基于比值线性化的高适应性光栅细分方法研究

电子细分是光栅位移传感器实现纳米级分辨率测量的关键,细分误差由细分方法和光栅信号质量共同决定。针对信号实时修正方法在小步距测量和一些复杂工况环境下的局限性,从提高细分方法对非理想光栅信号的适应性的角度出发,提出了基于信号比值线性化的新型电子细分方法,构建了2种实时补偿信号以进一步提高信号的线性度;详细阐述了所提细分方法的细分原理,分析了其在非理想光栅信号输入情况下的细分误差;实现了最高为0.003μm的理论细分精度和0.08μm的实际细分精度。数值仿真和对比实验结果表明,所提细分方法对非理想光栅信号的适应性明显优于常用的反正切细分法和正余弦绝对值相减细分法。     

用改进的小脑模型神经网络实现光栅信号连续细分

光栅测量系统常用硬件进行信号细分，存在细分数不高、硬件复杂等问题。为此，研究了一种神经网络细分方法，即利用一种改进的小脑模型神经网络的泛化能力实现光栅信号的连续细分。该小脑模型神经网络采用直接权地址映射技术，将光栅样本信号直接映射到联想存储器，对非样本信号经联想泛化即可实现连续细分。仿真实验结果表明，仅用少量样本即可达到很高的细分精度。     

两种莫尔条纹信号对光栅系统测量精度的影响

对两种莫尔条纹（光闸条纹和横向条纹）信号的谐波含量以及它们对光栅系统测量精度的影响进行了分析，并用实测数据加以比较，指出在光栅系统中，取横向莫尔条纹信号的谐波量小，正弦性好，细分误差小，在其它参数相同的情况下，可提高系统的测量精度2倍左右。     

炫耀光栅位移测量系统的精度试验

炫耀光栅位移测量系统采用尤学倍增与电子细分相結合的方法,使系统的分辨率达到0.4微米。与金属线纹尺比较测量,系统在300毫米内的极限误差为1.5微米。试验证明炫耀光栅位移测量系统具有较高的精度,如能进一步提高标尺光栅的刻划精度,并相应地提高分辨率,系统的实际精度还可以提高。     

光栅微振动测试传感技术的研究

提出了一种微弱地震波信号新的检波理论和方法．设计了一种测微振动传感器，采用光栅谐振子为敏感元件，将地震的机械信号转换成光调制信号——莫尔条纹，再用光电器件将其转换成数字化电信号．采用80倍细分技术，使传感器分辨力达到125nm；用MATLAB软件对PIC单片机记录的信号进行振动过程重构．结构中设计了横向限振片和光栅调节器，对光栅间隙、光栅间纵横向夹角进行调节，有效地限制了光栅间隙及角度的变化产生的误差．同时也校正了光栅的累积误差．有效地抑制了横向振动，提高了系统精度，同时可降低对该系统的横向限振要求，有利于结构设计．由于该传感器直接将地震的机械信号以脉冲数字信号输出，避免了器件精度不同对传感器输出数据质量的影响，简化了组装工艺．重点介绍了光栅数字地震检波器的结构设计、理论分析、辨向细分技术等．最后给出了测试数据和结论．     

用神经网络技术实现的高精度光栅测量装置

描述了一种用神经网络技术实现的高精度光栅测量装置。该装置采用BP神经网络对光栅信号进行细分，在仅用7个训练样本的情况下，细分精度可达0.18μm，使装置的分辨率得到很大提高，同时，简化了硬件设计，提高了系统的可靠性。     

非同步采样法的光栅纳米测量

光栅计量技术向纳米测量发展需要更高的光栅细分倍数.基于所研制的快速互补函数细分算法,细分精度即仅取决于测量启停时刻的光栅信号采样精度,选用市场上货源丰富的、低成本的扫描式多通道模数转换卡,即可实现光栅纳米测量所需的高倍数细分.在测量的启停时刻对光栅信号进行精确测量,在测量过程中对光栅信号进行快速跟踪测量.实验表明,当采用15位200 kHz的A/D转换卡对20 μm栅距的光栅信号进行非同步采样时,可以得到小于1 nm的测量分辨率及大于120 mm/s的测量速度.     

