Predictive maintenance is particularly important for the normal operation of coal mining machines,which can effectively prevent major safety accidents and casualties,for this reason,a predictive maintenance method for key parts of coal mining machines based on qualitative and quantitative analysis of monitoring data is proposed.The offline and online monitoring data can be used to qualitatively analyze the position,attitude,trajectory and other information of the key parts of the coal mining machine,and then construct a prediction model combined with an auto-encoder(AE)and a deep bi-directional gated recirculation unit(Bi-GRU)in order to predict the remaining service life of the key parts of the coal mining machine,as well as quantitatively analyze it.Taking the coal mining machine rocker arm as an example,the remaining service life is predicted based on the AE Bi-GRU prediction model,and the results show that the AE Bi-GRU prediction model effectively improves the fault prediction accuracy of the coal mining machine rocker arm.
关键词
采煤机摇臂/定性定量分析/监测数据
Key words
coal mining machine rocker arm/qualitative and quantitative analysis/monitoring data
维普
万方数据