Research On The Methods Of Heating Parameter Prediction And Regulation Based On Heat Consumer’s Demand

On-demand heating is an important technological means to improve energy utilization efficiency,improve user’s satisfaction,reduce heating consumption and alleviate air pollution in heating season in heating cities.It is also one of the main tasks of heating system under the goal of “double carbon”.In order to achieve this goal,the heat source,heating substation and heat user should coordinate their actions,and the heating parameters should be changed based on the the user’s demand.But in actual engineering,due to the lack of indoor temperature that can accurately reflect the user’s heating effect,the lack of professional interpretable and high precision parameter prediction model and accurate regulation model in the heating substation,and the deviation of the actual heating parameters from the demand value,most heating systems have not realized on-demand heating.Therefore,this paper carries out relevant research on the above issues,and the main research content and achievements are as follows:(1)In view of the random installation of typical indoor temperature monitoring,and the low accuracy of calculation methods may lead to misjudgment of heating effect by operation managers.Taking the residential building whose heating terminal is radiator as an example,the correlation analysis method and clustering algorithm are used to analyze the indoor temperature changes of neighboring households and households in different locations respectively.Based on the analysis results,the installation position,number and comprehensive indoor temperature calculation model of typical indoor temperature monitoring points are determined.The case application shows that compared with the traditional average method,the calculated result of the proposed model is closer to the actual temperature,and can better reflect the real heating effect of the building,and it is less affected by the instability of communication.(2)In view of the historical operation data is directly used for the training of heating parameter prediction model,without considering the rationality,and the prediction model lacks professional interpretation,which may lead to the poor effect in practical engineering.Based on the heating professional mechanism and the characteristics of the actual operation data,the evaluation method of the historical data is put forward.The core of this method is to evaluate the historical data of the heating station and obtain the relationship between the outdoor temperature and heating parameters in different indoor temperature interval.The historical data with and without evaluation method were used to train the prediction models,and the prediction accuracy and professional mechanism were compared.The results show that the prediction model based on the data evaluation method has higher accuracy and stronger professional interpretability.(3)In view of in the existing heating parameter prediction model,only outdoor meteorological parameters,historical operation data and indoor temperature are considered,heat user behavior is not taken into account.There may be poor adaptability for the system that users participate in the regulation.Taking the on-off control system as an example,the main influencing factors of heat user behavior,including outdoor temperature and solar radiation intensity,are analyzed.On this basis,the heat user behavior is introduced into the heating parameter prediction model of heat substation.Compared with the traditional prediction model,the prediction error of the proposed model is reduced.In addition,the prediction error of the model is further reduced and the accuracy is significantly improved after the clustering method is adopted to identify the heat user behavior.(4)In view of the existing feedforward prediction model of heating station,which does not consider factors such as building thermal inertia and indoor temperature,which may lead to high heating consumption and poor indoor temperature stability.Sensitivity analysis method is used to establish the secondary supply temperature feedback prediction model based on building thermal inertia,the cross-correlation analysis method was used to determine the adjustment cycle and time of heating parameters,and the professional knowledge was used to establish the correction model of solar radiation intensity,outdoor temperature uncertainty and heat user behavior on heating parameters,which was used to modify the prediction model,so as to realize the closed-loop control.The case application results show that,compared with the traditional empirical adjustment method,the model can not only guarantee the operation stability of pipe network and indoor thermal comfort,but also has significant energy saving effect.(5)In view of the problem that heat source heating parameters of large heating system determined by empirical value method and design heat index method often do not match the demand value.The operation model of heat substation is established with professional knowledge,the pressure variation of system under different heat source heating parameters is studied.The monitoring parameters of heat source were analyzed by clustering method,and the operation strategies of different heating period were identified,and the rationality of the clusters was diagnosed by professional knowledge.On this basis,the association analysis combined with professional knowledge was used to identify the heat source parameters successfully,and the on-demand heating parameters of each cluster were obtained.