Data-driven Fault Diagnosis Method Research Articles

Data-driven fault diagnosis of PEMFC water management with segmented cell and deep learning technologies

Timely and effective fault diagnosis is crucial to guarantee the reliability and durability of proton exchange membrane fuel cell (PEMFC). Segmented cell technology has been proven to be a powerful tool for exploring the internal characteristics of fuel cells, yet it is seldom applied for fault diagnosis. This paper proposes a data-driven fault diagnosis method that combines segmented cell and deep learning technology for PEMFC water management fault diagnosis. The dynamic characteristics of the current distribution of the fuel cell under different levels of faults are explored with segmented cell technology. A dual-input convolutional neural network is applied to combine current distribution data and sensor data for a more comprehensive system status for fault diagnosis. The results show that the model can efficiently identify flooding and drying faults with more than 98.5% accuracy, faster and more accurately than other algorithms. This work highlights the significance of segmented cell technology for the comprehension of water management inside PEMFC and provides a new concept for PEMFC fault diagnosis.

Data-driven fault diagnosis method for abnormal RF oscillation of gyrotrons

Gyrotrons are vacuum electronic devices capable of generating high-power, high-frequency THz radiation. With the increasing utilization of gyrotrons in thermonuclear fusion experimental devices, achieving higher output performance and ensuring stable and reliable operation have become key development goals. However, occasional radio frequency (RF) oscillation faults inside the gyrotron during long-pulse operation hinder its stable and reliable operation. Due to the difficulty in directly observing abnormal changes within the gyrotron resonator during operation, this paper explores a data-driven approach to diagnose RF oscillation faults in the gyrotron for the first time and proposes a classification model that combines K-Nearest Neighbors and Random Forest (KNN-RForest) algorithms for fault identification. Compared with seven baseline models, the results show that our proposed KNN-RForest model has a better classification performance. It is verified that data-driven methods can effectively identify RF oscillation faults in gyrotrons. Finally, the state probabilities of the gyrotron under different power levels were predicted using the KNN-RForest model, demonstrating the application of the KNN-RForest model in identifying gyrotron RF oscillation faults, which may be helpful for setting stable operating parameters for the gyrotron to reduce the likelihood of RF oscillation faults.

Fault diagnosis of injection molding machine non-return valve based on data-driven model

Non-return valve (NRV) is one of the key components in determining the consistency of the quality of the products molded by injection molding machine. Wear on the NRV affects the quality of the molded product. Nevertheless, detecting wear on the NRV can be challenging and disassembly of the machine is the only diagnostic method, which can have a negative impact on productivity. In this paper, a data-driven fault diagnosis method is proposed, which uses Stacked Auto Encoder (SAE) to analyze the pressure, torque, and displacement signals of the injection molding machine and combined with XGBoost (Extreme Gradient Boosting) to diagnose the faults of the NRV. The experimental results indicate that the SAE-XGBoost method accurately predicts NRV failures. Compared to using only XGBoost for prediction, the accuracy has improved from 97.5% to 99.6%. Eventually, the SAE-XGBoost model is integrated into the control program of the injection molding machine in the form of functional modules. Throughout the production process, the model adeptly monitors and identifies the production profile, promptly dispatching warning messages to users when diagnosing NRV wear. This facilitates intelligent diagnosis of the service status of injection molding machine components, which will have a positive influence on improving the production efficiency and intelligence of injection molding machines. The results of this study represent a synergistic application of artificial intelligence and time-domain statistical features in the realm of fault diagnosis for injection molding machines. This has the potential to significantly broaden the scope of AI utilization within the domain of injection molding processes, thereby advancing the intelligent technology associated with injection molding machines.

