Fault Diagnosis Task Research Articles

Fault Description Based Attribute Transfer for Zero-Sample Industrial Fault Diagnosis

In this article, a challenging fault diagnosis task is studied, in which no samples of the target faults are available for the model training. This scenario has hardly been studied in industrial research. But it is a common problem that massive fault samples are not available for the target faults, which limits the successes of conventional data-driven approaches in practical application. Here, we introduce the idea of zero-shot learning into the industry field, and tackle the zero-sample fault diagnosis task by proposing the fault description based attribute transfer method. Specifically, the method learns to determine the fault categories using the human-defined fault descriptions instead of the collected fault samples.The defined description consists of arbitrary attributes of the faults, including the fault positions, the consequences of the fault, and even the cause of the fault, etc. For the attribute knowledge of target faults, they can be prelearned and transferred from some readily available faults occurred in the same process. Afterwards, the target faults can be diagnosed based on the defined fault descriptions without the need for any additional data based training. Besides, the supervised principle component analysis is adopted in our method to extract the attribute related features to offer an effective attribute learning. We analyze and interpret the feasibility of the fault description based method theoretically. Also, the zero-sample fault diagnosis experiments are designed and conducted on the benchmark Tennessee-Eastman process and the real thermal power plant process to validate the effectiveness. The results show that it is indeed possible to diagnose target faults without their samples.

Locally linear back-propagation based contribution for nonlinear process fault diagnosis

This paper proposes a novel locally linear backpropagation based contribution (LLBBC) for nonlinear process fault diagnosis. As a method based on the deep learning model of auto-encoder (AE), LLBBC can deal with the fault diagnosis problem through extracting nonlinear features. When the on-line fault diagnosis task is in progress, a locally linear model is firstly built at the current fault sample. According to the basic idea of reconstruction based contribution (RBC), the propagation of fault information is described by using back-propagation (BP) algorithm. Then, a contribution index is established to measure the correlation between the variable and the fault, and the final diagnosis result is obtained by searching variables with large contributions. The smearing effect, which is an important factor affecting the performance of fault diagnosis, can be suppressed as well, and the theoretical analysis reveals that the correct diagnosis can be guaranteed by LLBBC. Finally, the feasibility and effectiveness of the proposed method are verified through a nonlinear numerical example and the Tennessee Eastman benchmark process.

Fault detection and recognition of multivariate process based on feature learning of one-dimensional convolutional neural network and stacked denoised autoencoder

Multivariate process pattern recognition (MPPR) is essential towards continuous quality control task. A challenging problem is to extract effective features from complex process signals with high-dimensional and nonlinear characteristics. This affects effectiveness of various classifiers in process fault detection and diagnosis significantly. In this paper, we propose a hybrid deep learning model (i.e. 1-DCNN + SDAE) that integrates one-dimensional convolutional neural network (1-DCNN) and stacked denoising auto-encoders (SDAE) to extract high level features from complex process signals. In comparison with two-dimensional images, one-dimensional process signals allow not only to extract spatial features, but also reduce calculation cost. 1-DCNN is capable of extracting representative features from one-dimensional process signals and then improves MPPR performance of classifiers significantly. SDAE is embedded after fully connected layer of 1-DCNN for further dimension reduction and feature extraction. 1-DCNN + SDAE preserves advantages of 1-DCNN and SDAE for feature learning from high-dimensional data. This makes it be flexible for those process fault detection and diagnosis tasks. The effectiveness of 1-DCNN + SDAE is validated on a complex numerical process, two process benchmarks i.e. Tennessee Eastman process (TEP) and Fed-batch fermentation penicillin process (FBFP), and a real-life manufacturing case of industrial conveyor belt. The experimental results illustrate effectiveness of the proposed method for feature learning and fault diagnosis on multivariate manufacturing processes. The comparison between 1-DCNN + SDAE and other typical DNNs on these processes, indicates the effectiveness of the proposed method for process fault detection and diagnosis. This study will provide the guidance for development of hybrid deep learning-based multivariate control models.

