Fault Diagnosis Framework Research Articles

Advanced framework for intelligent fault diagnosis in rotary machinery with out-of-distribution recognition

Abstract Fault diagnosis plays a crucial role in maintaining mechanical equipment reliability. Deep Neural Networks (DNNs) have exhibited superior performance in fault diagnosis under closed sample space assumptions. However, the deployment of neural network models in practical industrial environments faces significant challenges due to the emergence of out-of-distribution (OOD) data, particularly novel fault categories that deviate substantially from the initial training assumptions. To address these limitations, we propose a robust fault diagnosis framework incorporating OOD sample detection and adaptive learning capabilities. The framework utilizes a novel Mixture of Experts model (MoEFormer) as the feature extraction mechanism, which enables fine-grained feature representation while enhancing computational efficiency. Furthermore, we introduce a Gaussian Density Peak Clustering (GDPC) algorithm for OOD data identification and autonomous model retraining. This integrated framework enables real-time adaptation and online diagnosis in industrial scenarios where novel fault categories may emerge. Experimental validation was conducted using the bearing fault dataset from the University of Ottawa and the gear fault dataset from Xi'an Jiaotong University. The results demonstrate that our proposed method outperforms existing state-of-the-art approaches in identifying unknown out-of-distribution faults and achieves superior diagnostic performance. The adaptively updated model attained a diagnostic accuracy of 96.0%, representing a significant improvement of approximately 30 percentage points compared to non-adaptive methods.

Explainable artificial intelligence of tree-based algorithms for fault detection and diagnosis in grid-connected photovoltaic systems

A grid-connected photovoltaic system integrates solar panels with the utility grid through a power inverter unit, allowing them to operate in parallel with the grid. Commonly known as grid-tied or on-grid solar systems, these configurations enable panels to feed electrical energy back into the grid, offering simplicity, low operating and maintenance costs, and reduced electricity bills. Despite these advantages, this environmentally friendly energy solution is still susceptible to downtimes and faults. This study utilizes advanced machine learning tree-based algorithms for fault detection and diagnosis in such systems with the goal of maintaining reliability, improving performance, and ensuring optimal energy generation. Specifically, the research investigates the effectiveness of Extra Trees as a fault detection and diagnosis algorithm through an efficient two-phase framework that consists of a binary fault detection phase followed by a multi-class fault diagnosis phase, achieving respective accuracies of 99.5% and 98.7%. In addition, the study underscores the importance of oversampling in improving results, particularly for imbalanced datasets. Moreover, explainable artificial intelligence is employed to enhance transparency in the model’s output and sensitivity to specific features in a given order. Remarkably, the findings align directly with results obtained from techniques such as feature importance averaging and incremental feature accuracy tracking. The research unveils a highly scalable, lightweight, and simple framework for fault detection and diagnosis in grid-connected photovoltaic systems.

Small-Sample-Oriented Multicondition Fault Diagnosis Framework Based on Classifier- Free Denoising Diffusion Implicit Model With Multiclass Contrastive Learning

Small-Sample-Oriented Multicondition Fault Diagnosis Framework Based on Classifier- Free Denoising Diffusion Implicit Model With Multiclass Contrastive Learning

Denoising Diffusion Implicit Model Combined with TransNet for Rolling Bearing Fault Diagnosis Under Imbalanced Data.

Data imbalances present a serious problem for intelligent fault diagnosis. They can lead to reduced diagnostic precision, which can jeopardize equipment reliability and safety. Based on that, this paper proposes a novel fault diagnosis method combining the denoising diffusion implicit model (DDIM) with a new convolutional neural network framework. First, the Gramian angular difference field (GADF) is used to generate 2D images, which are then augmented using DDIM. Next, by utilizing the weight-sharing properties of a convolutional neural network and the self-attention mechanism along with the global data processing capabilities of Transformers, a TransNet model is constructed. The augmented data are input into the model for training to establish a fault diagnosis framework. Finally, the method is validated and analyzed using the CWRU bearing dataset and the Nanchang Railway Bureau dataset. The results show that the proposed method achieves over 99% recognition accuracy on the two datasets. Meanwhile, the proposed model provides better generalization performance and recognition accuracy than existing fault diagnosis methods.