Hierarchical multihead self-attention for time-series-based fault diagnosis

Fault diagnosis is important for maintaining the safety and effectiveness of chemical process. Considering the multivariate, nonlinear, and dynamic characteristic of chemical process, many time-series-based data-driven fault diagnosis methods have been developed in recent years. However, the existing methods have the problem of long-term dependency and is difficult to train due to the sequential way of training. To overcome these problems, a novel fault diagnosis method based on time-series and the hierarchical multi-head self-attention (HMSAN) is proposed for chemical process. First, a sliding window strategy is adopted to construct the normalized time-series dataset. Second, the HMSAN is developed to extract the time-relevant features from the time-series process data. It improves the basic self-attention model in both width and depth. With the multi-head structure, the HMSAN can pay attention to different aspects of the complicated chemical process and obtain the global dynamic features. However, the multiple heads in parallel lead to redundant information, which cannot improve the diagnosis performance. With the hierarchical structure, the redundant information is reduced and the deep local time-related features are further extracted. Besides, a novel many-to-one training strategy is introduced for HMSAN to simplify the training procedure and capture the long-term dependency. Finally, the effectiveness of the proposed method is demonstrated by two chemical cases. The experimental results show that the proposed method achieves a great performance on time-series industrial data and outperforms the state-of-art approaches.

Gas Path Fault Diagnosis of Turboshaft Engine Based On Novel Transfer Learning Methods

Abstract Data-driven fault diagnosis method is widely used in the field of engine health management, which uses engine sensor data as input and engine faulty components as output for component-level fault diagnosis of the engine. The application premise of the general data-driven fault diagnosis method is that all data come from the same working conditions, that is, they belong to the same distribution. However, this assumption is not valid in the actual engine fault diagnosis, because the engine state will change with the increase of running time. In the meantime, collecting engine data is usually expensive, time-consuming, and laborious. To solve these problems, extreme learning machine (ELM) based two transfer learning methods for fault diagnosis of turboshaft engines are proposed in this paper. One is joint solving ELM (JSELM), which regards the information of the target domain and source domain as similar and different parts respectively, and knowledge is extracted from them at the same time. The other is model transfer-based ELM (MTELM), which uses the idea of pretraining. First, a general ELM classifier is trained with the source domain data and then fine-tuned with the target domain data. Both methods have a good real-time performance as the traditional ELM. When there are a few data in the target domain, they achieve much better classification accuracy than traditional ELM. Finally, experiments are carried out with turboshaft engine simulation data. The results show that both methods are effective, especially MTELM, which has better classification accuracy.

Industrial process fault diagnosis based on feature enhanced meta-learning toward domain generalization scenarios

With the rise of the Internet of Things, computing science, and artificial intelligence, data-driven fault diagnosis methods for industrial processes have attracted considerable attention. However, these methods typically assume that the distribution of the test data (target domains) is similar to that of the training data (source domains), making it difficult to deploy them in actual scenarios because the distribution of the collected data varies with operating conditions. Faced with this challenge, a novel fault diagnosis method based on feature enhanced meta-learning (FEML) is proposed to learn domain generalization knowledge effectively. It aims to jointly extract domain-invariant information that can be transferred across domains, and domain-specific information related to classification labels. First, feature extractors based on gated recurrent units (GRUs) are designed to capture domain-invariant and domain-specific representations with time-sequence characteristics from industrial process data. Second, a meta-learning strategy is introduced to enhance the domain-specific representation, thus improving the generalization ability of the model without accessing the target domains. Finally, the Tennessee Eastman process (TEP), Fed-Batch Fermentation Penicillin process (FBFP), and Three-Phase Flow process (TPFP) are used to verify the effectiveness of FEML. The results illustrate that the FEML is robust and outperforms other related methods. The diagnostic accuracies for TEP, FBFP, and TPFP are 87.11%, 95.94%, and 83.07%, respectively.

