摘要
鉴于传感器采集的数据格式复杂、信息量大,不能有效地对传感器故障进行自动检测和隔离,从而影响评估的准确性,并产生错误预警信息,提出了一种基于几何后非线性独立元分析方法(Geometric Post Nonlinear ICA,gpICA)的传感器故障检测与隔离算法。该算法通过引入几何后非线性混合模型,将非线性数据线性化,再利用快速独立元分析(FastICA)对故障进行检测。通过计算监测数据对监控统计量的贡献度,基于贡献度分析法得以确定具体故障传感器,最终利用MATLAB软件进行数值模拟,实现了模拟故障传感器的检测和隔离。该算法相比传统的线性ICA故障检测具有更高的故障检测率,更适用于桥梁健康监测系统的故障检测与隔离。
Due to the complex format of the data collected by the sensors and the large amount of information,the failure of sensors could not be automatically detected and isolated effectively,thus the accuracy of the evaluation will be affected and the false warning information will be generated.A kind of sensor fault detection and isolation algorithm based on(Geometric Post Nonlinear ICA,gp ICA)is proposed.The algorithm linearizes the nonlinear data by introducing a geometric post nonlinear hybrid model,and then uses the fast independent element analysis(Fast ICA)to detect faults.By calculating the contribution of the monitoring data to the monitoring statistics,the isolation of fault sensor is realized based on the contribution analysis method.Using MATLAB software for the numerical simulation,the detection and isolation of simulated fault sensors are realized.This algorithm has a higher fault detection rate than the traditional linear ICA fault detection,and is more suitable for the fault detection and isolation of bridge health monitoring systems.
作者
王兵见
陈可
陈麒元
WANG Bingjian;CHEN Ke;CHEN Qiyuan
出处
《城市道桥与防洪》
2022年第8期166-169,M0018,共5页
Urban Roads Bridges & Flood Control
关键词
传感器
故障
独立元
健康监测
桥梁
sensor
failure
independent element
health monitoring
bridge