Dilated dynamic supervised contrastive learning framework for fault diagnosis under imbalanced dataset conditions

Deep learning-based fault diagnosis methods have achieved significant results. However, under actual working conditions, the inability of equipment to operate for long periods in a fault state results in a much smaller amount of collected fault data compared to normal data. This leads to existing fault diagnosis models being able to learn only limited diagnostic knowledge, causing a substantial decline in diagnostic accuracy. Therefore, a dilated dynamic supervised contrast learning framework for imbalanced data scenarios is proposed. In this framework, a dilated convolution network is designed as a feature extractor to enhance the feature extraction ability by changing the dilation rate. Then, the dynamic supervision loss function is designed, and the sample compensation factor of each category is given according to the frequency of sample used in the training process to enhance the contribution of the minority class and optimize the feature extractor. Finally, a dynamic cross-entropy loss function is designed to train the network and classifier to achieve accurate classification of faults. Experiments on two open-source datasets show the effectiveness of the proposed model framework.

Adaptive evolutionary neural architecture search based on one-dimensional convolutional neural network for electric rudder fault diagnosis

Abstract The electric rudder is the core actuator of the flight control system. Fault diagnosis of rudders is essential for the production and repair of rudders. While existing methods for rudder fault diagnosis are effective, the manual design of neural network models is a time-consuming and challenging process. Therefore, this paper proposes a fault diagnosis framework for the electric rudder based on an adaptive evolutionary neural architecture search (AENAS-FD). AENAS-FD employs an adaptive strategy to guide the evolution of a one-dimensional convolutional neural network towards achieving optimal diagnostic accuracy. This adaptive strategy adjusts the relevant parameters of the genetic operator based on the relationship between individual and population fitness. This leads to improved algorithm search performance and mitigates premature convergence. The experiments on the real electric rudder dataset demonstrate that AENAS-FD can generate superior network architectures for diagnosing rudder faults, exhibiting better diagnostic accuracy when compared to manually designed networks.

A novel adaptive cost-sensitive convolution neural network based dynamic imbalanced fault diagnosis framework for manufacturing processes

Abstract Due to the influences of sensor faults, communication lines, and human factors, it is difficult to collect and label fault data in large quantities, resulting in the imbalance between normal and fault data, and between fault and fault data. Those kinds of data imbalances violate the assumption of relatively balanced distribution of most traditional fault diagnosis methods. Associated with those trends, some imbalanced fault diagnosis methods have been put forward. However, most of those methods only consider that the proportion of various samples remains unchanged, that is, the imbalance rate is stable. In the actual manufacturing processes, the industrial data flows are fast, continuous, and dynamically changing. The imbalance rates of all kinds of samples often change continuously, showing the dynamic imbalanced characteristic. To solve this problem, a novel adaptive cost-sensitive convolution neural network based dynamic imbalanced fault diagnosis framework is designed for manufacturing processes. More specifically, a new adaptive cost-sensitive convolutional neural network is firstly constructed by coordinating the cross entropy loss function with a specific cost sensitive index, of which the dynamic imbalance rates and the diagnosis performance indicators are comprehensively considered. Subsequently, a dynamic time factor is reasonably designed and introduced to make the diagnosis model pay more attention to identification of new fault data in the industrial data flow, aiming at improving the fault diagnosis performance. Finally, sufficient simulation experiments are conducted by a typical manufacturing process, the hot rolling process, to demonstrate the superiority of the proposed framework compared with some classical algorithms.

A New Universal Cross-Domain Bearing Fault Diagnosis Framework With Dynamic Distribution Adaptation Guided by Metric Learning

A New Universal Cross-Domain Bearing Fault Diagnosis Framework With Dynamic Distribution Adaptation Guided by Metric Learning

Digital twin-assisted fault diagnosis framework for rolling bearings under imbalanced data

The application of deep learning-based fault diagnosis methods is constrained by the imbalanced data. Recently, many studies have suggested integrating dynamic model responses into the training process to address data imbalances. However, significant distribution discrepancies exist between dynamic model responses and real measured data, resulting in suboptimal performance. To address this challenge, this research proposes a digital twin-assisted framework for rolling bearings fault diagnosis under imbalanced data, which minimizes the distribution discrepancies between dynamic model responses and real measured data through information and feature transfer. Firstly, a Digital Twin-assisted Data Fusion Strategy (DTDFS) is proposed to facilitate information transfer from physical entities to dynamic models, generating digital twin data for data augmentation. Subsequently, a Frequency Filter Subdomain Adaptation Network (FFSAN) is proposed to achieve feature transfer between twin data and measured data. Finally, experimental results and engineering applications demonstrate that the proposed framework significantly outperforms existing imbalanced fault diagnosis methods, which is crucial to the application of deep learning-based fault diagnosis in industrial settings.

