Wind Turbine Vibration Signals Research Articles

Offshore field experimentation for novel hybrid condition monitoring approaches

This study details the development of a fully automated pipeline for the condition monitoring of wind turbine drive trains. Vibration data is collected using hardware designed and manufactured in-house and used directly to monitor the condition of the drive trains. The complex nature of wind turbine vibration signals, due to the large number of components and highly variable operating conditions, makes drive train condition monitoring a challenging task. This paper details the full data measurement and analysis flow from sensor to insights and proposes a hybrid automated pipeline with signal processing and data-driven techniques to address the complexity of dealing with wind turbine vibration data. The vibration signals are directly employed to estimate the wind turbine’s instantaneous angular speed to compensate for any rotation speed fluctuations. Pre-processing is performed on the speed-independent signals to evaluate condition indicators in both the time and spectral domain for the vibration signals and their envelopes. Machine learning is then employed to distinguish the healthy state of the machine from a faulty one using the computed condition indicators. Besides the scalar indicators, also two-dimensional vibration decompositions such as the cyclic spectral correlation maps are used as inputs to the machine learning pipeline. This comprehensive and automated approach ensures both an early and reliable fault detection. Experimental results demonstrate that the fully automated hybrid pipeline can effectively be used for fleet-based health tracking of offshore wind turbine drivetrains.

Fault diagnosis of offshore wind turbines based on component separable synchroextracting transform

The fault diagnosis of wind turbines is crucial for wind power generation. However, the violent variation in wind power and directions of offshore wind turbines often results in high nonstationary vibration signals, which poses a challenge for effective fault recognition. Time-frequency analysis (TFA) is a common method to reveal the time-varying frequency components caused by faults. In this paper, we propose a component separable synchroextracting transform (CSSET) by exploring the unique modulation characteristics of wind turbine vibration signals. For the method, a fundamental frequency is firstly estimated by a time-frequency ridge optimization method, and the potential fault characteristic frequency (FCF) set is constructed according to the modulation characteristics of the wind turbine vibration signal. Next, the instantaneous amplitudes (IAs) of these components in the set are estimated by a time-varying bandpass filter. Then, the instantaneous frequency (IF) of the reconstructed single component is extracted by the synchronous extraction operator (SEO), and a high-precision time-frequency representation (TFR) is obtained. The proposed method takes full advantage of the physical modulation characteristics of vibration signals and the frequency components which are most relevant to fault information are only preserved. The effectiveness of the proposed method is verified by analyzing simulation and experimental signals.

Filtering method of coupled components in wind turbine signals

To study whether the coupled components in vibration signals affect the accuracy of judging the wind turbine running state, an adaptive filter was used to filter the coupled components, with different filters performing differently. The vibration signal subjected to a stationary random process was taken as the object of research. The probability distribution that the non-coupled part of the simulated signals obey was determined by the collected mechanical vibration signals. The frequencies of coupled components in the simulated signals were determined by a one-and-a-half-dimensional spectrum of their corresponding wind turbine vibration signals. The simulated signal, including both non-coupled and coupled components, was input to the adaptive filter to train it, and afterwards the adaptive filter was used to filter the coupled components in vibration signals. The experimental results show that this method may be more effective in judging the mechanical operation states,the correct rate of fault diagnosis has been improved by more than 10%. This method can help improve the power generation efficiency of wind turbines.

Investigating parallel multi-step vibration processing pipelines for planetary stage fault detection in wind turbine drivetrains

This paper proposes a signal processing approach for wind turbine gearbox vibration signals based on employing multiple analysis pipelines. These so-called pipelines consist of combinations of various advanced signal processing methods that have been proven to be effective in literature when applied to wind turbine vibration signals. The performance of the pipelines is examined on vibration data containing different wind turbine gearbox faults in the planetary stages. Condition indicators are extracted from every pipeline to evaluate the fault detection capability for such incipient failures. The results indicate that the multipronged approach with the different pipelines increases the reliability of successfully detecting incipient planetary stage gearbox faults. The type, location, and severity of the fault influences the choice for the appropriate processing method combination. It is therefore often insufficient to only utilize a single processing pipeline for vibration analysis of wind turbine gearbox faults. Besides investigating the performance of the different processing techniques, the main outcome and recommendation of this paper is thus to employ a diversified analysis methodology which is not limited to a sole method combination, to improve the early detection rate of planetary stage gearbox faults.

