可见-红外波段光全散射法颗粒粒径测量范围的研究

在光全散射法颗粒粒径测量中,采用独立模式算法在可见-红外波段对粒径测量范围进行了深入的研究。通过对多种R-R分布函数反演结果进行分析,比较,以确定光全散射法颗粒粒径测量范围。同时在消光系数计算中,采用修正的消光系数代替原始的消光系数,以便得到更准确的粒径测量范围。仿真计算结果表明,在可见-红外光谱区,相对折射率m=1.235时的粒径测量范围为0.05～18 μm。在此区域内测量粒径分布,反演结果与真值基本吻合。随着颗粒相对折射率、波长范围的变化,粒径测量范围也随之发生改变。采用限制最小二乘算法能够最大限度地克服随机噪声对测量结果的影响,使测量的精度和可靠性明显提高。对测量的消光值加入±1%随机噪声时,颗粒粒径分布的反演均能得到满意的结果。     

Simultaneous Measurement of Particle Size and Particle Velocity by the Spatial Filtering Technique

The objective of this study was to compare the measuring results of a fiber‐optical probe based on a modified spatial filtering technique with given size distributions of different test powders and also with particle velocity values of laser Doppler measurements. Fiber‐optical spatial filtering velocimetry was modified by fiber‐optical spot scanning in order to determine simultaneously the size and the velocity of particles. The fiber‐optical probe system can be used as an in‐line measuring device for sizing of particles in different technical applications. Spherical test particles were narrow‐sized glass beads in the range 30–100 μm and irregularly shaped test particles were limestone particles in the range 10–600 μm. Particles were dispersed by a brush disperser and the measurements were carried out at a fixed position in a free particle‐laden air stream. Owing to the measurement of chord lengths and to the influence of diffraction and divergent angle, the probe results show differences from the given test particle sizes. Owing to the particle‐probe collisions, the mean velocity determined by the probe is smaller than the laser Doppler mean velocity.     

Fire Smoke Particle Size Measurement Based on the Multiwavelength and Multiangle Light Scattering Method

A multiwavelength and multiangle mathematical model （MWMA） is developed, and the corresponding influence factors of the model are analysed by the numerical calculation, Combined with measurement data of radiation flux and the curve in theoretical calculation, particle sizes of different materials can be obtained, it is found that smoke particle size gradually decreases from land plaster, cotton wick, beech wood, polyurethane, sandalwood, decahydronaphthalene, to N-heptane,     

基于模式搜索的光谱消光粒径分布反演算法的研究

发展快速有效的反演算法用于粒径分布的重建是颗粒测量领域重要的研究课题之一。在光谱消光法粒径测量中,提出将模式搜索算法用于独立模式下粒径分布的重建,同时引入Tikhonov平滑泛函构建算法的目标函数,为保证搜索过程的快速性和准确性,设计了关于初始点的优选策略。利用该算法测量国家标准颗粒体积平均直径的相对误差为3.14%,不超过国家标准物质局给出的±8%的范围,且粒径分布宽度合理,没有明显展宽和伪峰现象。与Phillips-Twomey方法和遗传算法的对比结果表明,在综合考虑反演精度和反演时间条件下,该算法具有明显优势,更适合于快速准确的现场测量,具有较好的应用前景。     

Study on the Projection Non-mode Algorithm of Laser Particle Size Measurement

１ＩｎｔｒｏｄｕｃｔｉｏｎＬａｓｅｒｌｉｇｈｔｓｃａｔｅｒｉｎｇｍｅａｓｕｒｅｍｅｎｔ，ｗｈｉｃｈｉｓｂａｓｅｄｏｎｔｈｅＭｉｅｓｃａｔｅｒｉｎｇａｎｄＦｒａｕｎｈｏｆｅｒｄｉｆｒａｃｔｉｏｎｔｈｅｏｒｙ，ｉｓｗｉｄｅｌｙｕｓｅｄｏｗｉｎｇｔｏｉｔｓ...     

结合吸收光谱与互相关法的燃气速度测量方法研究

针对可调谐半导体激光器吸收光谱(TDLAS)基于多普勒效应测速方法在燃气流速测量中频移量小、误差较大的问题,提出了结合固定波长吸收光谱法与互相关法的燃气速度测量方法。考虑碳氢燃料燃烧产物特点,选取H₂O分子7 185.597 cm-1吸收谱线,通过布置上下游两束固定波长吸收光谱测点,分析两信号的互相关特性来计算得到燃气速度。利用平面火焰炉实验系统对该方法测量燃气速度开展实验研究,获得了变工况下燃气速度随时间的变化情况。在相同工况下开展数值计算,将测量结果与数值模拟计算结果进行对比,相对偏差不超过8%。同时将该方法初步应用于煤油燃料火箭基组合循环发动机(RBCC)的高速羽流速度测量,获得了上下游探测器脉动信号,通过互相关分析计算得到了羽流速度,验证了该方法的可行性。实验结果表明,该燃气速度测量方法具有测量范围宽、测量精度高,环境干扰小等优点。提出的方法为发动机燃气速度测量提供一种简单可靠的测量方法。     

