Abstract
AbstractBearings are one of the core components used in the mechanical equipment. However, mechanical failures caused by rolling bearing failures account for around 20%–40%. The convolutional neural network model is effective at detecting the fault of rolling bearings, but it suffers overfitting problem. In this paper, we propose a combined network model for the fault detection of rolling bearings by combining the convolutional neural network model and the random forest algorithm. Experiment results show that the combined network model can achieve the expected results in the classification accuracy of rolling bearing mechanical faults.KeywordsRolling bearingsFault detectionConvolutional neural networkRandom forestCombined network model
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have