An Active Learning Method Based on Uncertainty and Complexity for Gearbox Fault Diagnosis

Abstract

It is crucial to implement an effective and accurate fault diagnosis of a gearbox for mechanical systems. However, being composed of many mechanical parts, a gearbox has a variety of failure modes resulting in the difficulty of accurate fault diagnosis. Moreover, it is easy to obtain raw vibration signals from real gearbox applications, but it requires significant costs to label them, especially for multi-fault modes. These issues challenge the traditional supervised learning methods of fault diagnosis. To solve these problems, we develop an active learning strategy based on uncertainty and complexity. Therefore, a new diagnostic method for a gearbox is proposed based on the present active learning, empirical mode decomposition-singular value decomposition (EMD-SVD) and random forests (RF). First, the EMD-SVD is used to obtain feature vectors from raw signals. Second, the proposed active learning scheme selects the most valuable unlabeled samples, which are then labeled and added to the training data set. Finally, the RF, trained by the new training data, is employed to recognize the fault modes of a gearbox. Two cases are studied based on experimental gearbox fault diagnostic data, and a supervised learning method, as well as other active learning methods, are compared. The results show that the proposed method outperforms the two common types of methods, thus validating its effectiveness and superiority. INDEX TERMS Active learning, gearbox fault diagnosis, uncertainty and complexity, supervised learning.

A Linear Narrowband Filtering Method Based on Migration-Enhanced Singular Value Decomposition and Its Application to Reduce Noise

“…SVD, as a mathematical method, has been introduced in signal processing [38], [39]. It has been combined with other processing methods, such as wavelet analysis and empirical mode decomposition (EMD), to improve the effect of signal processing [40], [41].…”

Design, Prototyping and Testing of a Rotating Electrical Machine With Linear Geometry for Shipboard Applications

“…On the order side, we can notice that in some cases, such as in shipboard applications, electric motors and generators need to be fit into small and unusually shaped vanes leading to very strict constraints on the electric machine height, width and length. Possible ways to cope with space constraints is to use highspeed machines with gear-boxes [3], which are, however, know to suffer from reliability, and maintenance issues [3], [4]. Another solution is represented by hydraulic actuators which, however, may suffer from heavy maintenance, working-fluid leakages [5] and cumbersome equipment for pressurized oil supply.…”

Active Informative Pairwise Constraint Formulation Algorithm for Constraint-Based Clustering

“…The answer of pairs considered by those above-mentioned formulation algorithms are provided by active learning which is mostly applied in labeling and sampling of image processing [14]- [17], natural language processing [18], [19] and so on [20], [21]. According to sampling strategy, active learning can be divided into three categories: uncertainty sampling [14]- [17], [22], [23], query-bycommittee [24]- [27], expected error minimization [28]- [31].…”