基于时间序列分析的光栅信号时域细分方法（英文）

针对光栅细分倍数和精度受到莫尔条纹信号质量限制的问题,提出一种基于时间序列分析的光栅信号时域细分方法.首先利用等空间间隔的栅距与采样时间周期的对应关系,将空间栅距转换为时间序列.然后利用时间序列分析中的自回归模型进行建模预测,并采用最大似然估计法计算模型参数.最后在预测时间内发送出代表位移的细分脉冲.实验表明,细分倍数为400或800时的角位移细分误差均为±2.4″,且光栅运动加速度及其变化率越小,细分误差越小.此细分方法一定程度上降低了对光栅信号质量的依赖,具有一定的应用价值.     

光栅位移测量系统的误差自修正方法研究

文中介绍的误差自修正方法是通过光栅位移测量系统中单片机对光栅传感器的多个零位信号进行计数,并根据测量值和系统设定值得到的误差函数自动进行误差修正.实验结果表明,该方法对光栅位移测量系统的误差既可自动进行有效的修正,又可提高系统的测量精度.     

动态测量光栅测长仪的位置误差

介绍了一种光栅测量系统动态位置精度评定的方法,以现有的光栅测长仪为研究对象,高精度HP5529A双频激光干涉仪作为校准系统.利用光栅测长仪编码器产生的A-quad-B脉冲信号触发激光干涉仪进行同步动态数据采集,由激光干涉仪软件来获取和显示位置误差数据,并绘制更真实详细的动态误差图.最后对动态误差实验结果进行分析和讨论.     

Pitch-modulated phase grating and its application to displacement encoder

Abstract

A binary phase grating with modulated pitch is investigated for a simple displacement encoder. The grating consists of a binary phase grating to eliminate the zeroth-order diffraction, and the pitch of the grating is modulated to compensate the higher harmonics of the encoder displacement signal. Therefore, an undistorted sinusoidal signal as a function of displacement is obtained by simply superimposing a conventional binary grating on the pitch-modulated phase gratings for any air gap between the gratings. The characteristics of the proposed gratings and the encoder signal are investigated by the Fresnel diffraction theory. The proposed grating has been fabricated lithographically, and the signal was examined experimentally. Considering these results, the proposed technique can suppress interpolation error and will be useful for an encoder in precision machining.     

大量程光栅位移测量累积误差的修正

提出了一种自动测量光栅栅距修正累积误差的方法。栅距测量是基于高阶累积量估计光栅传感器输出的两路莫尔条纹信号的时间延迟而得到的,该方法能够实现每个栅距的测量,通过对每个光栅栅距的误差进行修正来减少累积误差,为大量程高精度测量奠定了基础。实验采用长为500 mm的50线/mm的光栅传感器,该传感器包含栅线25 000条,实现栅距测量分辨力为3 nm,达到了纳米级测量。该方法抗干扰能力强,适合在生产现场应用。     

Theoretical error analysis of the sampling moiré method and phase compensation methodology for single-shot phase analysis

Recently, a rapid and accurate single-shot phase measurement technique called the sampling moiré method has been developed for small-displacement distribution measurements. In this study, the theoretical phase error of the sampling moiré method caused by linear intensity interpolation in the case of a mismatch between the sampling pitch and the original grating pitch is analyzed. The periodic phase error is proportional to the square of the spatial angular frequency of the moiré fringe. Moreover, an effective phase compensation methodology is developed to reduce the periodic phase error. Single-shot phase analysis can perform accurately even when the sampling pitch is not matched to the original grating pitch exactly. The primary simulation results demonstrate the effectiveness of the proposed phase compensation methodology.     

