中国安全科学学报 ›› 2017, Vol. 27 ›› Issue (1): 65-70.doi: 10.16265/j.cnki.issn1003-3033.2017.01.012

• 安全工程技术科学 • 上一篇    下一篇

尾段残差修正GM(1,1)模型在管道腐蚀预测中的应用

张新生 教授, 赵梦旭, 王小完   

  1. 西安建筑科技大学 管理学院,陕西 西安 710055
  • 收稿日期:2016-10-08 修回日期:2016-12-28 发布日期:2020-11-23
  • 作者简介:张新生 (1978—) ,男,河南驻马店人,博士,教授,主要从事管道风险评估理论、建模与方法、智能信息处理等方面的教学和科研工作。E-mail: xinsheng.zh@outlook.com。
  • 基金资助:
    国家自然科学基金资助(61271278);陕西省社会科学基金资助(2015R13);陕西省自然科学基金资助(2016JM6023);陕西省教育厅自然专项基金资助(16JK1465)。

Application of modified empennage residual error GM(1,1) model in prediction of pipeline corrosion

ZHANG Xinsheng, ZHAO Mengxu, WANG Xiaowan   

  1. School of Management, Xi'an University of Architecture and Technology, Xi'an Shaanxi 710055, China)
  • Received:2016-10-08 Revised:2016-12-28 Published:2020-11-23

摘要: 为解决海底油气管道由于外表面破损引起腐蚀加快,进而导致管道腐蚀失效的问题,基于传统灰色模型,建立尾段残差修正GM(1,1)模型,用以预测管道剩余寿命。首先,检验管道腐蚀深度数据的光滑性和准指数规律性,建立灰色微分方程;然后利用最小二乘法求出方程参数值,用传统灰色模型预测腐蚀深度,并对残差进行修正,从而得到一个完整的用于海底管道腐蚀趋势预测的尾段残差修正GM(1,1)模型,并对预测结果进行后验差检验。最后以某一海底管道试验段为例,预测管道剩余寿命。结果表明,传统管道腐蚀深度预测灰色模型预测相对误差为36.7%,尾段残差修正GM(1,1)模型预测的相对误差为3.79%,后者预测精度等级更高。

关键词: 海底管道, 尾段残差修正GM(1,1)模型, 坠物风险, 腐蚀, 剩余寿命预测

Abstract: In order to solve the problem that the corrosion of the offshore oil and gas pipeline is accelerated by the damage to the outer surface, the paper was aimed at building a modified empennage residual error GM(1,1) model based on the traditional gray model to predict the residual life of pipelines. Firstly, a gray differential equation was established after testing the smoothness and quasi-exponential regularity of pipeline corrosion depth data are tested. Secondly, a complete modified gray model of submarine pipeline corrosion was obtained by calculating the gray differential equation parameters with a least square method and correcting the residual error through predicting the corrosion depth. The model was used to predict the remaining life of a certain test submarine pipeline section as an example. The results show that the relative error of gray model prediction is 36.7% and that of forecasting GM (1,1) is 3.79%, so the prediction accuracy is higher and the life prediction is more reasonable.

Key words: submarine pipeline, modified empennage residual error GM(1,1) model, falling object risk, corrosion, remaining life prediction

中图分类号: