Planetary Gearbox Fault Diagnosis Research Articles

High-order synchrosqueezing wavelet transform and application to planetary gearbox fault diagnosis

The synchrosqueezing transform (SST) is a powerful tool for time-frequency analysis of signals with slowly varying instantaneous frequency (IF). However, the SST and its extensions provide poor time-frequency resolution for signals with wide frequency range and fast varying IF. In this paper, a new SST method called high-order synchrosqueezing wavelet transform is proposed to achieve a highly energy-concentrated time-frequency representation (TFR) for nonstationary signals with wide frequency range and fast varying IF. This method uses high-order group delay and chirp rate operators to obtain the accurate estimation of instantaneous frequency. The proposed method can effectively improve the energy concentration of the TFR and remain invertible simultaneously. The numerical simulations investigate the performance and noise robustness of the proposed method when analyzing a typical amplitude-modulated and frequency-modulated (AM-FM) multicomponent signal. Finally, the application of planetary gearbox fault diagnosis in the variable operating condition verifies the effectiveness of the proposed method.

Multiple Wavelet Coefficients Fusion in Deep Residual Networks for Fault Diagnosis

Wavelet transform, an effective tool to decompose signals into a series of frequency bands, has been widely used for vibration-based fault diagnosis in machinery. Likewise, the use of deep learning algorithms is becoming popular to automatically learn discriminative features from input data for the sake of improving diagnostic performance. However, the fact that no general consensus has been reached as to which wavelet basis functions are useful for diagnosis motivated this investigation of methods to fuse multiple wavelet transforms into deep learning algorithms. In this paper, two methods—i.e., multiple wavelet coefficients fusion in deep residual networks by concatenation and multiple wavelet coefficients fusion in deep residual networks by maximization—were developed to capture discriminative information from diverse sets of wavelet coefficients for fault diagnosis. The efficacy of the developed methods was verified by applying them to planetary gearbox fault diagnosis.

The Optimized Deep Belief Networks With Improved Logistic Sigmoid Units and Their Application in Fault Diagnosis for Planetary Gearboxes of Wind Turbines

Efficient and accurate planetary gearbox fault diagnosis is the key to enhance the reliability and security of wind turbines. Therefore, an intelligent and integrated approach based on deep belief networks (DBNs), improved logistic Sigmoid (Isigmoid) units, and impulsive features is proposed in this paper. The vanishing gradient problem is an inherent drawback of conventional Sigmoid units, and it usually occurs in the backpropagation process of DBNs, resulting in that the training is considerably slowed down and the classification rate is reduced. To solve this problem, Isigmoid units are designed to combine the merits of unsaturation from leaky rectified linear (LReL) units. The results of handwritten digit recognition experiments show the superiority of Isigmoid over Sigmoid on convergence speed and classification accuracy. Since impulses contain much useful fault information, especially for early failures, an integrated approach using the optimized Morlet wavelet transform, kurtosis index, and soft-thresholding is applied to extract impulse components from original signals to improve the diagnosis accuracy. Then, the features extracted from original signals and impulsive signals are employed to train and test the DBNs with Isigmoid, Sigmoid, and LReL units for comparison. Finally, the results of planetary gearbox fault diagnosis show that Isigmoid has higher comprehensive performance than conventional sigmoid and LReL.

Time-Frequency demodulation analysis via Vold-Kalman filter for wind turbine planetary gearbox fault diagnosis under nonstationary speeds

Abstract Wind turbine planetary gearbox fault diagnosis under nonstationary speeds is a challenging topic, because of the high complexity and strong time variability of vibration signals. In order to resolve time-varying gear fault features, a quality time-frequency analysis method is in demand. Time-frequency representations based on Hilbert transform and analytic signal approach have fine time-frequency resolution, and are free from both outer (cross-term) and inner (auto-term) interferences, thus providing an effective approach to nonstationary signal analysis. However, they rely on accurate instantaneous frequency estimation, and thereby are subject to mono-component constraint. To address this issue, Vold-Kalman filter is exploited to construct time-frequency representation, by virtue of its capability to decompose the multi-component vibration signal of rotating machinery into constituent mono-component harmonic waves. Even so, intricate time-varying sidebands inherent with raw planetary gearbox vibration signals in joint time-frequency domain are still a hurdle, because they do not link to gear fault frequency directly. To solve this problem, the proposed time-frequency analysis method is further extended to generate time-varying amplitude and frequency demodulated spectra, inspired by the fact that gear fault frequency is manifested straight by the amplitude and frequency modulating frequencies. The proposed method is illustrated by numerical simulation, and is further validated using lab experimental signals of a wind turbine planetary gearbox. Both the localized and distributed faults on gears are successfully diagnosed under nonstationary speeds.

Induction Motor Stator Current AM-FM Model and Demodulation Analysis for Planetary Gearbox Fault Diagnosis

Induction motor-planetary gearbox drivetrains are widely used for industrial productions, including machine tools in manufacturing systems. For fault diagnosis of planetary gearboxes in such electromechanical systems, motor current signal analysis provides an effective alternative approach, because motor current signals have easier accessibility and are free from time-varying transfer path effects. Planetary gearbox faults generate load torque oscillations, leading to both amplitude modulation and frequency modulation (AM-FM) effects on induction motor current signals. To thoroughly understand gear fault features in current signals, an AM-FM current signal model is derived through mechanical–magnetic–electric interaction analysis, explicit equation of Fourier spectrum is derived, and sidebands characteristics are summarized. To avoid an intricate sideband analysis, amplitude and frequency demodulation analyses are proposed, explicit equations of corresponding demodulated spectra are derived, and gear fault features are summarized. The theoretical derivations are validated through lab experiments. Localized fault on the sun, planet, and ring gears are all successfully diagnosed using the proposed method.

Weighted sparse representation based on failure dynamics simulation for planetary gearbox fault diagnosis

The most common and effective way of rotating machinery diagnostics is to extract fault impact features from the vibration signals and to carry out further processing. As for planetary gear sets, because of the simultaneous mesh of multiple gears and the effect of the carrier rotation, the fault features are submerged in the strong harmonic signals and other noise. Due to the lack of prior information on the faults, the conventional fault diagnostic methods often fail to achieve satisfactory results. To address this problem, we give the prior information of a chipped planetary gear set by dynamics simulation, and utilize the information in the time domain to improve the diagnosis performance. Firstly, a pure torsional lumped parameter model is used to simulate the system vibration response with different degrees of failure. Through statistical analysis, the margin factor, the most sensitive indicator, is selected as prior information to reflect the local gear fault among several indexes. Finally, a weighted sparse representation method based on the prior information provided above is proposed to extract the impact features. Moreover, it is found that the extracted components have a strong–weak cyclic impact feature in the chipped planetary gear sets. The features and effectiveness of the method are verified by an experiment on a planetary gearbox test rig. To validate the superiority of the proposed method, comparisons are made among several state-of-the-art feature extraction methods.

Deep residual learning with demodulated time-frequency features for fault diagnosis of planetary gearbox under nonstationary running conditions

Due to the tough and time-varying working conditions, fault diagnosis technique is of critical significance for drive-chain system in rotating machines. In recent years, many statistical and spectral feature extraction methods have been developed and applied, but unfortunately, they are incapable of dealing with mechanical behaviors under varying running conditions. Besides, the lack of specific dynamical knowledge also becomes an obstacle for effective diagnosis through direct spectral analysis. Accordingly, a data-driven fault diagnosis method based on time-frequency analysis and deep residual network is proposed in this research. Firstly, a deep residual network is pre-trained on spectral features extracted under fixed rotating speeds. For the transient signals, an accurate phase function is constructed via probabilistic instantaneous angular speed (IAS) estimation algorithm based on time-frequency representations. Then the generalized demodulation operator is utilized to remove rotating speed fluctuation. Afterwards, several groups of instantaneous features demodulated from time-frequency representations are input to the deep residual network to test the performance of proposed method under nonstationary running conditions. The diagnosis results of a planetary gearbox test rig are compared with other traditional methods; the comparisons show that the proposed data-driven fault diagnosis method achieved significant improvement on incipient fault detection accuracy under varying rotating speed.

Vibration based condition monitoring and fault diagnosis of wind turbine planetary gearbox: A review

Abstract As one of the most immensely growing renewable energies, the wind power industry also experiences a high failure rate and operation & maintenance cost. Therefore, the condition monitoring and fault diagnosis of a wind turbine (WT) generator set are highly needed. Among different components of a WT generator set, WT planetary gearbox plays a crucial role in transmission and leads to relatively higher failure rate and longer downtime. Towards this, a number of studies have been reported in both the academic journals and conference proceedings. This paper provides a systemic and pertinent state-of-art review on WT planetary gearbox condition monitoring techniques on the topics of fundamental analysis, signal processing, feature extraction, and fault detection. Moreover, a few valuable open issues are pointed out and potential research directions are suggested.

Intelligent fault diagnosis of rotating machinery based on one-dimensional convolutional neural network

Fault diagnosis of rotating machinery plays a significant role in the reliability and safety of modern industrial systems. The traditional fault diagnosis methods usually need manually extracting the features from raw sensor data before classifying them with pattern recognition models. This requires much professional knowledge and complex feature extraction, only to cause results in a poor flexibility of the model, which only applies to the diagnosis of a fault in particular equipment. In recent years, deep learning has developed rapidly, and great achievements have been made in image analysis, speech recognition and natural language processing. However, its application in fault diagnosis of rotating machinery is still at the initial stage. In order to solve the problem of end-to-end fault diagnosis, this paper focuses on developing a convolutional neural network to learn features directly from the original vibration signals and then diagnose faults. The effectiveness of the proposed method is validated through PHM (Prognostics and Health Management) 2009 gearbox challenge data and a planetary gearbox test rig. Compared with the other three traditional methods, the results show that the one-dimensional convolutional neural network (1-DCNN) model has higher accuracy for fixed-shaft gearbox and planetary gearbox fault diagnosis than that of the traditional diagnostic ones.

An enhanced convolutional neural network with enlarged receptive fields for fault diagnosis of planetary gearboxes

Due to the complicated structure and tough working environment of planetary gearboxes, intelligent identification of the health states based on the raw vibration signal is still a huge challenge in equipment maintenance. Aiming at this issue, an enhanced convolutional neural network (ECNN) with enlarged receptive fields was proposed in this paper. First, a one-dimensional convolutional layer was applied to enlarge receptive field preliminarily and capture the fault information within each group of adjacent points in the vibration signal. Then, several fused dilated convolutional layers were constructed to enlarge the receptive field further and capture the long distance dependencies of the raw signal comprehensively. At last, the raw vibration signals were directly fed into the developed ECNN to train the fault diagnosis model, and evaluate the ECNN model with the testing data. The experimental results demonstrated that the developed method can enhance the fault feature learning ability by enlarging the receptive fields twice, and achieved higher diagnosis accuracies than the traditional deep learning methods in fault diagnosis of planetary gearboxes.

Self-Powered Wireless Sensor for Fault Diagnosis of Wind Turbine Planetary Gearbox

This paper proposes a wind turbine planetary gearbox (PGB) fault diagnosis method based on a self-powered wireless sensor. The proposed wireless sensor, which consists of a piezoelectric energy harvester, a power management circuit, a microcontroller unit (MCU), a radio-frequency (RF) module, and an accelerometer, can acquire the vibration signals of wind turbine PGB by the accelerometer. The piezoelectric energy harvester utilizing vibration environment is optimized as a power supply for the proposed wireless sensor, including the MCU, RF module, and accelerometer. An ac–dc converter combined with a low-dropout voltage regulator is developed to provide stable dc voltage for the proposed wireless sensor. Stacked denoising autoencoder (SDAE) shows excellent performance in learning robust features from the noised signal. Thus, in this paper, the SDAE method is adopted to learn robust and distinguishable features from measured signals. Then, the least squares support vector machine (LSSVM) is employed to classify features extracted by the SDAE. Both the SDAE and LSSVM are optimized by quantum particle swarm optimization (QPSO). The experimental results show that the presented power supply can generate 3.3-V dc voltage, which ensures regular operation of the rest of the wireless sensor. The proposed wireless sensor can achieve a reliable communication distance of 40.8 m in the test environment. Furthermore, the SDAE approach and LSSVM show excellent performance in feature extraction and fault diagnosis, respectively. The experimental results indicate that the proposed method is effective in terms of fault diagnosis for the wind turbine PGB.

