旋转机械状态监测中提高FFT线性谱精度的一种方法

针对旋转机械状态监测中振动信号一般以离散频率成分为其主要频率构成的特点，讨论一种改善ＦＦＴ线性谱精度的方法。应用该方法对代表某谐和分量的谱峰参数进行适当修正，能获得该分量较精确的频率、幅值和相位。其中相位分析误差小于５°     

数字滤波对瞬态冲击脉冲信号的影响

讨论了冲击脉冲类型的瞬态信号用数字信号方法处理时,数字滤波对处理结果的幅值及波形将产生影响。并对这种影响进行了研讨。经用矩形窗和Hanning窗对半周正弦脉冲信号进行数字滤波可知,采用具有过渡带的窗并提高采样频率可减小滤波后结果的幅值误差和波形失真。还对某型火箭发射时燃气流冲击载荷的测试结果进行了分析。     

消除负频率影响的频谱校正

采用比值法校正低频成分的参数精度比较低，其原因是模型中忽略了负频率成分的贡献。为解决这一问题，基于包含负频率贡献的频谱公式，建立新的校正方案，给出矩形窗和汉宁窗对应的校正公式，采用数值仿真对校正公式进行考核。结果表明，提出的校正公式正确，尤其是汉宁窗的频率定位精度、幅值相对精度和相位精度几乎到双精度运算的精度下限。     

计及负频率的极高频信号离散频谱校正新方法

对于机械振动与故障诊断等领域中常见的极高频(接近奈奎斯特频率)信号,传统的离散频谱校正方法存在着较大误差,负频率成分干涉严重是影响其频谱分析精度的重要因素。为提高极高频信号的频谱分析与校正精度,给出了一种计及负频率影响的离散频谱校正新方法。该方法基于Blackman窗,依据离散频谱的周期性并利用局部谱峰附近的3条谱线,建立包含正负频率贡献的离散频谱校正模型,通过对模型的求解获得频率、幅值和相位校正公式。采用频段内扫描的方式对频谱校正公式进行了仿真验证,结果表明所提方法可有效降低负频率成分对极高频信号频谱的干涉影响,提高其频率、幅值和相位校正精度。     

自于自回归（AR）谱估计的高精度超声晶粒尺寸估计技术的研究

由于ＤＦＴ谱估计方法的谱“泄漏”，导致基于这种谱的分析超声晶粒尺寸估计结果误差较大。本文采用自回归谱估计方法，克服了传统ＤＦＴ方法的缺点，精确地重构材料特征函数功率谱，提高了晶粒尺寸估计精度，尤其在材料晶粒局部特性定量评价方面，本文方法具有很高的实用价值。     

基于双窗全相位FFT双谱线校正电力谐波分析

基于快速傅里叶变换的电力谐波分析在非同步采样情况下存在频谱泄露,影响测量精度。为抑制频谱泄露对谐波分析的影响,分析对比了双窗全相位FFT与传统加窗FFT在抑制频谱泄露方面的特点,并通过计算证明了双窗全相位FFT在电力谐波分析中的独特优势。针对全相位频谱分析通常采用的时移相位差法存在计算量大、实时性差及存在相位模糊现象等不足,提出了一种基于双窗全相位FFT双谱线校正的谐波分析算法,利用基波频点附近的两条谱线幅值计算基波频率,进而推导了简洁实用的谐波幅值校正公式。该算法实现方便,且实时性优于相位差法。通过与传统加Hanning窗、Nuttall窗插值FFT的仿真对比验证了新算法具有更高精度,基于该算法的实验结果也验证了算法的有效性。     

基于遗传算法的信号识别技术

将遗传算法应用于识别气浮式动平衡测量装置准静态信号的参数 ,提出了二次磨光法的数据预处理方法。首先采用二次磨光法消除坏点并进行快速傅立叶变换以确定信号的频率成分 ,并设定各频率成分信号的幅值和初相位的取值范围 ,然后以实测信号与识别结果的差值定义适应度 ,最后经过基本遗传操作和迭代得出最优解。仿真结果和试验实例表明 ,提出的方法精度很高 ,幅值识别误差不超过 1% ,初相位误差小于 2° ,满足了实际工程需求。     