Meta-learning-based fault diagnosis method for rolling bearings under cross-working conditions

Abstract Accurate prediction of bearing failures is crucial for reducing maintenance costs and enhancing production efficiency in rotating machinery. However, the variable speed conditions and complex working environments encountered during operation pose significant challenges to fault diagnosis. Problems such as domain shift and insufficient sample quantity may occur during fault diagnosis under cross-working conditions, which can decrease the accuracy and generalization of deep learning algorithms. In this paper, we introduce a fault diagnosis framework grounded in meta-learning. Centered on a dual-channel feature fusion network and employing a meta-learning training paradigm, the framework not only performs well in cross-condition fault diagnosis tasks but also demonstrates superior performance in few-shot learning scenarios. Firstly, dual-channel network is used to extract the classification features of different domains, and the features are fused. Next, training is conducted using a meta-learning strategy to acquire prior knowledge, enabling rapid model adaptation to cross-working conditions and addressing the challenge of limited training samples. Finally, two public rolling bearing data sets are used to demonstrate the efficacy of the proposed method across different operational conditions. Prior to this, we selected the appropriate sample length and fusion domain through experimental validation. The proposed method also has good fault diagnosis accuracy in cross-device tasks. The experimental results verify the effective classification capability and robustness of the proposed method. Furthermore, comparisons with other meta-learning approaches confirm the superior performance of our method. The ablation experiments validated the importance and irreplaceability of each component of the proposed method.

A Novel Multi-Task Self-Supervised Transfer Learning Framework for Cross-Machine Rolling Bearing Fault Diagnosis

In recent years, intelligent methods based on transfer learning have achieved significant research results in the field of rolling bearing fault diagnosis. However, most studies focus on the transfer diagnosis scenario under different working conditions of the same machine. The transfer fault diagnosis methods used for different machines have problems such as low recognition accuracy and unstable performance. Therefore, a novel multi-task self-supervised transfer learning framework (MTSTLF) is proposed for cross-machine rolling bearing fault diagnosis. The proposed method is trained using a multi-task learning paradigm, which includes three self-supervised learning tasks and one fault diagnosis task. First, three different scales of masking methods are designed to generate masked vibration data based on the periodicity and intrinsic information of the rolling bearing vibration signals. Through self-supervised learning, the attention to the intrinsic features of data in different health conditions is enhanced, thereby improving the model’s feature expression capability. Secondly, a multi-perspective feature transfer method is proposed for completing cross-machine fault diagnosis tasks. By integrating two types of metrics, probability distribution and geometric similarity, the method focuses on transferable fault diagnosis knowledge from different perspectives, thereby enhancing the transfer learning ability and accomplishing cross-machine fault diagnosis of rolling bearings. Two experimental cases are carried out to evaluate the effectiveness of the proposed method. Results suggest that the proposed method is effective for cross-machine rolling bearing fault diagnosis.

A Multiscale Adaptive Fusion Network for Modular Multilevel Converter Fault Diagnosis

Modular Multilevel Converters (MMCs) play a crucial role in new energy grid connection and renewable energy conversion systems due to the significant merits of good modularity, flexible scalability, and lower operating loss. However, reliability is a significant challenge for MMCs, which consist of a large number of Insulated Gate Bipolar Transistors (IGBTs). Failures of the IGBTs in submodules (SMs) are a critical issue that affect the performance and operation of MMCs. The insufficient ability of convolutional neural networks to learn key fault features affects the accuracy of MMC fault diagnosis. To resolve this issue, this paper proposes a novel deep fault diagnosis framework named the Multiscale Adaptive Fusion Network (MSAFN) for MMC fault diagnosis. In the proposed MSAFN, the fault features of the raw current in an MMC are extracted by employing multiscale convolutional neural networks (CNNs) firstly, and then a channel attention mechanism is added to adaptively select the channel containing key features, so as to improve the fault diagnosis ability of the MMC in a noisy environment. Finally, the adaptive size of a one-dimensional CNN is adopted to adjust the weight of the feature channels of different scales, which are adaptively fused for fault diagnosis. Experimental validation is performed on two different MMC datasets. Experimental results confirm that the introduction of an attention mechanism of the multiscale feature adaptive fusion channel improves the recognition accuracy of the model by an average of 15.6%. Moreover, comparative experiments under different signal-to-noise ratios (SNRs) demonstrate that the MSAFN maintains accuracy levels above 96.7%, highlighting its excellent performance, particularly under noisy conditions.

