Research On Technology Of Pipeline Leak Detection And Localization Based On Steady-state Modeling Method

As a significant way of transportation, pipeline transport has important strategic position in the national economy. However, pipeline leak occurs frequently, which not only causes severe economic losses, but also courses environmental pollution. Therefore, pipeline leak detection and location have significant meaning and value. At present, pipeline leak detection method based on steady-state modeling method has made some progress. As distributed parameter systems, the pipes use a number of bent pipes which influences the distribution of friction coefficient. The uneven distribution of friction coefficient causes low precision of leak detection. Therefore, the study of correction method of local resistance coefficient of bend pipe, eliminating the affect of uneven distribution of friction, and improving accuracy of leak detection and location based on steady-state model have important theoretical significance and application value.On the basis of detailed analysis of the pipeline fluid dynamics, this paper presents a method of pipeline leak detection and location based on stable model; analyze the impact of uneven distribution of the pipe friction on pipeline leak detection and location, combine with simulation study of finite element analysis software ANSYS, given long-distance pipeline section method and the friction coefficient calculation method. Based on a lot of simulation experiments, the estimating method based on least squares of frictional resistance factor and pipeline friction coefficient correction method combined with leak posteriori information are given, and improved pipeline leak detection and location method based on Steady-state Model. At last, position correction formula is given.The experimental results show that the improved steady-state modeling method can effectively reduce the pipeline location error caused by bent pipes. The improved leak detection and location method based on steady-state model can improve pipeline leak location accuracy. The relative positioning error is less then 0.5%.