Wind Turbine Planetary Gearbox Fault Diagnosis via Proportion-Extracting Synchrosqueezing Chirplet Transform

Wind turbine planetary gearboxes usually work under time-varying conditions,leading to nonstationary vibration signals.These signals often consist of multiple time-varying components with close instantaneous frequencies.Therefore,high-quality time-frequency analysis(TFA)is needed to extract the time-frequency feature from such nonstationary signals for fault diagnosis.However,it is difficult to obtain high-quality timefrequency representations(TFRs)through conventional TFA methods due to low resolution and time-frequency blurs.To address this issue,we propose a new TFA method termed the proportion-extracting synchrosqueezing chirplet transform(PESCT).Firstly,the proportion-extracting chirplet transform is employed to generate highresolution underlying TFRs.Then,the energy concentration of the underlying TFRs is enhanced via the synchrosqueezing transform.Finally,wind turbine planetary gearbox fault can be diagnosed by analysis of the dominant time-varying components revealed by the concentrated TFRs with high resolution.The proposed PESCT is suitable for achieving high-quality TFRs for complicated nonstationary signals.Numerical and experimental analyses validate the effectiveness of the PESCT in characterizing the nonstationary signals from wind turbine planetary gearboxes.

An Improved Second-Order Multisynchrosqueezing Transform for the Analysis of Nonstationary Signals

Second-order multisynchrosqueezing transform(SMSST),an effective tool for the analysis of nonstationary signals,can significantly improve the time-frequency resolution of a nonstationary signal.Though the noise energy in the signal can also be enhanced in the transform which can largely affect the characteristic frequency component identification for an accurate fault diagnostic.An improved algorithm termed as an improved second-order multisynchrosqueezing transform(ISMSST)is then proposed in this study to alleviate the problem of noise interference in the analysis of nonstationary signals.In the study,the time-frequency(TF)distribution of a nonstationary signal is calculated first using SMSST,and then aδfunction is constructed based on a newly proposed time-frequency operator(TFO)which is then substituted back into SMSST to produce a noisefree time frequency result.The effectiveness of the technique is validated by comparing the TF results obtained using the proposed algorithm and those using other TFA techniques in the analysis of a simulated signal and an experimental data.The result shows that the current technique can render the most accurate TFA result within the TFA techniques employed in this study.

基于Synchrosqueezing小波变换的谐波和间谐波检测方法

非线性电力元件的广泛使用使电力系统的谐波和间谐波污染越来越严重。为准确计算谐波和间谐波的参数特征,以有效克服噪声影响,提出基于Synchrosqueezing小波变换的谐波和间谐波的一种检测方法。首先对电力系统信号进行连续小波变换;然后确定同步挤压阈值,对连续小波变换结果进行同步挤压,并利用同步挤压结果计算电力信号主频率;最后,设置提取频率区间,将电力信号分解为一组内蕴模态类函数分量(IMT),并结合Hilbert变换及最小二乘拟合,精确计算噪声背景下谐波和间谐波的幅值与频率。通过模拟信号和实测信号对所提方法有效性进行了分析,实验结果表明,与Prony和HHT方法相比,本文方法通过同步挤压有效抑制了噪声干扰,谐波和间谐波的检测精度有较好的提高。

High impedance fault detection in distribution network based on S-transform and average singular entropy

When a high impedance fault(HIF)occurs in a distribution network,the detection efficiency of traditional protection devices is strongly limited by the weak fault information.In this study,a method based on S-transform(ST)and average singular entropy(ASE)is proposed to identify HIFs.First,a wavelet packet transform(WPT)was applied to extract the feature frequency band.Thereafter,the ST was investigated in each half cycle.Afterwards,the obtained time-frequency matrix was denoised by singular value decomposition(SVD),followed by the calculation of the ASE index.Finally,an appropriate threshold was selected to detect the HIFs.The advantages of this method are the ability of fine band division,adaptive time-frequency transformation,and quantitative expression of signal complexity.The performance of the proposed method was verified by simulated and field data,and further analysis revealed that it could still achieve good results under different conditions.