A globally optimized fault diagnosis model based on generative flow model for imbalanced data

Abstract In the actual scenario of fault diagnosis based on deep learning, the diagnosis accuracy is often affected by the lack of fault state data, so the processing of imbalanced data is always a significant challenge. generative adversarial networks (GAN) and denoising diffusion probability models (DDPM) are widely used for data augmentation. However, GAN often shows sensitivity and instability in the training process, and the sample generation speed of DDPM is slow due to the steps requiring multiple iterations–both of which are limiting factors. To solve these problems, we introduce the generative flow network with invertible 1 × 1 convolutions (GLOW) into fault diagnosis. The GLOW model is optimized by maximum likelihood estimation and does not require multiple iterations to generate samples, avoiding the problems faced by GAN and DDPM. In order to generate balanced data explicitly, we propose a condition GLOW (CGLOW) to provide class-balanced samples in real time throughout the framework. On the other hand, using the reversibility of CGLOW, we design an end-to-end fault diagnosis framework that is globally optimized to mitigate the decline in diagnostic accuracy caused by the separation of generation and diagnosis and simplify the steps of fault diagnosis. In addition, to accommodate the non-stationary characteristics of fault signals, we propose a new data transformation method to improve the feature mining ability of the model and the diagnostic accuracy. Finally, we conduct extensive experiments to validate the superiority of the proposed approach. The experimental results demonstrate that our method outperforms existing ones.

Contrastive learning-enabled digital twin framework for fault diagnosis of rolling bearing

Abstract Rolling bearings are essential components in various industrial machines, and their failures can lead to significant downtime and maintenance costs. Traditional data-driven fault diagnosis methods often require extensive fault datasets for training, which may not always be available in critical industrial scenarios, limiting their practicality. Digital twins, virtual representations of physical entities reflecting their operational conditions, offer a promising solution for the fault diagnosis of rolling bearings with limited fault data. In this paper, we propose a novel digital twin-driven framework to address the challenge of limited training data in rolling bearing fault diagnosis. Firstly, a virtual bearing simulation model is used to generate the simulated data. Subsequently, a transformer-based network is introduced to learn the discrepancy features from the raw data. Then, a maximum mean discrepancy loss and a supervised contrastive learning loss for raw and augmentation data are established to achieve global domain alignment and instance-based domain alignment. Finally, an unsupervised contrastive learning loss for the augmentation data of the target domain is established to further improve the diagnostic performance. In five cross-domain fault diagnosis tasks representing real industrial scenarios set, the average diagnostic accuracy of the proposed method is 84.39%, which is more than 10% higher than the two existing advanced domain adaptation methods. The experimental results demonstrate that the proposed method achieves high diagnostic performance in real industrial scenarios where labeled data is lacking. This shows its significant benefits for monitoring the condition of critical bearings.

A Cooperative Convolutional Neural Network Framework for Multisensor Fault Diagnosis of Rotating Machinery

A Cooperative Convolutional Neural Network Framework for Multisensor Fault Diagnosis of Rotating Machinery

Transient gas path fault diagnosis of aero-engine based on domain adaptive offline reinforcement learning

Real-time measurement parameters are crucial for diagnosing faults in aero-engine gas path performance, ensuring engine reliability, and mitigating potential economic losses. Traditional aero-engines performance diagnosis was mainly based on the measurements of steady-state condition and lacked the utilization of data under transient conditions. Gas path diagnosis of aero-engines under transient conditions is crucial for early fault detection and safety of flight within the envelope. The challenge lies in the inconsistent distribution of performance deviations caused by variable operating conditions, especially with complex fault types, which can undermine diagnostic credibility. To improve reliability of gas path diagnosis under transient conditions, an offline reinforcement learning fault diagnosis framework based on a transient aero-engine performance model is proposed. To address the issue of variable operating conditions during transient states, a domain adaptive approach is utilized to reconstruct the measurement baseline and facilitate the transfer of different performance deviation distributions. Additionally, by adding spool acceleration as a measurement parameter, the multi-component fault coupling is solved. Finally, validation with actual operating data simulates fault cases, demonstrating the proposed method's efficacy in quantitatively detecting gradual, sudden, and multiple component faults under transient conditions with high accuracy and efficiency. The method proposed in this study achieves a computational speed improvement by 64% compared to the conventional method, achieving a time of 0.13 seconds, with an average error of less than 0.00389%. Additionally, it demonstrates strong robustness in the presence of noise, with an average error of less than 0.03125%. This proposed method improves real-time fault detection under transient conditions for its higher accuracy and efficiency, and therefore significantly enhance gas path health monitoring and diagnosis capability.

