Study On Settlement Monitoring And Risk Control Of Oil And Gas Pipeline

Quantitative and risk control of pipeline settlement is the key and difficult point of pipeline integrity research.Pipeline settlement caused by external factors is the frontier of current research,and it is also a hot area of oil and natural gas industry.Based on the basic data of stations and pipelines,the risk assessment and prediction models of pipelines are established by using the methods of FLAC^3D,ANSYS finite element analysis and BP neural network.The types and characteristics of pipelines under different working conditions are summarized,and the corresponding suggestions are put forward.Finally,several models are evaluated to find out the factors that have great influence on pipeline settlement and the maximum values of load and falling objects.Finally,after the simulation and evaluation are completed,field experiments are used to verify the accuracy of the model and the reliability of the results.In this paper,it is found that:1)the settlement deformation of pipeline is closely related to the elastic modulus of surrounding soil,and it is approximately inversely proportional.The larger the elastic modulus of soil,the smaller the maximum stress and settlement of pipeline,and the change is non-linear.2)When the pile height is small,the maximum tensile stress and settlement of pipeline are approximately linear ratio with the increase of the upper surcharge.Case enlargement.With the increase of pile height,the linear effect of pile soil on pipeline deformation is weakened.3)Under the impact of falling objects,the maximum impact force of falling objects affects pipeline and soil layer,and the calculation model of impact force and stress is proposed.In the process of pipeline impact,with the increase of soil shear modulus and Poisson's ratio,the maximum Mises stress of pipeline will be generated,which will reduce the impact destructiveness to a certain extent.4)The stress of pipeline in the soil collapse around pipeline is much greater than that of pipeline in the non-collapse area,which can increase the amplitude of 71 MPa under certain conditions,greatly increasing the risk of pipeline;5)Based on neural network model;For the ANASYS and FLAC3D numerical calculation results,the corresponding neural network architecture is established.The stress characteristics of pipeline and the loads of pipeline,soil,surcharge and layout are non-linear.The non-linear relationship is fitted by machine learning of neural network.In this paper,the subsidence monitoring test of oil and gas pipelines is further carried out,and the geological hazard monitoring system of pipelines is established.The real-time data acquisition of mechanical characteristics of pipelines under the influence of static loads,falling objects and other dynamic loads based on the Internet of Things is realized.Compared with the theoretical model,the accuracy of the model is verified.