Abstract
Effective intelligent fault diagnosis of rotating machinery using its vibrational signals has a considerable influence on certain analysis factors such as the reliability, performance, and productivity of a variety of modern manufacturing machines. Traditional intelligent approaches lack generalization schemes and add the burden of extracting features from data-driven cases. On the other hand, the Deep Learning (DL) studies have reported capabilities higher than the expectations of the researchers’ objectives. In this context, this paper proposes a new deep architecture based on Stacked Variant Autoencoders for multi-fault machinery identification with imbalanced samples. The proposed model starts with a Variational Autoencoder (VAE) for facilitating data augmentation of small and imbalanced data samples using Gaussian distribution. After the preparation of suitable samples based on quality and size, the preprocessed vibration signals obtained are injected into the deep framework. The proposed deep architecture contains two subsequent unsupervised Sparse Autoencoders (SAE) with a penalty term that helps in acquiring more abstract and essential features as well as avoiding redundancy. The output of the second SAE is integrated on a supervised Logistic Regression (LR) with 10 classes. This is utilized for the proposed classifier training to achieve accurate fault identification. Experimental results show the efficiency of the proposed model which achieved an accuracy of 93.2%. In addition, for extensive comparative analysis issue, the Generative Adversarial Network (GAN) and triNetwork Generative Adversarial Network (tnGAN) were both implemented on the vibrational signal data, where the proposed method reported better results in terms of training and testing time and overall accuracy.
Highlights
Defects in rotating machinery are considered a core challenge during the analysis of causes of low productivity for certain systems and applications
Based on the results reported in the previous research, the autoencoders, with their variety of functions, recorded the highest classification accuracy compared to Artificial Neural Network (ANN), convolutional neural network (CNN), and other deep learning models
The output features of the highest layer are investigated through the classification process by augmenting a Logistic Regression classifier after the high-level feature representations of input data are extracted through the first Sparse Autoencoders (SAE) above the stacked sparse autoencoder’s last hidden layer to fine-tune the deep learning architecture and boost learned features in a supervised manner
Summary
Defects in rotating machinery are considered a core challenge during the analysis of causes of low productivity for certain systems and applications. Zhu et al [12] proposed a deep learning model based on a convolutional neural network (CNN) that can efficiently and accurately recognize vibration faults by automatically extracting rotor vibration features. They diagnosed the faults and reported enhanced results compared to traditional methods. Liu et al [26] built a stacked autoencoder based on a deep learning model for solving early detection of gearbox faults from data-driven analysis Their model directly extracts salient features from frequency-domain signals, which saves the effort of handcrafting features. Will remain near the first critical speed as the speed increases [43]
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call