Representation Of Faults Research Articles

Cross-modal zero-sample diagnosis framework utilizing non-contact sensing data fusion

Gearboxes, fundamental components in the domains of manufacturing, transportation, and aerospace apparatus, are highly susceptible to impairments. The emerging technique of non-contact sensing measurement holds promise as a valuable tool for real-time monitoring and diagnosis of gearboxes, especially in challenging, dynamic environments. However, historical non-contact diagnostic techniques often face difficulties due to load-variation-induced interference, which leads to inconsistent diagnostic results. Moreover, acquiring sufficient non-contact sensing data for specific rare or severe gearbox faults can be exceedingly challenging and, at times, unattainable. This data scarcity limits the application of existing supervised and semi-supervised non-contact diagnostic methods. To address these limitations, this paper develops a novel diagnostic model, called Non-Contact Sensing Data Fusion Driven Cross-Modal Zero-sample Diagnostic Network. Specifically, we first develop a novel central fusion module to facilitate feature-level fusion at both global and local scales utilizing infrared thermal and acoustic data, thereby establishing a rich and comprehensive fault representation. Following that, we construct an end-to-end cross-modal zero-sample diagnostic architecture to mine and fuse complementary fault information. This approach enhances zero-sample diagnostic performance through efficient information utilization and cross-modal feature fusion. Afterward, we adopt a cross-modal self-enhancing fusion strategy during the training phase, which improves the information fusion performance of intra-class features from various modalities. This strategy effectively reduces misclassification risks and augments the robustness of the proposed zero-sample diagnostic network against interference from load variations. Comprehensive experimental results confirm that our proposed NCFZD sets the new state-of-the-art in multiple non-contact, zero-sample diagnostic scenarios, reaching HM scores of 86.06 %, 81.53 %, and 62.56 % respectively. By incorporating the information fusion theory, this model advances gearbox diagnostics in non-contact sensing and contributes to the broader field of information fusion theory by demonstrating its practical application in real-world problem-solving scenarios.

Graph Feature Fusion-Driven Fault Diagnosis of Complex Process Industrial System Based on Multivariate Heterogeneous Data

The stable operation of the process industrial system, which is integrated with various complex equipment, is the premise of production, which requires the condition monitoring and diagnosis of the system. Recently, the continuous development of deep learning (DL) has promoted the research of intelligent diagnosis in process industry systems, and the sensor system layout has provided sufficient data foundation for this task. However, these DL-driven approaches have had some shortcomings: (1) the output signals of heterogeneous sensing systems existing in process industry systems are often high-dimensional coupled and (2) the fault diagnosis model built from pure data lacks systematic process knowledge, resulting in inaccurate fitting. To solve these problems, a graph feature fusion-driven fault diagnosis of complex process industry systems is proposed in this paper. First, according to the system’s prior knowledge and data characteristics, the original multisource heterogeneous data are divided into two categories. On this basis, the two kinds of data are converted to physical space graphs (PSG) and process knowledge graphs (PKG), respectively, according to the physical space layout and reaction mechanism of the system. Second, the node features and system spatial features of the subgraphs are extracted by the graph convolutional neural network at the same time, and the fault representation information of the subgraph is mined. Finally, the attention mechanism is used to fuse the learned subgraph features getting the global-graph representation for fault diagnosis. Two publicly available process chemistry datasets validate the effectiveness of the proposed method.

TFPred: Learning discriminative representations from unlabeled data for few-label rotating machinery fault diagnosis

Recent advances in intelligent rotating machinery fault diagnosis have been enabled by the availability of massive labeled training data. However, in practical industrial applications, it is often challenging and costly to annotate a large amount of data. To address the few-label fault diagnosis problem, a time–frequency prediction (TFPred) self-supervised learning framework is proposed to extract latent fault representations from unlabeled fault data. Specifically, the TFPred framework consists of a time encoder and a frequency encoder, with the frequency encoder to predict the low-dimensional representations of time domain signals generated by the time encoder with randomly augmented data. Subsequently, the pre-trained network is hooked with a classification head and fine-tuned with limited labeled data. Finally, the proposed framework is evaluated on a run-to-failure bearing dataset and a hardware-in-the-loop high-speed train simulation platform. The experiments demonstrate that the self-supervised learning framework TFPred achieved competitive performance with only 1% and 5% labeled data. Code is available at https://github.com/Xiaohan-Chen/TFPred.

