretreatment part,the recognition method of outlier based on relative variance rate of time series and the method of filling up vacancy based on the nearest neighbor are introduced,which provide data basis for anomaly detection and load forecasting;In the anomaly detection section,on the basis of local outlier factor(LOF),the local outlier coefficient(LOC)is introduced to analyze monitoring data of the chilled water storage which can improve the performance;In the load forecasting section,the analog complexing algorithm is used to find the most similar analog sequences from the historical data to generate the prediction.Finally,by integrating all algorithms above,the chilled water storage monitoring system based on time series analysis is built on the platform of Matlab.Then the model is validated by using real-time monitoring data of a software park in Beijing,and the conclusion is consistent with the actual situation.It is proved that time series analysis is feasible in the chilled water storage monitoring system.In addition,based on the above algorithm,the function interface can be built to monitor and display the running state of the chilled waterstorage in real time.The chilled water storage analysis system integrating monitoring and forecasting is formed.Finally,the economic analysis of the summer operation is made,and the conclusion is drawn that the system saves 210 thousand yuan compared with the conventional system.The application of time series makes a new way for the monitoring of water storage air conditioning system,and overcomes the inefficiency of manual detection.The analysis method of chilled water storage monitoring system is a new attempt and exploration for the field of the chilled water storage.