Multi-fault diagnosis of rotating machinery based on deep convolution neural network and support vector machine

Because multi-fault vibration signals in rotating machinery are often more complicated than single faults, human-designed fault feature sets are not yet able to respond adequately to the complex task of fault diagnosis. To address this, a fault diagnosis method, based on deep convolution neural network (DCNN) and support vector machine (SVM), is proposed. The DCNN, a data-driven deep-learning model, is applied to extract the fault features automatically. The fault-feature information can be extracted adaptively according to the minute differences in local fault signals. First, the envelope spectrum obtained by Hilbert transform was fed into the DCNN to extract the fault features, and then combined with SVM (the classifier) to diagnose multi faults of bearings and rotors. Then, 16 frequency-domain features were extracted by the DCNN, and the average accuracy was found to be 90.29%, which showed that machine wisdom alone was insufficient for this task. Therefore, two additional time-domain features extracted by human experts were adopted together with the machine-designed features to create the method that we call the semi-DCNN method. The theoretical mechanism of the proposed method was clear: it utilized both machine wisdom and human experience in designing the fault features. The effectiveness of the new method was proved by the fact that average diagnostic accuracy increased to 98.71% during diagnosis of composite faults in N205 bearing and rotor.

Multi-Fault Bearing Classification Using Sensors and ConvNet-Based Transfer Learning Approach

The development of sensor technology and modern computing allows the fault diagnosis of rotating machinery to exploitunder differentworkingconditions.As an effect, digital health monitoring of machine is performed based on the actual data values obtained from the multisensor, and also, it supports to design the data-rich datasets using multi-sensor. Thus, it motivates to develop the datadriven approaches for fault diagnosis based on multi-sensor data. However, they require a large amount of historical evidence for the development, and in a real-world situation, usually sufficient data is not available for building the datadriven model. As a result, they become less competent to perform the task of fault diagnosis. Therefore, in this paper, the transfer of knowledge from data-rich datasets is proposed to identifying the fault in new working conditions of the related domain. To transfer the experience, firstly the data has transformed to kurtogram, which is the fusion ofmulti-level kurtosis values and then convolutional neural network-based transfer learning approach has presented for fault diagnosis under various operating conditions. Also, the simultaneous change in speed and load in source and target domain has analyzed to explore the effect of the proposed method on performance of fault diagnosis. Further, the issue of negative transfer of knowledge has investigated by determining the association of speed as well as a load between the source and target domain. The proposed method is tested using two vibration datasets, and results demonstrate the decent performance of fault diagnosis using transfer learning approach.

Discriminative dictionary learning based sparse representation classification for intelligent fault identification of planet bearings in wind turbine

Planet bearing fault identification is an attractive but challenging task in wind turbine condition monitoring and fault diagnosis. Traditional fault characteristic frequency identification based diagnostic strategies are not sufficient for reliable planet bearing fault detection, due to complex physical configurations and modulation characteristics in planetary drivetrains. In this paper, we propose a discriminative dictionary learning based sparse representation classification (DDL-SRC) framework for intelligent planet bearing fault identification. Our framework could learn a reconstructive and discriminative dictionary for signal sparse representation and an optimal linear classifier for classification tasks simultaneously, which bridges the gap between dictionary learning and classifier training in traditional SRC methods. Specifically, the optimization objective introduces a novel regularization term called ‘discriminative sparse codes error’ and incorporates it with the reconstruction error and classification error. Thus, the dictionary learned by our framework possesses not only the reconstructive power for sparse representation but also the discriminative power for classifier training. The optimization formulation is efficiently solved using K-SVD and orthogonal matching pursuit algorithms. The experiment validations have been conducted for demonstrating the effectiveness and superiority of the proposed DDL-SRC framework over the state-of-the-art dictionary learning based SRC and deep convolutional neural network methods for intelligent planet bearing fault identification.