TSoft-Net: A novel transfer soft thresholding network based on self-attention for intelligent fault diagnosis of rotating machinery

In the industrial scenes, the machinery operates under diverse working conditions and generates varying levels of noise, which can hinder the performance of the intelligent fault diagnosis model that is trained in the laboratory. This is because the different working conditions and environmental noise change the distribution of laboratory vibration signal. To tackle this issue, we proposed a novel transfer soft thresholding network (TSoft-Net) based on self-attention for intelligent fault diagnosis of rotating machinery, which has good anti-noise performance and can be transferred to different working conditions with excellent accuracy. This paper constructs a soft residual block for extracting the fault representation and enhancing the residual learning. In this block, we propose a residual factor to learn and enhance domain-invariant fault representation. Furthermore, a representation soft fusion block is built for extracting and fusing the different scale fault representation. In this block, we propose a scale-attention weight to dynamically fuse the different scale fault representation. The experiments show that the TSoft-Net, compared with six existing methods on eight sub-datasets, has better anti-noise ability and achieves an accuracy of up to 100% on the target domain.

Machine Fault Diagnosis through Vibration Analysis: Continuous Wavelet Transform with Complex Morlet Wavelet and Time–Frequency RGB Image Recognition via Convolutional Neural Network

In pursuit of advancing fault diagnosis in electromechanical systems, this research focusses on vibration analysis through innovative techniques. The study unfolds in a structured manner, beginning with an introduction that situates the research question in a broader context, emphasising the critical role of fault diagnosis. Subsequently, the methods section offers a concise summary of the primary techniques employed, highlighting the utilisation of short-time Fourier transform (STFT) and continuous wavelet transform (CWT) for extracting time–frequency components from the signal. The results section succinctly summarises the main findings of the article, showcasing the results of features extraction by CWT and subsequently utilising a convolutional neural network (CNN) for fault diagnosis. The proposed method, named CWTx6-CNN, was compared with the STFTx6-CNN method of the previous stage of the investigation. Visual insights into the time–frequency characteristics of the inertial measurement unit (IMU) data are presented for various operational classes, offering a clear representation of fault-related features. Finally, the conclusion section underscores the advantages of the suggested method, particularly the concentration of single-frequency components for enhanced fault representation. The research demonstrates commendable classification performance, highlighting the efficiency of the suggested approach in real-time scenarios of fault analysis in less than 50 ms. Calculation by CWT with a complex Morlet wavelet of six time–frequency images and combining them into a single colour image took less than 35 ms. In this study, interpretability techniques have been employed to address the imperative need for transparency in intricate neural network models, particularly in the context of the case presented. Notably, techniques such as Grad-CAM (gradient-weighted class activation mapping), occlusion, and LIME (locally interpretable model-agnostic explanation) have proven instrumental in elucidating the inner workings of the model. Through a comparative analysis of the proposed CWTx6-CNN method and the reference STFTx6-CNN method, the application of interpretability techniques, including Grad-CAM, occlusion, and LIME, has played a pivotal role in revealing the distinctive spectral representations of these methodologies.

Bayesian Hierarchical Graph Neural Networks With Uncertainty Feedback for Trustworthy Fault Diagnosis of Industrial Processes.

Deep learning (DL) methods have been widely applied to intelligent fault diagnosis of industrial processes and achieved state-of-the-art performance. However, fault diagnosis with point estimate may provide untrustworthy decisions. Recently, Bayesian inference shows to be a promising approach to trustworthy fault diagnosis by quantifying the uncertainty of the decisions with a DL model. The uncertainty information is not involved in the training process, which does not help the learning of highly uncertain samples and has little effect on improving the fault diagnosis performance. To address this challenge, we propose a Bayesian hierarchical graph neural network (BHGNN) with an uncertainty feedback mechanism, which formulates a trustworthy fault diagnosis on the Bayesian DL (BDL) framework. Specifically, BHGNN captures the epistemic uncertainty and aleatoric uncertainty via a variational dropout approach and utilizes the uncertainty information of each sample to adjust the strength of the temporal consistency (TC) constraint for robust feature learning. Meanwhile, the BHGNN method models the process data as a hierarchical graph (HG) by leveraging the interaction-aware module and physical topology knowledge of the industrial process, which integrates data with domain knowledge to learn fault representation. Moreover, the experiments on a three-phase flow facility (TFF) and secure water treatment (SWaT) show superior and competitive performance in fault diagnosis and verify the trustworthiness of the proposed method.

A Diffuse Interface Method for Earthquake Rupture Dynamics Based on a Phase‐Field Model

AbstractIn traditional modeling approaches, earthquakes are often depicted as displacement discontinuities across zero‐thickness surfaces embedded within a linear elastodynamic continuum. This simplification, however, overlooks the intricate nature of natural fault zones and may fail to capture key physical phenomena integral to fault processes. Here, we propose a diffuse interface description for dynamic earthquake rupture modeling to address these limitations and gain deeper insight into fault zones' multifaceted volumetric failure patterns, mechanics, and seismicity. Our model leverages a steady‐state phase‐field, implying time‐independent fault zone geometry, which is defined by the contours of a signed distance function relative to a virtual fault plane. Our approach extends the classical stress glut method, adept at approximating fault‐jump conditions through inelastic alterations to stress components. We remove the sharp discontinuities typically introduced by the stress glut approach via our spatially smooth, mesh‐independent fault representation while maintaining the method's inherent logical simplicity within the well‐established spectral element method framework. We verify our approach using 2D numerical experiments in an open‐source spectral element implementation, examining both a kinematically driven Kostrov‐like crack and spontaneous dynamic rupture in diffuse fault zones. The capabilities of our methodology are showcased through mesh‐independent planar and curved fault zone geometries. Moreover, we highlight that our phase‐field‐based diffuse rupture dynamics models contain fundamental variations within the fault zone. Dynamic stresses intertwined with a volumetrically applied friction law give rise to oblique plastic shear and fault reactivation, markedly impacting rupture front dynamics and seismic wave radiation. Our results encourage future applications of phase‐field‐based earthquake modeling.

Spatial weighted graph-driven fault diagnosis of complex process industry considering technological process flow

Each chemical process industry system possesses unique process knowledge, which serves as a representation of the system’s state. As graph-theory based methods are capable of embedding process knowledge, they have become increasingly crucial in the field of process industry diagnosis. The fault representation ability of the diagnosis model is directly associated with the quality of the graph. Unfortunately, simple fully connected graphs fail to strengthen the internal connections within the same process but weaken the interactive connections between different processes. Moreover, each node in the graph is considered equally important, making it impossible to prioritize crucial system monitoring indicators. To address the above shortcomings, this paper presents a spatial weighted graph (SWG)-driven fault diagnosis method of complex process industry considering technological process flow. Initially, the physical space sensor layout of the technological process flow is mapped into the spatial graph structure, where each sensor is regarded as a node and these nodes are connected by the k nearest neighbor algorithm. Subsequently, according to the mechanism knowledge, the sensors in the process are divided into different importance categories and weight coefficients are assigned to their nodes. The similarities between these weighted nodes are calculated, and the resulting edge information are used to construct the SWGs. Finally, the SWGs are input to a graph convolutional network, facilitating fault representation learning for fault diagnosis of complex process industry. Validation experiments are conducted using public industrial datasets, and the results demonstrate that the proposed method can effectively integrate the process knowledge to improve the fault diagnosis accuracy of the model.

A dual-view network for fault diagnosis in rotating machinery using unbalanced data

Abstract Data-driven intelligent methods have demonstrated their effectiveness in the area of fault diagnosis. However, most existing studies are based on the assumption that the distributions of normal and faulty samples are balanced during the diagnostic process. This assumption significantly decreases the application range of a diagnostic model as the samples in most real-world scenarios are highly unbalanced. To cope with the limitations caused by unbalanced data, this paper proposed an original dual-view network (DVN). Firstly, an interactive graph modeling strategy is introduced for relationship information modeling of multi-sensor data. Meanwhile, the graph convolution operation is used as the baseline for feature extraction of the constructed interactive graph to mine for fault representations. Secondly, an original dual-view classifier consisting of a binary classifier and a multi-class classifier is proposed, which divides fault diagnosis into two stages. Specifically, in the first stage, the binary classifier performs the binary inference from the view of fault detection. In the second stage, the multi-class classifier performs the full-state inference from the view of fine-grained fault classification. Then, based on the dual-view classifier, a weight activation module is designed to alleviate training bias toward majority classes by sample-level re-weighting. Finally, the diagnosis results can be obtained according to the output of the multi-class classifier. Fault diagnosis experiments using two different datasets with varying data unbalance ratios were conducted to validate the effectiveness of the proposed method. The superiority of the proposed DVN is verified through comparisons with state-of-the-art methods. The effectiveness of the DVN is further validated through ablation studies with some ablative models. The DVN code is available at: https://github.com/CQU-ZixuChen/DualViewNetwork.

A Novel Method for Bearing Fault Diagnosis under Variable Speed Based on Envelope Spectrum Fault Characteristic Frequency Band Identification.

Rolling element bearing (REB) vibration signals under variable speed (VS) have non-stationary characteristics. Order tracking (OT) and time-frequency analysis (TFA) are two widely used methods for REB fault diagnosis under VS. However, the effect of OT methods is affected by resampling errors and close-order harmonic interference, while the accuracy of TFA methods is mainly limited by time-frequency resolution and ridge extraction algorithms. To address this issue, a novel method based on envelope spectrum fault characteristic frequency band identification (FCFBI) is proposed. Firstly, the characteristics of the bearing fault vibration signal's envelope spectrum under VS are analyzed in detail and the fault characteristic frequency band (FCFB) is introduced as a new and effective representation of faults. Then, fault templates based on FCFB are constructed as reference for fault identification. Finally, based on the calculation of the correlation coefficients between the envelope spectrum and fault templates in the extended FCFB, the bearing fault can be diagnosed automatically according to the preset correlation coefficient criterion. Two bearing VS experiments indicate that the proposed method can achieve satisfactory diagnostic accuracy. The comparison of OT and TFA methods further demonstrates the comprehensive superiority of the proposed method in the overall consideration of accuracy, diagnostic time, tachometer dependency, and automatic degree.

A Novel Data-Driven Mechanical Fault Diagnosis Method for Induction Motors Using Stator Current Signals

Most of the mechanical fault diagnosis methods of induction motors (IMs) are based on vibration signals. However, vibration sensors are expensive and require direct contact with IMs. This article proposes a data-driven mechanical fault diagnosis method for IMs using stator current signals, which is more convenient and low cost, since no additional sensors are required. Aiming at the weak representation of mechanical faults in current signals, an intelligent noise elimination method based on noise reconstruction model is proposed to improve the signal-to-noise ratio. Through the automatic feature extraction and classification of the residual current envelope spectrum, high diagnosis accuracy can be obtained even if some differences have existed among samples. The effectiveness of the proposed method is verified by high accuracy diagnosis results on two experimental platforms. The results show that the average diagnostic accuracy of the proposed algorithm for one bearing fault and two eccentricity faults can reach 96%. Even if the fault type and the fault degree are distinguished at the same time, an accuracy of 90% can be achieved for six kinds of bearing faults.

Geomechanical assessment of a large-scale CO2 storage and insights from uncertainty analysis

Geological storage of CO2 provides a feasible solution to curb CO2 emissions to the atmosphere and relieve climate change. However, as CO2 is injected into the reservoir, the increased pore pressure may result in fault reactivations or damaging geomechanical changes to the caprock, which could be detrimental to the containment of injected CO2. On the other hand, uncertainty is inevitable due to the limited information of the complex subsurface system. In this study, we develop a workflow to assess the geomechanical impact of CO2 geological storage and include management of sources of uncertainties associated with the subsurface system. We utilize the hypothetical large-scale carbon sequestration situated in the Southern San Joaquin Valley in California, USA as a case study. We firstly set up a three-dimensional (3D) geomechanical model using a non-linear quasi-static finite element method. The applied workflow has features including: (1) Coulomb friction behavior to model fault slippage, (2) CAD representation of land surface and faults, (3) non-uniform pressure increase based on dynamic reservoir simulations, and (4) variable rock properties for formation layers. We perform uncertainty analyses using a response surface methodology and a Box-Behnken design to assess the significance of subsurface uncertainties to geomechanical responses. We find that the intact caprock will not have a shear failure even considering the subsurface uncertainties. However, the faults can reactivate and give potential flow pathways in the caprock. The presented workflow integrates advanced geomechanical models and uncertainty analysis, which can be readily applied to other CO2 storage projects to mitigate geomechanical sources of risk.

Deep causal factorization network: A novel domain generalization method for cross-machine bearing fault diagnosis

Domain generalization (DG) has attracted much attention in bearing fault diagnosis since it can generalize the prior diagnostic knowledge to invisible working conditions. However, previous DG-based approaches are easy to introduce personalized bias due to the specificity of the mechanical equipment, which worsens the generalization performance. To solve this problem, a deep causal factorization network (DCFN) is proposed for cross-machine bearing diagnosis without the involvement of target domain data in training. Specifically, by leveraging the structural causal model of bearing fault signal generation, DCFN defines the cross-machine generalized fault representations as causal factors and the domain-related representations as non-causal factors. Afterwards, taking advantage of the causal properties that can be preserved in data distribution shifts, causal task factorization and feature factorization modules are designed to reconstruct causal mechanisms. To separate the two types of underlying factors, causal task factorization aims to maximize the predicted output entropy of the domain classifier using learned causal factors as input, and simultaneously maximizes the entropy of the fault classifier using learned non-causal factors as input. Moreover, causal feature factorization highlights that the ideal causal factors desire cross-domain consistency and inter-dimensional independence, thereby learning causal factors with stable and sufficient distinguishing features. Finally, under broad bearing fault datasets including public, laboratory and simulation datasets, the effectiveness of the proposed DCFN is verified by comparing with various state-of -the-arts diagnosis methods.

A New Diagnosis Method with Few-shot Learning Based on a Class-rebalance Strategy for Scarce Faults in Industrial Processes

For industrial processes, new scarce faults are usually judged by experts. The lack of instances for these faults causes a severe data imbalance problem for a diagnosis model and leads to low performance. In this article, a new diagnosis method with few-shot learning based on a class-rebalance strategy is proposed to handle the problem. The proposed method is designed to transform instances of the different faults into a feature embedding space. In this way, the fault features can be transformed into separate feature clusters. The fault representations are calculated as the centers of feature clusters. The representations of new faults can also be effectively calculated with few support instances. Therefore, fault diagnosis can be achieved by estimating feature similarity between instances and faults. A cluster loss function is designed to enhance the feature clustering performance. Also, a class-rebalance strategy with data augmentation is designed to imitate potential faults with different reasons and degrees of severity to improve the model’s generalizability. It improves the diagnosis performance of the proposed method. Simulations of fault diagnosis with the proposed method were performed on the Tennessee-Eastman benchmark. The proposed method achieved average diagnosis accuracies ranging from 81.8% to 94.7% for the eight selected faults for the simulation settings of support instances ranging from 3 to 50. The simulation results verify the effectiveness of the proposed method.

Reducing the Probability of Failure in Manufacturing Equipment by Quantitative FTA Analysis

Abstract Fault Tree Analysis (FTA) is a method that directly focuses on the modes of failures. The FTA is a graphical representation of the major faults or critical failures associated with a product, as well as the causes for the faults and potential countermeasures. The aim of this research paper is to calculate the probability of the top event – the failure of the process using FTA and propose a technique to prioritize factors for action design and reduce the likelihood of a top event failure based on manufacturers' requirements. We have constructed a qualitative fault tree to produce office components packed and sealed in blister packs using a KOCH KBS-PL machine. We defined the top event G – the production of office components, packed and sealed in blister packs on the machinery KOCH KBS-PL. Then we defined events leading to top events down to individual failure factors. Based on the links between the fault tree and the probability of failure, we performed a quantitative analysis to determine the probability of failure of individual events. We found out that the probability of failure of G is 5.04%. Subsequently, we identified which factors most significantly reduce the resulting probability of failure of factor G. These are the factors: E – feed rate, F – cooling, AL – incorrect setting and D – break. It has been proven that by controlling these 4 factors, we can reduce the probability of failure of top event G to 2.36%, provided that effective measures are taken. The final proposal meets the requirements of several manufacturers for a fast, efficient, and cost-effective solution. We have created a proposal that saves time, has minimal software and hardware requirements, and is easy to use. The efficiency and effectiveness of the proposal was that we identified the weakest points in the fault tree that most significantly cause the top event to fail. This prioritized the factors for the design of the measures.

GridOnto: Knowledge Representation and Extraction for Fault Events in Power Grid

An accurate and comprehensive fault knowledge representation is indispensable for an automated and intelligent processing of power grid failures. Current knowledge graphs are incapable of capturing the complex relations among power grid failures. This paper extends the current knowledge graph representation mechanism through temporal, spatial, and causal representations to facilitate the knowledge representation of power grid faults, allowing for ontology modeling of power event elements and event relationships. During the modeling process, this paper proposes an extraction method which includes Bi-directional Encoder Representation from Transformers (BERT), Bi-directional Long Short-Term Memory (BiLSTM), and Conditional Random Field (CRF) for entities and relationships in power grid faults. The innovation of the method lies in the clever combination of the three, BERT learning semantic representation, BiLSTM further learning semantic features, and CRF joint modeling of labels to improve accuracy, and the results verify the effectiveness and practicality of the method presented in this paper.

Enhanced Bayesian sparse representation of mechanical fault signals by structural feature-oriented matching composite dictionary construction

The traditional orthogonal matching pursuit algorithm exploits only the overall sparsity of signals without considering the effects caused by structural characteristics. To this end, this paper proposes an enhanced Bayesian sparse representation (EBSR) of mechanical fault signals by structural feature-oriented matching redundant dictionary construction. First, an EBSR model which can improve sparse reconstruction results is proposed. The proposed model improves the recovery accuracy and robustness of the sparse representation by using the structural information as a priori information. Subsequently, a composite dictionary is designed combining a Sin-Chirplet dictionary with an Impulse dictionary, and a multi-group and multi-strategy grey wolf optimizer algorithm is employed to enable the composite dictionary match the structural features of the fault signal and reduce its redundancy degree. Finally, the optimized matching composite dictionary is introduced into the EBSR algorithm, endowing it with an efficient atom selection strategy and reducing the complexity of the sparse representation. The simulation and experimental results demonstrated that the proposed method can effectively reduce the interference from background noise and impurity frequencies, verifying the effectiveness and applicability of the proposed method for the sparse representation of mechanical faults.

Fast nonlinear convolutional sparse filtering: A novel early-stage fault diagnosis method of rolling bearing

Early fault diagnosis of rotating machinery has been widely studied. Among these methods, sparse optimization-based methods are drawing more and more attention. Obviously, the fault representation performance of the objective function is the key to ensure the successfully extraction of the weak signature. Inspired by this, to improve the noise adaptability and convergence robustness of diagnosis process, this paper proposes fast nonlinear convolutional sparse filtering (FNCSF) for early-stage fault diagnosis of rotating machinery. Generalized Sigmoid function is used to improve the representation and convergence stability of the objective function. The performance of fault feature measures with different scaling and translation factors is studied. The experimental and simulation signals are employed to verify the performance of the FNCSF. The results show that FNCSF is a promising tool for early fault diagnosis, which can learn the fault information without any prior knowledge. The nonlinear sparse representation can improve the gradient information in noise environment.

Fast nonlinear blind deconvolution for rotating machinery fault diagnosis

Sparse optimization based early fault diagnosis method is drawing more and more attention. In these methods, the objective function is usually a sparsity measure which can represent the impulsive signature produced by the mechanical impacts between rotating components. Therefore, the key point is the fault representation and convergence stability of the objective function. Inspired by this, fast nonlinear blind deconvolution algorithm is proposed for early fault diagnosis of rotating machinery. First, sigmoid function is developed to the generalized form to improve the fault representation ability of the objective function under noisy environment. The nonlinear mapping and sparse expression ability are discussed in detail. Then, Gaussian fitting window function and L1/2 penalty of filter are used to improve the distribution and performance of the weight vector. Finally, simulation data and early bearing fault data are employed to verify the effectiveness and determine optimal parameter setting of the proposed method. Based on the comparison results with the existing methods, the proposed method is found to be a promising method for the early-stage fault diagnosis, which can significantly improve the noise adaptability, computation effectiveness and robustness of fault diagnosis.

Nonlinear fast kurtogram for the extraction of gear fault features with shock interference

The extraction of gearbox fault features under shock interference is an exceedingly difficult and valuable subject. The effective usage of the resonance frequency band is one of the solutions for this subject. However, the existing fast kurtogram (FK) method is prone to misdiagnosis due to the sensitivity of this method to aperiodic shocks. To overcome the sensitivity of the FK method to irrelevant shock, this paper proposes a nonlinear fast kurtogram (NFK) method. First, Z-score normalization is performed on the signal. Then, Sigmoid is used to improve the fault representation under shock interference. Third, the signal is divided into different frequency bands, and the band with the largest kurtosis is selected for filtering. Finally, the gear fault is analyzed by the square envelope spectrum. Simulation and experimental verification are used to demonstrate the effectiveness of the proposed method. The results show that gear fault features can be extracted under shock interference by the NFK method.