基于遗传算法的信号识别技术

将遗传算法应用于识别气浮式动平衡测量装置准静态信号的参数 ,提出了二次磨光法的数据预处理方法。首先采用二次磨光法消除坏点并进行快速傅利叶变换以确定信号的频率成分 ,并设定各频率成分信号的幅值和初相位的取值范围 ,然后以实测信号与识别结果的差值定义适应度 ,最后经过基本遗传操作和迭代得出最优解。仿真结果和试验实例表明 ,该方法精度很高 ,幅值识别误差不超过 1 %,初相位误差小于 2°,满足了实际工程需求。     

基于双窗全相位FFT双谱线校正的电力谐波分析

基于快速傅里叶变换的电力谐波分析在非同步采样情况下存在频谱泄露,影响测量精度。为抑制频谱泄露对谐波分析的影响,分析对比了双窗全相位FFT与传统加窗FFT在抑制频谱泄露方面的特点,并通过计算证明了双窗全相位FFT在电力谐波分析中的独特优势。针对全相位频谱分析通常采用的时移相位差法存在计算量大、实时性差及存在相位模糊现象等不足,提出了一种基于双窗全相位FFT双谱线校正的谐波分析算法,利用基波频点附近的两条谱线幅值计算基波频率,进而推导了简洁实用的谐波幅值校正公式。该算法实现方便,且实时性优于相位差法。通过与传统加Hanning窗、Nuttall窗插值FFT的仿真对比验证了新算法具有更高精度,基于该算法的实验结果也验证了算法的有效性。     

“ZOOMBDFT”法高精度求系统阻尼比的探讨

提出了用“ＺＯＯＭＢＤＦＴ”（聚焦细化大容量数据谱分析）法高精度求系统阻尼比的途径和计算实例，文内指出了国内外常规１Ｋ字节ＦＦＴ计算系统阻尼比的不合理性，指出了采样时间长度Ｔ是影响频域用半功率带宽法求阻尼比精度的关键因子，并提出了保证阻尼计算精度的两个判别准则     

Better narrowband spectral measurements

Spectrum analysis based on traditional FFT techniques is a widely available, commonly understood and general purpose tool for use with mechanical systems. In applications such as system identification, modal analysis and rotating machinery signature analysis, relevant information content in signals is greatest near narrowband phenomena in the frequency domain. Effort should be focused on reducing bias error in these regions in preference to reducing variance error. Time aliasing methods of spectrum analysis can result in superior bias error reduction, producing an accurately sampled spectral estimate and permitting window function characteristics to be decoupled from the DFT resolution, or frequency line spacing. Windows created with time aliasing can be optimized by adjusting the amount of time or frequency localization, according to the type of spectrum being measured. Assuming conventional criteria, time aliased windows can achieve better mainlobe and sidelobe characteristics than other windows. An innovative approach, appropriate with continuous spectra, is to use the window as a discrete sampling function in frequency to eliminate bias error inherent in the effective convolution between window and signal transform. Both approaches have been successfully implemented and results are presented using time aliased spectral estimates showing more accurate measurements of narrowband phenomena when used in applications of rotating machinery vibration and system identification.     

Estimation of frequency components in stator current for the detection of broken rotor bars in induction machines

The limitation of data window length in induction machine broken rotor bar diagnostics is a real challenge in practice. Sideband frequencies which are used as broken rotor bar indicators are very close to the fundamental frequency and have low magnitude. Traditional spectral analysis approach such as Discrete Fourier Transform (DFT) can be inaccurate in these conditions due to its inherent drawbacks such as the requirement of long data window for high resolution and the side lobe leakage in frequency domain. In this paper, a high-resolution spectral analysis technique, Prony Analysis (PA), is proposed for broken rotor bar detection in induction machines. The method is described and demonstrated in detail, validated by experimental data, and compared with DFT. Results clearly indicate the advantages of PA over DFT in terms of maintaining a high resolution with a much shorter window and a better frequency estimate accuracy with the same window length.     

