The analysis of the simulation results showed that the Herd-Johnson method is suitable to establish examination plans. In order to receive more accurate results, sampling plans have to be varied according to the simulation results. During the time of early failures only few machine actually fail. Therefore the confidence interval of the parameter ß is big enough to consider conclusions concerning sampling sizes. Besides one can make safe statements for the guaranteed time of the use (one year), so the economical consequences on behalf of extending this period are now known.
In the present work, all steps involved in the measurement of Brinell hardness indentations were studied, with the main purpose of evaluating the contribution of each step to the total uncertainty of the process. For the measurement of the indentations an expert system using digital image processing was developed. This automatic procedure can be very helpful as it reduces by orders of magnitude the time needed to perform the measurements and makes the process more reproducible by reducing the influence of the human operator.
This paper addresses the problem of least-squares curve adjustment in the case of uncertain measurement data. Compact matrix analysis is used to derive an iterative solution for the parameters of the adjusted curve. An expression for the uncertainty matrix of these parameters is also given.
Precision design work and tight manufacturing tolerances demand highly accurate measurements. A Vector Network Analyzer (VNA) is one of the most important aids for linear RF analog designs. It can give accurate information about the passive network performance in both the frequency domain and the time domain. Designers heavily rely on these measurement results to locate the design problems and to evaluate the network performance. But the measurement errors always exist. In any high frequency measurement, there are measurement errors associated with the system that contribute uncertainty to...
An analysis of estimation errors of correlation measurements due to A/D conversion with a dither signal has been carried out. It has been shown that any random signal satisfying appropriate conditions can be used as dither. Gaussian noise, although not satisfying one of these conditions, may influence the improvement in the accuracy of digital measurement. It is also easier to generate than dither with rectangular or triangular envelope probability density function (PDF), and, moreover, it may occur in a system in a form of intrinsic noise described by the term 'self-dither'. Because the...
Measured data of blood chemical analysis have dispersion results in themselves, but if a statistical analysis is applied to the results, an effective factor is obtained on accuracy quality control. As a method, there is the one which confirms the existence of abnormal reaction products by Michalies-Menten factor. This factor expresses the relation between the change of concentration of reagent and measurement data, and also the factor discriminates the good or bad of blood test reagent. And in this study, the effect of deposit days of the factor was studied. Then in this study, the reagent...
Results of a method are presented to measure the gain factor of a single passive antenna without the need of any reference antenna(s). A set-up for insertion loss measurement and only the antenna itself in front of a totally reflecting surface is required simulating a second identical antenna speeding up calibration procedures considerably. Since altogether only the antenna to be calibrated and a complex network analyser is needed one can expect among other things much smaller measurement uncertainty, directly traceable results, as well as much shorter time and lower cost in routine...
The results of optimization of a measuring system applied to the identification of a human respiratory tract by a method of forcing non-periodic pressure changes are presented in the paper. The aim of the optimization is minimizing errors of determination of the parameters of a model describing a system being identified in the multidimensional technological space of the applied measuring tools and the parameters of the measurement experiment and identification algorithm. The results obtained in the simulation process showed that the optimum properties of the identification system can be...
Signal-processing applications frequently use least-square sine-fitting algorithms. The parameter estimation provided by these algorithms is exposed to errors due to different causes. Good results in amplitude estimation may be achieved when the frequency is unknown applying the new method presented in this paper.
The ISO-GUM (Guide for Uncertainty Measurement) is a milestone in the history of uncertainty analysis, but practical procedures have still not been completely defined. In order to clarify some operative aspects, this paper proposes a practical improvement to the GUM, presenting a step-by-step procedure that illustrates how to evaluate uncertainty, developing analysis in parallel with the measurement process of the parameter. Following the ISO-GUM baseline, the proposed procedure is subdivided into: a) modelling stage: before designing a measurement chain suitable for a certain parameter, the...