Wind Turbine Planetary Gearbox Fault Diagnosis Based on Self-Powered Wireless Sensor and Deep Learning Approach

In this paper, a novel and more effective fault diagnosis approach for wind turbine planetary gearbox (PGB) is proposed. In order to better detect the faults in the early stage of the faults of the wind turbine PGB, the corresponding maintenance measures can be carried out to prevent the faults from becoming more serious, so as to seriously affect the normal operation of the fan gearbox. The gear with lighter fault degree is used to simulate the early fault signal. Compared with the fault which has seriously affected the normal working condition, the fault characteristics of the early fault signal are more difficult to detect. So in this design, deep belief network (DBN) optimized by quantum particle swarm optimization (QPSO) algorithm is used to extract deeper and more identifiable features of slight fault signal. After optimization by QPSO algorithm, DBN can get a most suitable structure according to the actual working signal of fan gearbox. Then these extracted features are input into the least squares support vector machine (LSSVM) optimized by QPSO for fault diagnosis test. At the same time, the wireless sensor nodes using self-energy in vibration state are optimized. By using microcontroller unit (MCU) MSP430F149 and nRF24L01 radio frequency (RF) chip with lower energy consumption, the normal dormant state can be maintained, the power requirement of transmission mode can be met, the stability of the whole node can be improved, and the phenomenon of energy shortage caused by short-term fluctuation can be prevented. The comparative experiments in this paper show that this method has good effect on the fault diagnosis of wind turbine PGB.

Adaptive Estimation of Instantaneous Angular Speed for Wind Turbine Planetary Gearbox Fault Detection

Planetary gearbox faults are the leading causes of downtime in wind turbines (WTs). In recent years, numerous and various vibration-based approaches have been put forward for WT gearbox fault detection. In the vibration-based techniques, order tracking-based methods, which by identifying fault orders in gearbox drivetrains, are regarded as very promising and powerful techniques. In the currently available order tracking methods, auxiliary devices are required to accurately obtain the instantaneous angular speed (IAS) of drivetrain. To tackle this problem, instantaneous angular speed estimation from vibration signals has been studied and some tacho-less order tracking (TLOT) approaches have been developed. However, many vital parameters for IAS estimation in the currently available TLOT algorithms need to be manually selected, which raise the question of user-friendliness, even result in a false diagnosis. As mentioned earlier, aiming at the shortcomings, a novel TLOT method based on adaptive IAS estimation is proposed for WT planetary gearbox fault diagnosis. In the proposed method, the nonlinear mode decomposition (NMD) method is improved, and its computational burden is reduced. And, the tachometer information of the drivetrain is adaptively extracted by the improved NMD method from generator vibration signal for gearbox vibration signal resampling. A field test is conducted, and the vibration signal of WT planetary gearbox with the compound fault is used for further investigation. The experimental validation results demonstrate that the planetary gearbox compound fault can be successfully detected, and the proposed method outperforms the traditional method based on generalized demodulation.