Missing-Class-Robust Domain Adaptation by Unilateral Alignment

Domain adaptation aims at improving model performance by leveraging the learned knowledge in the source domain and transferring it to the target domain. Recently, domain adversarial methods have been particularly successful in alleviating the distribution shift between the source and the target domains. However, these methods assume an identical label space between the two domains. This assumption imposes a significant limitation for real applications since the target training set may not contain the complete set of classes. We demonstrate in this paper that the performance of domain adversarial methods can be vulnerable to an incomplete target label space during training. To overcome this issue, we propose a two-stage unilateral alignment approach. The proposed methodology makes use of the inter-class relationships of the source domain and aligns unilaterally the target to the source domain. The benefits of the proposed methodology are first evaluated on the MNIST$\rightarrow$MNIST-M adaptation task. The proposed methodology is also evaluated on a fault diagnosis task, where the problem of missing fault types in the target training dataset is common in practice. Both experiments demonstrate the effectiveness of the proposed methodology.

Imbalanced Fault Diagnosis of Rolling Bearing Using Enhanced Generative Adversarial Networks

Machinery fault diagnosis tasks have been well addressed when sufficient and abundant data are available. However, the data imbalance problem widely exists in real-world scenarios, which leads to the performance deterioration of fault diagnosis markedly. To solve this problem, we present a novel imbalanced fault diagnosis method based on the enhanced generative adversarial networks (GAN). By artificially generating fake samples, the proposed method can mitigate the loss caused by the lack of real fault data. Specifically, in order to improve the quality of generated samples, a new discriminator is designed using spectrum normalization (SN) strategy and a two time-scale update rule (TTUR) method is used to stabilize the training process of GAN. Then, an enhanced Wasserstein GAN with gradient penalty is developed to generate high-quality synthetic samples for the fault samples set. Finally, a deep convolutional classifier is constructed to carry out fault classification. The performance and effectiveness of the proposed method are validated on the Case Western Reserve University bearing dataset and rolling bearing dataset acquired from our laboratory. The simulation results show that the proposed method has a superior performance than other methods for imbalanced fault diagnosis tasks.

Deep Semi-supervised Domain Generalization Network for Rotary Machinery Fault Diagnosis under Variable Speed

In recent years, deep learning has become a promising tool for rotary machinery fault diagnosis, but it works well only when testing samples and training samples are independent and identically distributed. In practice, rotary machinery usually works under variable speed. The change of speed leads to the variation of samples’ distribution, which can significantly decrease the performance of the deep learning model. Scholars try to utilize transfer learning techniques for solving this problem. However, most exiting methods can just work well under target speed instead of all speed, while the target samples are always required in model training. In this article, a deep semisupervised domain generalization network (DSDGN) is proposed for rotary machinery fault diagnosis under variable speed, which can generalize the model to the fault diagnosis task under unseen speed. Under the setting of semisupervised domain generalization, only one fully labeled source (LS) domain data set and one totally unlabeled source (US) domain data set are available during training. To make full use of these data, the proposed method simultaneously utilizes Wasserstein generative adversarial network with gradient penalty (WGAN-GP)-based adversarial learning and pseudolabel-based semisupervised learning for training. The transmission and bearing fault diagnosis cases are utilized for evaluation. The comparative experiments indicate that the proposed method has a better performance than other state-of-the-art methods.

Learn Generalization Feature via Convolutional Neural Network: A Fault Diagnosis Scheme Toward Unseen Operating Conditions

In recent years, Convolutional neural networks (CNNs) have achieved start-of-art performance in the fault diagnosis field. If there is no available information on the unseen operating conditions, the model trained on the seen operating condition cannot perform well. One of the feasible strategies is to enhance the generalization ability of the network on various seen operating conditions. We introduce the center loss to the traditional CNN and build an end-to-end fault diagnosis framework (called CNN-C). By minimizing the intra-class variations, center loss cluster the learned features across various seen operating conditions. With the joint supervision of the center loss and the softmax loss, the learned features of the same class could minimize the domain difference across various seen operating conditions while the features of different classes are separable. The generalization ability of network is improved on unseen operating conditions. Compared with the shallow methods and traditional CNN, the proposed method is promising to deal with the fault diagnosis tasks of the bearing and gearbox.

