Fault Prediction of Shield Machine Based on Rough Set and BP Neural Network
Hui Han, Xiang Gao · 2017
In order to accomplish the fault prediction of complicated and enormous mechanical equipment, this paper proposed a fault prediction model for complicated mechanical equipment that based on rough sets theory and BP neural network . Firstly, the discretization of continuous data was implemented by the discretization algorithm based on dynamic hierarchical clustering in rough set theory; secondly, an improved algorithm based on discernibility function attribute reduction was used to reduce the decision table to remove redundant attributes of decision table, then the reduced data were input into BP neural network for fault prediction. Finally, case studies are conducted to predict the failure of cutterhead of shield machine.The experimental results showed that the prediction model can effectively reduced the input information, simplified the neural network architecture, shortened the training time and improved the efficiency and accuracy of fault prediction.