Optimization Design Of Complex Central Heating Network

Aiming at the complex branch central heating system,taking the sum of the annual construction cost,depreciation cost,heat loss cost and energy consumption cost of the heating system pipe network as the optimization objective,the design optimization of the heating pipe network is carried out on the premise of ensuring the reasonable and safe operation of the pipe network system without hydraulic imbalance.A new pipe network structure model is proposed to simplify the complex central heating pipe network into multiple groups of parallel pipe networks,which reduces the scale of pipe network model.Based on the simplified model of parallel pipe network,a unified formula of correlation matrix and basic circuit matrix is proposed,which improves the modeling efficiency of different forms of branch heating pipe network.According to graph theory and Kirchhoff's law,the matrix equations of water conservancy calculation in the operation of central heating network are listed.The matrix equations are transformed into ordinary equations by the basic idea of solving linear equations,and then the equations are calculated according to the principle of iterative method.The outdoor temperature,relative humidity,wind speed,wind direction,precipitation and other data of a residential area in Shijiazhuang City in 2018-2019 are selected to predict the building heat load in a short term.The parameters needed for prediction are determined by correlation analysis,the best prediction model is determined by BP neural network,and the accuracy of prediction is enhanced by using genetic algorithm to optimize BP neural network and time-delay neural network.The results show that the correlation analysis is helpful to determine the prediction model,and the genetic algorithm can improve the accuracy of BP neural network.When the number of hidden layers and node delay is 9 and 6 respectively,the prediction results of the model are the best,and the mean square error and the average absolute percentage error are 0.021 7 GW and 2.148 6 respectively.The time-delay neural network is used to predict the building heat load and determine the design flow of the pipe network system.The optimization method of the pipeline network structure model,the hydraulic calculation model and the cost optimization model is proposed.Taking the low temperature two times heating network of the district as an example,the economy,energy saving and practicability of the algorithm are proved.The results show that the cost is reduced by 24.68 %,the power consumption is reduced by 27.89 %,and the flow rate and hydraulic balance of the pipe network meet the requirements,which has good practical application value.