A Novel Deep Learning Method for Intelligent Fault Diagnosis of Rotating Machinery Based on Improved CNN-SVM and Multichannel Data Fusion

Abstract

Intelligent fault diagnosis methods based on deep learning becomes a research hotspot in the fault diagnosis field. Automatically and accurately identifying the incipient micro-fault of rotating machinery, especially for fault orientations and severity degree, is still a major challenge in the field of intelligent fault diagnosis. The traditional fault diagnosis methods rely on the manual feature extraction of engineers with prior knowledge. To effectively identify an incipient fault in rotating machinery, this paper proposes a novel method, namely improved the convolutional neural network-support vector machine (CNN-SVM) method. This method improves the traditional convolutional neural network (CNN) model structure by introducing the global average pooling technology and SVM. Firstly, the temporal and spatial multichannel raw data from multiple sensors is directly input into the improved CNN-Softmax model for the training of the CNN model. Secondly, the improved CNN are used for extracting representative features from the raw fault data. Finally, the extracted sparse representative feature vectors are input into SVM for fault classification. The proposed method is applied to the diagnosis multichannel vibration signal monitoring data of a rolling bearing. The results confirm that the proposed method is more effective than other existing intelligence diagnosis methods including SVM, K-nearest neighbor, back-propagation neural network, deep BP neural network, and traditional CNN.

A Data-Driven-Based Fault Diagnosis Approach for Electrical Power DC-DC Inverter by Using Modified Convolutional Neural Network With Global Average Pooling and 2-D Feature Image

“…The above methods have been applied to some extent, and they have achieved positive and considerable effects on some diagnosis problems. However, some shortcomings of these methods still exist [20], [21]. Model-based diagnosis methods need to establish a high-precision mathematical model that describes the fault evolution mechanisms.…”

“…However, establishing a high-precision complex multi-variable system fault model is very difficult. In addition, the established fault mathematical model usually aims to solve specific devices and fault problems, which is difficult to transplant to solve other devices and similar problems [20]. Furthermore, with the increase of the complexity of mechanical and electrical equipment, the mathematical model-based fault diagnosis methods are restricted in practical applications.…”

A Real-Time Fault Early Warning Method for a High-Speed EMU Axle Box Bearing

“…However, before leaving the factory, a high-speed EMU is equipped with many other sensors, such as sensors to monitor the axle box A fault early warning is a part of condition monitoring and different methods are used for the diagnosis of rotary machines, such as bearing, gears, and motors. According to data sources, they can be broadly classified as vibration-based data [1][2][3][4][5][6], acoustic-based data [7,8], temperature-based data [9,10], and fusion-based data [11][12][13][14][15][16][17].…”

“…First, the deviation index is calculated. The deviation index is defined using Equation (13). In this equation, x j t represents the predicted temperature of the MLSTM at time t, superscript j represents the index number arranged from small to large of the predicted temperature of different bearings at the same location and the same time t. The smaller index number gives the lower temperature, that is, x j−1 t < x j t .…”

“…First, the deviation index is calculated. The deviation index is defined using Equation (13). In this equation, j t…”