Time-frequency space vector modulus analysis of motor current for planetary gearbox fault diagnosis under variable speed conditions

Abstract Motor current signature analysis is a promising technique for electromechanical system fault detection, and has been studied under steady states. In this paper, a planetary gearbox fault diagnosis method under variable speed conditions using stator current signal is proposed. In order to thoroughly understand the frequency characteristics of current signals, an analytical amplitude modulation and phase modulation (AM-PM) current model considering gear fault modulation effects is presented. Two more aspects of endeavor are made to highlight gear fault signatures in stator current signals in context of inconspicuousness and time variability. Firstly, to address the sideband complexity and power supply dominance issues inherent with stator current signals, squared space vector modulus (SVM) together with its time-varying spectral characteristics in planetary gearbox fault cases under variable speed conditions are mathematically derived. Secondly, to reveal time-varying fault features in details, polynomial chirplet transform (PCT) is improved by iterative algorithm, and merits of fine time-frequency resolution and cross term free nature are achieved. The effectiveness of the proposed method is illustrated by numerical simulation, and is further validated by lab experiments on a real world 4 kW induction motor driven planetary gearbox test rig.

Velocity synchronous bilinear distribution for planetary gearbox fault diagnosis under non-stationary conditions

This paper proposes a time-frequency method, named velocity synchronous bilinear distribution (VSBD), for planetary gearbox fault diagnosis under non-stationary conditions. This method is based on the Cohen's class bilinear distribution (CCBD). The current CCBDs are able to provide a time-frequency representation (TFR) with good time-frequency resolution for multi-component signal with constant frequencies. However, their poor time-frequency resolution in dealing with time-varying frequencies has limited their applications. As such, the VSBD is proposed to solve this problem. This method firstly uses the generalized demodulation to demodulate the signal under analysis into several signals based on the shaft rotational speed to meet the constant-frequency requirement of the bilinear distribution. The CCBD is then applied to the demodulated signals to obtain the TFRs of the demodulated signals and finally the TFR of the original signal is restored from the demodulated TFRs. This method is free from smear effect and has improved time-frequency resolution in comparison with other existing time-frequency methods. This method only considers planetary gearbox systems running under time-varying speed and fixed external load conditions. Therefore, the industrial applications of the proposed method may be restricted to such conditions.

Planetary gearbox fault feature learning using conditional variational neural networks under noise environment

The features signals of early fault collected from planetary gearbox are usually weak. It is difficult to extract effective fault features from the collected vibration signals under noise environment. In this paper, a new feature learning method for fault diagnosis of planetary gearbox based on deep conditional variational neural networks (CVNN) is proposed. First, the new method utilizes multi-layer perceptron (MLP) to model the normal distribution features of frequency spectra from noisy vibration signals. Second, the new features are obtained by resampling normal distribution features in order to eliminate the effect of noise. Then the denoised features are compressed and reduced dimensionally by MLP. Third, the effective denoised features are input to classifier. Finally, the trained CVNN is applied for intelligent fault diagnosis of planetary gearbox. The experimental results confirm that CVNN method can extract effective fault features from noisy vibration signals, and it has higher accuracy of fault diagnosis than other methods in the case of low signal to noise ratio (SNR) values.

A multivariate encoder information based convolutional neural network for intelligent fault diagnosis of planetary gearboxes

Rotary encoder signal, as the built-in position information, possesses a wide variety of advantages over vibration signal and has aroused great interest in the field of health monitoring for rotating machinery. However, there are two major issues when attempting to detect and diagnose failures with encoder information. First of all, a series of proper signal processing methods need to be designed for fault feature extraction, which largely relies on the expert experience and domain knowledge. Furthermore, existing studies primarily concentrate on a single transform, such as instantaneous angular speed, which neglects the diversity of encoder information. In view of above deficiencies, a multivariate encoder information based convolutional neural network (MEI-CNN) is proposed for intelligent diagnosis in this paper. In this framework, three different types of dynamic encoder information are firstly acquired by analyzing and processing the raw position sequence, after that multivariate encoder information (MEI) data are constructed by data fusion. Finally, a concise and effective convolutional neural network is designed to extract discriminating features and provide diagnosis results. The proposed method not only overcomes drawbacks of traditional techniques based on vibration analysis, but also provides an intelligent way to achieve satisfactory diagnosis results. The effectiveness and superiority of MEI-CNN are validated by experimental data from a planetary gearbox test rig. The results also indicate that the proposed method may offer a promising tool for intelligent diagnosis of rotating machinery.

