In order to improve the accuracy of fault detection of jack hydraulic system,this paper proposes a fault detection method of jack hydraulic system based on multi-source data fusion,combined with long and short term memory network to improve the detection performance of multi-source data fusion technology.The test results show that when the number of iterations is 600,the improved multi-source data fusion fault detection model can get a better fault detection effect.The more serious the fault,the higher the detection accuracy,the detection accuracy of motor oil leakage fault is as high as 99.4%;Among the four algorithms,the fault detection accuracy of the improved multi-source data fusion model is the highest,which is 98.72%on the training set and 98.54%on the test set.
关键词
多源数据融合/长短期记忆网络/千斤顶/液压系统/故障检测
Key words
multi-source data fusion/long short-term memory network/jack/hydraulic system/fault detection
维普
万方数据