摘要
通过多个传感器采集了挖掘机故障和非故障状态下斗杆液压缸伸出这一过程的压力信号,运用动态PCA将采集的多维数据降至一维,经过多次试验建立训练样本和检验样本,利用GMM方法建立了挖掘机故障检测模型。实验表明此方法可以有效的用于挖掘机液压系统故障检测。
The pressure signals in the process of bucket arm cylinder stretching out are collected via several sensors. The collected data of many dimensions are reduced to that of one dimension through DPCA and the training set and the testing set are buit after many experiments. The fault detection model of the established is eatablished through GMM. The experiment indicates that this meth- od can be used to detect the fault in the excavator hydraulic system effectively.
出处
《机械制造与自动化》
2013年第3期135-138,共4页
Machine Building & Automation
