Rolling Bearing Fault Diagnosis Model for Running Gear of Metro Vehicle Based on Edge Computing
This article aims to improve the safety and reliability of metro operation by designing a rolling bearing fault diagnosis model for running gear of metro car based on edge computing.The wide application of edge computing in the industrial IoT shows its obvious advantages in real-time data processing and low-latency response,which is significant for the fault diagnosis of the running gear of metro car.Most of the conventional fault diagnosis methods rely on the human experience and the spectral analysis,while the modern artificial intelligent methods,like the Convolutional Neural Network(CNN)and the attention mechanism,demonstrate their distinguished performance in accuracy and efficiency of fault detection.The CNN and the attention mechanism are combined to design a bearing fault diagnosis model based on edge computing and its effectiveness has been verified by testing.The test result shows that this model is obviously advantageous in improving the accuracy of fault detection.
running gear of metro carrolling bearingfault diagnosisConvolutional Neural Networkattention mechanism
NETL
NSTL
万方数据
