High-resolution time-frequency representation (TFR) method is effective for signal analysis and feature detection. However, for variable speed bearing vibration signal, conventional TFR method is prone to blur and affect the accuracy of the instantaneous frequency estimation. Moreover, the traditional order tracking, relying on equi-angular resampling, usually suffers from interpolation error. To solve such problems, we propose a joint time-frequency (TF) squeezing method and generalized demodulation (GD) to realize variable speed bearing fault diagnosis. The method can represent the time-varying fault characteristic frequency precisely and be free from resampling. First, using fast spectral kurtosis to select the optimal-frequency band which is sensitive to rolling bearing fault, and extracting envelope by Hilbert transform within the selected optimal frequency band. Next, a high-quality TF clustering method based on short-time Fourier transform is applied to the TF analysis of the envelope to get a clear TFR, from which the frequency information for GD is obtained. Finally, processing the basic demodulator via the peak search through the TF analysis results in the TFR for GD to gain a resampling-free-order spectrum. Based on the more precise TF information from the clearer TFR, the bearing fault can be diagnosed via GD without tachometer or any resampling involved, avoiding the amplitude error and low computational efficiency of resampling. Simulation study and experimental signal analysis validate that the proposed method has better performance than those methods based on conventional TF analysis and resampling.

The vibration monitoring of gearboxes is an effective means of ensuring the long-term safe operation of rotating machinery. A gearbox may have more than one fault in actual applications. Therefore, gearbox compound fault diagnosis should be investigated. In this paper, a novel multiple enhanced sparse decomposition (MESD) method is proposed to address multiple feature extraction for gearbox compound fault vibration signals. Through this method, a novel MESD algorithm is utilized to simultaneously separate and extract the harmonic components and transient features of the gear and bearing from the compound fault signal. Three subdictionaries are specially constructed according to the gearbox failure mechanism to accurately extract each feature component. Meanwhile, the generalized minimax concave (GMC) penalty is used as sparse regularization to further ensure the accuracy of sparse decomposition. The simulation and engineering signals of the gearbox validate the performance of the proposed MESD method.

Machine learning models have been widely successful in the field of intelligent fault diagnosis. Most of the existing machine learning models are deployed in static environments and rely on precollected datasets for offline training, which makes it impossible to update the models further once they are established. However, in the open and dynamic environment in reality, there is always incoming data in the form of streams, including new categories of data that are constantly generated over time. In addition, the operating conditions of mechanical equipment are time-varying, which results in continuous stream data that are nonindependently and homogeneously distributed. In industrial applications, the diagnosis problem of nonindependent and identically distributed continuous streaming data is referred to as the cross-domain class incremental diagnosis problem. To address the cross-domain class incremental problem, a novel cross-domain class incremental broad network (CDCIBN) is proposed. Specifically, to solve the nonindependent identically distributed problem, a novel domain-adaptation learning loss function is first designed, which enables the conventional broad network to handle the category increment task well. Then, a cross-domain class incremental learning mechanism is designed, which learns new categories while retaining the knowledge of old categories well enough without replaying old category data. The effectiveness of the proposed method is evaluated through multiple mechanical failure increment cases. Experimental analysis demonstrates that the designed CDCIBN has significant advantages in the variable working condition class incremental application.

It confronts great difficulty to apply traditional artificial intelligence (AI) techniques to machinery prognostics and health management (PHM) in manufacturing systems due to the lack of abnormal samples corresponding to different fault conditions. This paper explores an unsupervised feature learning method for machinery health monitoring by proposing a generative adversarial networks (GAN) model that exploits the merits of the autoencoder (AE) and the traditional GAN. The major contribution is that the data distribution of the normal samples is accurately learned by the GAN model within both the signal spectrum and latent representation spaces. Specifically, the discriminative feature for machinery health monitoring is learned in an unsupervised manner by the proposed method in three steps. First, the proposed GAN model is trained by the normal samples of the inspected machine with the aim to correctly reconstruct the signal spectrum and its latent representation. Then, the trained model is applied to test the online samples of the same machine with unknown health conditions. Finally, the dissimilarity between the tested samples and their reconstructed ones in the latent representation space is taken as the discriminative feature. The feature value will increase significantly if a fault occurs in the inspected machine because the abnormal samples are never trained in the proposed GAN model. Experimental studies on three different machines are conducted to validate the proposed method and its superiority over the traditional methods in detecting abnormal points and characterizing fault propagation.

Intelligent fault diagnosis methods based on domain adaptation (DA) have been extensively employed for tackling domain shift problems, and the basic diagnosis tasks under time-varying working conditions were well achieved. Nevertheless, the existing methods usually focus on learning diagnosis knowledge from single-source domain while ignoring abundant underlying information in multisource domain. In practical scenarios, multiple source domains are available, and there are few studies in fault diagnosis based on multisource DA. To this end, a novel multisource domain feature adaptation network (MDFAN) is proposed for bearing fault diagnosis under time-varying working conditions. The proposed network first uses a feature extractor to learn transferable features from different pairs of source and target domains, and then, a domain-specific distribution alignment module is constructed, which adopts the intra-domain alignment strategy and the inter-domain alignment strategy to reduce the shift between all domain pairs. Meanwhile, considering the classifier prediction disagreement, a classifier alignment module is further designed to relieve the classifier prediction discrepancy and enhance the prediction consistency. Case studies on two real datasets with multiple sources demonstrate the effectiveness of the proposed network, and the comparison results show its robustness and superiority.