Simultaneous fault type and severity identification using a two-branch domain adaptation network

Abstract

Simultaneous fault type and severity identification is critical for timely maintenance actions to prevent accidents from industrial machinery. The former can indicate occurrences of specific faults, and the latter can track early fault evolutions. Existing methods generally assume training and testing data are drawn from the same data distribution. However, in real industries, due to the change of working conditions, domain shift phenomenon can be triggered. The existing intelligent diagnosis methods are less effective in such scenarios for lack of domain adaptation ability. To address such problems, a novel two-branch domain adaptation network is developed. A deep convolutional neural network with two branches, as the main hierarchical architecture, is designed to handle two recognition tasks. The maximum mean discrepancy based multi-kernel learning is embedded to reduce the distribution discrepancy between the source domain and target domain. As such, the domain-invariant features with a hierarchical structure can be effectively extracted and the fault types and fault severities can be recognized at the same time. Experiments on a bearing dataset, a gear dataset and a motor bearing dataset are carried out to validate the effectiveness of the proposed approach. The results demonstrate that the proposed method can effectively perform fault type and severity identification simultaneously and obviously outperforms other state-of-the-art methods.

Using Data From Similar Systems for Data-Driven Condition Diagnosis and Prognosis of Engineering Systems: A Review and an Outline of Future Research Challenges

“…Other approaches that include MMD terms in the training or fine-tuning loss function can be found in [50], [163], [205], [206], [207], [208], [209], and [210] on condition diagnosis of bearings and in [211] and [212] on gearbox condition diagnosis. Other condition diagnosis approaches are presented in [213] and [214] on bearings and gearboxes, [215] on bearings and a crack rotating machinery dataset, and [216] on induction motors. Yang et al [179], Yang et al [217], and Guo et al [218] added an MMD term and a pseudo label term to the loss function.…”

A novel deep convolutional neural network and its application to fault diagnosis of the squirrel-cage asynchronous motor under noisy environment

“…After obtaining the CWT images of motor vibration signals in directions of x-axis and y-axis, how to input the timefrequency information of two directions into the deep learning network is vital. Referring to Chen et al [26], two distinct approaches for information fusion can be employed. Firstly, the two types of information can be fused at the initial stages of the network, where the disparate information is amalgamated into novel information before being inputted into the network.…”

“…In the field of motor fault diagnosis, an increasing number of researchers have embraced CNN and obtained productive outcomes. Chen et al [26] designed a deep CNN with two branches to diagnose motor bearing faults. Amarouayache et al [27] proposed a fault diagnosis method for motors by combining ensemble empirical mode decomposition with CNN.…”

Clustering Federated Learning for Bearing Fault Diagnosis in Aerospace Applications with a Self-Attention Mechanism

“…Various research on essential diagnosis issues, such as deep learning methods [3,4], knowledge transfer [5][6][7][8][9], fault decoupling and detection [10][11][12], imbalance data augmentation, and model generalization [13][14][15][16], have been carried out. For example, Syed Muhammad Tayyab et al [17] used machine learning through optimal feature extraction and selection for intelligent fault diagnosis of machine elements.…”