Pre-Treatment of Weak Information of Status Feature of Wind Turbine Gearbox Based on Wavelet New Threshold Transform

The early stage of the running stability degradation of wind turbine drive system is very subtle and always covered up by variable working condition or other nondegradation signals; what is more, the wind turbine vibration signals collected by sensors usually contain a lot of noises, thus they are of low signal-noise ratio and high complexity. Wavelet analysis is superior to Fourier analysis in the processing of complex and nonstationary signal because the former can focus on any detail of the analytic target. To make up the deficiencies of hard and soft threshold functions and integrate their advantages, a new threshold function algorithm has been designed. The failure data and field test of wind turbine gearbox proved the effectiveness of the method.

A novel wind turbine fault diagnosis method based on intergral extension load mean decomposition multiscale entropy and least squares support vector machine

Abstract Aimed at the non-stationary and nonlinear characteristics of wind turbine vibration signals, a novel fault diagnosis method based on integral extension load mean decomposition multiscale entropy and least squares support vector machine was proposed in this paper. At first, the raw vibration signals monitored from the wind turbine were divided into groups for the pre-process. Then the signals were processed in groups with integral extension load mean decomposition method and Product Functions were obtained. The characteristic parameters were achieved by multiscale entropy method of processing main Product Functions, which described the signal characteristics. Finally, the characteristic parameters were entered into least squares support vector machine, and least squares support vector machine was trained. Next the trained least squares support vector machine was tested and the pattern was classified. The method can not only extract characteristic parameters effectively, but also classify the fault type accurately. The effectiveness and availability of the proposed method were proved in the wind turbine measured data experiment.

Fault diagnosis method for spherical roller bearing of wind turbine based on variational mode decomposition and singular value decomposition

For the non-stationary characteristics of the vibration signal of wind turbine’s roller bearing in fault condition, a bearing fault diagnosis method based on variational mode decomposition (VMD) and singular value decomposition (SVD) is proposed. The VMD method is used to decompose wind turbine’s roller bearing’s fault vibration signal into several components. These components are regard as initial feature vector matrix. The singular value decomposition of the matrix is done. The obtained singular value is used as the extracted bearing fault feature vectors. The probabilistic neural network is used as pattern recognition classifier to determine the working state and fault type of wind turbine roller bearings. The result of case study showed that the proposed method can effectively identify the working state and fault type of wind turbine roller bearings.

Fault diagnosis of wind turbine bearing based on variational mode decomposition and Teager energy operator

Vibration signal of wind turbine has the non-linear and non-stationary characteristic, thus it is difficult to extract the fault feature. In this study, a novel method based on variational mode decomposition (VMD) and Teager energy operator (TEO) is proposed to diagnose the bearing faults of wind turbine. First, vibration signal is decomposed into several intrinsic mode function (IMF) components by means of VMD, which is a recently proposed signal decomposition method. Then, the most sensitive IMF component is selected according to kurtosis criterion. Moreover, TEO is applied to the most sensitive IMF in order to highlight impact signal. Finally, spectrum is obtained by applying Fourier transform to Teager energy of the selected IMF, thus extracting the fault feature to diagnose bearing fault. The effectiveness of the proposed method for fault diagnosis is validated by simulation and experimental signal analysis results, and comparison studies show its advantage over empirical mode decomposition and conventional spectrum analysis for wind turbine bearing fault diagnosis.

Vibration condition monitoring system for wind turbine bearings based on noise suppression with multi-point data fusion

A noise suppression method for feature frequency extraction that is supplemented with multi-point data fusion was investigated in consideration of issues involving wind turbine vibration signals subject to high noise disturbance. The difficulty of extracting early weak fault features was examined as well. First, a de-noising and feature extraction method that uses EMD-Correlation was developed by adding empirical mode decomposition (EMD) and autocorrelation de-noising to the wavelet package transform under the effects of white noise and short-term disturbance noise in wind turbine vibration signals. Second, an EMD-Correlation analysis model for feature frequency extraction supplemented with multi-point data fusion was established with reference to adaptive resonance theory-2 to highlight the feature frequency of a possible early weak fault. Third, the results obtained with the actual and simulated fault vibration signals of wind turbine bearing faults and the outcomes of comparing the different feature frequency extraction methods show that the proposed method of EMD-Correlation that is supplemented with multi-point data fusion can not only effectively reduce white noise and short-term disturbance noise but can also extract the feature frequency of early weak faults. Finally, prototype hardware and software were developed for a wind turbine condition monitoring system based on the aforementioned fault feature extraction algorithms and tested in an actual wind turbine generation system.