Experimental Study on Particle Size Distribution and Concentration Using Transmission Fluctuation Spectrometry with the Autocorrelation Technique

Based on the statistical characteristics of the transmission fluctuations in the particle suspension, transmission fluctuation spectrometry with autocorrelation (TFS‐AC) is described theoretically, with the assumptions of geometric ray propagation and completely absorbent particles in the suspension. The experiments presented here are realized in a focused Gaussian beam with the TFS‐AC technique. The acquisition of transmission fluctuation signals is achieved by using a high‐resolution digital oscilloscope. The transition function of TFS‐AC is obtained by varying the autocorrelation time. With a modified iterative Chahine inversion algorithm, solving a linear equation retrieves information on the particle size distribution and particle concentration. Some experimental results on spherical and non‐spherical particles are presented and discussed. The experiments cover a particle size range from 1μm to 1000 μm and a particle concentration of up to 12 %.     

Development of a Method for the Measurement of Particle Size and Velocity based on Projection onto a Variable Frequency Grating

Irregular particles down to approximately 3.8 μm have been sized by imaging them onto a variable frequency grating using laser illumination. Velocity can also be determined. The method was successful for certain particle types that were rough or irregular, partly absorbing or translucent. It was not successful for transparent spheres that display glare spots in the image, or other particles that produce localised regions of high brightness.     

Particle Size Distribution by Combined Elastic Light Scattering and Turbidity Measurements. A Novel Method to Estimate the Required Normalization Factor

Turbidity (T) and Elastic Light Scattering (ELS) measurements can be combined to estimate the particle size distribution (PSD) of a polymer latex. To this effect a “normalization factor” must be employed to adequately scale both measurements. In a real application, such factor is a priori unknown, and it must be estimated for determining the PSD. In a previous publication, a method was proposed to simultaneously estimate the normalization factor and the PSD by solving an inverse nonlinear problem. In this work, a simpler technique is presented which estimates the normalization factor and the PSD by sequentially solving two inverse linear problems. To evaluate the proposed technique, a synthetic example based on a polystyrene latex sample with a bimodal PSD was solved. It is shown that, for slight noise measurements, the errors in the estimation of the normalization factor are quite small and have a low effect on the PSD recovery.     

PSP技术在叶栅叶片表面压力测量中的应用

PSP(Pressure Sensitive Paint)测量技术以不改变叶片表面结构及可以进行全域压力测量的独特优势而受到广泛关注,但是受到几何结构的限制,有关内流场的PSP测量实验难度非常大.本文对ISSI公司所生产PSP测量系统开展了校准试验,完成了三个来流马赫数下叶栅叶片吸力面上静压分布的测量,并与传统压力扫描...     

Characterization of Pyrogenic Powders with Conventional Particle Sizing Technique: I. Prediction of Measured Size Distributions

Pyrogenic powders consist of fractal like aggregates with nanosized primary particles. The formation of such aggregates, their hydrodynamic behavior and their optical properties are in principle well understood. Even so, there is only little experience in interpreting results from particle sizing of such materials. Dramatic differences in size distribution obtained from different measurement techniques give frequently rise to confusion on the “true” aggregate size. However, such differences can be attributed to the different particle properties used for size measurement and to the different types of quantities, by which the frequency of the individual size fractions are weighted. For two conventional sizing techniques, Dynamic Light Scattering and Optical Centrifugation Analysis, the influence of the structural properties on the relevant optical and hydrodynamic aggregate properties is discussed on the basis of virtual aggregates as well as of empirical data for pyrogenic powders. Finally measurable size distributions are predicted in a case study.     

基于散射光偏振分布差的颗粒尺寸及折射率测量

提出一种基于颗粒散射光强正交分布差的颗粒尺寸和折射率同时测量方法.该方法利用颗粒散射光的垂直/平行分量和预设折射率,通过改进的Chahine算法反演得到粒径分布.根据所得粒径分布,计算得到平行/垂直分量,并与测量的平行/垂直分量比对,计算其拟合残差.遍历可能的折射率,使拟合残差趋于无穷小时,所对应的折射率即为颗粒的折射率,对应的粒度分布即为样品粒度分布.对聚丙乙烯标准颗粒、碳化硅及石墨样品进行测量,测量结果显示:对无吸收颗粒,折射率测量准确,吸收性颗粒虚部测量准确,使用所得到折射率测量值可得到准确的粒度分布.     