Fault diagnosis method of rolling bearing based onthreshold denoising synchrosqueezing transform and CNN

The rolling bearing vibration signal is non-stationary and is easily disturbed by background noise,so it is difficult to accurately diagnose bearing faults.A fault diagnosis method of rolling bearing based on the time-frequency threshold denoising synchrosqueezing transform(TDSST)and convolutional neural network(CNN)is proposed.Since the traditional methods of wavelet threshold denoising and wavelet adjacent coefficient denoising are greatly affected by the estimation accuracy of noise variance,a time-frequency denoising method based on the STFT spectral correlation coefficient threshold optimization is adopted,which is combined with a synchrosqueezing transform.The ability of the TDSST to reduce noise and improve time-frequency resolution was verified by simulated impact fault signals of rolling bearings.Finally,the CNN is utilized to diagnose the time-frequency diagrams obtained by the TDSST.The diagnostic results of the rolling bearing experimental data show that the proposed method can effectively improve the accuracy of diagnosis.When the SNR of the bearing signal is larger than 0 dB,the accuracy is over 95%,even when the SNR reduces to-4 dB,the accuracy is still around 80%.Moreover,the standard deviation of multiple test results is small,which means that the method has good robustness.

Spectrum Sensing and AM-FM Decomposition through Synchrosqueezing

In this paper we have accomplished one of the tasks of cognitive radio i.e. dynamic spectrum sensing by using wavelet based Synchrosqueezing transform [1], a novel technique, which was proposed to analyze a signal in time-frequency plane. The distinctive feature of this transform compared to other techniques is that it enables us to decompose amplitude and frequency modulated signals and allows individual reconstruction of these components. The objective is also to separate the occupied band into amplitude modulated and frequency modulated bands.

Recognition of Hybrid PQ Disturbances Based on Multi-Resolution S-Transform and Decision Tree

Aiming at the problems of multiple types of power quality composite disturbances,strong feature correlation and high recognition error rate,a method of power quality composite disturbances identification based on multiresolution S-transform and decision tree was proposed.Firstly,according to IEEE standard,the signal models of seven single power quality disturbances and 17 combined power quality disturbances are given,and the disturbance waveform samples are generated in batches.Then,in order to improve the recognition accuracy,the adjustment factor is introduced to obtain the controllable time-frequency resolution through multi-resolution S-transform time-frequency domain analysis.On this basis,five disturbance time-frequency domain features are extracted,which quantitatively reflect the characteristics of the analyzed power quality disturbance signal,which is less than the traditional method based on S-transform.Finally,three classifiers such as K-nearest neighbor,support vector machine and decision tree algorithm are used to effectively complete the identification of power quality composite disturbances.Simulation results showthat the classification accuracy of decision tree algorithmis higher than that of K-nearest neighbor and support vector machine.Finally,the proposed method is compared with other commonly used recognition algorithms.Experimental results show that the proposedmethod is effective in terms of detection accuracy,especially for combined PQ interference.

基于振动信号和同步压缩小波变换的电动机测速方法及实验分析

针对部分工业现场电动机无法安装测速传感器,提出一种基于振动信号和同步压缩小波变换的电动机测速方法。对电动机振动信号进行Hilbert变换得到包络信号,解调出其中与转速相关的振动分量;利用同步压缩小波变换分析计算出包络信号的时频图;引入脊线的四分位距和方差对代价函数法进行改进,并利用该方法提取基频振动分量的时频脊线,得到电动机的转速曲线,达到无转速计测速的目的。实验与仿真分析表明,所提方法无论在稳态还是在非稳态工况下都能准确检测出电动机转速,且误差不超过5%。该方法的研究与实践过程可加深学生对调制解调、时频分析等理论知识的理解,提高理论联系实际的能力。

