柴油运动粘度检测能力验证的统计分析

采用稳健统计技术对全国77家实验室车用柴油运动粘度能力验证结果进行统计分析,结果显示满意的为68家,结果可疑或离群的为9家。并对结果出现可疑或离群的原因进行了分析。     

塑料薄膜阻隔性能检测能力现状

阻隔性能检测应用范围广泛,LABTHINK兰光2014年春季塑料薄膜阻隔性能实验室间比对以塑料薄膜为试样,对现阶段的阻隔性能检测水平进行验证。通过对结果的统计分析,可以看出,现阶段我国塑料薄膜阻隔性能整体检测水平尚属理想,但是检测结果仍具有一定的离散性,部分实验室可疑或离群。在目前阻隔性能检测缺乏统一校准规范的情况下,建议实验室积极进行实验室间数据比对,以使自身实验室数据保持在合适的水平。     

稳健统计与格拉布斯准则在能力验证结果分析中的应用

分别运用稳健统计（四分位和迭代）法、格拉布斯准则剔除离群值后用经典统计法对热轧带肋钢筋的6组数据进行了统计分析。结果表明:数据组中Z比分值大于5的数据,为离群值。数据组总体服从正态分布或接近正态分布的,3种统计方法结果基本吻合;不存在离群数据的情况下,建议采用稳健统计（四分位或迭代）法进行统计;存在离群数据的情况下,应采用格拉布斯准则剔除离群值后用经典统计法进行统计。对于数据明显偏离正态分布的,不存在离群数据的情况下,建议采用迭代稳健统计法进行统计;存在离群数据的情况下,应采用格拉布斯准则剔除离群值后用经典统计法进行统计。     

实验室间比对的能力验证及稳健统计技术第四讲能力验证试样的均匀性和稳定性检验

能力验证是利用实验室间比对确定实验室的检测或校准能力。因此，用于能力验证试样的均匀性和稳定性是进行能力验证的基础，所有参加能力验证实验室所用的试样都应均匀一致，稳定可靠，这就能确保能力验证中出现任何可疑结果或离群结果皆不能归因于试样之间的变异，而是实验室的能力差异所引起。为此，能力验证提供者必须选择材质均     

转盘型吸烟机与直线型吸烟机卷烟主流烟气测试比对

为了对转盘型吸烟机与直线型吸烟机主流烟气测试结果进行比对,选用两种类型的主流吸烟机进行卷烟抽吸,首先对测试数据进行z比分数计算,剔除可疑结果和离群结果,再进行双样本T检验,从概率统计学上发现焦油、烟气烟碱、一氧化碳用不同类型的吸烟机测试结果部分具有显著性差异,部分无显著性差异,最后对不同类型的吸烟机测试结杲进行常规直观性分析,所有的相对相差绝对值均小于5%,故认为两种不同类型的吸烟机对焦油、烟气烟碱、烟气一氧化碳测试结果具有良好的一致性。     

用Excel判断正态样本离群值

<正>测量结果中离群的异常值是由测量结果中粗大误差引起的。通常是由测量过程中不可重复的突发事件所致。一般是源于测量过程中的电子噪声或机械噪声、测量条件的突然改变、操作人员在读数和书写方面的疏忽,以及错误地使用测量设备等。如果在测量结果中混有离群的异常值,必然会歪曲测量结果,剔除离群值将使测量结果更符合客观事实。     

基于非线性数据变换的离群点检测算法

为了提高高维数据集合离群数据挖掘效率,在分析了传统的离群数据挖掘算法优点和缺点的基础上,提出了一种离群点检测算法,首先将非线性问题转化为高维特征空间中的线性问题,然后利用非线性数据变换进行维数约减,对所得数据对象每个投影分量逐个判断数据点是否是离群点,通过实验证明该算法不仅可用于线性可分数据集的离群点检测,而且可用于线性不可分数据集的离群点检测,表明了算法的优越性。     

实验室间比对结果分析统计方法的探讨

对实验室间比对结果分析的统计方法进行了探讨。比较了常用的统计方法——稳健统计法和经典统计法的计算过程，对几次实验室间的比对结果分别用这两种统计方法进行了分析，同时结合格拉布斯检验法对离群值进行判断，从而探讨了两种统计方法的适用范围。认为当比对数据较多时，可采用稳健统计法给出严格的评价结论；当比对数据较少时，应先利用格拉布斯检验法对离群值进行剔除，同时要考虑方法标准或产品标准规定的允许误差，然后对余下的数据用经典统计法进行分析和评价。     

薄板拉伸能力验证计划与技术分析

介绍了由宝钢股份公司组织的"BSTC T002金属材料拉伸试验"能力验证计划,以及围绕金属薄板所开展的"两步法"性能均匀性检验.基于稳健统计技术与Z比分数,对15家实验室获得的数据进行了技术分析,讨论了影响试验结果可靠性的测量因素.稳健化置信椭圆揭示出造成PT计划离群结果的主要因素应为系统误差(实验室间变异).     

