Nowadays, using vibration data in conjunction with pattern recognition methods is one of the most common fault detection strategies for structures. However, their performances depend on the features extracted from vibration data, the features selected to train the classifier, and the classifier used for pattern recognition. Deep learning facilitates the fault detection procedure by automating the feature extraction and selection, and classification procedure. Though, deep learning approaches have challenges in designing its structure and tuning its hyperparameters, which may result in a low generalization capability. Therefore, this study proposes an ensemble deep learning framework based on a convolutional neural network (CNN) and Dempster–Shafer theory (DST), called CNN-DST. In this framework, several CNNs with the proposed structure are first trained, and then, the outputs of the CNNs selected by the proposed technique are combined by using an improved DST-based method. To validate the proposed CNN-DST framework, it is applied to an experimental dataset created by the broadband vibrational responses of polycrystalline nickel alloy first-stage turbine blades with different types and severities of damage. Through statistical analysis, it is shown that the proposed CNN-DST framework classifies the turbine blades with an average prediction accuracy of 97.19%. The proposed CNN-DST framework is benchmarked with other state-of-the-art classification methods, demonstrating its high performance. The robustness of the proposed CNN-DST framework with respect to measurement noise is investigated, showing its high noise-resistance. Further, bandwidth analysis reveals that most of the required information for detecting faulty samples is available in a small frequency range.

Vibration-based quality monitoring of manufactured components often employs pattern recognition methods. Albeit developing several classification methods, they usually provide high accuracy for specific types of datasets, but not for general cases. In this paper, this issue has been addressed by developing a novel ensemble classifier based on the Dempster-Shafer theory of evidence. In the proposed procedure, prior to DST combination, three steps should be taken: (i) selection of proper classifiers by maximizing the joint mutual information between predicted and target outputs, (ii) optimal redistribution of the classifiers’ outputs by considering the distance between the predicted and target outputs, (iii) utilizing five different weighting factors to enhance the fusion performance. The effectiveness of the proposed framework is validated by its application to 13 UCI and KEEL machine learning datasets. It is then applied to two vibration-based datasets to detect defected samples: one synthetic dataset generated from the finite element model of a dogbone cylinder, and one real experimental dataset generated by collecting broadband vibrational response of polycrystalline Nickel alloy first-stage turbine blades. The investigation is made through statistical analysis in presence of different noise levels. Comparing the results with those of five state-of-the-art fusion techniques reveals the good performance of the proposed ensemble method.

Measurements are not exactly accurate, and measurement errors could lead to a biased trained classifier, and finally to a wrong classification of the parts. This paper extends the recently proposed (Integrated) Mahalanobis Classification System with the concept of Interval Mahalanobis distance (IMD) in order to account for measurement uncertainty. This novel Integrated Interval Mahalanobis Classification System (IIMCS) is applied to an experimental case study of complex shaped metallic turbine blades with various damage types. The turbine blades have been vibrationally tested in a wide frequency range. The IIMCS selects a subset of optimal features that contribute the most to the system under the framework of Binary Particle Swarm Optimization, and determines the optimal decision threshold based on Particle Swarm Optimizer. A Monte Carlo method (MCM) is implemented to account for measurement uncertainty, and as such yields an indicator of reliability, implying the confidence level of the classification results. The obtained results illustrate a high performance of the IIMCS for classifying turbine blades based on vibrational response data with measurement uncertainty.