In response to the challenges of complex operating environments and the difficulty of identifying vibration signals in hy-draulic pumps of mining excavators,a hydraulic pump anomaly detection method for excavator traveling hydraulic systems based on pressure load spectrum was proposed.The typical operating conditions for the traveling hydraulic system were divided according to the characteristics of excavator operation.Then,the outlet pressure of the traveling hydraulic system hydraulic pump,pilot valve pressure,en-gine speed and torque percentage were selected as pressure load spectrum,the data cleaning,interpolation alignment and smoothing fil-tering were performed.Finally,an anomaly detection model for hydraulic pumps was established based on pressure load spectrum and LSTM networks.The outlet pressure for pump 1 and 2 were predicted separately using the model.A sliding cosine similarity measurement technique was used to assess the differences in the prediction results.The assessment results were combined with the 1δ criterion to real-ize the anomaly detection of hydraulic pumps for mining excavator walking.The experimental results demonstrate that the proposed meth-od can effectively detect anomalies in hydraulic pumps,which can provide early warning for faults in the traveling hydraulic system of mining excavators.
关键词
载荷谱/液压泵/出口压力/异常检测/行走液压系统
Key words
load spectrum/hydraulic pump/outlet pressure/anomaly detection/traveling hydraulic system
维普
万方数据