A Multi-Information Fusion Algorithm to Fault Diagnosis of Power Converter in Wind Power Generation Systems

Power electronics-based converters are the major and most vulnerable components in wind power generation systems. Converter faults will affect power quality,damage expensive equipment such asgenerators, or even pose a massive threat to the entire power grid. Fault diagnosis is considered as a powerful means to improve system reliability and reduce maintenance costs. Existing data-driven fault diagnosis methods related to improving feature extraction approaches to obtain high diagnostic accuracy are mostly based on single-scale feature of signals; this neglects the potentially valuable information of other scales. This paper proposes an algorithm of Dempster-Shafer and Deng entropy fusion multi-scale approximate entropy (DSDEMAE) for wind power converter fault diagnosis. Firstly, it calculates the multi-scale approximate entropy of fault signals, mining more potentially valuable information. Secondly, Dempster-Shafer is used to fuse features of various scales and effectively handles the conflicts and uncertainties between different features. Finally,Deng entropyis adopted to measure the uncertainty to adjust weight distribution, reducing the influence of highly conflicting features and mutually benefitting features of different scales. Extensive experimental results on simulated and experimental data demonstrate the effectiveness of the proposed algorithm. Compared with advanced methods, this method can diagnose the faults with higher accuracy and stronger robustness.

Residual attention temporal recurrent network for fault diagnosis of gearboxes under limited labeled data

Data-driven fault diagnosis methods have significantly contributed to the rapid development of prognostics and health management of gearboxes. However, these methods require sufficient labeled data for model training. In real industrial scenarios, labeled data is usually scarce. Thus, a residual attention temporal recurrent network is proposed for fault diagnosis of gearboxes under limited labeled data. More specifically, a fault diagnosis framework with self-supervised sample generation learning is explored in this paper. First, a small amount of labeled data is utilized to self-supervise the generation of new samples with similar semantics in adversarial generation modeling. Second, a temporal recurrent network with residual attention is constructed to extract fault features of vibration signals using two-stage temporal and recursive operations. In addition, the construction of the residual attention mechanism can facilitate the network to have a deeper semantic understanding of the vibration signal. Finally, the model can guarantee a more robust classifier in the proposed fault diagnosis framework using a small number of labeled samples. The extensive validation of the proposed method is performed on the gearbox dataset. The experimental results show that the proposed method can achieve competitive fault diagnosis performance under limited labeled data.

Fault detection and diagnosis of energy system based on deep learning image recognition model under the condition of imbalanced samples

Faults in energy systems impact the reliability of the energy supply and cause energy waste. Data-driven fault detection and diagnosis (FDD) methods can detect and diagnose system faults by mining historical operational data. However, the quantitative imbalance between fault and normal samples degrades the performance of the FDD method. Therefore, this study proposes a novel deep learning-based FDD method under the condition of imbalanced samples that converts the time-series signals into an image signal to extract the timing and coupling features, and then applies the improved conditional variational autoencoder-generative adversarial network (CVAE-GAN) to generate fault samples for balancing the training sample set. Subsequently, according to the framework of image recognition, a two-dimensional convolutional neural network was adopted to identify the image samples and achieve FDD. The experimental results showed that, under the condition of imbalanced samples, the proposed method could increase diagnosis accuracy by an average of 5.71% compared with other common data-driven methods. After the samples were augmented with the improved CVAE-GAN, the accuracy improved by an average of 3.79%. Consequently, the feasibility and superiority of the proposed method were demonstrated.

Fault diagnosis of marine machinery via an intelligent data-driven framework

Large amounts of ship accidents caused by marine machinery indicate the significance of fault diagnosis for marine machinery. Since numerous onboard sensor signals contain feature information that reflects the status of marine machinery and considering the time-consuming and experience-dependent of previous data-driven fault diagnosis methods, an intelligent data-driven framework based on a multi-head attention neural network (MANet) is proposed for the fault diagnosis of marine machinery in this work. MANet integrates the multi-head attention (MHA) mechanism, convolutional layers, and residual (Res) structure. Therein, the global correlation features in long-term operating signal sequences extracted by MHA can effectively characterize the status variation of marine machinery; the local fault features in operating signals extracted by convolutional layers can effectively characterize status information at certain moments for marine machinery; the application of Res structure in MANet is conducive to alleviating the gradient propagation anomaly. With a case study for the fault diagnosis of marine machinery, the superiority of MANet in terms of the accuracy and reliability of fault diagnosis is validated by comparing it with two recently published methods. Moreover, the results under different cases of engine loads and application scenarios validate that MANet possesses promising generalization performance for fault diagnosis. Hence, the proposed method provides a new approach for the real-time fault diagnosis of marine machinery in engineering applications.

