Industrial Fault Diagnosis Research Articles

A Novel Hybrid Signal Decomposition Technique for Transfer Learning Based Industrial Fault Diagnosis

In the fourth industrial revolution, data-driven intelligent fault diagnosis for industrial purposes serves a crucial role. In contemporary times, although deep learning is a popular approach for fault diagnosis, it requires massive amounts of labelled samples for training, which is arduous to come by in the real world. Our contribution to introduce a novel comprehensive intelligent fault detection model using the Case Western Reserve University dataset is divided into two steps. Firstly, a new hybrid signal decomposition methodology is developed comprising Empirical Mode Decomposition and Variational Mode Decomposition to leverage signal information from both processes for effective feature extraction. Secondly, transfer learning with DenseNet121 is employed to alleviate the constraints of deep learning models. Finally, our proposed novel technique surpassed not only previous outcomes but also generated state-of-the-art outcomes represented via the F1 score.

Industrial fault diagnosis based on active learning and semi-supervised learning using small training set

Industrial fault diagnosis has been investigated for many years, and many approaches have been proposed to identify industrial faults. However, the size of the actual training set is usually small, which severely degrades the performance of existing fault diagnostic models. To solve this problem, a new fault diagnosis method was proposed based on active and semi-supervised learning. First, uncertain unlabelled samples were selected by estimating the first two values in the class probability distribution of the samples. They were labelled by experts to update the performance of the models learned from a small training set. Second, heterogeneous classifiers were adopted to increase the diversity of the base classifiers, and noise samples were deleted using a sample pruning operation. The weights of the base classifier were designed for ensemble learning based on the test error rates. An evaluation using the Case Western Reserve University and Intelligent Maintenance Systems data showed that the performance of the proposed method was better than those of the other methods in the experiment. The experimental results showed that this study provided a promising and useful methodology for fault diagnosis under a small training set.

Fault diagnosis of a rolling bearing based on the wavelet packet transform and a deep residual network with lightweight multi-branch structure

Deep residual networks (DRNs) are a state-of-the-art deep learning model used in the data-driven fault diagnosis field. Their especially deep architectures give them sufficient capacity to deal with very complex diagnosis issues. However, a neural network with excellent performance usually requires hundreds of thousands of parameters, which is unaffordable for use in current industrial machines due to their limited computational resources. To enable practical applications for fault diagnosis, developing deep learning methods that can perform powerfully and have an economical computational burden is necessary. This study proposes a novel bearing fault diagnosis method based on the wavelet packet transform (WPT) and a lightweight variant of DRN called a multi-branch deep residual network (MB-DRN) in order to resolve the above issues. WPT is utilized to map raw signals into the time-frequency domain, from which the MB-DRN can extract a set of robust features more easily. Additionally, MB-DRN builds several small-sized convolutional layer branches in each building block to increase the network non-linearity, the construction of layer branches can be achieved freely and this design strategy largely saves the parameter usage while approaching a stronger model’s capacity. Two rolling bearing datasets with variable operating conditions were conducted on the proposed method to validate performance. The results verify the necessity of the WPT-based data processing method and show that MB-DRN can outperform the accuracies of standard DRN with only one quarter of the parameter amount, revealing the significant potential of the proposed method for realistic industrial fault diagnosis applications.

Auxiliary Information-Guided Industrial Data Augmentation for Any-Shot Fault Learning and Diagnosis

The label scarcity problem widely exists in industrial processes. In particular, samples of some fault types are extremely rare; even worse, the samples of certain faults cannot be accessed, but they may appear in the actual process. These two kinds of challenges together can be termed as any-shot learning problem in industrial fault diagnosis. In this article, taking the advantages of generative adversarial network, a generative approach is proposed to tackle the any-shot learning problem, which generates the abundant samples for those rare and inaccessible faults, and trains a strong diagnosis model. To reach this, an attribute space is built to introduce the auxiliary information, which achieves the diagnosis of unseen faults and makes the generated samples more resembled to the real data. Besides, an auxiliary loss of triplet form is introduced as a joint training loss term, further improving the quality of augmented data and diagnosis accuracy. Finally, the performance of model is verified by the experiments of a hydraulic system and Tennessee-Eastman process, the results of which show that our method performs excellently for both zero-shot and few-shot fault diagnosis problems.

An Improved Non-negative Matrix Factorization Method for Dynamic Industrial Fault Diagnosis

Most classical fault diagnosis methods such as principal component analysis (PCA), are extracting comprehensive information to represent data features in fault diagnosis. In comparison, Non-Negative Matrix Factorization (NMF) is a method for dimension reduction and feature extraction, and because its characteristic matrix has sparsity, this method is superior in the ability of extracting the local feature and suppressing noises; however, the NMF method is not applicable for dynamic industrial processes. In this paper, we introduce the past information of industrial processes for fault diagnosis, proposing Canonical Variate Analysis - Non-Negative Matrix Factorization (CVA-NMF) methods to improve the dynamic performance of NMF. The experimental results via TE process indicate that the proposed approach could handle a dynamic production process such as Fault 2 and Fault 5 and retain the superior performance of NMF.

