Research On Resistance Characteristics And Optimization Of Operation Energy Consumption Of Long Distance Heating Network

Under the background of reducing coal consumption and implementing carbon neutralization,long distance heating network technology has become the key technology to solve the increasingly serious pollution control pressure in winter.With the rapid expansion of heating radius,there are bottleneck problems of transportation capacity and operation problems of optimal regulation.Taking Yinchuan long distance heating network project as the object,this paper studies the resistance characteristics of long distance heating network,and puts forward the corresponding optimization strategy for the actual operation energy consumption.Friction resistance coefficient is a key parameter in the calc ulation of resistance loss of long distance heating network,which is usually calculated by Colebrook-White formula.Colebrook-White formula is an implicit calculation formula,and there is no explicit calculation formula for long distance heating network.According to the literature,when the water supply and return temperature is 20～130 ℃,the pressure is 0.5～2.5 MPa,and the pipe diameter is DN1200～DN1600,i.e.1/3200≤ε/D≤1/2400,within the range of common operating parameters of long distance heating network,20 empirical approximate formulas are calculated.The results show that the calculation stability and accuracy of Biberg formula and Serghides-a formula are very high,and they are suitable for the calculation of friction resistance coefficient of long distance heating network.Based on Serghides-a formula,the calculation model of resistance loss of long distance heating network is established,and the resistance characteristics of long distance heating network are analyzed theoretically.The results show that the pipe diameter,circulating water velocity and temperature of heat ing network have a certain influence on the resistance of long distance heating network.The resistance loss of long distance heating network increases with the increase of heating network temperature and decreases with the increase of pipe diameter and flow v elocity.When the temperature of the heating network is at the minimum design return water temperature of 20 ℃,the change of resistance loss of the long distance heating network is more obvious.In order to reduce the operation energy consumption of long distance heating network,artificial intelligence method is used for modeling.Firstly,the actual operation data are cleaned and 1955 groups of data are obtained.Then the actual operation characteristics of four water supply pumps and four return pumps are studied,and the distribution of current and frequency is obtained.Taking the total current of the supply and return water pump as the output parameter,30 parameters are selected as the input parameters by correlation analysis.Then,a neural network model with 3 hidden layers and 9 nodes per hidden layer is constructed using Keras,the deep learning framework.The average relative error of the model on the test set is 0.85%,and the generalization ability is strong,which can accurately describe the operation energy consumption of long distance heating network.The corresponding operation optimization GUI software is developed.Taking the operation frequency of the water pump as the operation value and the minimum total current of the supply and return water pump as the optimization objective,the particle swarm optimization algorithm is used to optimize the operation energy consumption of the long distance heating network.When the load of the initial station of the heating network is 1200 MW,the total current of the supply and return water pump can be reduced by 28.4 A.When the load of the initial station of the heating network is 800 MW,the total current of the water supply and return pump decreases by 68.89 A.The operation energy consumption under other loads is reduced to varying degrees,which has guiding significance for the actual operation of long distance heating network.