Method Of Data Reverse Identification For Flow And Quality Operation And Adjustment Strategy Of Central Heating System

The identification of the current operation strategy of a heating system is the basic prerequisite of operation optimization.And the application of heat metering system in the management of building services makes a large number of heating operation data to be recorded and in which the information of system operation regulation is often hidden.Therefore,it undoubtedly is the most convenient and useful program to obtain the operation strategy of heating system through the analysis of these data.However,it’s difficult for the traditional data analysis method to handle these massive and multidimensional time series data.Well the data mining technology which combine the traditional data analysis methods with the complex algorithm of processing massive data provides a good tool for extract useful information from the messy data.Consequently,this paper propose a method to recognize the operation pattern of heating system based on the attention attracting and fast developing data mining from operation data.The mining task is gradually decomposed.Firstly,the wavelet decomposition method and PauTa criterion are used to improve data quality by pretreatment.Secondly,the Unit Root Test method,Clustering Analysis and Decision Analysis method are combined to mining the flow adjustment regulation.Thirdly,the Unit Root Test method,Clustering Analysis,Fourier transform method and Decision Analysis method are combined to identify the supply water temperature regulation mode.Then the correlation analysis is proposed to study the relationship between variables to obtain more system regulation knowledge.Moreover,this paper develops the technological process of the data pretreatment of heating operation data,the identification of flow regulation pattern,and the identification of supply water temperature regulation pattern and the method of correlation analysis of system variable.Finally,the analysis modules are integrated into a data mining framework for the identification of operation pattern based on heating operation data.The data mining framework is applied to make case study of two actual heating station,the results show that the framework can identify the operation regulation mode of the heating system,the rationality and universality of the data mining framework is verified effectively.So the research results of this paper can be used to mining much more heating systems’ operation pattern in the future.Afterwards the work efficiency of the relevant researchers can be improved and there is certain practical significance.