基于瞬变流和遗传算法的管道泄漏辨识  

Leak detection in pipelines based on hydraulic transient and genetic algorithm analysis

作　　者：郭新蕾[1] 杨开林[1] 

机构地区：[1]中国水利水电科学研究院水力学研究所,北京100038

出　　处：《计算力学学报》2009年第5期664-669,共6页Chinese Journal of Computational Mechanics

基　　金：国家自然科学基金(50679085);国家社会公益专项基金(126301041003);中国水利水电科学研究院专项基金(水集05KY01)资助项目

摘　　要：输水系统管道的泄漏影响瞬变水击波波形的畸变和衰减特性,对泄漏的定位和定量是泄漏辨识研究的重要环节。考虑水力瞬变中非恒定摩阻的影响,将泄漏参数反映在建立的管道瞬变流时域数学模型中,给出离散网格和求解方法,分析不同泄漏参数对瞬变水击波的影响,用遗传算法结合瞬变流时域模型的特征线法最小化管道测点压力实测和计算响应值的均方误差来辨识泄漏参数,其中实测信号首先经过预滤波和小波去噪处理。通过模型试验对瞬变流数学模型和泄漏反问题分析模型作了率定和验证,结果表明此法能有效的辨识出小泄漏参数,并且检测信号主动,阀门扰动可操作性强。Leaks in pipeline systems contribute to the damping and aberration of hydraulic transient events. Considering the effect of unsteady friction and leaks, the inverse transient model in time domain is established. The discrete grid and its calculation method are presented. The impact of changing leak size and position on the damping of pipeline transient are studied. The genetic algorithm is integrated into the method by minimizing the mean square error between modeled piezometric pressure magnitudes and measured date at points to identify the parameters of the leak. The time domain signals were first filtered using pre-filter and discrete wavelets. The model is applied to physical test with the calibrated unsteady friction, the results proved that the inverse transient model with genetic algorithm analysis is effective to identify leaks that may be difficult to find out by other methods as they are small and the maneuver of valve operation is simple.

关 键 词：瞬变流 泄漏检测 非恒定摩阻 遗传算法 反问题分析 

分 类 号：TV131[水利工程—水力学及河流动力学]