Intelligent fault diagnosis of rotating machinery under variable working conditions based on deep transfer learning with fusion of local and global time–frequency features

With the development of deep learning methods, the data-driven fault diagnosis methods have attracted a great deal of interest. However, as for the data-driven fault diagnosis methods, technology has to overcome various difficulties in the practical industrial scenarios, such as variable working conditions, insufficient effective samples, and environmental noise interference. Combining with the time–frequency analysis of vibration signals, a domain adaptation fault diagnosis model based on ResNet and Transformer (DAFDMRT) is proposed in this work, aiming to solve the problems encountered by current rotating machinery fault diagnosis methods in the field of application. Firstly, the vibration signal is processed by wavelet packet transform and the time–frequency information grayscale maps is constructed. Next, a deep fusion feature extraction network combining ResNet and Transformer encoder, is designed for the extraction and fusion of the local and global features of multi-scale time–frequency information. Finally, the multi-kernel maximum mean discrepancy is applied to measure and minimize the distribution difference between the deep features of source and target domain, thereby improving the diagnostic performance of the diagnosis model in variable working conditions. In this work, comparative experiments are conducted as for bearing and gearbox datasets under variable working conditions. The results indicate that DAFDMRT can show excellent performances in terms of fault diagnosis and generalization ability.

Koopman-based deep iISS bilinear parity approach for data-driven fault diagnosis: Experimental demonstration using three-tank system

Precise and timely fault diagnosis is crucial in many practical systems and control processes. Particularly due to the increasing amount of available data collected by sensors, data-driven fault diagnosis has been a hot research topic in the prognosis and health management of industrial systems. In this paper, a reliable, accurate, and interpretable data-driven fault detection and isolation method is proposed for the general class of nonlinear systems. The proposed approach is based on the integration of the bilinear Koopman model realization, deep learning, and a bilinear parity-space framework. This work leverages the potential of neural networks to investigate lifting functions and bilinear Koopman realization simultaneously. Furthermore, to enhance the stability of the realized model, the input-to-state stability constraint is enforced on the training algorithm, ensuring the realized model is integral input-to-state stable. This method does not require any prior knowledge regarding the system dynamics and only uses the data collected from the normal operation of the system. In addition, it is capable of diagnosing all types of sensor and actuator faults, whether they are additive or multiplicative. The effectiveness of the proposed method is demonstrated experimentally using a laboratory setup of the three-tank system. Lastly, a comparison is conducted to demonstrate the advantages of the proposed method over another recent data-driven fault diagnosis method.

Fusing logic rule-based hybrid variable graph neural network approaches to fault diagnosis of industrial processes

It is practical that industrial processes are regarded as hybrid systems involving both continuous and discrete variables. Particularly, process discrete variables are usually rather crucial to process normal operations. However, existing data-driven fault diagnosis methods have been rarely applied for industrial processes concerning hybrid system characteristics. In response to this issue, this paper proposes a novel industrial process fault diagnosis approach based on fusing logic rule-based hybrid variable graph neural networks (GNNs) able to learn hybrid variable spatial structures. Therein, continuous variables samples are regarded as learning objects, while logic rule information is utilized to guide the learning process. Specifically, a mixed 0–1 integer linear programming algorithm is used to extract logic rules with fault labels from discrete variable data. Subsequently, a rule classification method is used to classify the logic rules with fault labels into independent and interdependent rule classes which are filed in a rule class library using as the basis for topological connections among data samples. In addition, an edge pruning-based graph optimization is employed to reduce redundant connections of the nodes. It is found that the established hybrid variable GNNs is able to learn the spatial topological structure of hybrid industrial process variables, providing rich heuristic knowledge for process fault classifications so as to effectively improve the fault diagnose performance. Simulation experiments are carried out in an industrial coal gasification process platform and a storage tank systems simulation platform, achieving satisfactory results.

