论文部分内容阅读
针对现行管道泄漏检测系统仅分析单段管道,以致误报率高且存在漏报的问题,从成品油管网全局出发,建立泄漏检测方法,通过引入现场操作信息,实现了辨识压力异常原因的功能。以模糊神经网络作为主要分类方法,以工况调整和负压波出现的位置、时间和变化量等为模糊神经网络的输入,对其进行训练,从而识别负压波出现的原因,以此屏蔽工况调整对管道泄漏检测系统的影响。通过在华北成品油管网鲁皖一期管道的试验应用,验证了该方法的可行性。
Aiming at the problem that the existing pipeline leak detection system only analyzes single-section pipe so that the false alarm rate is high and there is a problem of omission, a leak detection method is established based on the overall situation of the product pipeline network. By introducing the field operation information, the function of recognizing the cause of abnormal pressure is realized . Taking the fuzzy neural network as the main classification method, the position, time and variation of working condition adjustment and negative pressure wave appear as the input of fuzzy neural network, which are trained to identify the cause of the negative pressure wave so as to shield Effect of Condition Adjustment on Pipeline Leakage Detection System. The feasibility of this method is verified through the experimental application of the first-phase pipeline of Ru-an oil pipelines in North China.