Fault detection in switching process of a substation using the SARIMA\u2013SPC model

To detect substation faults for timely repair, this paper proposes a fault detection method that is based on the time series model and the statistical process control method to analyze the regulation and characteristics of the behavior in the switching process. As the first time, this paper proposes a fault detection model using SARIMA, statistical process control (SPC) methods, and 3σ criterion to analyze the characteristics in substation’s switching process. The employed approaches are both very common tools in the statistics field, however, via effectively combining them with industrial process fault diagnosis, these common statistical tolls play excellent role to achieve rich technical contributions. Finally, for different fault samples, the proposed method improves the rate of detection by at least 9% (and up to 15%) than other methods.

A Fine-Grained Adversarial Network Method for Cross-Domain Industrial Fault Diagnosis

While machine-learning techniques have been widely used in smart industrial fault diagnosis, there is a major assumption that the source domain data (where the diagnosis model is trained) and the future target data (where the model is applied) must have the same distribution. However, this assumption may not hold in real industrial applications due to the changing operating conditions or mechanical wear. Recent advances have embedded the adversarial-learning mechanism into deep neural networks to reduce the distribution discrepancy between different domains to learn domain-invariant features and perform fault diagnosis. However, they only aligned the distributions of domains and neglected the fault-discriminative structure underlying the target domain, which leads to a decline in the diagnostic performance. In this article, a new method termed the fine-grained adversarial network-based domain adaptation (FANDA) is proposed to address the cross-domain industrial fault diagnosis problem. Different from the existing domain adversarial adaptation methods considering the domain discrepancy only, the features in FANDA are learned by competing against multiple-domain discriminators, which enable both a global alignment for two domains and a fine-grained alignment for each fault class across two domains. Thus, the fault-discriminative structure underlying two domains can be preserved in the adaptation process and the fault classification ability learned on the source domain can remain effective on the target data. Experiments on a mechanical bearing case and an industrial three-phase flow process case demonstrate the effectiveness of the proposed method.

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.

Application of fault monitoring and diagnosis in process industry based on fourth order moment and singular value decomposition

AbstractPrincipal component analysis (PCA) and partial least squares (PLS) have been frequently used for process industry monitoring; however, their application on industrial sites is limited because they cannot be used to process data with non‐Gaussian distribution. Independent component analysis (ICA) has become a powerful modelling method for non‐Gaussian process monitoring. However, the ICA‐based modelling method has been found to contribute to double the amount of data loss in feature extraction. There are two reasons for this. First, when the PCA algorithm is used to whiten the original data, the smaller principal component is discarded. Second, when selecting independent components, some smaller independent components will be discarded according to the evaluation index. The abovementioned two data feature extraction methods may discard useful information for fault monitoring, which will inevitably lead to inaccurate fault monitoring. To solve this problem, a fault monitoring and diagnosis method based on fourth order moment (FOM) analysis and singular value decomposition (SVD) is proposed. First, the fourth order moments of each process variable were constructed separately. Then, the data space of the fourth order moments was decomposed by singular value decomposition to establish the global monitoring statistics. Finally, the contribution diagram was drawn and the fault diagnosis was performed based on the global monitoring results. The proposed method was applied to the Tennessee Eastman (TE) simulation platform, and its effectiveness and feasibility were verified by a comparison with PCA and ICA.

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.

Fault diagnosis and prioritisation in sugar industry: a fuzzy-AHP approach

The main aim of this study is to develop a fuzzy analytic hierarchy process (FAHP) model to prioritise the faults of the sugar industry. Although the concept of fault diagnosis has been discussed in previous studies, fault diagnosis of sugar industry is not investigated yet. To fill this gap, this study is carried out in three main phases as follows. Firstly, the general faults of industries are developed from the previous literature. Secondly, the developed faults are investigated with regard to their availability in sugar industry. Finally, a FAHP model is applied to prioritise the faults of sugar industry. According to the results, old technology, improper financing, poor research and development and lack of creativity and innovation are four main faults of sugar industry, respectively. Some managerial implications are suggested to deal with these faults and improve the performance of this industry.

Integrated Wireless Technologies with Computer for Industrial Machinery Fault Diagnosis: Challenges Comparison and Characteristics: A Review

Integrated Wireless Technologies with Computer for Industrial Machinery Fault Diagnosis: Challenges Comparison and Characteristics: A Review

A deep Boltzmann machine and multi-grained scanning forest ensemble collaborative method and its application to industrial fault diagnosis

The essence of big data-based intelligent industrial fault diagnosis lies in the process of machine learning and feature engineering. Deep learning methods can discover the complex relationship between data and potential faults, and outperform the traditional machine learning methods. The gcForest is able to generate a deep forest ensemble, which allows gcForest to do representation learning and fault classification. At each node of the random tree, gcForest selects the one with the best gini value from candidates for splitting. However, most of the data acquired from industrial scene is with continuous and unstructured attributes, accordingly the node-splitting procedure will be generally intractable. We present a novel approach with the combination of deep Boltzmann machine and multi-grained scanning forest ensemble, to effectively deal with industrial fault diagnosis based on big data. At first, we use deep Boltzmann machine to turn all features of data to be processed by forests into binary, and then utilize multi-grained scanning forest ensemble to process them in every layer of deep Boltzmann machine. By means of the collaborative method, we can address the aforementioned issues. The experimental results and analysis on industrial fault diagnosis under different experimental conditions, show that the fault classification accuracy of the proposed approach is competitive to other popular deep learning algorithms, but also takes much less time than gcForest.

