Abstract: How to effectively identify the threats to security of oil and gas pipelines, such as manual excavation, mechanical excavation and vehicle passing, is of vital importance to the coherent Rayleigh-based pipeline fiber security warning system. This paper proposes a method of pipeline security threat event recognition based on optimized support vector machine (SVM). The method is used to analyze the vibration signals along the pipeline collected by the optic fiber safety warning system, and extract the normalized energy and signal duration of each frequency band as the eigenvectors (proper vectors). The penalty factors and kernel function parameters of the SVM are optimized by the artificial bee colony algorithm,and then the optimized SVM is used to classify the features. This method was tested by the safety experiment of the product oil pipeline of Dagang - Zaozhuang, and the correct identification rate of 90. 7% was obtained. All these proved the effectiveness of the method and its engineering application value.

摘要： 如何有效识别人工挖掘、机械挖掘和车辆经过等油气管道安全威胁事件, 对基于相干瑞利的管道光纤安 全预警系统是至关重要的问题。对此提出一种基于优化支持向量机的管道安全威胁事件识别方法。该方法对管道光纤 安全预警系统采集到的管道沿线振动信号进行分析, 提取各频段的归一化能量与信号持续时间作为特征向量, 利用人 工蜂群算法对支持向量机的惩罚因子和核函数参数进行优化, 采用优化后的支持向量机对特征进行分类。通过港枣成 品油管线安全实验对该方法测试, 获得了90. 7%的识别正确率, 证明了该方法的有效性和工程应用价值。

关键词: 油气管道, 安全预警, 支持向量机, 人工蜂群算法, 参数优化