基于多元同步压缩变换的电力系统强迫振荡源定位

为实现电力系统强迫振荡源的快速、准确定位,该文提出一种基于多元同步压缩变换(multivariate synchrosqueezingtransform,MSST)的电力系统强迫振荡源定位方法。该方法首先利用电力系统的广域量测信息构建发电机的多通道量测信息矩阵,采用MSST对多通道量测信息矩阵同步分解得到对应的三维MSST系数矩阵;然后通过能量权重筛选出表征强迫振荡模式的MSST系数矩阵;进一步,构建基于MSST的发电机强迫振荡耗散能量计算模型,通过筛选出的表征发电机强迫振荡模式的MSST系数矩阵计算各发电机的耗散能量;然后,依据所提的强迫振荡源判据,定位出系统的强迫振荡源;最后,通过WECC-179节点测试系统仿真数据和辽宁电网PMU实测数据验证了所提方法的准确性和有效性。

电力系统强迫振荡源定位的时-频域耗散能量流方法

准确定位强迫振荡源对电力系统的安全稳定运行意义重大。然而,由于强迫振荡模式的可观性和振荡时变特征,传统方法难以从多通道量测信息中有效提取振荡分量,从而降低了基于耗散能量流的强迫振荡源定位方法的定位精度。为此,提出了一种基于耗散能量流的电力系统强迫振荡源时-频域定位方法。首先,根据节点各量测通道间信息相关性,利用同步压缩短时傅里叶变换处理节点多通道量测信息,构建节点统一时-频系数矩阵;然后,根据强迫振荡分量的能量特性,利用时-频域能量筛选并同步提取时-频系数矩阵中的时-频域强迫振荡分量;进一步,根据测量信息的时-频域特性,在传统时域强迫振荡耗散能量流计算模型的基础上推导出基于同步压缩短时傅里叶变换的时-频域耗散能量流计算模型,并根据系统强迫振荡期间的时-频域耗散能量流能量特性定位强迫振荡源;最后,将所提方法应用于WECC 179节点测试系统、WECC 240节点测试系统的仿真振荡场景以及美国New England的实际振荡事件,所得结果表明所提时-频域定位方法可快速、精准定位强迫振荡源。

基于同步挤压变换和深度迁移学习的GIS隔离开关故障诊断

为实现气体绝缘组合电器(GIS)隔离开关机械故障的智能诊断,基于GIS隔离开关分合闸过程中的振动信号,提出了基于深度迁移学习的GIS隔离开关机械故障诊断方法。首先应用二阶同步挤压傅里叶变换(FSST2)获取GIS隔离开关振动信号的时频分布,然后基于深度迁移学习的思想构建预训练模型并进行优化,建立了GIS隔离开关机械故障智能诊断模型。对某GR角型GIS隔离开关正常和典型机械故障状态下的振动信号的分析结果表明,基于FSST2得到的GIS隔离开关振动信号时频表示具有较好的能量聚集性,所建立的GIS隔离开关机械故障智能诊断模型识别准确率高且模型复杂度低,可用于GIS隔离开关机械故障的高效诊断。

基于同步压缩广义S变换的电力系统次/超同步振荡检测

为实现电力系统次/超同步振荡的快速、准确辨识,提出了一种基于同步压缩广义S变换(synchrosqueezing generalized S transform, SSGST)和改进稀疏时域法(improved sparse time domain method,ISTD)结合的次/超同步振荡辨识方法。该方法首先利用能量比函数对电力系统广域量测信息实时检测,当检测到信号能量发生突变时,利用SSGST对检测到的振荡信号分解得到相应的SSGST时频系数矩阵;然后通过改进的脊线提取方法在时频域实现对各振荡分量的最优轨迹搜索;进一步,结合最优轨迹时频索引重构各振荡分量的时域分量,并利用ISTD辨识方法计算出各振荡分量的频率和阻尼比系数;最后,通过自合成模拟信号、双馈风电场经串补并网系统仿真信号和某实际风电场实测数据验证了所提方法的准确性和有效性。