Understanding and Learning Discriminant Features based on Multiattention 1DCNN for Wheelset Bearing Fault Diagnosis

Recently, deep-learning-based fault diagnosis methods have been widely studied for rolling bearings. However, these neural networks are lack of interpretability for fault diagnosis tasks. That is, how to understand and learn discriminant fault features from complex monitoring signals remains a great challenge. Considering this challenge, this article explores the use of the attention mechanism in fault diagnosis networks and designs attention module by fully considering characteristics of rolling bearing faults to enhance fault-related features and to ignore irrelevant features. Powered by the proposed attention mechanism, a multiattention one-dimensional convolutional neural network (MA1DCNN) is further proposed to diagnose wheelset bearing faults. The MA1DCNN can adaptively recalibrate features of each layer and can enhance the feature learning of fault impulses. Experimental results on the wheelset bearing dataset show that the proposed multiattention mechanism can significantly improve the discriminant feature representation, thus the MA1DCNN outperforms eight state-of-the-arts networks.

Multiple wavelet regularized deep residual networks for fault diagnosis

As an emerging deep learning method, deep residual networks are gradually becoming popular in the research field of machine fault diagnosis. A significant task in deep residual network-based fault diagnosis is to prevent overfitting, which is often a major reason for low diagnostic accuracy when there is insufficient training data. This paper develops a multiple wavelet regularized deep residual network (MWR-DRN) model that uses one wavelet basis function (WBF) as the primary WBF and other WBFs as the auxiliary WBFs. “Regularized” means that a constraint or restriction is applied to yield a high performance on the testing data. To be specific, the developed MWR-DRN model is trained not only by the 2D matrices from the primary WBF, but also by the 2D matrices from the auxiliary WBFs using a stochastic selection strategy. Experimental results validate the effectiveness of the developed MWR-DRN in improving diagnostic accuracy.

An approach for fault detection and location in solar PV systems

The task of fault detection and diagnosis in large-scale photovoltaic (PV) plants is expected to be a major challenge as more and more plants with increasingly large capacities continue to come into existence. To maintain safety, reliability, and productivity of large-scale PV plants it is essential to develop approaches that allow automatic detection and location of any mal-operation among thousands of PV modules. This paper proposes an approach to detect PV plant faults through the generation of fault indicator signals called ‘residuals’ for each string and the comparison of residuals with a threshold value. Furthermore, a regression-based approach is proposed to estimate fault location as a function of fault current and irradiance level measurements. The proposed approach is demonstrated by specifically focusing on intra-string line-line faults. Various line-line fault case studies are verified through simulations and validated on an experimental setup in a solar PV plant. Initially the approach is developed for a fixed size PV array, but subsequently extensions of the proposed approach to any array size are developed and presented. The generalisation potential of the approach is demonstrated through successful validation on multiple PV array configurations.

Refined composite multivariate multiscale symbolic dynamic entropy and its application to fault diagnosis of rotating machine