应用稳健统计法对实验室间比对实验结果评价

采用稳健统计法统计,以直方图、正态分布图和尤登图对统计结果进行表示,对实验室间比对实验结果进行结果评价。统计结果表明,采取的统计分析是有效的、合理的,并有助于离群实验室发现问题,找出自身差距,进行整改。     

PMP与CCEM衰减关键比对的链接

APMP 衰减关键比对APMP.EM.RF-K19.CL将比对测量结果与CCEM衰减关键比对进行了链接。在链接计算中,应用加权平均方法计算链接修正值;应用中位数方法剔除链接数据中的统计离群值,并考虑了链接实验室CCEM比对原数据有效性问题。链接计算结果表明链接修正值可靠,评定的不确定度合理。     

Pyroceram 9606, A Certified Ceramic Reference Material For High-Temperature Thermal Transport Properties: Part 2—Certification Measurements

Following the characterization of the batch of Pyroceram 9606 material, a number of the partners in the European Commission (EC) supported program carried out certification measurements of thermal conductivity and thermal diffusivity. Six laboratories undertook thermal-diffusivity measurements using either the flash or the modulated beam methods. Eight laboratories measured the thermal conductivity, using either the steady-state guarded-hot-plate method or one of the transient hot-wire methods. Results from each series of measurements were provided in a standard format as an aid to simplify the statistical analysis of the data. The results were corrected to the nominal measured temperature and for change in dimension, analyzed separately, and presented in a standard format. Outliers were identified and rejected where appropriate, based on both statistical and technical evidences. The individual data sets were combined, and the grand mean data for each property analyzed further to provide the certified values together with their uncertainty limits. Finally, using the specific heat capacity and density values obtained from the characterization tests, values of thermal conductivity were calculated from the measured thermal diffusivity. The difference between the calculated and certified values is less than 2.7 %, which is well within the uncertainty limit assigned for the certified thermal property values.     

A comparative study of statistical methods used in analyzing the proficiency testing results of yield stress

The proficiency test (PT) is a powerful tool to help a laboratory to demonstrate its competence. The statistical analysis used plays very important role in the PT results. This paper demonstrates two PT numerical examples for tensile test using different statistical methods in their analysis. The study shows that zeta score and the normalized error value (E_n) give representative impression about the consistency of the results with regarding the claimed laboratories uncertainties. The robust z-score value gives opportunity to avoid the effect of the outlier values regardless of the claimed laboratory uncertainty.     

Precision of method for determining resistance of bituminous mixtures to de-icing fluids

This paper presents the results of a Round-Robin test to estimate the precision of European method EN-12697-41 “Test methods for hot mix asphalt-Resistance to de-icing fluids”. The purpose of the project was to determine precision data according to ISO 5725, ASTM E691 and ASTM C802. The examined test method is intended for use in requirements specifications for airfield de-icing chemicals and/or as a tool for development of such products. Precision statistics, repeatability and reproducibility standard deviations, are based on observed values from six laboratories and six levels, each level comprising four samples. From a general statistical analysis, which was conducted in addition to precision determination, it could be concluded that the most damaging de-icing agents (treatments) were identified by all participating laboratories both in terms of absolute values and by ranks.     

