The Research Of Subsea Pipeline Leak Detection And Location System Based On SPRT And Pressure Gradient Method

As one of the most important method of oil and gas transportation,pipelines are widely used for safety and economic reasons.However,many factors such as the aging of material itself or sabotage by people may lead to leakage of pipelines,which will cause economic losses and environment pollution.In order to alert and locate leakage in time,the method of leak detection and location is studied in this paper,especially pipelines of gas-liquid two-phase flow.A software of leak detection and location is developed based on above theory.The accuracy of the theory and software is verified by experiments and simulations.The main contents of this paper include:(1)This paper summarizes the theory of pipeline leak detection based on sequential probability ratio test(SPRT),puts forward the parameters of leakage judgment and its threshold.Usurally,SPRT is used conbine with mass balance principle for dynamic detiction,in this part,a method used for static detection,conbined SPRT with the pressure balance is generalized.(2)Based on pressure gradient method of pipeline leak location theory.The calculation method of pressure gradient and leakage location under single-phase flow is summarized.For the case of two-phase flow,three models are discussed in detail(Xiao-Shoham-Brill,Beggs-Brill and Oliemans).The judgement of the each flow pattern in different model,and the pressure gradient calculation method of each flow regime are presented.(3)A leak detection software is developed based on the principle of leak detection and location.This software can monitor the operation state of the pipeline in real time,analyze the data and show the calculation result in visual graph.When a leak occurs,the program can alert and estimate size of leakage as well as leakage percentage.The location of leakage can also be targeted.The composition of the fluid,layout of the pipeline and the environment condition can be customized and three models can be choosen.(4)The accuracy of the theory and software is verified by experiments and simulations.For the positioning part,the difference of three different models is discussed.Finally,the accuracy of the localization under longer distance is verified by comparison with other software.