A Novel Data-Driven Fault Feature Separation Method and Its Application on Intelligent Fault Diagnosis Under Variable Working Conditions

Abstract

As mechanical fault diagnosis enters the era of big data, the traditional fault diagnosis methods under variable working condition are difficult to be applied because of the massive computation cost and excessive reliance on human labor. For the application of intelligent fault diagnosis under variable working conditions, the crucial difficulty is that the variable speed or load can cause smearing and skewing of classable feature. It is the key to break the predicaments by extracting the features which are irrelevant to the working conditions and contain fault information. This paper propose a new intelligent fault diagnosis framework under variable working conditions called Data-driven Fault feature Separation Method (DFSM) which can eliminate the working condition features from all the information and employ the rest fault information for diagnosis. In our DFSM, classification loss ensures the basic classified ability, first. Second, uncorrelation loss increases the discrepancy between fault features and working condition features. Then, reference loss guides the working condition encoder only extracting the working condition features. Finally, autoencoder loss ensures that all the information has been extracted. It should be noticed that our DFSM is trained only using the dataset under a certain working condition and can diagnose faults with high accuracy under variable even time-varying working condition, which means that it is easy to be applied. The experimental results of rolling bearing dataset show that the proposed DFSM can not only break the limitation of existing methods, but also achieve a superior performance comparing with related method. Besides, through the visual understanding, the proposed DFSM is certified that it is able to eliminate the working condition information and extract precise classable feature for fault diagnosis. INDEX TERMS Data-driven, Feature separation, Intelligent fault diagnosis, Variable working conditions.

Deep domain adaptation with adversarial idea and coral alignment for transfer fault diagnosis of rolling bearing

“…One is normal data, denoted as N; the other three types of data are fault data, namely roller fault, inner race fault and outer race fault denoted RF, and OF. Among them, for each type we took faults with different damage degrees (7,14 and 21 Mils). Therefore, the above can be summarized as we have a total of ten different types of data, which are simply labeled N, RF1, RF2, RF3, IF1, IF2, IF3, OF1, OF2, OF3, corresponding to the ten kinds of category labels (C0-C9).…”

“…At the same time, with the development of networks, information technology and various monitoring technologies, the data generated by mechanical equipment shows exponential growth [4,5]. The traditional fault diagnosis method is unable to cope with such a huge amount of data due to its own disadvantages, so the intelligent fault diagnosis method emerged [6,7]. The intelligent fault diagnosis method is expected to overcome some difficulties that exist in traditional fault diagnosis methods and solve the problem of complex system fault diagnosis [8].…”

Transfer Diagnosis Model of Internal Combustion Engine With Embedded Vibration Signal Impact Decomposition

“…Shen et al [25] proposed a deep subdomain adaptation network (DSAN) to extract features of multiscale vibration signals using local maximum mean discrepancy (LMMD) loss to reduce the distance between source and target domains and achieve transfer diagnosis of wind turbine system faults. Li et al [26] introduced an adversarial DA method based on conditional adversarial DA (CDAN), which enforces better intraclass compactness and interclass separability of labelrelated prediction results to improve model generalization, enabling transfer diagnosis of typical bearing faults under variable conditions. The above studies have shown that constructing feature distribution constraint functions effectively obtains domain-invariant features under variable operating conditions.…”

A Novel 1D-CNN-Based Diagnosis Method for a Rolling Bearing with Dual-Sensor Vibration Data Fusion

“…For example, Zhang et al [22]constructed a multimode CNN model to improve the fault detection accuracy of rolling bearings under variable working conditions. Li et al [23] proposed a data-driven fault feature separation method to eliminate the working condition information and extract the precise classable feature for fault diagnosis. Su et al [24]proposed a hierarchical branch CNN scheme that considered polluted data and variations in the working environment conditions.…”