Assessment and prediction of leakage degree of expansion joints in underground network heating pipe network based on LS-SVM and ARIMA model

The study shows that the leakage of the expansion joints in the underground heating primary pipe network will lead to the fluctuation of soil temperature and conductivity around the leakage point. Based on the online monitoring of changes of physical quantities around the expansion joint, this paper established the assessment model of temperature and conductivity as the input parameters and the degree of expansion joints leakage as the output parameters, which took the radial basis function as kernel function and made use of PSO algorithm to achieve the optimization of the kernel function parameters. Then, the temperature and conductivity were predicted based on the ARIMA model, and the LS-SVM assessment of expansion joint leakage was realized according to the prediction results. Both simulation and field experiments show that the prediction model based on ARIMA has higher accuracy, which has identified the foundation for on-line prediction and accurate assessment of the expansion joint leakage in the underground heating primary pipe network.