Research On Method Of Hydraulic Balance And Thermal Regulation Of Heating System

In recent years,while benefiting from the rapid economic and technological development,China is also perplexed by the increasingly severe energy crisis and environmental pollution problems.China’s heating enterprises largely maintain the traditional central heating method,instead of effectively taking advantage of heating data,which results in loads of waste in the heating process.The advance of information technology and Internet of Things has brought changes to the traditional heating industry.Combined with database,large-scale data storage and data analysis technology,heating mines and analyzes some important information hidden behind a large number of data and uses machine learning method to build models,monitor and improve the heating process for the realization of precise and personalized heating to further the development of heating enterprises and improve the quality of heating,so as to achieve the purpose of saving energy and reducing environmental pollution.Focusing on the data analysis of heating system,this paper can be mainly divided into three parts: hydraulic calculation of heating system,thermal calculation and heating data analysis system.First of all,this paper aims to solve the problem of hydraulic balance of heating system by obtaining the optimization objective of optimal flow distribution.The difficulty lies in the fact that in terms of engineering,the heating system has a long and complicated pipe network,which leads to restrictions in measuring the pipe network and high measurement cost.Therefore,the expected calculation,prediction and optimization cannot be achieved due to limited measurement data.In terms of calculation,the objective function of hydraulic calculation is complex and the solution process is cumbersome.Thus,it is difficult to get a reasonable flow distribution method through formula calculation.This paper analyzes the hydraulic calculation differences and relations between looped network and branch network of the heating system,elaborates on the method and relevant formula of flow distribution in pipe network,and puts forward the idea of reverse hydraulic calculation with resistance coefficient of pipe section as the target to be optimized.In addition,the modeling work is carried out,with the hydraulic calculation optimization model being established and the objective function and target for optimization being set.Bayesian Optimization is then used to optimize the resistance coefficient with the valve opening adjusted to achieve the goal of hydraulic balance.Secondly,the intelligent heating system needs to accurately predict the temperature and heat load in the heating system through thermal calculations.The difficulty lies in the fact that in terms of engineering,it is complicated to collect data of the heating system for it is greatly affected by noise.What ’ s more,the operation of the heating system is also under the influence of the time delay.In terms of calculation,there are many non-linear relationships and complex functions,which are difficult to solve with traditional calculation methods.This paper proposes a thermal calculation method based on the BP neural network to solve the time delay model.Taking heating parameters as input data,this model uses a fully connected BP neural network to determine the optimal network weight coefficients through continuous learning,which is used to guide heating process.At the same time,considering the impact of time delay on the heating system,the pruning algorithm is used to reduce the complexity of the solution.Furthermore,the thermal calculation model based on neural network is simulated with the combination of measured data and simulation data,and the algorithm in this paper is compared with the support vector machine model and the least square support vector machine model.Finally,the heating data analysis system is introduced in this paper.In particular,a heating data analysis system based on online analysis technology is proposed,which can effectively reduce energy consumption,improve environmental protection,enhance communication between heating providers and users,and further the ability of urban heating management compared with the traditional central heating method.