同步挤压S变换与τ⁃p变换联合的微地震信号消噪方法

微地震监测是指导页岩气开采水力压裂作业和评价压裂效果的常用手段。地面监测所采集的微地震信号能量弱、信噪比低,微地震事件识别困难,严重影响定位的准确性。针对低信噪比地面微地震监测资料,联合使用同步挤压S变换、谱分解和τ⁃p变换,提出一种新的消噪方法。首先对监测资料进行时差校正,将微地震信号的同相轴校平;之后使用同步挤压S变换对校平后的资料进行谱分解获取单频切片;再对每个单频切片进行τ⁃p变换,并根据τ⁃p变换的结果获取微地震信号位置;最后根据信号的位置在时频域完成消噪。实际低信噪比地面微地震监测数据的处理结果表明,新方法可以获得理想的消噪结果。

基于AVMD-ASWT-PCNN的滚动轴承故障识别方法

针对传统方法直接舍弃高频分量导致信号降噪不充分和信号在时、频域表征效果不好的问题,提出一种基于自适应变分模态分解融合自适应同步压缩小波变换(AVMD-ASWT)的少噪声时频图像生成方法,在此基础上结合动态惯性权重粒子群优化卷积神经网络(PCNN)实现滚动轴承故障的识别。采用AVMD-ASWT算法对轴承振动信号进行二次处理,同时引入互信息熵-相关系数准则,获得高分辨率的少噪声时频图像。将少噪声时频图像作为网络模型的输入进行故障识别,同时采用动态惯性权重粒子群优化算法(PSO)对卷积神经网络模型(CNN)参数进行优化,可解决模型结构难以确定的问题,模型识别正确率和识别速度均有明显提升。工程实例表明:运用AVMD-ASWT方法得到的时频图像具有更高的分辨率,显著降低了信号中噪声的影响,且提出的PCNN模型的故障识别正确率达99%以上。

结合SLMSST和DO提取时变结构瞬时频率

为提升局部最大同步挤压变换估算瞬时频率的精度,本文结合2阶局部最大同步挤压变换(Second-order Local Maximum Synchrosqueezing Transform,SLMSST)和动态规划(Dynamic Optimization,DO)方法提出一种识别时变结构瞬时频率的新方法。该方法首先通过引入2阶瞬时振幅与相位得到精度更高的2阶瞬时频率估算位置。其次,搜索频率方向上时频系数的局部最大值所对应的2阶瞬时频率位置并根据这些位置对时频系数进行重排,从而得到2阶局部最大同步挤压变换后的瞬时频带。再次,运用动态规划法在限定频带范围内提取瞬时频率曲线。通过一组数值算例和一个时变拉索试验验证了所提新方法的有效性,研究结果表明:相比既有的局部最大同步挤压变换算法,2阶局部最大同步挤压变换和动态规划的联合算法不仅具有较好的精度,而且具有更好的时频聚集性。

Fractional S-transform-part 2:Application to reservoir prediction and fluid identification

The fractional S-transform （FRST） has good time-frequency focusing ability. The FRST can identify geological features by rotating the fractional Fourier transform frequency （FRFTfr） axis. Different seismic signals have different optimal fractional parameters which is not conducive to multichannel seismic data processing. Thus, we first decompose the common-frequency sections by the FRST and then we analyze the low-frequency shadow. Second, the combination of the FRST and blind-source separation is used to obtain the independent spectra of the various geological features. The seismic data interpretation improves without requiring to estimating the optimal fractional parameters. The top and bottom of a limestone reservoir can be clearly recognized on the common-frequency section, thus enhancing the vertical resolution of the analysis of the low-frequency shadows compared with traditional ST. Simulations suggest that the proposed method separates the independent frequency information in the time-fractional-frequency domain. We used field seismic and well data to verify the proposed method.

