| The defect geometry of oil and gas pipelines mostly appears in the form of irregular cluster corrosion.However,most of the existing methods for evaluating the residual strength of corroded pipelines with multiple defects are based on a single defect,which results in the evaluation result of the failure pressure of the corroded pipeline is often weaker than that of the actual corroded pipeline.Reasonable determination of the failure pressure of multi-defect corroded pipelines is the key to ensuring the structural integrity of oil and gas pipelines in service.The influence of the interaction of defects and failure pressure under multiple corroded pipelines is studied,and a prediction method for failure pressure of corroded pipelines with multiple defects is established.(1)The finite element model of pipeline with double defects is established.For symmetric and asymmetric longitudinally,symmetric and asymmetric circumferentially and diagonally aligned defects,the influence of the interaction between defects on the failure process and the failure pressure of the pipe is investigated.The influence of defect spacing,defect length and depth on the interaction coefficient between double defects and the failure pressure of the pipe are analyzed.The rule of interaction between double-defect is proposed.The results show that the interaction between axially composite adjacent defects is the most significant.and the interaction between defects is the most intense when the longitudinal overlap area is the sum of the depth of the double defects.The location of the maximum von Mises stress moves with the relative position of the defect for the double defects with interaction.However,the failure position of the pipeline is located in the center of the more severely corroded defect when the distance between adjacent defects is far enough.The effect of defect spacing on the failure pressure of the pipeline is nonlinear.The increase in defect length will weaken the interaction between defects and the failure pressure of the pipeline,and the increase in defect depth will enhance the interaction between defects and reduce the failure pressure of the pipeline.In addition,according to the results of double-defect cases with different defect geometries,the longitudinally critical distance of defects with interaction based on different defect lengths is proposed,and the circumferentially critical distance of defects is determined to be 0.18πD.(2)The nonlinear finite element model of the three-defects corroded pipeline is constructed.Based on the example parameter with the most significant interaction among the defects in the double defects,the interaction between the three defects is studied.It is obtained that the ratio of the failure pressure of the three-defects pipeline to the failure pressure of the double-defects pipeline is 0.9446,and the error is 5%.That indicates that when the number of defects exceeds two,the influence of the third defect on the failure pressure of the pipeline can be ignored.It provides a basis for determining the failure pressure of pipelines with multiple defects.(3)Based on the XGBoost learning algorithm,the prediction model of the interaction coefficient between double defects is established using the calculation results of all double-defect cases as the data set.The variables of the predicted model are defect spacing,defect size and pipe size.The prediction results are compared with blasting tests and common evaluation standards.The average error between the interaction coefficient of double defects obtained by the prediction model and the actual interaction coefficient is 3.27%,which is the highest level of agreement with actual results.(4)Combining the prediction model of the interaction coefficient between double defects and the prediction formula of failure pressure of pipelines with single defects,the prediction process of failure pressure for pipelines with multiple defects is proposed.In addition,the corresponding failure pressure prediction software for pipelines with multiple defects is developed,which provides a convenient and fast calculation method for the residual strength evaluation of corroded pipelines with complex defects. |