Accurate and efficient identification of various fault categories, especially for the big data and multisensory system, is a challenge in rotating machinery fault diagnosis. For the diagnosis problems with massive multivariate data, extracting discriminative and stable features with high efficiency is the significant step. This paper proposes a novel feature extraction method, called Refined Composite multivariate Multiscale Symbolic Dynamic Entropy (RCmvMSDE), based on the refined composite analysis and multivariate multiscale symbolic dynamic entropy. Specifically, multivariate multiscale symbolic dynamic entropy can capture more identification information from multiple sensors with superior computational efficiency, while refine composite analysis guarantees its stability. The abilities of the proposed method to measure the complexity of multivariate time series and identify the signals with different components are discussed based on adequate simulation analysis. Further, to verify the effectiveness of the proposed method on fault diagnosis tasks, a centrifugal pump dataset under constant speed condition and a ball bearing dataset under time-varying speed condition are applied. Compared with the existing methods, the proposed method improves the classification accuracy and F-score to 99.81% and 0.9981, respectively. Meanwhile, the proposed method saves at least half of the computational time. The result shows that the proposed method is effective to improve the efficiency and classification accuracy dealing with the massive multivariate signals.

Fault diagnosis for space utilisation

The space application task is to carry out various scientific experiments and applied research by using the ability of space experiment of spacecraft. In the past 20 years, >50 space application studies have been carried out in Chinese manned space flight application system, >500 units have been involved in the previous flight missions, and fruitful results have been achieved. The white paper ‘Chinese spaceflight in 2016’ pointed out that in the next 5 years, Chinese satellite system will enhance the level and basic ability to construct the satellite system. Chinese manned space station project is scheduled to be completed ∼2022 and it will plan to operate >10 years. The space station, based on the world-wide integrated information network, has a large number of payloads and will become a national space laboratory. Space activities are full of risks and challenges. On the basis of a great deal of literatures, the method of avoiding space risk in the field of spaceflight is discussed. Aiming at the fault diagnosis task for space utilisation, the intelligent methods of deep learning including deep belief network, convolutional neural network and generative adversarial network are discussed.

A slow independent component analysis algorithm for time series feature extraction with the concurrent consideration of high-order statistic and slowness

For the process data analytics, numerous statistical methods are designed to extract informative features to reveal latent characteristics and correlation patterns among the variables. As a classic statistical method, independent component analysis extracts mutually independent components by searching for the maximum direction of non-Gaussianity from the mixed signals. However, the algorithm does not work for the latent Gaussian components, and the disorder of extracted features also brings trouble to the practical applications. In the present work, a slow independent component analysis algorithm is proposed to tackle the problems by concurrently considering the high-order statistic and slowness. It constructs a dual-objective optimization criterion function, simultaneously incorporating independence and slow varying characteristics into the feature extraction procedure, which ties together the ideas of independent component analysis and slow feature analysis under the same mathematical umbrella. Through adjusting the controllable sub-optimization objective weight, the proposed slow independent component analysis adds insight into the different roles of independence and slow varying characteristics in calibration modeling, and, thus, provides possibilities to combine the advantages of independent component analysis and slow feature analysis. Compared with the traditional method, the new algorithm no longer relies solely on higher-order statistic but also utilizes the slow varying characteristic to extract latent Gaussian or non-Gaussian components from mixed signals. Besides the unsupervised paradigm, the proposed algorithm is extended to the supervised form for classification task by utilizing the temporal structure as well as the label information. The effectiveness of the proposed algorithm is illustrated by blind source separation task and industrial fault diagnosis task using real-world datasets including speech signals and multiphase flow facility.

Intelligent Fault Identification Based on Multisource Domain Generalization Towards Actual Diagnosis Scenario

The data-driven diagnosis methods based on conventional machine-learning techniques have been widely developed in recent years. However, the assumption of conventional methods that the training and test data should be identically distributed is usually unsatisfied in actual diagnosis scenario. While there are several existing works that have been studied to construct diagnosis models by transfer learning methods, most of them are only focused on learning from a single source. Actually, how to discover effective and general diagnosis knowledge from multiple related source domains and further generalize the learned knowledge to new target tasks is crucial to data-driven fault diagnosis. To this end, this paper proposes a novel intelligent fault identification method based on multiple source domains. First, the method describes the discriminant structure of each source domain as a point of Grassmann manifold using local Fisher discriminant analysis. Through preserving the within-class local structure, local Fisher discriminant analysis can learn effective discriminant directions from multimodal fault data. Second, the mean subspace of source domains is computed on the Grassmann manifold through Karcher mean. The mean subspace can be viewed as a representation of the general diagnosis structure that can facilitate the construction of the diagnosis model for the target domain. Experiments on bearing fault diagnosis tasks verify the effectiveness of the proposed method.

