为提升海洋环境下深水区立管腐蚀速率的预测精度，建立基于改进秃鹰搜索算法(Improved Bald Eagle Search，IBES)的EDGM(1，1，p)腐蚀速率预测模型。通过Sine混沌映射初始化种群、莱维飞行策略、折射反向学习策略和柯西高斯变异策略提高秃鹰搜索算法的寻优能力和收敛速度;利用IBES算法优化EDGM(1，1，p)中的参数p，建立IBES-EDGM(1，1，p)模型以提高立管腐蚀速率的预测精度。以南海某海洋深水区立管数据为基础进行腐蚀速率预测，分析对比3种模型的预测结果。结果表明，优化后的模型与原模型相比误差更小，且预测精度得到了提高，能够更准确地预测深海立管的腐蚀速率，为后续管道系统的维修和更换提供理论参考。
Prediction of corrosion rates of risers in deep water in the marine environment
Corrosion of oil and gas pipelines is a major cause of pipeline failure,and although research on corrosion prediction of buried and subsea pipelines is relatively mature,little research has been conducted on marine risers.In this paper,the corrosion of risers in deep water under a marine environment is studied.To improve the prediction accuracy of the corrosion rate of risers in deep water under the marine environment,an EDGM(1,1)corrosion rate prediction model based on an Improved Bald Eagle Search(IBES)algorithm is established.The initial population layout of the Bald Eagle Search(BES)algorithm is optimized by initializing the population with sine chaotic mapping to enhance the optimization capability.A Lévy flight strategy is used to optimize the problem of the population converging too early and falling into a local optimum when selecting spatial solutions for the Condor Search algorithm.To address the problem of the search algorithm falling into a local optimum late in the iteration,the refractive backward learning strategy and the Corsi-Gaussian variational strategy are used to optimize the search algorithm to improve the optimization capability and convergence speed of the Condor search algorithm.The IBES algorithm is used to optimize the parameter ρ in EDGM(1,1,ρ)to establish the IBES-EDGM(1,1,ρ)model to improve the prediction accuracy of the riser corrosion rate.The corrosion rate prediction was carried out based on riser data from a deep-water marine area in the South China Sea,and the prediction results of the three models were analyzed and compared.The results show that the IBES-EDGM(1,1,ρ)model has an EMAE of 0.030 7,ERMSE of0.009 8,and FD of 0.945 0,which are all better than the EDGM(1,1)and BES-EDGM(1,1,ρ)models.The constructed IBES-EDGM(1,1,ρ)model has less error and improved prediction accuracy compared to the conventional model and can predict the corrosion rate of deep-sea risers more accurately.In the absence of a large amount of corrosion data,this model can be used to predict the corrosion rate of key corrosion areas of risers in deep-sea areas,which can provide a theoretical reference for subsequent risk warning assessment and maintenance of marine pipeline systems.
safety engineeringmarine environmentdeep-water risercorrosion rateImproved Bald Eagle Search(IBES)algorithmEDGM(1,1)model
NETL
NSTL
万方数据
维普