Multidisciplinary Framework for Vibration-Based Gear Fault Diagnosis Using Experiments, Modeling, and Machine Learning

Vibration-based gear diagnosis is crucial for ensuring the reliability of rotating machinery, making the monitoring of gear health essential for preventing costly downtime and optimizing performance. This study proposes a multidisciplinary framework to enhance gear diagnosis, that aligns with the new era of digital twins by integrating experiments, dynamic modeling, physical preprocessing, and machine learning. Within this framework, we focus on three core procedures: domain adaptation to reduce discrepancies between measured data and synthetic data generated by dynamic models; physical preprocessing, grounded in in-depth investigations dictating signal processing and feature engineering techniques; and learning algorithms, encompassing the process of training AI-based models. We demonstrate this framework through a comprehensive case study of localized tooth fault diagnosis, using controlled-degradation tests and realistic simulations. First, we detect faults using unsupervised learning algorithms; then, we use zero-shot-learning for classifying between localized and distributed faults; finally, we adopt a one-shot-learning strategy for severity estimation. Above all, this hybrid framework bridges the gaps between physical-based and AI-based approaches by combining physical knowledge and advanced algorithmics with machine learning. This contributes to the PHM field by offering valuable insights into integrating different aspects of research, thereby enhancing performance in gear diagnosis tasks.

Global-Local Continual Transfer Network for Intelligent Fault Diagnosis of Rotating Machinery

Existing fault diagnosis methods face three fundamental challenges when deployed in the dynamic environment, including insufficient continuous diagnostic capability, poor model generalization performance, and inadequate data privacy protection. To address these problems, we develop a novel continual fault diagnosis framework named Global-Local Continual Transfer Network (GLCTN) for fault classification of unlabeled target samples under different working conditions without any source samples. Specifically, a consistency loss and a mutual information loss are introduced in the proposed GLCTN to transfer the learned diagnostic knowledge. Moreover, a dual-speed optimization strategy is utilized to preserve the acquired diagnostic knowledge and to endow the model with the ability to acquire new knowledge. Validation experiments conducted on an automobile transmission dataset demonstrate that the proposed GLCTN achieves satisfactory diagnostic performance on multiple continuous transfer diagnostic tasks.

An Intelligent Optimized Digital Twins Framework for Fault Diagnosis in Complex Control Systems

An Intelligent Optimized Digital Twins Framework for Fault Diagnosis in Complex Control Systems

Knowledge-informed FIR-based cross-category filtering framework for interpretable machinery fault diagnosis under small samples

Relying on sufficient training data, the existing fault diagnosis methods rarely focus on the methodological interpretability and the data scarcity in real industrial scenarios simultaneously. Motivated by this issue, we deeply reexamined the intrinsic characteristics of fault signals and the guiding significance of classical signal-processing methods for feature enhancement. From the perspective of multiscale modes, this study tailors multiple learnable knowledge-informed finite impulse response (FIR) filtering kernels to extract sensitive modes for explainable feature enhancement. On this foundation, a knowledge-informed FIR-based cross-category filtering (FIR-CCF) framework is further proposed for interpretable small-sample fault diagnosis. With the consideration of the mode complexity, a cross-category filtering strategy is explored to further enhance feature expressions for identifying single state. To be special, this strategy divides a multi-class recognition process into multiple two-class recognition task. A multi-task learning is then presented where multiple binary-class base learners (BCBLearners) that consists of a feature extractor and a two-class classifier is established to seek discriminate mode features for each type of state. Eventually, all feature extractors are fixed and a multi-class classifier is established and to fuse all mode features for high-precision multi-class identification via ensemble learning. As a variant of signal-processing-collaborated deep learning frameworks, the FIR-CCF method fully exploits the strengths of signal-processing methods in interpretability and feature extraction. Three experimental cases highlight the superiority and significant improvement of the FIR-CCF framework over other five state-of-the-art diagnosis methods and three ablation models. Specially, extensive visualization is implemented to place in-depth insight into how the FIR-CCF framework works. It can be also foreseen that the signal-processing-collaborated deep learning framework shows enormous potential in interpretable fault diagnosis for knowledge-informed artificial intelligence. Related source codes will be available at: https://github.com/BITS/FIR-CCF-main.