Robust and sparse canonical correlation analysis for fault detection and diagnosis using training data with outliers

A well-known shortcoming of the traditional canonical correlation analysis (CCA) is the lack of robustness against outliers. This shortcoming hinders the application of CCA in the case where the training data contain outliers. To overcome this shortcoming, this paper proposes robust CCA (RCCA) methods for the analysis of multivariate data with outliers. The robustness is achieved by the use of weighted covariance matrices in which the detrimental effect of outliers is reduced by adding small weight coefficients on them. The RCCA is then extended to the robust sparse CCA (RSCCA) by imposing the l1-norm constraints on canonical projection vectors to obtain the sparsity property. Based on the RCCA and RSCCA, a robust data-driven fault detection and diagnosis (FDD) method is proposed for industrial processes. A residual generation model is built using projection vectors of the RCCA or RSCCA. The robust squared Mahalanobis distance of the residual is used for fault detection. A contribution-based fault diagnosis method is developed to identify the faulty variables that may cause the fault. The performance and advantages of the proposed methods are illustrated with two case studies. The results of two case studies prove that the RCCA and RSCCA methods have high robustness against outliers, and the robust FDD method is able to yield reliable results even if using the low-quality training data with outliers.

IRMSwin-T: A lightweight shifted windows transformer based on inverted residual structure and residual multi-layer perceptron for rolling bearing fault diagnosis.

The data-driven fault diagnosis method has achieved many good results. However, classical convolutional and recurrent neural networks have problems with large parameters and poor anti-noise performance. To solve these problems, we propose a lightweight shifted windows transformer based on inverted residual structure and residual multi-layer perceptron (IRMSwin-T) for fault diagnosis of rolling bearings. First, the original data are expanded by using overlapping sampling technology. Then, the collected one-dimensional vibration signals are vector serialized by using the patch embedding strategy. Finally, the IRMSwin-T network is developed to extract features of vector sequences and classify faults. The experimental results showed that compared with mainstream lightweight models, the IRMSwin-T model in this paper has fewer parameters and higher diagnostic accuracy.

Rolling bearing fault diagnosis based on multi-scale weighted visibility graph and multi-channel graph convolution network

Current data-driven fault diagnosis methods are prone to overfitting and a decrease in accuracy when working with only a limited number of labeled samples. Additionally, existing graph neural network-based fault diagnosis methods often fail to comprehensively utilize both global and local features. To address these challenges, we propose a rolling bearing fault diagnosis method based on multi-scale weighted visibility graph and a multi-channel graph convolutional network (MCGCN). Our approach converts vibration signals into multiple weighted graphs from the perspective of geometric meaning and extracts local node feature information and global topology information of graphs using MCGCN. Experimental results demonstrate that our method achieves excellent performance under both sufficient and limited data conditions, providing a promising approach for real-world industrial bearing fault diagnosis.

Data-Driven Model Space Method for Fault Diagnosis of High-Speed Train Air Brake Pipes

A data-driven fault diagnosis method is proposed in this study to address the challenge of handling a large volume of pressure data in the air brake pipe of high-speed trains. The suggested method utilizes a BP (back propagation) neural network to transform the time series pressure data into model elements in the model space, ensuring simplicity and stability. Various fitting functions, including Fourier basis, Gaussian basis, polynomial basis, sine basis, and others, are employed to accurately fit the pressure curve of the air brake pipe. The fault diagnosis process involves two steps: classifying the fault based on an optimal approximation equation and diagnosing it by analyzing the topological relationship of the model elements in the model space. The proposed method achieves an average fault diagnosis accuracy of 89.8%, with high accuracy rates for different fault states: 98% for normal state, 88% for blockage state, 84% for leakage state, and 96% for compressor fault state. Compared to the hidden Markov model method, the proposed method improves the average diagnostic accuracy by 2% for known working conditions and 4.87% for all working conditions, demonstrating its effectiveness and reliability. The fault diagnosis of the air brake tube in high-speed trains is of great significance, which aims to realize accurate fault diagnosis and prediction through sensor data monitoring and signal processing technology, so as to ensure the safe operation of high-speed trains. These studies provide an important theoretical and practical basis for the improvement and application of fault diagnosis methods.

