Imbalanced Fault Diagnosis of Rotating Machinery Based on Deep Generative Adversarial Networks with Gradient Penalty

Abstract

In practical industrial application, the fault samples collected from rotating machinery are frequently unbalanced, which will create difficulties when it comes to diagnosis. Besides, the variation of working conditions and noise factors will further reduce the diagnosis’s accuracy and stability. Considering the above problems, we established a model based on deep Wasserstein generative adversarial network with gradient penalty (DWGANGP). In this model, the unbalanced fault data set will first be trained by the sample generation network to generate synthetic samples, which will be used to restore the balance. A one-dimensional convolutional neural network with a specific structure is then used as the fault diagnosis network to classify the reconstructed equilibrium samples. The experimental results show that the proposed sample generation network can generate high-quality synthetic samples under highly imbalanced data, and the diagnostic network has a fast training convergence. Compared to the combination methods of support vector machines, back propagation neural network and deep belief network, our method has a 74% average accuracy in all unbalanced experimental conditions, which has 64%, 69% and 87% averages leading, respectively.

Generation of Time-Series Working Patterns for Manufacturing High-Quality Products through Auxiliary Classifier Generative Adversarial Network

“…Unlike the existing studies in this paper, we propose the AC-GAN model to generate the optimal working patterns based on the historical data from melting and casting processes. Although some studies used GAN-based models in the manufacturing field, most have focused on detecting defect products based on image datasets [ 9 , 10 ] and non-time-series datasets [ 11 , 12 , 13 ]. In other words, generating working patterns using GAN models in the manufacturing field has not been studied so far.…”

“…However, generating TSWP using GAN models in the manufacturing field has not been studied so far. Many studies have used GAN-based models in the manufacturing field; however, most cases have focused on detecting defect products based on image datasets [ 9 , 10 ] and non-time-series datasets [ 11 , 12 , 13 ]. This study uses historical data from melting and casting processes to generate TSWP using a generative deep learning approach.…”

Research on an Improved Auxiliary Classifier Wasserstein Generative Adversarial Network with Gradient Penalty Fault Diagnosis Method for Tilting Pad Bearing of Rotating Equipment

“…Some equipment will not even have any kind of fault in the whole life-cycle. The lack of equipment-fault sample data is one of the challenging problems faced by fault diagnosis technology based on artificial intelligence and big data [18].…”

Intelligent Defect Diagnosis of Rolling Element Bearings under Variable Operating Conditions Using Convolutional Neural Network and Order Maps

“…Their health status has a significant effect on the performance and availability of industrial machinery. As they are one of the most vulnerable components of rotating machines, bearing faults are a major cause of machine defects [1]. Bearing failure may cause serious damage to a machine and may result in the unavailability of important machinery, leading to financial loss and serious safety hazards.…”