基于粒子滤波的管道泄漏检测与定位方法  

Pipeline Leak Detection and Location Method Based on Particle Filter

作　　者：贺利乐[1] 陈奕昕 贺宁 吴文[2] HE Li-le;CHEN Yi-xin;HE Ning;WU Wen(School of Mechanical and Electrical Engineering,Xi’an University of Architecture and Technology,Xi’an 710055,China;Key Laboratory of Road Construction and Equipment of MOE,Chang’an University,Xi'an 710055,China)

机构地区：[1]西安建筑科技大学机电工程学院,陕西西安710055 [2]长安大学道路施工技术与装备教育部重点实验室,陕西西安710055

出　　处：《控制工程》2021年第4期787-798,共12页Control Engineering of China

基　　金：陕西省教育厅基础科学研究专项计划资助项目(18JK0438);陕西省科协青年人才托举计划资助项目(20180112);道路施工技术与装备教育部重点实验室(长安大学)开放基金资助项目(300102258501)。

摘　　要：针对长距离管道中可能发生的较小泄漏,提出一种基于粒子滤波的泄漏检测与定位方法。首先,基于特征线法建立了管道流体的数学模型,并将其转化为状态空间模型,基于状态空间模型使用粒子滤波对管道瞬态流动条件下的压头值和流量值进行状态估计。在泄漏定位方面,提出一种估计—隔离—再估计的定位策略和一种基于重采样的泄漏量估计方法,该方法能够通过判断粒子滤波估计值和输油管道沿线测量值的匹配程度,获得泄漏点位置、泄漏量和泄漏系数。仿真结果显示,所述方法对长距离管道中发生的较小泄漏的定位误差低于1%,能够有效对管道泄漏进行定位。For a small leak in the long distance pipeline, this paper proposes a leak detection and location method based on particle filter. Firstly, the hydraulic model of pipeline is derived through the method of characteristic(MOC) and converted into the state space model. Based on the state space model, particle filter(PF) is used to estimate the pressure head value and flow value of the pipeline under the condition of transient flow. As to the leak location, an estimation-isolation-reestimation location strategy and a resample-based leak estimation method are addressed to determine the matching degree between the particle filter estimation value and the measurements and then obtain the leak location, the leak flow rate as well as the leak coefficient. Finally, the simulation results show that the small leak location error is less than 1% in the long distance pipeline, and the proposed method can locate the pipeline leak effectively.

关 键 词：管道 泄漏检测与定位 粒子滤波 状态估计 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]