Research On Prediction Of External Corrosion Depth Of Overhead Pipeline In Marine Atmospheric Environment

With the continuous development of offshore oil and gas resources,the construction of overhead pipelines has developed rapidly.Offshore oil and gas pipelines are the lifeline of offshore oil and gas,and many overhead pipelines in China are often distributed near seaports due to transportation tasks,so as to facilitate the transportation of petroleum and natural gas chemical products.However,under the action of high temperature,high humidity,high salt fog concentration and strong ultraviolet light,overhead pipelines exposed to marine atmospheric environment for a long time are prone to corrosion deterioration,and pipeline corrosion failure will cause a series of problems.Therefore,in this paper,the corrosion depth of overhead pipelines in marine atmospheric environment is predicted and studied.The research content of this paper mainly includes three parts: the first part,the influencing factors of overhead pipeline corrosion in marine atmospheric environment are quite complicated,so the grey correlation analysis is carried out on the influencing factors of overhead pipeline corrosion in marine atmospheric environment.Because of the defects of the traditional grey correlation method,the solution of its correlation coefficient is limited,and it is impossible to analyze the weight of each corrosion influencing factor.In this paper,the calculation process of grey correlation coefficient is improved,and the entropy weight method is combined with grey correlation degree,and then the corrosion index system of overhead pipeline is established by using the improved grey correlation method.In the second part,there are few corrosion data of overhead pipelines in marine atmospheric environment,so the DGM(1,1)pipeline corrosion depth prediction model is established,and the DGM(1,1)model is improved by exponential transformation,variable weight weakening buffer operator and metabolism.For the optimization of DGM(1,1,λ)model parameters,the IWOA algorithm is used for optimization,and then the IWOA-DGM(1,1,λ)model is compared with DGM(1,1)model and WOA-DGM(1,1,λ)model.In the third part,the multi-variable NSGM(1,N)model is used to predict the corrosion depth of pipelines.Aiming at the defects of the model,the real-number gray generation operator is used instead of cumulative generation to improve the accuracy of the model,and the background value is optimized to improve the stability of the model.The performance of INSGM(1,N)model is compared with other models.Therefore,the corrosion depth of overhead pipelines in marine atmospheric environment is studied and analyzed in many aspects.The research results show that the improved univariate and multivariate corrosion depth prediction models show good prediction effect and model characteristics,and the statistical evaluation indexes of the two optimized and improved combined models are superior to other comparative models.The improved discrete grey prediction model makes up for the deficiency of GM(1,1)model in long-term prediction,and the improved multivariate grey prediction model overcomes the parameter defects and structural defects of GM(1,N)model,and improves the prediction accuracy.Two optimized and improved combined models enrich the prediction method of corrosion depth of offshore overhead pipelines,and provide theoretical reference for corrosion maintenance and protection of overhead pipelines in marine atmospheric environment.