基于局部最大值同步压缩广义S变换识别结构瞬时频率

为提升土木工程结构的瞬时频率识别精度,本文提出局部最大值同步压缩广义S变换方法.该方法首先对响应信号进行广义S变换;然后针对时频系数求取关于时间的偏导,进而估算信号的瞬时频率;最后采用局部最大值同步压缩算子来对时频系数进行重排,以此得到响应信号的瞬时频带.通过一组两层剪切框架模型的数值算例和一个七层钢筋混凝土剪力墙振动台试验验证该方法的有效性和可行性.研究结果表明,与既有的局部最大值同步压缩变换相比,本文方法不仅提升了瞬时频率的识别精度,而且改善了响应信号瞬时频带的时频聚集性.

最大系数多重同步挤压变换识别结构瞬时频率

针对多重同步挤压变换及其改进算法存在的未重排点问题,提出了一种基于最大系数的多重同步挤压变换(maximum coefficient based multi-synchrosqueezing transform,简称MCMSST)方法来识别时变结构非平稳响应信号的瞬时频率(instantaneous frequency,简称IF)。首先,引入傅里叶频谱来辅助多分量响应信号选取截止频率;其次,对响应信号进行短时傅里叶变换(short time fourier transform,简称STFT),针对短时傅里叶变换系数求取针对时间的偏导,从而获得估算的瞬时频率;然后,在对瞬时频率的估算值进行多次迭代的基础上,仅保留时频系数模值最大处所对应的估算瞬时频率,并将其余位置的瞬时频率值归零;最后,对时频系数模值最大处所对应的瞬时频率进行时频重排即可得到细化后的瞬时频带。由于基于MCMSST方法提取的是瞬时频带,故采用时频系数模极大值法在限定的频带范围内提取瞬时频率曲线。通过2组数值算例和1个铝合金悬臂梁质量突变试验,验证了所提方法的有效性。研究结果表明,MCMSST方法不仅彻底解决了未重排点问题,而且提高了瞬时频率的识别精度和抗噪能力。

同步挤压小波变换在矿山微震监测系统中的应用

微震监测技术是一种保证矿山安全、高效、可持续发展的重要地压监测手段。矿山传感器获取的微震数据容易受到宽频带非平稳噪声的影响,导致微震监测缺乏可用的高质量数据。本文介绍了一种新的频谱分析方法——同步挤压小波变换,它提供了一种将数据同时分解到时域和频域的方法,且比小波变换等方法的时频分辨率更高,可以在时频谱上更为清晰地展示微震信号。同时,本文还比较了Morlet小波基和Bump小波基对微震波形的影响程度,并通过现场微震监测波形验证了Morlet小波基的相对可靠性。采用基于噪声水平的硬阈值滤波方法对其进行分别处理,结果表明,基于Morlet小波基的同步挤压小波变换在从原始数据中提取微震信号方面具有更大的实用价值,能够有效地提高信号的信噪比。

基于改进多重同步挤压样条Chirplet变换的结构瞬时频率识别

为提高样条Chirplet变换的结构频率识别精度,提出了一种基于改进多重同步挤压样条Chirplet变换(improved multi-synchrosqueezing spline-kernelled Chirplet transform, IMSSSCT)的瞬时频率识别方法。首先通过引入三参数高斯窗函数和能量集中度原理对样条Chirplet变换进行改进,然后结合多重同步挤压算法进一步提高改进样条Chirplet变换(improved spline-kernelled Chirplet transform, ISCT)的时频能量聚集度。通过单自由度Duffing系统和三层剪切框架结构的数值仿真及悬臂梁结构试验对所提方法进行了验证。研究结果表明,IMSSSCT能有效识别非线性结构和时变结构的瞬时频率。