Amplitudes of characteristic frequencies for fault diagnosis of planetary gearbox

Frequency contents have been widely investigated to understand the vibration behaviors of planetary gearboxes. Appearances of certain sideband peaks in the frequency spectrum may indicate the occurrence of gear fault. However, analyzing too many sidebands will create problems and uncertainty of fault diagnoses. To this end, it is of vital importance to focus on those sidebands, as well as their amplitudes, which are directly induced by the gear faults. The Sideband Energy Ratio (SER) method, which synthesize the amplitudes of characteristic frequencies and meshing frequency, has shown its effectiveness in fault diagnosis of fixed-shaft gearboxes. However, for planetary gearboxes, the effectiveness and theoretical explanation behind this method still needs to be explored. In this paper, we first explored the amplitudes of characteristic frequencies based on a phenomenological model. Our investigation demonstrated that monitoring the amplitude of a single frequency component is inadequate for fault diagnosis of planetary gearbox. Second, the theoretical explanation of SER for a planetary gearbox is explored. Finally, a modified SER, namely the Modified Sideband Energy Ratio, is proposed to deal with the problem of rotating speed fluctuation. Experimental studies are provided to demonstrate the effectiveness of the proposed method.

Planetary gearbox fault diagnosis based on data-driven valued characteristic multigranulation model with incomplete diagnostic information

There are many uncertain factors that may result in incomplete diagnostic information of planetary gearboxes, such as sensor malfunctions, communication lags, and data discretization, etc. Therefore, incomplete diagnostic information of planetary gearboxes may simultaneously contain two categories of unknown attribute values. However, existing fault diagnosis methods of planetary gearboxes are hard to realize fault diagnosis using incomplete diagnostic information that simultaneously contains two categories of unknown attribute values. To overcome this issue, a fault diagnosis method of planetary gearboxes based on data-driven valued characteristic multigranulation model with incomplete diagnostic information is proposed. First, a calculation method of characteristic similarity degrees among cases is introduced, and a data-driven valued characteristic relation is defined. The data-driven valued characteristic relation is used to analyze and process incomplete diagnostic information that simultaneously contains two categories of unknown attribute values. Then, a data-driven valued characteristic multigranulation model is defined according to multigranulation model. An attribute reduction algorithm based on pessimistic data-driven valued characteristic multigranulation model is employed to extract fault diagnosis decision rules. Finally, naive Bayesian classifier is constructed to identify planetary gearbox conditions. The effectiveness of this method is validated and the advantages are investigated using a fault diagnosis experiment of planetary gearbox. Experimental results demonstrate that this method can accurately determine indiscernibility relation among cases, reduce computational complexity, and enhance fault diagnosis accuracy.

An intelligent diagnosis scheme based on generative adversarial learning deep neural networks and its application to planetary gearbox fault pattern recognition

Abstract Planetary gearbox has complex structures and works under various non-stationary operating conditions. The vibration signals of planetary gearbox are complicated and usually polluted by noise and interference. It is difficult to extract effective features of early faults. In addition, there are only a small number of fault samples for planetary gearbox fault diagnosis. All of these increase the difficulty of planetary gearbox fault diagnosis. Aiming at these problems, a novel fault diagnostic method is proposed which combines Generative Adversarial Networks (GAN) and Stacked Denoising Autoencoders (SDAE). The generator of GAN can generate new samples which has similar distribution with original samples from planetary gearbox vibration signals. Then, generated samples are transformed to the discriminator together with original samples which expand the sample size. SDAE is used as the discriminator of GAN which can automatically extract effective fault features from input samples and discriminate their authenticity and fault categories. Through novel adversarial machine learning mechanism, the generator and discriminator are concurrently optimized to enhance the quality of generation samples and the ability of fault mode classification. The experimental results show that the developed SDAE-GAN method for planetary gearbox has good anti-noise ability and achieve better fault diagnosis performance in the case of small samples.