Correlation Analysis Of Water Temperature Data Of The Second Network Of The Heat Exchange Station Of The Heating Syste

Heating project is an extremely important cause of people’s livelihood,which is related to the quality of life of the broad masses of people.In recent years,the mainstream of world development is to improve energy efficiency and promote energy conservation and emission reduction.The regulation of heating system usually includes one network regulation and two network regulation.Compared with the primary network,the secondary pipe network has a wider distribution range and has greater energy-saving potential.The water temperature of the second network includes the water supply temperature and the return water temperature of the second network,and the difference between them is called the temperature difference between the supply and return water of the second network,which is the direct embodiment of the heat release and heat release efficiency of the second network.It is of great significance in the intelligent water temperature control of the second network of heating system.Therefore,the study of the correlation between the temperature difference of the second network and other factors is of great reference significance to realize the accurate regulation and control of the water temperature of the second network.Based on the heating and weather data recorded by a heating company in Changchun City,this paper probes into the correlation between the temperature difference between the supply and return water of the second network and other factors from three dimensions: linear correlation,causality and multifractal crosscorrelation,and establishes a BP neural network model to fit the temperature difference between the supply and return water of the second network determined by other factors,the purpose of this model is to explore whether there is some unknown nonlinear relationship between the whole influencing factors and the temperature difference between the supply and return water of the second network,so as to lay a theoretical foundation for the accurate regulation and control of the water temperature of the second network.Through empirical analysis,this paper finds that,from the point of view of linear correlation,the temperature difference between the supply and return water of the second network has a strong linear correlation with the instantaneous heat of the first network,the instantaneous flow of the first network,heat metering and outdoor temperature.From the point of view of causality,the variable frequency is a white noise sequence,so it is not studied.Among other factors,the Granger cause of the first network supply and return water pressure difference and the air humidity is the Granger cause of the second network supply and return water temperature difference,all other series are the Granger cause of the second network supply and return water temperature difference,which are helpful to explain the future change of the second network temperature difference.From the multifractal point of view of correlation,except that the four series of the first network supply and return water pressure difference,the second network instantaneous flow,the frequency conversion number and humidity do not meet the non-stationary research conditions,there is a multifractal cross-correlation between the other series and the temperature difference between the supply and return water of the second network.In addition to power consumption,the multifractal cross-correlation between other factors and the temperature difference between supply and return water of the second network shows long-term and continuous characteristics.Through the established model,we can find that there is a close non-linear relationship between the whole influencing factors and the temperature difference between the supply and return water of the second network.And the BP neural network model optimized by PSO algorithm is more accurate and efficient,and is more suitable for the modeling of this kind of problems,so it can provide some reference for the intelligent regulation of water temperature in the second network.