Overview of Various Industrial Fault Diagnosis Methods

Overview of Various Industrial Fault Diagnosis Methods

DPLS-MSOM Modeling for Visual Industrial Fault Diagnosis and Monitoring Based on Variation Data from Normal to Anomalous

Process variables obtain significant dynamic variation characteristics from the moment faults are introduced through a continuous period of time. Therefore, these data possess not only strong corresponding fault information but also robust dynamic characteristics that can be effectively used for fault diagnosis. Accordingly, the dynamic partial least squares (DPLS) and self-organizing map (SOM) combination methodology is utilized for visual fault diagnosis and monitoring. Different types of fault data (from normal to anomalous) are collected and trained by DPLS in the form of a dynamic data matrix to reveal the most discriminated orientations for the different statuses. A multilayer SOM is then trained to model the relationship with different status data under such orientations to the regular positions on a two-dimensional map. The trained map can be used for deciding the real operating state of the online observations. Excellent experimental results on the Tennessee Eastman chemical process have demonstr...

Optimal Sensor Network Upgrade for Fault Detection Using Principal Component Analysis

The efficiency of a fault monitoring system critically depends on the structure of the plant instrumentation system. For processes monitored using principal component analysis, the multivariate statistical technique most used for fault diagnosis in industry, an existing strategy aims at selecting the set of instruments that satisfies the detection of a given set of faults at minimum cost. It is based on the minimum fault magnitude concept. Because that procedure discards lower-cost feasible solutions, in this work, a new optimal selection methodology is proposed whose constraints are straightaway defined in terms of the principal component analysis’s statistics. To solve the optimization problem, a level traversal search with cutting criteria is enhanced taking into account that the fault observability is a necessary condition for fault detection when statistical monitoring techniques are applied. Furthermore, observability and detection degree concepts are defined and considered as constraints of the opt...

An industrial Fault Diagnosis System based on Bayesian Networks

This paper presents a DC motor fault diagnosis system based on Bayesian networks. This was done by the design of a new electromechanical test bed allowing the collection of functioning data from a real world industrial Direct current (DC) Motor. The data collection will help in the construction of Bayesian networks models. These data are collected from sensors measuring different types of variables that are directly related to the industrial system. Without doing any mathematical modeling that describes the physical properties of the studied DC motor, the proposed tool provides with the help of Bayesian networks parameters and structure learning algorithms, the base to construct a fault diagnosis tool that can be extended to a fault prognosis tool.

A Modeling and Probabilistic Reasoning Method of Dynamic Uncertain Causality Graph for Industrial Fault Diagnosis

Online automatic fault diagnosis in industrial systems is essential for guaranteeing safe, reliable and efficient operations. However, difficulties associated with computational overload, ubiquitous uncertainties and insufficient fault samples hamper the engineering application of intelligent fault diagnosis technology. Geared towards the settlement of these problems, this paper introduces the method of dynamic uncertain causality graph, which is a new attempt to model complex behaviors of real-world systems under uncertainties. The visual representation to causality pathways and self-relied “chaining” inference mechanisms are analyzed. In particular, some solutions are investigated for the diagnostic reasoning algorithm to aim at reducing its computational complexity and improving the robustness to potential losses and imprecisions in observations. To evaluate the effectiveness and performance of this method, experiments are conducted using both synthetic calculation cases and generator faults of a nuclear power plant. The results manifest the high diagnostic accuracy and efficiency, suggesting its practical significance in large-scale industrial applications.

Application of the Technology of Multi-Source Information Fusion in Industrial Monitoring and Fault Diagnosis

nformatization construction is a major issue of current social construction, and the technology of multi-source information fusion is an integral part of the information construction. This paper provides an overview of the technology of the multi-source information fusion, gives an introduction of the application of the technology of multi-source information fusion in industrial monitoring and fault diagnosis, and gives the direction of development and trends of the technology in the future according to the main problems in the research of the technology of multi-source information fusion.

Outline of a fault diagnosis system for a large-scale board machine

Global competition forces the process industries to continuously optimize plant operation. One of the latest trends in efficiency and plant availability improvement is to set up fault diagnosis and maintenance systems for online industrial use. This paper presents a methodology for developing industrial fault detection and diagnosis (FDD) systems. Since model- or data-based diagnosis of all components cannot be achieved online on a large-scale basis, the focus must be narrowed down to the most likely faulty components responsible for abnormal process behavior. One of the key elements here is fault analysis. The paper describes and briefly discusses also other development phases, process decomposition and the selection of FDD methods. The paper ends with an FDD case study of a large-scale industrial board machine including a description of the fault analysis and FDD algorithms for the resulting focus areas. Finally, the testing and validation results are presented and discussed.