Proficiency Testing in Chemical Analysis of Iron Ore: Comparison of Statistical Methods for Outlier Rejection

Participation in proficiency testing (PT) is an important task to meet the requirements of ISO/IEC 17025 in the area of quality assurance of laboratory results. A PT program in the field of chemical analysis of iron ore was organized by CSIR-National Metallurgical Laboratory, Jamshedpur (nodal laboratory) and CSIR-National Physical Laboratory, New Delhi (PT Coordinator) during November 2011–January 2012. Twenty-two (22) laboratories in India participated in the PT program. The results of participating laboratories were first analyzed to identify the distribution patterns and the presence of outliers. Several parametric and robust statistical methods were used to identify the outliers. Correct outlier rejection is of utmost importance because the choice of the outlier test method influence the consensus value and standard deviation which in turn determine the Z-score of a laboratory result in a PT program. In the present study, five parametric outlier tests were compared: Dixon’s Q test, Grubbs single test, double test, t test, and Z-scores. In addition three robust tests as alternative to parametric tests were chosen: box plot, Huber test and MAD-based test. It was observed that multiple outlier test methods should be used to identify the outliers in a PT program especially when the number of participating laboratories is less. They complement each other and helps give diverse information and better overview of the data set. Among the 22 participating laboratories, Z-scores of 4 laboratories for analysis of total iron fall outside the acceptable limit of ±2. Similarly, for analysis of alumina and silica, five laboratories had unacceptable Z-scores.     

Modelling and Statistical Treatment of Residual Stress Distributions in an Edge Welded Stainless Steel Beam

Abstract: Experimental, finite element analysis and statistical studies of residual stresses in edge welded type 316H stainless steel beams are presented. The experimental stress measurements were previously performed by different laboratories using neutron and synchrotron X‐ray diffraction. An analytical model to describe the magnitude and distribution of the residual stresses is presented. Results from finite element analyses are also provided. It is found that there is general agreement between the trends in the residual stresses derived from analysis and measurement. However, the scatter in results is substantial and a statistical framework for treating residual stresses using Bayesian statistics based on experimental and simulation results is described. The Bayesian analytical approach that uses the analytical model as a reference permits the model to be applied to circumstances outside the experimental conditions.     

Teaching Statistical Experimental Design Using a Laboratory Experiment

An experimental project introduces the concepts of statistical experimental design to undergraduates in a laboratory setting. A safe, inexpensive and easily operable experiment uses a gas chromatograph to give quantitative results and to allow students to concentrate on applying statistical skills without being impeded by complex equipment or experimental methods. One of the unique aspects of the experiment is a trade-off between the two most significant variables, forcing students to compromise in the selection of optimum conditions. Such compromises are typical in many real-world industrial situations. The experiment has been used for several years in the undergraduate chemical engineering laboratories at the University of North Dakota. Keywords: Statistics, Experimental Design, Laboratory.     

Preparation and Certification of Novel Reference Material for Smoke Density Measurements

A novel type of reference material has been prepared and characterized for smoke density chamber calibration in the flame condition by Fire Protection Laboratory, National Institute of Standards, Egypt. The new reference material was fabricated by mixing modified magnesium hydroxide nanoparticles (MH-NP-OP) with acrylonitrile–butadiene–styrene copolymer (ABS). The reference material was certified for the physical properties of maximum specific optical density (D_m) and corrected maximum specific optical density (D_m·corr) using the analysis of two different competent laboratories. Preparation, characterization, homogeneity testing and certified value assignment for the developed reference material have been studied. It was clearly demonstrated that the smoke measurement of the two laboratories are harmonic which is an obvious prerequisite for the confidence assessment of the reference material. A metrological approach was followed to detect the statistical biases between different laboratories data to achieve an appropriate accuracy in smoke measurements. The obtained measurement results were statistically analyzed and the certified values of D_m and D_m·corr for reference material were estimated as 510.35 and 472.95, respectively as well as their expanded relative uncertainties at confidence level 95% were recorded 2.78 and 2.7%, respectively. The developed reference material are expected to be used to assist in validating smoke measurements of different materials to realize their fire hazard when be used in building.     

Statistical Analysis of a Round-Robin Measurement Survey of Two Candidate Materials for a Seebeck Coefficient Standard Reference Material

In an effort to develop a Standard Reference Material (SRM™) for Seebeck coefficient, we have conducted a round-robin measurement survey of two candidate materials—undoped Bi₂Te₃ and Constantan (55 % Cu and 45 % Ni alloy). Measurements were performed in two rounds by twelve laboratories involved in active thermoelectric research using a number of different commercial and custom-built measurement systems and techniques. In this paper we report the detailed statistical analyses on the interlaboratory measurement results and the statistical methodology for analysis of irregularly sampled measurement curves in the interlaboratory study setting. Based on these results, we have selected Bi₂Te₃ as the prototype standard material. Once available, this SRM will be useful for future interlaboratory data comparison and instrument calibrations.