Signals captured in rotating machines to obtain the status of their components can be considered as a source of massive information. In current methods based on artificial intelligence to fault severity assessment, features are first generated by advanced signal processing techniques. Then feature selection takes place, often requiring human expertise. This approach, besides time-consuming, is highly dependent on the machinery configuration as in general the results obtained for a mechanical system cannot be reused by other systems. Moreover, the information about time events is often lost along the process, preventing the discovery of faulty state patterns in machines operating under time-varying conditions. In this paper a novel method for automatic feature extraction and estimation of fault severity is proposed to overcome the drawbacks of classical techniques. The proposed method employs a Deep Convolutional Neural Network pre-trained by a Stacked Convolutional Autoencoder. The robustness and accuracy of this new method are validated using a dataset with different severity conditions on failure mode in a helical gearbox, working in both constant and variable speed of operation. The results show that the proposed unsupervised feature extraction method is effective for the estimation of fault severity in helical gearbox, and it has a consistently better performance in comparison with other reported feature extraction methods.
This paper addresses the development of a random forest classifier for the multi-class fault diagnosis in spur gearboxes. The vibration signal's condition parameters are first extracted by applying the wavelet packet decomposition with multiple mother wavelets, and the coefficients' energy content for terminal nodes is used as the input feature for the classification problem. Then, a study through the parameters' space to find the best values for the number of trees and the number of random features is performed. In this way, the best set of mother wavelets for the application is identified and the best features are selected through the internal ranking of the random forest classifier. The results show that the proposed method reached 98.68% in classification accuracy, and high efficiency and robustness in the models.
At present, countless approaches to fault diagnosis in reciprocating machines have been proposed, all considering that the available machinery dataset is in equal proportions for all conditions. However, when the application is closer to reality, the problem of data imbalance is increasingly evident. In this paper, we propose a method for the creation of diagnoses that consider an extreme imbalance in the available data. Our approach first processes the vibration signals of the machine using a wavelet packet transform-based feature-extraction stage. Then, improved generative models are obtained with a dissimilarity-based model selection to artificially balance the dataset. Finally, a Random Forest classifier is created to address the diagnostic task. This methodology provides a considerable improvement with 99% of data imbalance over other approaches reported in the literature, showing performance similar to that obtained with a balanced set of data.
Signal acquisition from mechanical systems working in faulty conditions is normally expensive. As a consequence, supervised learning-based approaches are hardly applicable. To address this problem, a one-shot learning-based approach is proposed for multi-class classification of signals coming from a feature space created only from healthy condition signals and one single sample for each faulty class. First, a transformation mapping between the input signal space and a feature space is learned through a bidirectional generative adversarial network. Next, the identification of different health condition regions in this feature space is carried out by means of a single input signal per fault. The method is applied to three fault diagnosis problems of a 3D printer and outperforms other methods in the literature.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.