基于信号延拓的采样信号频谱泄漏抑制

提出一种基于信号数据延拓的采样信号离散傅里叶变换频谱泄漏抑制方法,分析了经典谱估计方法中频谱泄漏产生的原因,得出含噪声周期信号的无泄漏条件,进而提出频谱泄漏抑制方法:对采样信号作周期估计和整数周期数据延拓处理,使DFT计算窗内信号的边界连续或近似连续,以达到抑制频谱泄漏的目的。对通信、电力系统中常出现的加性高斯白噪声恶化信号的仿真表明:该方法可以显著减少频谱能量泄漏,获得更高质量的频谱。     

THE AUTOCOHERENT SPECTRUM: A USEFUL SPECTRAL ESTIMATOR FOR VIBRATION ANALYSIS OF ROTATING MACHINERY ACCURATE ESTIMATION AND CANCELLATION OF PURE TONES

This paper describes a new spectral estimator based on the Fourier transform, the autocoherent spectrum (ACS), which enables the extraction of the pure sinusoidal components with a greater sensitivity and an improved frequency resolution. It's statistical properties are studied, and it is shown how to remove pure tones in a spectrum, while keeping it's broadband components. After detection of lines, their exact frequencies are computed by means of the autocoherent spectrum phase. Then, the amplitude and phase associated to each pure tone are determined from the Fourier transform of each sequence according to the exact frequency calculated and the weighing window used. The removal of the pure tones is then achieved by correcting the Fourier channels containing them, and also some neighbouring channels which are effected by the side-effects of the window used.     

PHASE DIFFERENCE CORRECTION METHOD FOR PHASE AND FREQUENCY IN SPECTRAL ANALYSIS

A new method, phase difference corrections method is developed to correct the frequency and phase of spectrum peak. The continuous time-domain signal is separated into two segments and fast Fourier translation (FFT) is carried out for them, respectively. The frequency and phase are corrected using the phase difference of corresponding discrete spectral lines. Furthermore, the amplitude can also be rectified using the formula of window function spectrum. This method, with good adaptability, high speed and accuracy, is theoretically simple. It can resolve the frequency by means of phase difference directly without the formula of window function. Simulation shows that the single-component frequency, phase and amplitude of theoretical signal can be corrected satisfactorily, with frequency error less than 0.0002 frequency resolution, phase 0.1° and amplitude 0.0002. If the signal involves noise, the mean corrected errors are less than 0.001 frequency resolution, 1° for phase, and 0.01 for amplitude, respectively, and the maximum corrected errors of one segment are less than 0.01 frequency resolution, 1° and 0.03, respectively.     

短时AR谱分析及在内燃机故障诊断中的应用

针对利用传统短时Fourier变换(STFT)进行时频分析时不可能同时得到任意高时域分辨率和频域分辨率的问题,提出对传统的短时Fourier变换进行改进,在短时Fourier变换的计算中,利用自回归(AR)谱估计代替离散Fourier变换(DFT),得到了一种基于AR谱估计的短时AR谱分析方法。应用该方法对内燃机气阀机构的故障进行了分析,结果表明短时AR谱分析的估计性能大大改善,能够较好地给出信号的时频分布表示。     

信号频谱的误差分析及MATLAB实现

介绍了利用DFT分析信号频谱的基本流程,重点阐述了利用DFT分析信号频谱时产生误差的原因,即存在的频谱混叠、频谱泄漏及栅栏效应,并提出了相应的改进方法。实例列举了MATLAB环境下DFT分析信号频率分辨率的影响因素,通过与理论分析的对比,解释了提高频率分辨率与计算分辨率的方法。     

