The growing complexity of machinery and the increasing demand for operational efficiency and safety have driven the development of advanced fault diagnosis techniques. Among these, convolutional neural networks (CNNs) have emerged as a powerful tool, offering robust and accurate fault detection and classification capabilities. This comprehensive review delves into the application of CNNs in machine fault diagnosis, covering its theoretical foundation, architectural variations, and practical implementations. The strengths and limitations of CNNs are analyzed in this domain, discussing their effectiveness in handling various fault types, data complexities, and operational environments. Furthermore, we explore the evolving landscape of CNN-based fault diagnosis, examining recent advancements in data augmentation, transfer learning, and hybrid architectures. Finally, the future research directions and potential challenges to further enhance the application of CNNs for reliable and proactive machine fault diagnosis are highlighted.

In this work, a novel bearing fault identification scheme making use of deep learning has been proposed. Initially, the raw vibration signal is passed through a time-varying filter based empirical mode decomposition (TVF-EMD) to obtain different modes. Filter parameters of TVF-EMD are optimized by a newly developed optimization algorithm i.e. ameliorated African vulture optimization algorithm The Kernel estimate for mutual information has been considered as the fitness index for the developed optimization algorithm. The mode having the least value of fitness index is known as a prominent mode from which sensitive features representing different bearing conditions are extracted. These extracted features help in preparing the data matrix which is further utilised to build fuzzy-based classification models. The results obtained revealed that the linguistic hedge neuro-fuzzy classifier obtained maximum performance with the least computational time. The comparison of the developed method has also been done with other classification models viz., KNN, SVM, ELM and random forest that revealed the superiority of the developed method.

The bearing vibration signal possesses nonlinear and non-stationary characteristics; hence; it is difficult to diagnosis the faults in the bearing under different working conditions. In this article, a new scheme has been proposed based on complete ensemble empirical mode decomposition with adaptive noise and corrected conditional entropy to recognize the different class of faults in bearing. The mode with minimum corrected conditional entropy is treated as a prominent mode from which sensitive features are extracted. A filter-based feature selection scheme is used for the same and for ranking the features based on variance to reduce the redundancy of data set. This data set is made input to support vector machine. The performance of the support vector machine classifier is improved by optimizing its parameters to obtain maximum classification accuracy. To address this issue, an evolutionary algorithm (diversity-driven multi-parent evolutionary algorithm) is used. With optimized support vector machine parameters, the support vector machine is trained to build a classification model with 10-fold cross-validation. After training, the built model is tested against test data set for fitness evaluation. The support vector machine classifier gives 100% accuracy at regularization and kernel parameter’s value of 1.3343 and of 782.6329, respectively, with 27.93 s of training time for a single iteration.