A novel wind turbine bearing fault diagnosis method based on Integral Extension LMD

This paper proposed a novel wind turbine ball bearing fault diagnosis method based on Integral Extension Local Mean Decomposition (IELMD). Wind turbine vibration signal has the characteristic of non-Gaussian and non-stationary. Some typical time–frequency analysis methods cannot achieve ideal effects. A new method named LMD can deal with non-stationary signals and can extract obvious features. The IELMD method is proposed based on the integral local waveform matching of the right and left side of the original signal, in order to suppress the end effect of LMD method itself. Firstly, all the extreme points are scanned out, in which the left three points and right three points are specially marked. Secondly, two characteristic waveforms and two matching waveforms are established at both the left and right side. Finally, both the left and right matching waveforms are extended to finish the LMD process. The wind turbine bearing fault diagnosis experimental can improve the efficiency and validity of the novel IELMD method.

Study on Unified Chaotic System-Based Wind Turbine Blade Fault Diagnostic System

At present, vibration signals are processed and analyzed mostly in the frequency domain. The spectrum clearly shows the signal structure and the specific characteristic frequency band is analyzed, but the number of calculations required is huge, resulting in delays. Therefore, this study uses the characteristics of a nonlinear system to load the complete vibration signal to the unified chaotic system, applying the dynamic error to analyze the wind turbine vibration signal, and adopting extenics theory for artificial intelligent fault diagnosis of the analysis signal. Hence, a fault diagnostor has been developed for wind turbine rotating blades. This study simulates three wind turbine blade states, namely stress rupture, screw loosening and blade loss, and validates the methods. The experimental results prove that the unified chaotic system used in this paper has a significant effect on vibration signal analysis. Thus, the operating conditions of wind turbines can be quickly known from this fault diagnostic system, and the maintenance schedule can be arranged before the faults worsen, making the management and implementation of wind turbines smoother, so as to reduce many unnecessary costs.

Research on Vibration Signal Feature Extraction Method to the Wind Turbine Generator

Aiming at the safe operation of the wind turbine, a feature extraction method of vibration signal based on the principle of blind source separation was proposed. Blind source and the current state of fault signal was separated and predicated by Using historical data of wind turbine vibration signal as the observation noise, and then fault diagnosis signal mechanical operation was analyzed, the HMM/SVM series fault diagnosis models structure was proposed. By calculating the matching degree of unknown signal and wind power equipment in the state using HMM, the features for SVM was formed to achieve the finally discriminant. The experimental results showed that the fault diagnosis method can precisely and effectively complete the wind power equipment, higher than pure HMM or SVM diagnostic accuracy.

Application of the state deterioration evolution based on bi-spectrum in wind turbine

Concerning the problem of large rotating machinery like wind turbine which runs in low speed and non-stationary state, this research mainly focuses on separating fault trend feature from non-fault feature and the method of state deterioration evolution based on bi-spectrum. Firstly, the experimental signal such as low-speed startup vibration signal of rotor test rig in the normal state and a plurality of unbalanced state have been collected. Bi-spectrum method is applied to extract fault feature which submerged in complex background noise. On the basis of bi-spectrum analysis, the fault feature evolutionary matrix is defined to represent the state of equipment deterioration. The eigenvalues of fault feature evolutionary matrix are computed and fitted to a normal distribution curve, from which the mean value and variance are taken as fault feature parameters. Fault feature parameters are verified effectively by experiments. Finally, depending on fault feature parameters, graphical representation of state deterioration evolution is established. It is beneficial to provide guidelines for equipment deterioration trend. This method is applied to analyze the real vibration signal of wind turbine with the type of WD646/600 KW, and actual equipment condition verified the effectiveness of the proposed method.

Research on Compression and Reconstruction of Wind Turbine Vibration Signal Based on Discrete Cosine Transform

The method of compression and reconstruction for wind turbine vibration signal based on Discrete Cosine Transform is presented. Two group wind turbine vibration signals are compressed and reconstructed by DCT. The compression ratio and the compression reconstruction accuracy are investigated. The relationship between compression ratio, NSRMSE and the number of the reserved data is clarified. The result facilitates the application of compression and reconstruction for wind turbine vibration signal based on DCT.

