Fault Diagnosis Of ELM Shield Machine Based On Reverse Feature Elimination

With the continuous reduction of living space per capita in the world,human beings have invented various devices to expand living space,among which the most representative machine is shield machine.Shield machine is a kind of complex large-scale tunneling equipment with multiple systems and multiple driving sources.However,due to its complex structure and relatively closed working environment,the shield machine is prone to various failures in the process of work.The closed structure makes the repair work of the shield machine very difficult.Therefore,we need a method that can predict the location of the fault in the first time or even before the fault occurs,to improve work efficiency and reduce economic losses.In this paper,neural network and shield machine are combined to construct a new fault diagnosis system of shield machine based on neural network.In order to improve the accuracy and efficiency of shield machine fault diagnosis,a fault diagnosis method based on ELM is proposed.In view of the characteristics of shield machine operation data with many dimensions and large quantity,the reverse feature elimination(RFE)method is introduced to reduce the dimension of data,eliminate the redundant dimension and remove the correlation between features.Considering the slow speed and low efficiency of neural network fault diagnosis,the ELM neural network classifier model is built based on the limit learning mechanism for shield machine fault diagnosis.The simulation results based on the field construction data show that the method improves the accuracy and efficiency of shield machine fault diagnosis significantly,and has good engineering application value.