Application Research of Differential Evolution Algoritm in Resistance Coefficient Identification of Heating Pipeline

Abstract

The district heating system is an important heating mode in the northern cities of China. In recent years, the scale of the district heating system is expanding day by day, the pipe network structure is more and more complex. The problem of hydraulic imbalance of the pipe network is gradually emerging, therefore, it is urgent to establish an accurate and perfect hydraulic simulation model of heating network to assist operation management. Pipe network simulation modeling is one of the important prerequisites to solve the hydraulic imbalance problem of heating pipe network. However, with the increase of service time, the actual resistance coefficient of heating network becomes difficult to obtain, which is one of the key reasons for the low accuracy of pipe network simulation model. In order to overcome this difficulty, this paper proposes to use the resistance coefficient identification model based on the differential evolution algorithm (DEA) to identify the resistance coefficient of the heating pipe network. Based on graph theory, network matrix and the law of conservation of mass, the hydraulic model of the heating pipe network is built, and the nodal pressure method is used to solve the model. On the basis of comprehensive consideration of the mainstream intelligent algorithm, the differential evolution method is selected as the algorithm to identify the resistance coefficient of pipeline. In order to verify the identification effect, the feasibility of the model was verified by calculating the data of three different operating conditions of the practical engineering named “K district heating system”. The results demonstrated that the relative errors of the identified resistance coefficients are all within 10, and 98% of the identified values are less than 5%.