Automated and real-time anomaly indexing for district heating maintenance decision support system

Faults are estimated to cause a significant amount of energy consumption and waste in district heating systems (DHS), and lowering this is critical to reduce costs and energy waste, but also to increase the security of supply. There are multiple examples of data-driven fault detection and diagnosis methods being employed in DHS to detect faults, however, it has been observed that a great number of faults are detected and it is not sustainable to investigate all of them. This paper, therefore, proposes implementing Chernoff bound, which is a method that can determine the fault probability level of detected faults to filter out insignificant faults. This methodology is integrated into a general continuous fault detection maintenance framework which can utilize multiple fault detection and diagnosis methods. The Chernoff bound methodology is modified for better generalization and implementation, with regard to the input data and other methods employed. The methodology has been applied to a DHS case and has proven to be able to filter out insignificant faults while still retaining a known fault, and identifying other anomalies in the operation. This framework and the introduction of Chernoff bound as a means to rank the confidence of the various detection of faults, can help maintenance crews allocate their time and effort in a better way, to only focus on the most potential critical faults. The data used as a case study is of 1-hour frequency, and some, but not all faults can be observed in such data. The optimal data is hypothesized to be of higher frequency, increasing the capabilities of the framework.

Digital twin-driven partial domain adaptation network for intelligent fault diagnosis of rolling bearing

Fault diagnosis of rolling bearings has attracted extensive attention in industrial fields, which plays a vital role in guaranteeing the reliability, safety, and economical efficiency of mechanical systems. Traditional data-driven fault diagnosis methods require obtaining a dataset of full failure modes in advance as the training data. However, this kind of dataset is not always available in some critical industrial scenarios, which impairs the practicability of the data-driven fault diagnosis methods for various applications. A digital twin, which establishes a virtual representation of a physical entity to mirror its operating conditions, would make fault diagnosis of rolling bearings feasible when the fault data are insufficient. In this paper, we propose a novel digital twin-driven approach for implementing fault diagnosis of rolling bearings with insufficient training data. First, a dynamics-based virtual representation of rolling bearings is built to generate simulated data. Then, a Transformer-based network is developed to learn the knowledge of the simulated data for diagnostics. Meanwhile, a selective adversarial strategy is introduced to achieve cross-domain feature alignments in scenarios where the health conditions of the measured data are unknown. To this end, this study proposes a digital twin-driven fault diagnosis framework by using labeled simulated data and unlabeled measured data. The experimental results show that the proposed method can obtain high diagnostic performance when the real-world data is unlabeled and has unknown health conditions, proving that the proposed method has significant benefits for the health management of critical rolling bearings.

Attention-based long short-term memory fully convolutional network for chemical process fault diagnosis

A correct and timely fault diagnosis is important for improving the safety and reliability of chemical processes. With the advancement of big data technology, data-driven fault diagnosis methods are being extensively used and still have considerable potential. In recent years, methods based on deep neural networks have made significant breakthroughs, and fault diagnosis methods for industrial processes based on deep learning have attracted considerable research attention. Therefore, we propose a fusion deep-learning algorithm based on a fully convolutional neural network (FCN) to extract features and build models to correctly diagnose all types of faults. We use long short-term memory (LSTM) units to expand our proposed FCN so that our proposed deep learning model can better extract the time-domain features of chemical process data. We also introduce the attention mechanism into the model, aimed at highlighting the importance of features, which is significant for the fault diagnosis of chemical processes with many features. When applied to the benchmark Tennessee Eastman process, our proposed model exhibits impressive performance, demonstrating the effectiveness of the attention-based LSTM FCN in chemical process fault diagnosis.