A new noise-controlled second-order enhanced stochastic resonance method with its application in wind turbine drivetrain fault diagnosis

Condition monitoring of a wind turbine is important to extend the wind turbine system's reliability and useful life. However, in many cases, to extract feature components becomes challenging and the applicability of information drops down due to the large amount of noise. Stochastic resonance (SR), used as a method of utilising noise to amplify weak signals in nonlinear systems, can detect weak signals overwhelmed in the noise. Therefore, a new noise-controlled second-order enhanced SR method based on the Morlet wavelet transform is proposed to extract fault feature for wind turbine vibration signals in the present study. The second-order SR method can obtain better denoising effect and higher signal-to-noise ratio (SNR) of resonance output by means of twice integral transform compared with the traditional SR method. Morlet wavelet transform can obtain finer frequency partitions and overcome the frequency aliasing compared with the classical wavelet transform. Therefore, through Morlet wavelet transform, the noise intensity of different scales can be adjusted to realize the resonance detection of weak periodic signal whatever it is a low-frequency signal or high-frequency signal. Thus the method is well-suited for enhancement of weak fault identification, whose effectiveness has been verified by the practical vibration signals carrying fault information. Finally, the proposed method has been applied to extract feature of the looseness fault of shaft coupling of wind turbine successfully.

Research on Wavelet Compression and Reconstruction of Wind Turbine Vibration Signal

The method of wavelet compression and reconstruction is studied. The wind turbine vibration signal is compressed and reconstructed by wavelet transformation. The compression and reconstruction of wind turbine vibration signal are achieved with different wavelet base. The compression ratio and the compression reconstruction accuracy are analyzed. The method and the result facilitate the choice of wavelet base for transmission of wind turbine vibration signal.

A new wind turbine fault diagnosis method based on the local mean decomposition

This paper proposed a novel wind turbine fault diagnosis method based on the local mean decomposition (LMD) technology. Wind energy is a renewable power source that produces no atmospheric pollution. The condition monitoring and fault diagnosis in wind turbine system are important in avoiding serious damage. Vibration analysis is a normal and useful technology in wind turbine condition monitoring and fault diagnosis. However, the relatively slow speed of the wind turbine components set a limitation in early fault diagnosis using vibration monitoring method. The traditional time-frequency analysis techniques have some drawbacks which make them not suitable for the nonlinear, non-Gaussian signal analysis. LMD is a new iterative approach to demodulate amplitude and frequency modulated signals, which is suitable for obtaining instantaneous frequencies in wind turbine condition monitoring and fault diagnosis. The experiment analysis of the wind turbine vibration signal proves the validity and availability of the new method.

State Characteristic Parameters Extraction Method of Wind Turbine Blades Based on the Length Fractal Dimension

According to characteristic of wind turbine vibration signal, base on the length fractal dimension principle, carry out quantitative description of length fractal characteristics to nonlinear signals on the fault generated by the wind turbine blades. The test and calculation results show that when the different wind turbine blades failures occur, the length fractal dimension value appear clear rules. Therefore, we can use the length fractal dimension effectively extract the fault character parameters of wind turbine blade.

Feature extraction method of wind turbine based on adaptive Morlet wavelet and SVD

Analyzing the vibration signals of wind turbine usually requires feature extraction. However, in many cases, to extract feature components becomes challenging and the applicability of information drops down due to the large amount of noise. In this paper, a new denoising method based on adaptive Morlet wavelet and singular value decomposition (SVD) is applied to feature extraction for wind turbine vibration signals. Modified Shannon wavelet entropy is utilized to optimize central frequency and bandwidth parameter of the Morlet wavelet so as to achieve optimal match with the impulsive components. The time-frequency resolution can be adapted to different signals of interest. Then, an improved matrix construction method is used to construct matrix of the wavelet coefficient, and the scale periodical exponential (SPE) spectrum is obtained by SVD for selecting the appropriate transform scale. Experimental analysis and application into signal denoising indicate that the proposed method has better denoising performance than other wavelet transforms. The results of the experimental analysis in rolling bearing and the application in planetary gearbox show that the proposed method is an effective approach to detecting the impulsive feature components hidden in vibration signals and performs well for wind turbine fault diagnosis.