基于PSO-LSSVM时序预测模型的管网漏失信号识别

为解决城市供水管网的漏失问题,基于在供水管网各测压点收集的压力数据,构建粒子群(PSO)算法优化LSSVM的时序预测模型来预测压力监测点下一时刻压力值,并提出了城市供水管网漏失识别模型,通过监测点压力值与预测值的残差值是否在阈值范围内来判断管网是否处于正常工况。测试分析结果表明,改进的时序预测模型预测精度较高,可确定各压力监测点阈值,识别管网是否发生漏失事故,为相似工程提供借鉴。

Identification of Pipeline Leakage Signal Based on PSO-LSSVM Time Series Prediction Model

In order to solve the leakage problem of urban water supply network, based on the pressure data collected from each pressure measuring point of water supply network, the time series prediction model based PSO algorithm optimized LSSVM was established to predict the pressure value of the pressure monitoring point at the next moment, and the leakage identification model of urban water supply network was put forward. Whether the residual value between the pressure of the monitoring point and the predicted value is within the threshold range or not was used to judge whether the pipe network was in normal working condition. The test and analysis show that the improved time series prediction model has high prediction accuracy, which can determine the threshold of each pressure monitoring point and identify the leakage accident in the pipe network, so as to provide relevant reference for the similar projects.