西南石油大学学报(自然科学版) ›› 2021, Vol. 43 ›› Issue (1): 149-156.DOI: 10.11885/j.issn.1674-5086.2020.06.29.03

• 石油与天然气工程 • 上一篇    下一篇

输气管道泄漏模型不确定性因素敏感性分析

安建川   

  1. 中国石油西南油气田分公司输气管理处, 四川 成都 610213
  • 收稿日期:2020-06-29 出版日期:2021-02-10 发布日期:2021-01-23
  • 通讯作者: 安建川,E-mail:anjc@petrochina.com.cn
  • 作者简介:安建川,1970年生,男,汉族,四川广安人,高级工程师,主要从事天然气管道管理方面的研究工作。E-mail:anjc@petrochina.com.cn

Sensitivity Analysis for Uncertain Factors of Gas Pipeline Leakage Model

AN Jianchuan   

  1. Gas Management Office, PetroChnia Southwest Oil&Gas Field Company, Chengdu, Sichuan 610213, China
  • Received:2020-06-29 Online:2021-02-10 Published:2021-01-23

摘要: 输气管道泄漏模型中不确定性因素的分析是影响管道泄漏风险评价不确定的主要原因。为此,在确定起点压力、泄漏孔面积、气体温度和泄漏点距起点距离4项为影响输气管道泄漏速率计算的主要不确定性因素基础上,考虑多因素相互关联、共同影响模型计算精度的问题,采用多因素敏感度整体分析方法分别抽取数量为500,1 000,2 000和3 000的样本各3组,研究各不确定性因素对泄漏速率的影响程度,分析表明,不同样本量下计算得到的各因素Spearman秩相关系数排序基本一致,说明泄漏孔面积对泄漏速率的影响最大,其次是气体温度,起点压力和泄漏点距起点距离对泄漏速率的影响较小;样本数量为500时得到的不确定性因素Spearman秩相关系数排序与样本数量为1 000,2 000和3 000时得到的不确定性因素Spearman秩相关系数排序趋势相同,说明在现有因素取值区间内,抽样数量大于500时的模型计算结果已具备可靠性,能够满足后续风险分析的精度要求。

关键词: 输气管道, 泄漏模型, 不确定性因素, 敏感性整体分析, 风险评价

Abstract: The main cause for affects the uncertainty of pipeline leakage risk assessment is the analysis for uncertain factors of gas pipeline leakage model. For this purpose, on the one hand, four factors which were inlet pressure, leakage hole area, gas temperature and distance from the leakage hole to the pipe inlet were determined as the main uncertain factors affecting the calculation of gas pipeline leakage rate; on other hand, the multi-factor sensitivity analysis method was used to study the effects of uncertainty factors on the leakage rate by extracting twelve groups of samples with 500, 1 000, 2 000 and 3 000 samples. The results showed that:the rank of Spearman rank correlation coefficient of uncertain factors calculated under different sample sizes is basically consistent, that is, the leakage hole area has the greatest influence on the leakage rate; the gas temperature the second, the inlet pressure and the distance from the leakage hole to the pipe inlet has the third influence; within the existing value range of uncertain factors, the calculated results of the gas pipeline leakage model could meet the accuracy requirements of subsequent pipeline leakage risk analysis when the number of samples is greater than 500, because of the same rank trend of Spearman rank correlation coefficient among different samples.

Key words: gas pipeline, leakage model, uncertainty factors, multi-factor sensitivity analysis, risk assessment

中图分类号: