Gearbox Fault Research Articles

Intelligent Detection Method of Gearbox Based on Adaptive Hierarchical Clustering and Subset.

Deep learning uses mechanical time-frequency signals to train deep neural networks, which realizes automatic feature extraction and intelligent diagnosis of fault features and gets rid of the dependence on a large number of signal processing technology and experience. Aiming at the problem of misclassification of similar samples, a fault diagnosis algorithm based on adaptive hierarchical clustering and subset (AHC-SFD) is proposed to extract features and applied to gearbox fault diagnosis. Firstly, the adaptive hierarchical clustering algorithm is used to analyze the characteristics of different data, and then the data set is clustered into multiple feature groups; finally, according to the feature group, the SubCNN model is established for multiscale feature extraction, so as to carry out fault diagnosis. The test results show that the fault recognition rate achieved by the proposed method is more than 99.7% on the gearbox dataset, and the method has better generalization ability.

SRMANet: Toward an Interpretable Neural Network with Multi-Attention Mechanism for Gearbox Fault Diagnosis

Deep neural network (DNN), with the capacity for feature inference and nonlinear mapping, has demonstrated its effectiveness in end-to-end fault diagnosis. However, the intermediate learning process of the DNN architecture is invisible, making it an uninterpretable black-box model. In this paper, a stacked residual multi-attention network (SRMANet) is proposed as a means of feature extraction of vibration signals, and visualizing the model training process, designing Squeeze-excitation residual (SE-Res) blocks to obtain additive features with minimal redundancy and sparsity. This study recommends the use of the attention fusion unit to ensure the interpretability of the model and ultimately to obtain representative features. By feeding the output gradient of the attention layer back to the original signal, the key feature components in the time domain signal can be effectively captured. Finally, the interpretability, identification accuracy and adaptability of the model under different operating conditions are verified on 12 different fault tasks in the planetary gearbox.

Infrared Thermography Smart Sensor for the Condition Monitoring of Gearbox and Bearings Faults in Induction Motors.

The monitoring of machine conditions is very important from the viewpoints of productivity, economic benefits, and maintenance. Several techniques have been proposed in which sensors are the key to providing relevant information to verify the system. Recently, the smart sensor concept is common, in which the sensors are integrated with a data processing unit executing dedicated algorithms used to generate meaningful information about the system in situ. Additionally, infrared thermography has gained relevance in monitoring processes, since the new infrared cameras have more resolution, smaller dimensions, reliability, functionality, and lower costs. These units were firstly used as secondary elements in the condition monitoring of machines, but thanks to modern techniques for data processing, the infrared sensors can be used to give a first, or even a direct, diagnosis in a nonintrusive way in industrial applications. Therefore, in this manuscript, the structure and development of an infrared-thermography-based smart sensor for diagnosing faults in the elements associated with induction motors, such as rolling bearings and the gearbox, is described. The smart sensor structure includes five main parts: an infrared primary sensor, a preprocessing module, an image processing module, classification of faults, and a user interface. The infrared primary sensor considers a low-cost micro thermal camera for acquiring the thermal images. The processing modules and the classification module implement the data processing algorithms into digital development boards, enabling smart system characteristics. Finally, the interface module allows the final users to require the smart sensor to perform processing actions and data visualization, with the additional feature that the diagnosis report can be provided by the system. The smart sensor is validated in a real experimental test bench, demonstrating its capabilities in different case studies.

Enhanced hierarchical symbolic sample entropy: Efficient tool for fault diagnosis of rotating machinery

Intelligent fault diagnosis of rotating machinery is a key topic for industrial equipment maintenance and fault prevention. In this study, an intelligent diagnosis approach of rotating machinery via enhanced hierarchical symbolic sample entropy (EHSSE) is proposed. Firstly, a novel indicator termed symbolic sample entropy (SSE) is proposed for complexity measure and representation of fault information. By using symbolic dynamic filtering, the raw continuous time-series will be discretized into symbolic data, and analysis of symbolic data is less sensitive to measurement noise, resulting in superior robustness. Secondly, SSE is combined with enhanced hierarchical analysis to further extract fault characteristics hidden in both low- and high-frequency components. To study the performance of SSE and EHSSE, multiple simulated signals and experimental studies are constructed and three widely used entropy methods are employed to present a comprehensive comparison. The comparison results show that EHSSE performs best in diagnosing various faults of planetary gearbox and rotor system with highest identification accuracy compared with other entropy-based approaches.

Fault diagnosis of industrial robot gearbox using joint interclass and intraclass mappings

Due to complex structure and harsh operating conditions, it remains challenging to accurately diagnose industrial robot gearbox fault patterns. For this reason, a joint interclass and intraclass mappings (JIIM) strategy is presented in this paper to enhance the performance of industrial robot gearbox fault diagnosis. To this end, an echo state network (ESN) is first adopted to map the training data into their class centers for minimizing the interclass distance. Meanwhile, the maximization of interclass distance in the data space is achieved by equalizing the distance between class centers. For the mapped data, another ESN is proposed to classify the fault patterns. The proposed JIIM method is evaluated by fault-diagnosis experiments for industrial robot gearboxes. Improvements owing to minimizing the intraclass distance and equalizing the interclass distance are discussed in detail. Results show that the presented JIIM can efficiently improve the performance of industrial robot gearbox fault diagnosis.

Investigation on Dynamic Characteristics of Spur Gear Transmission System with Crack Fault

To study the dynamic characteristics of spur gears with crack faults, it is essential to establish the meshing stiffness model accurately. In this paper, the time-varying meshing stiffness (TVMS) model and load-sharing ratio (LSR) model of spur gears were established by the potential energy method, and the correctness of the TVMS model was verified by KISSsoft. The influence of gear crack depth and crack angle on TVMS and LSR was analyzed by introducing gear crack into the TVMS model. The multi-degree of freedom dynamics model of spur gears with centralized parameters was established. The time-domain characteristics of vibration of different crack fault systems were discussed, and statistical indicators were introduced to quantitatively evaluate the gear crack faults. The results show that TVMS decreases due to crack propagation, and the vibration impact increases with crack propagation. The sensitivity of crack depth propagation to gear fault characteristics is higher than that of crack angles; the statistical indicators can be used to monitor the propagation of early crack faults. The research results can provide a reference for the research of gearbox dynamic characteristics and fault diagnosis.

Sparse random projection-based hyperdisk classifier for bevel gearbox fault diagnosis

The fault diagnosis of bevel gearbox is of great significance. At present, the commonly used methods are based on pattern recognition, such as support vector machine, convex hull classifier and hyperdisk classifier. However, the number of elements in the kernel matrix of these kernel function-based classification methods increases squarely with the data size, resulting in intolerable training time. Based on this, a sparse random projection-based hyperdisk classifier model is proposed. The proposed method has the following novelties: First, based on sparse random projection and the geometrical characteristics of the hyperdisk model, a method is designed to efficiently screen out the core samples, and these samples are given different weights in this process. Second, the proposed method introduces slack variables and the dynamic penalty parameter to obtain a hyperdisk model with more reasonable boundary. Last, a strategy is developed to minimize the adverse effects of imbalanced training data. The effectiveness and applicability of the proposed method are verified on bevel gearbox fault data. The experimental results show that compared with other classifiers, the proposed method can greatly reduce the training time while guaranteeing a high classification accuracy. What’s more, it has better performance and efficiency in fault diagnosis with imbalanced training data.

A deep boosted transfer learning method for wind turbine gearbox fault detection

Deep learning methods have become popular among researchers in the field of fault detection. However, their performance depends on the availability of big datasets. To overcome this problem researchers started applying transfer learning to achieve good performance from small available datasets, by leveraging multiple prediction models over similar machines and working conditions. However, the influence of negative transfer limits their application. Negative transfer among prediction models increases when the environment and working conditions are changing continuously. To overcome the effect of negative transfer, we propose a novel deep transfer learning method, coined deep boosted transfer learning, for wind turbine gearbox fault detection that prevents negative transfer and only focuses on relevant information from the source machine. The proposed method is an instance-based deep transfer learning method that updates the weights of the source and the target machine training samples separately. The weights of different source training samples are gradually decreased to reduce the impact on the final model. The proposed method is verified by the Case Western Reserve University bearing and real field wind farm datasets. The results show that the proposed method ignores negative transfer and achieves higher accuracy compared to standard deep learning and deep transfer learning methods.

Gearbox Fault Diagnosis Based on Multi-Sensor and Multi-Channel Decision-Level Fusion Based on SDP

In order to deal with the shortcomings (such as poor robustness) of the traditional single-channel vibration signal in the comprehensive monitoring of the gearbox fault state, a multi-channel decision-level fusion algorithm was proposed based on symmetrized dot pattern (SDP) analysis, with the visual geometry group 16 network (VGG16) fault diagnosis model. Firstly, the SDP method was used to convert the vibration signal of a single multi-channel sensor into an imaging arm. Secondly, the obtained image arm was input into the VGG16 convolutional neural network in order to train the fault diagnosis model that can be obtained. Then, the SDP images of the signals that were to be measured from multiple multi-channel sensors were input into the fault diagnosis model, and the diagnosis results of multiple multi-channel sensors could then be obtained. Experimentally, it was demonstrated that the diagnostic results of multi-channel sensors one, two, and three were more accurate than those of single-channel sensors one, two, and three, by 3.01%, 16.7%, and 5.17%, respectively. However, the fault generation was not generated in a single direction, but rather multiple directions. In order to improve the comprehensiveness of the raw vibration data, a fusion method using DS (Dempster–Shafer) evidence theory was proposed in order to fuse multiple multi-channel sensors, in which the accuracy achieved 99.93% when sensor one and sensor two were fused, which was an improvement of 8.88% and 1.02% over single sensors one and two, respectively. When sensor one and sensor three were fused, the accuracy reached 99.31%, which was an improvement of 8.31% and 6.17% over single sensors one and three, respectively. When sensor two and sensor three were fused, the accuracy reached 99.91%, which was an improvement of 1.00% and 6.74% over single sensors two and three, respectively. When three sensors were fused simultaneously, the accuracy reached 99.99%, which was 8.93%, 1.08%, and 6.81% better than single sensors one, two, and three, respectively. Therefore, it can be proved that the number of sensor channels has a great influence on the diagnosis results.

Intelligent fault diagnosis of helical gearboxes with compressive sensing based non-contact measurements

Helical gearboxes play a critical role in power transmission of industrial applications. They are vulnerable to various faults due to long-term and heavy-duty operating conditions. To improve the safety and reliability of helical gearboxes, it is necessary to monitor their health conditions and diagnose various types of faults. The conventional measurements for gearbox fault diagnosis mainly include lubricant analysis, vibration, airborne acoustics, thermal images, electrical signals, etc. However, a single domain measurement may lead to unreliable fault diagnosis and the contact installation of transducers is not always accessible, especially in harsh and dangerous environments. In this article, a Compressive Sensing (CS)-based Dual-Channel Convolutional Neural Network (CNN) method was proposed to accurately and intelligently diagnose common gearbox faults based on two complementary non-contact measurements (thermal images and acoustic signals) from a mobile phone. The raw acoustic signals were analysed by the Modulation Signal Bispectrum (MSB) to highlight the coupled modulation components relating to gear faults and suppress the irrelevant components and random noise, which generates a series of two-dimensional matrices as sparse MSB magnitude images. Then, CS was used to reduce the image redundancy but retain key information owing to the high sparsity of thermal images and acoustic MSB images, which significantly accelerates the CNN training speed. The experimental results convincingly demonstrate that the proposed CS-based Dual-Channel CNN method significantly improves the diagnostic accuracy (99.39% on average) of industrial helical gearbox faults compared to the single-channel ones.

Deep subclass alignment transfer network based on time–frequency features for intelligent fault diagnosis of planetary gearboxes under time-varying speeds

Vibration signals of planetary gearboxes have complex components and time-varying characteristics. As the unstable operation of planetary gearboxes leads to unbalanced data distribution within vibration signals, it is difficult to extract gearbox fault information hidden in a large amount of data. Therefore, fault diagnosis of planetary gearboxes under nonstationary conditions is highly challenging. For the past few years, intelligent diagnosis methods have been extensively studied in the fault diagnosis field. However, inappropriate signal representations, inadequate training samples, and data differences increase the difficulty of diagnosing planetary gearbox faults. To address the above issues, this paper proposes an intelligent diagnostic framework based on time–frequency features and a deep residual joint subclass alignment transfer network (DSATN) for planetary gearbox fault diagnosis under nonstationary conditions. One-dimensional vibration signals are converted into time–frequency representation through signal processing techniques to reflect the variation of vibration frequency components within the time–frequency domain with time. During network training, the DSATN evaluates the data distributions between relevant subclasses in source and target tasks by using the local maximum mean discrepancy. Also, it utilizes a nonlinear transformation to align the global data distributions between both tasks, thus improving the generalization of the trained model for small sample sets. The proposed method is validated through planetary gearbox experiments and achieves good fault classification in the time–frequency domain of nonstationary vibration signals. Different gear and planet bearing fault categories are successfully identified.

An improved chaotic recognition method for weak signal frequency and its application to fault diagnosis of planetary gearboxes

Planetary gearboxes are very prone to failure when they are used in low-speed and heavy-load conditions for a long time. Due to strong background noise in the industrial field and the signal decay in the process of fault signal transmission, the weak fault characteristics of planetary gearboxes are submerged by noise and difficult to extract and identify. Aiming at this problem, an improved chaos detection method for weak signal frequency recognition is proposed. In this paper, the maximum weighted kurtosis is selected as the fitness function, and the Aquila optimizer algorithm is used to find the optimal modal number K and penalty factor α of variational modal decomposition. According to the grey relational degree, an inherent modal function (IMF) containing rich fault features is selected to reduce the noise of the original signal. Further, a double-coupled Duffing oscillator equation is constructed, and a quantitative criterion for chaotic oscillator phase transition based on the standard deviation weighted-average Euclidean distance is proposed. Based on the reverse detection method, the sideband growth of the most sensitive IMF after standardization is detected to determine the health state of the planetary gearbox. Detecting the fault characteristic frequency in the most sensitive IMF after normalization and Hilbert transform processing is based on the forward detection method in order to determine the fault type of the planetary gearbox. The effectiveness of the proposed method is verified by simulation and experiment. The results show that the proposed method successfully identifies the weak fault characteristics of the planetary gearbox, which fully shows that this method has an excellent diagnostic effect for planetary gearbox faults with rich frequency fault characteristics, and provides a new method for the diagnosis and identification of weak faults in planetary gearboxes in engineering practice.

An intelligent fault diagnosis for machine maintenance using weighted soft-voting rule based multi-attention module with multi-scale information fusion

The ability of engineering systems to process multi-scale information is a crucial requirement in the development of an intelligent fault diagnosis model. This study develops a hybrid multi-scale convolutional neural network model coupled with multi-attention capability (HMS-MACNN) to solve both the inefficient and insufficient extrapolation problems of multi-scale models in fault diagnosis of a system operating in complex environments. The model's capabilities are demonstrated by its ability to capture the rich multi-scale characteristics of a gearbox including time and frequency multi-scale information. The capabilities of the Multi-Attention Module, which consists of an adaptive weighted rule and a novel weighted soft-voting rule, are respectively integrated to efficiently consider the contribution of each characteristic with different scales-to-faults at both feature- and decision-levels. The model is validated against experimental gearbox fault results and offers robustness and generalization capability with F1 value that is 27% higher than other existing multi-scale CNN-based models operating in a similar environment. Furthermore, the proposed model offers higher accuracy than other generic models and can accurately assign attention to features with different scales. This offers an excellent generalization performance due to its superior capability in capturing multi-scale information and in fusing advanced features following different fusion strategies by using Multi-Attention Module and the hybrid MS block compared to conventional CNN-based models.

Gearbox fault diagnosis under nonstationary condition using nonlinear chirp components extracted from bearing force

The main challenges in gearbox condition monitoring and fault diagnosis are the heavy noises caused by harsh operating environments and the complicatednonstationary condition during the runtime. It is extremely difficult and even impossible to obtain sensitive faultcharacteristics through direct measurements. Thus, it is ofgreat significance of faultcharacteristic inversion based on the structure vibration transfer paths for complicated mechanical systems fault diagnosis. This paper proposes a gearbox fault diagnosis method under nonstationary condition through using nonlinear chirp components extracted from bearing force (BFNCC). In the proposed method, the bearing dynamic forces are first identified based on the vibration transfer function matrix, and the intrinsic signal components are then decomposed through a nonlinear chirp mode decomposition method. The proposed method improved the quality of vibration signals by removing the effect of the structure vibration transfer paths. In comparison with order tracking, the BFNCC is capable of clearly reflecting the magnitude increase phenomenon under gear fault condition. Numerical simulation and experimental studies show that the proposed method is effective for gear fault diagnosis under nonstationary condition. The proposed method is demonstrated to be superior to the existing approaches and robust to the location of measuring points, and thus shows potential in machinery fault diagnosis under the interference of complicated structure transfer paths.

A dynamic-model-based fault diagnosis method for a wind turbine planetary gearbox using a deep learning network

The planetary gearbox is a critical part of wind turbines, and has great significance for their safety and reliability. Intelligent fault diagnosis methods for these gearboxes have made some achievements based on the availability of large quantities of labeled data. However, the data collected from the diagnosed devices are always unlabeled, and the acquisition of fault data from real gearboxes is time-consuming and laborious. As some gearbox faults can be conveniently simulated by a relatively precise dynamic model, the data from dynamic simulation containing some features are related to those from the actual machines. As a potential tool, transfer learning adapts a network trained in a source domain to its application in a target domain. Therefore, a novel fault diagnosis method combining transfer learning with dynamic model is proposed to identify the health conditions of planetary gearboxes. In the method, a modified lumped-parameter dynamic model of a planetary gear train is established to simulate the resultant vibration signal, while an optimized deep transfer learning network based on a one-dimensional convolutional neural network is built to extract domain-invariant features from different domains to achieve fault classification. Various groups of transfer diagnosis experiments of planetary gearboxes are carried out, and the experimental results demonstrate the effectiveness and the reliability of both the dynamic model and the proposed method.

Petrochemical Gearbox Fault Location and Diagnosis Method Based on Distributed Bayesian Model and Neural Network

<p>Increasing attention has been paid to the economic losses and personnel injuries caused by petrochemical gearbox faults. As a result, petrochemical enterprises started to pay huge attention on fault diagnosis technology to solve the fault diagnosis problem. Petrochemical gearboxes are characterized by many fault types, feature variables, and many-to-many relationships between the various fault parameters, which pose huge challenges in the fault diagnosis of petrochemical units. This paper proposes a petrochemical gearbox fault location and diagnosis method based on a distributed Bayesian model and neural network. The proposed approach is based on sample feature information and Bayesian network prior probability to construct a basic framework for petrochemical gearbox fault location. Neural network technology is used to to diagnose fault types. It is helpful to build a long-term fault diagnosis and monitoring system for rotating machinery of petrochemical units.</p> <p> </p>

Multiscale Weighted Morphological Network Based Feature Learning of Vibration Signals for Machinery Fault Diagnosis

Vibration signals are widely utilized for machinery fault diagnosis. However, the fault-related components (i.e., impulse) in vibration signals are often buried by strong background noises due to complex working conditions. Thus, it is challenging to directly extract discriminate features from vibration signals. In this article, a novel deep neural network (DNN), multiscale weighted morphological network (MWMNet), is proposed to extract impulses from vibration signals. First, a novel morphological layer is smoothly embedded in DNN as a signal processing layer to extract impulses and filter out the noise. Second, multiple branches with different structure element (SE) scales are employed to respectively extract impulses. An adaptive weighted fusion is utilized to enhance the scales that provide strong impulsive components. Third, a morphological operator, called average-hat transform, is adopted in MWMNet, and both positive and negative impulses can be extracted from vibration signals. The effectiveness of MWMNet is validated by the experiments of gearbox fault diagnosis and bearing fault diagnosis. The experimental results show that MWMNet can learn the fault-related features and filter out noise from vibration signals well. The comparison results illustrate that MWMNet has a better fault diagnosis performance than those DNNs, e.g., one-dimensional convolutional neural network, densely connected convolutional network, and residual network.

A low-frequency fault detection method for low-speed planetary gearbox based on acoustic signals

In some practical application of construction machinery and equipment, the planetary gearbox in the walking parts is wrapped with tires, and its speed is very low. When this planetary gearbox fails, the fault frequency is very low (less than 1 Hz) due to its complex transmission relationship. Because the planetary gearbox has the characteristics of low speed, low fault frequency and surrounded by tires, it is difficult to detect its fault condition in practical engineering. This work proposes a new low-frequency fault detection method. First, the pre-processing is used for the measured acoustic signals. Second, the pre-processed acoustic signals are decomposed into components by the multi-resolution singular value decomposition. Third, the autoregressive model is constructed to select fault-related component. Finally, the output component is further analyzed by envelope spectrum analysis. The results indicate that the proposed method can accurately detect the characteristic information of low frequency faults of planetary gearbox in practical engineering.

Gearbox Fault Identification Model Using an Adaptive Noise Canceling Technique, Heterogeneous Feature Extraction, and Distance Ratio Principal Component Analysis.

Using an adaptive noise canceling technique (ANCT) and distance ratio principal component analysis (DRPCA), this paper proposes a new fault diagnostic model for multi-degree tooth-cut failures (MTCF) in a gearbox operating at inconsistent speeds. To account for background and disturbance noise in the vibration characteristics of gear failures, the proposed approach employs ANCT in the first stage to optimize vibration signals. The ANCT applies an adaptive denoising technique to each basic frequency segment in the whole frequency response of vibrations. Following that, a novel DRPCA is used to extract the discriminating low-dimensional features. The DRPCA initially determines each feature’s relative proximity to fault categories by computing the average Euclidian distance ratio between similar and dissimilar classes. The most discriminatory features with the lowest dimensions are selected, as determined by principal component analysis (PCA). The new DRPCA is created by combining distance ratio–based feature inspection with PCA. The optimal feature set containing the most discriminative features is then fed to the support vector machine classifier to identify multiple failure categories. The experimental results indicate that the proposed model outperforms the state-of-art approaches and offers the highest identification accuracy.

Fault Diagnosis Method of Planetary Gearbox Based on Compressed Sensing and Transfer Learning

This paper suggests a novel method for diagnosing planetary gearbox faults. It addresses the issue of network bandwidth limitation during wireless data transmission and the problem of relying on expert experience and insufficient training samples in traditional fault diagnosis. The continuous wavelet transform was combined with the AlexNet convolutional neural network using transfer learning and the compressed theory of sense. The original vibration signal was compressed and reconstructed using the compressed sampling orthogonal matching pursuit reconstruction algorithm. A continuous wavelet transform was used to convert the compressed signal into a time–frequency image. The pretrained AlexNet model was selected as the migration object, the network model was fine-tuned and retrained, and the trained AlexNet model was used to diagnose the fault using the model-based migration method. It was demonstrated by the experimental results when the compression ratio CR = 0.5. Compared to other network models, the classification accuracy rate is 97.78%. This method has specific reference value and application prospects and good feature extraction and fault classification capabilities.