Periodic feature oriented adapted dictionary free OMP for rolling element bearing incipient fault diagnosis

Abstract To address the troublesome rolling element bearing incipient fault diagnosis task, it is of great importance to extract the periodic impact components embedded in heavy noise usually. The newly developed sparse representation approach, named adapted dictionary free orthogonal matching pursuit (ADOMP), not only provides a more flexible template – Asymmetric Gaussian Chirplet Model (AGCM), but also breaks through the dependence on essential dictionaries for analytic and learning-based sparse representation approaches. Therefore, ADOMP possesses the potential advantages in analyzing bearing incipient fault vibration signals. Considering the significant influence caused by a manual-selected stopping parameter in standard ADOMP, two pieces of prior knowledge are firstly exploited from the vibration signals. Then through establishing a harmonic product spectrum-based kurtosis to directly evaluate the obviousness of extracted periodic features and guide a satisfactory selection for the stopping parameter, the periodic feature oriented ADOMP (PF-ADOMP) is developed. The subsequent refinement by virtue of the exploited time-invariance can further highlight the periodic feature, which is beneficial to the detection of bearing incipient faults. The effectiveness of PF-ADOMP has been validated by analyzing the data from both simulation and bearing life accelerated tests. Meanwhile, the comparisons with respect to the automatic oscillatory behavior-based signal decomposition (Auto-OBSD) method have also been favorably conducted, which show the superiority of the PF-ADOMP in extracting rolling element bearing incipient fault features.

A novel gas turbine fault diagnosis method based on transfer learning with CNN

A transfer learning method based on CNN and SVM is investigated for gas turbine fault diagnosis. The excellent classification ability of CNNs is attributed to their ability to learn rich feature representations from a large number of annotated samples. This property, however, currently prevents application of CNNs to problems with fewer samples. This paper shows how feature representations learned with CNN on large-scale annotated gas turbine normal dataset can be efficiently transferred to fault diagnosis task with limited fault data. A feature mapping method to extract the feature representations for fault dataset by reusing the internal layers of CNN trained on the normal dataset is designed, and SVM is used for fault diagnosis. The influence of gas turbine performance parameters arrangement order on proposed method is theoretically analyzed. Finally, the proposed method is validated by the real-life operation data of a gas turbine sample fleet. The experimental results show that despite difference in the two datasets, the transferred feature representations lead to significant improved results for fault diagnosis as well as obviously weaken the individual difference and data noises. The experimental results also confirm that the proposed method has excellent ability for fault diagnosis under small sample condition.

Advancement of Fault Diagnosis and Detection Process in Industrial Machine Environment

Machine fault diagnosis is a very important topic in industrial systems and deserves further consideration in view of the growing complexity and performance requirements of modern machinery. Currently, manufacturing companies and researchers are making a great attempt to implement efficient fault diagnosis tools. The signal processing is a key step for the machine condition monitoring in complex industrial rotating electrical machines. A number of signal processing techniques have been reported from last two decades conventionally and effectively applied on different rotating machines. Induction motor is the one of widely used in various industrial applications due to small size, low cost and operation with existing power supply. Faults and failure of the induction machine in industry can be the cause of loss of throughput and significant financial losses. As compared with the other faults with the broken rotor bar, it has significant importance because of severity which leads to a serious breakdown of motor. Detection of rotor failure has become significant fault but difficult task in machine fault diagnosis. The aim of this paper is indented to summarizes the fault diagnosis techniques with the purpose of the broken rotor bar fault detection. Keywords: machine fault diagnosis, signal processing technique, induction motor, condition monitoring.