The Effect of Polydispersity in Primary Particle Size on Measurement of the Fractal Dimension of Aggregates

The use of the mass fractal dimension has become a popular method of characterising the structure of aggregates of fine particles. This parameter is often inferred from scattering experiments that exhibit a power law correlation between scattered intensity and the scattering vector. In this paper we demonstrate deviations from this behaviour that occur when the particles making up the aggregate are not monodisperse, even though the aggregate maintains the same fractal structure as observed in the monodisperse case. We have performed light scattering experiments with aggregating colloidal haematite and performed DLCA computer simulations to explain the observed behaviour. The behaviour can influence the determination of the mass fractal dimension, as can other factors such as scattering effects from primary particles.     

傅里叶变换红外光谱结合衰减全反射技术用于地贫筛查指标的无试剂定量分析

利用傅里叶变换红外光谱(FTIR)和衰减全反射(ATR)结合偏最小二乘(PLS)回归,建立地中海贫血(地贫)筛查指标平均红细胞血红蛋白含量(MCH)、平均红细胞体积(MCV)和总血红蛋白(Hb)的同时定量分析方法。收集人外周血样品380个,根据地贫血液学指标筛查标准,地贫阳性180个、阴性200个。从全体样品中随机选取150个为检验集(阴性64,阳性86),余下230个为建模集(阴性136、阳性94);再将建模集随机划分为定标集(阴性68、阳性47,共计115)、预测集(阴性68、阳性47,共计115),共200次。实验比较的结果表明,中红外指纹区(1 600～900 cm-1)PLS模型的预测效果显著优于全扫描谱区(4 000～600 cm-1)PLS模型,并且模型复杂性明显降低。基于中红外指纹区PLS模型,MCH,MCV,Hb最优PLS因子个数分别为10, 10, 6;建模预测均方根误差(M_SEP_Ave)分别为2.19 pg,5.13 fL,8.0 g·L-1;建模预测相关系数(M_R_{P, Ave})分别为0.902,0.898,0.922;检验预测均方根误差(V_SEP)分别为2.22 pg,5.38 fL,7.7 g·L-1;检验预测相关系数(V_R_P)分别为0.900,0.895,0.929;地贫筛查灵敏度、特异性分别达到100.0%和95.3%。结论：FTIR/ATR光谱结合PLS方法可以提供一种无需试剂、快速简便的大人群地贫筛查新技术。     

On-line Measurement of Solids Concentration or the Mean Particle Size in a Saturated Slurry Containing Background Particles Using a Turbidity Method

An on-line double-sensor turbidimeter for estimating the solids concentration or the mean particle size in a saturated slurry system in the presence of foreign insoluble particles is proposed. The correlation representing the operation of the double-sensor turbidimeter for on-line implementation is also developed. The technique relies on measuring the transmittance of an infrared light beam through the suspension, once in the presence of soluble particles and a second time when the soluble particles have completely dissolved. The device can be easily implemented on a crystallization system for monitoring and control applications. The maximum errors for solids concentration and mean particle size measurements were 13% and 7%, respectively.     

粉末样品颗粒大小对花椒挥发油近红外光谱定量预测的影响研究

为了快速有效评定花椒质量等级,应用近红外光谱分析技术,采用偏最小二乘法,对141份花椒样品粉碎成八种不同颗粒大小的粉末,对近红外光谱分别建立挥发油含量预测模型,当粉末样品颗粒大小为40目时,建立的模型最优,交叉验证测定系数r2141为0.9364,交叉验证误差均方根RMSECV141为0.421。使用105份40目粉末样品近红外光谱所建立的模型对36份样品的预测集进行预测,光谱预处理采用Meancentering vector normalization,谱区在6100.1~5774.2cm-1及4601.6~4424.2cm-1,则预测测定系数r326为0.9862,预测集验证误差均方根RMSEP36为0.192,预测相对标准差RSD36为4.95%,预测相对分析误差RPD36为8.517。研究结果表明,对花椒进行近红外光谱扫描前,粉碎到40目时所建立的近红外光谱模型最佳,使用近红外光谱技术快速有效检测花椒挥发油含量是可行的。     

线性回归粗差点判别在大学物理实验数据处理中的应用

大学物理实验数据处理过程中变量之间的关系大多可以转化为一元线性回归问题,但实验测量数据中可能存在的粗差点会降低其结果的准确性。本文将一元线性回归曲线拟合与C语言程序相结合,采用线性回归粗差点判别法,实现粗差点的剔除,有效地提高判别结果的准确性。