摘要
针对海底多相流管道内腐蚀速率的预测问题,首先对影响该种类型管道内腐蚀速率的相关因素进行了分析,对PCA算法、PSO算法和SVM算法分别进行了介绍,提出了可用于海底多相流管道内腐蚀速率预测的PCA-PSO-SVM组合模型,在此基础上使用PCA-PSO-SVM组合模型对44组海底多相流管道内腐蚀速率的影响因素和管道内腐蚀速率数据进行了学习训练,对10组数据进行了预测,并将该组合模型与PCA-GA-SVM模型、PCA-LS-SVM模型和PCA-CV-SVM模型3种预测模型的预测结果进行了对比,以验证所提方法的可靠性和可行性。结果表明:温度对海底多相流管道内腐蚀速率的影响相对较大,压力对其的影响相对较小;使用PCA-PSO-SVM组合模型对海底多相流管道内腐蚀速率预测的平均绝对误差仅为1.848%,模型训练时间仅为3.17 s,这两项数据均小于其他预测模型,表明针对海底多相流管道内腐蚀速率的预测问题,PCA-PSO-SVM组合模型具有可靠性和可行性。
In order to predict the internal corrosion rate of submarine multiphase flow pipelines,this paper analyzes the related factors affecting the internal corrosion rate of this type of pipelines,and introduces PCA algorithm,PSO algorithm and SVM algorithm respectively.Then the paper proposes the PCA-PSO-SVM combination model for prediction of internal corrosion rate of submatine multiphase flow pipelines.Based on the model,the paper studies 44 groups of the influencing factors and internal corrosion rate data of submarine multiphase flow pipelines,and forecasts with 10 groups of data.Finally,the paper compares the prediction results with those from the PCA-GA-SVM model,the PCA-LS-SVM model and the PCA-CV-SVM model to verify the reliability and feasibility of the proposed method.The results show that the temperature has a relatively large influence on the internal corrosion rate of the submarine multiphase flow pipelines.The influence of pressure on the internal corrosion rate of the submarine multiphase flow pipeline is relatively small.On average,the prediction error of internal corrosion rate of the submarine multiphase flow pipelines PCA-PSO-SVM combination model is only 1.848%,and the model training time is only 3.17 s,both of which are smaller than other models.The research proves that the PCA-PSO-SVM combination model has strong reliability and feasibility for the prediction of the internal corrosion rate of submarine multiphase flow pipelines.
作者
王盼锋
王寿喜
马钢
全青
WANG Panfeng;WANG Shouxi;MA Gang;QUAN Qing(College of Petroleum Engineering,Xi'an Shiyou University,Xi'an 710065,Cina)
出处
《安全与环境工程》
CAS
北大核心
2020年第2期183-189,共7页
Safety and Environmental Engineering
基金
国家自然科学基金青年基金项目(51704236)。