成像式光栅自准直测角方法研究

针对目前光电自准直仪难以做到既有高分辨力同时又有大的测量范围的不足,提出一种成像式光栅自准直测角方法。该方法将成像式光栅方法、自准直测角技术相结合,利用标尺光栅像和指示光栅形成的光栅副反映被测物体的角度变化。试验结果表明,该系统接收到的光栅信号稳定,满足光栅计数的要求,成像式光栅自准直测角方法原理可行。该方法避免了光栅副间隙带来的误差,具有大量程及较高的分辨力。     

Low energy impact damage monitoring of composites using dynamic strain signals from FBG sensors - Part I: Impact detection and localization

Fiber reinforced composite materials risk to suffer from subsurface, barely visible, damage induced by transverse relatively low energy impacts. This two-paper series presents a method for the localization of an impact and identification of an eventual damage using dynamic strain signals from fiber Bragg grating (FBG) sensors. In this paper, the localization method allowing to predict the impact position based on interpolation of a reference data set is developed and validated. The data utilized in the method are the arrival times of the asymmetric zero order Lamb waves at the different sensors. A high rate interrogation method based on intensity modulation of the Bragg wavelength shift is used to acquire the FBG signals. The localization method allows to predict the impact position with a good accuracy and therefore the inspection of the laminate can be limited to this region.     

一种钢琴全频段的基频识别方法

对钢琴全频段音符频率进行了研究,提出一种基于自相关和插值的钢琴基频识别方法。采用三电平中心削波的自相关计算得出函数值,用求最大值的方法选取代表基频整数倍周期的点数位置,运用抛物线插值进行位置修正,结合最小二乘计算得出基频估计,低中频的识别误差在±1音分内。对于高频音符,采用内插算法提高采样率以降低识别误差,使得识别结果准确可靠,满足精度要求。     

基于EMD算法的光栅莫尔条纹信号去噪方法研究

针对计量光栅莫尔条纹信号的质量问题,提出了基于经验模态分解（EMD）算法对非平稳光栅莫尔条纹信号模型的去噪方法。建立非平稳的光栅时变信号模型,利用EMD算法不需要定义滤波器参数的自适应性优点,对添加不同噪声的多组光栅信号模型进行了滤波分析的仿真实验,其信噪比和均方根误差两项指标优于均值滤波、小波阈值去噪方法。对两路正余弦理想信号添加高次谐波分量,通过对比EMD算法抑制高次谐波前后的李萨如图形,验证了该方法在去噪过程中对光栅莫尔条纹信号正弦性误差补偿的良好效果。     

Microwave measurements of effective dielectric constant ofsemiconductor waveguides via periodic-structure photoexcitation

In this paper a new method for measurements of the effective dielectric constant of layered or full-substrate semiconductor waveguides at microwave frequencies is presented. The light-induced generation of a photoconductivity grating within the excited waveguide section is utilized. Incoming swept-frequency signals produce a stop-band reflection spectrum which exhibits a dominant major-lobe peak value at the grating center frequency. The effective dielectric constant is calculated from a simple analytical expression by substituting the measured center frequency. Theoretical background, error estimation, and an example are presented     

Pound-Drever-Hall error signals for the length control of three-port grating coupled cavities

Gratings enable light coupling into an optical cavity without transmission through any substrate. This concept reduces light absorption and substrate heating and was suggested for light coupling into the arm cavities of future gravitational wave detectors. One particularly interesting approach is based on all-reflective gratings with low diffraction efficiencies and three diffraction orders (three ports). However, it was discovered that, generally, three-port grating coupled cavities show an asymmetric resonance profile that results in asymmetric and low quality Pound-Drever-Hall error signals for cavity length control. We experimentally demonstrate that this problem is solved by the detection of light at both reflection ports of the cavity and the postprocessing of the two demodulated electronic signals.