相位式激光测距谱分析鉴相的无偏改进

针对相位式激光测距中的鉴相误差,建立了谱分析鉴相的误差模型,提出了在数据预处理中引入希尔伯特变换来消除相位差估计偏差的测量方法.分析了传统谱分析鉴相的偏差和方差,指出这些测量偏差受初相位的影响,在高速测距中不容忽略.提出了一种无偏改进方法,通过窗函数法设计正弦信号的简易希尔伯特变换器,将离散傅里叶变换的对象转换为解析信号,在仅增加4次加减法运算和2次移位操作的情况下,实现了近似无偏谱分析鉴相.仿真分析和实验验证结果表明,鉴相均值与真实相位差相同;当信噪比为40 dB时,每秒百万次高速鉴相的误差为0.1°;当调制频率为100 MHz时,测距精度达到0.4 mm.实验表明,将希尔伯特变换应用于谱分析鉴相,可实现高准确度相位差测量,并可应用于高速相位式激光测距.     

On asymmetric analysis windows for detection of closely spaced signal components

Discrimination capacity, i.e. the capacity to resolve signal components, is probably one of the most important properties of the analysis windows. For components separated by more than a DFT bin the discrimination capacity is determined by the bandwidth and shape of the window's spectral main lobe and the side lobe fall-off rate. The new classical paper of Harris is a milestone where a plethora of windows is discussed and compared through the aforementioned parameters. However, those parameters fail to describe situations given by components clustered within a DFT bin bandwidth. In those cases the discrimination capacity is almost completely determined by the spectral phase of the analysis window, while the shape of the window's magnitude spectrum has little effect. This paper is devoted to a specific class of asymmetric analysis windows capable of resolving components separated by less than a DFT bin. Those windows have recently been proposed and analytically discussed. However, a number of important properties have not been tackled. Herein, the asymmetric analysis windows are thoroughly described and compared to the classical analysis windows in various aspects. The main benefits of the proposed windows are extended discrimination capacity, robustness against additive noise and simple generation. These properties make them very good candidates for narrow-band spectrum analysis of mechanical systems. Results for some real-world examples are presented, showing the capability of the new windows to resolve closely spaced vibration modes when there is a lack of spectral resolution due to the finite register length.     

Broken Rotor Bar Fault Detection of Induction Motors Using a Joint Algorithm of Trust Region and Modified Bare-bones Particle Swarm Optimization

A precise detection of the fault feature parameter of motor current is a new research hotspot in the broken rotor bar(BRB) fault diagnosis of induction motors. Discrete Fourier transform(DFT) is the most popular technique in this field, owing to low computation and easy realization. However, its accuracy is often limited by the data window length, spectral leakage, fence e ect, etc. Therefore, a new detection method based on a global optimization algorithm is proposed. First, a BRB fault current model and a residual error function are designed to transform the fault parameter detection problem into a nonlinear least-square problem. Because this optimization problem has a great number of local optima and needs to be resolved rapidly and accurately, a joint algorithm(called TR-MBPSO) based on a modified bare-bones particle swarm optimization(BPSO) and trust region(TR) is subsequently proposed. In the TR-MBPSO, a reinitialization strategy of inactive particle is introduced to the BPSO to enhance the swarm diversity and global search ability. Meanwhile, the TR is combined with the modified BPSO to improve convergence speed and accuracy. It also includes a global convergence analysis, whose result proves that the TR-MBPSO can converge to the global optimum with the probability of 1. Both simulations and experiments are conducted, and the results indicate that the proposed detection method not only has high accuracy of parameter estimation with short-time data window, e.g., the magnitude and frequency precision of the fault-related components reaches 10~(-4), but also overcomes the impacts of spectral leakage and non-integer-period sampling. The proposed research provides a new BRB detection method, which has enough precision to extract the parameters of the fault feature components.