基于支持向量机的油气管道安全监测信号识别方法.docxVIP

第 42 卷 第 5 期 2009 年 5 月 天 津 大 学 学 报 Journal of Tianjin University Vol.42 No.5 May 2009 基于支持向量机的油气管道安全监测信号识别方法 曲志刚，封 皓，靳世久，曾周末，周 琰 (天津大学精密测试技术及仪器国家重点实验室，天津 300072) 摘 要 ：针对基于经验风险最小化原则的传统学习方法的不足，提出了一种基于支持向量机的油气管道安全识别方 法．通过基于 Mach-Zehnder 光纤干涉仪原理的分布式光纤振动信号传感器获取管道沿线振动信号，利用基于小波包 分解的“能量-模式”方法提取振动信号的特征向量，并利用支持向量机根据振动信号的特征对其进行识别，判断管道 沿线是否有异常事件发生． 利用现场实验数据对该方法进行验证分析． 结果表明，该方法识别正确率较高， 实时性好． 关键词：支持向量机；油气管道安全；识别；分布式光纤；小波包 中图分类号：TN247；TN911.6 文献标志码：A 文章编号 ：0493-2137(2009)05-0465-06 An SVM-Based Recognition Method for Safety Monitoring Signals of Oil and Gas Pipeline QU Zhi-gang，FENG Hao ，JIN Shi-jiu，ZENG Zhou-mo，ZHOU Yan (State Key Laboratory ofPrecision Measurement Technology and Instrument，Tianjin University，Tianjin 300072，China) Abstract：An SVM-based recognition method for the safety of oil and gas pipeline was proposed due to limitation of the traditional learning methods based on empirical risk minimization. The vibration signals along the pipelines were obtained with the distributed optical fiber vibration sensor on the basis of Mach-Zehnder optical fiber interferometer theory. Then the eigenvectors of vibration signals were extracted through the energy-pattern method based on wavelet packet decomposition. At last the vibration signals were recognized by support vector machine(SVM)through the eigenvectors with a view to de- tecting whether abnormal events happened along the pipelines. The data obtained at the experimental site were used to evalu- ate the method, and the results show that the method is of high accuracy and excellent real-time performance. Keywords ：support vector machine；safety of oil and gas pipeline；recognition；distributed optical fiber；wavelet packet 油气管道由于自身具备的诸多优势，已成为主要 的油气资源运输手段．但由于种种自然或人为原因， 管道泄漏事故时有发生，有时会伴随着巨大的生命财 产损失和环境污染． 目前，常用的油气管道泄漏检测装置主要通过管 道输送压力、流量等参数的变化来判断[1]．该类方法 受输送物质特性及输送工况等诸多因素影响[2]，且报 警发生在泄漏之后． 针对可造成管道泄漏的诸多因素，笔者研制出一 种基于 Mach-Zehnder 光纤干涉仪的分布式光纤油气 管道泄漏检测及预警系统，通过获取并分析和识别管 收稿日期： 2008-05-06 ；修回日期：2008- 10-31.