高速电视摄象系统的象质评价和动态计测精度

本文对高速摄象系统(包括电视摄象系统和光机式摄象系统)的象质评价方法予以讨论,并用实验方法对目前广泛使用的Kodak SP-2000、NAC HSV-200等高速电视摄象系统及我们自己开发的一维扫描电视摄象系统的象质,在新的定义下给出了评价,对它们的动态精度作了对比性测试,同时和光机式高速摄象系统作了比较。     

Guaranteed-accuracy approximation of reachable sets for a linear dynamic system subject to impulse actions

A method is proposed for approximating the reachable set of a dynamic system with a state space dimension no higher than six-eight considered on a finite time interval. The system is governed by linear differential equations with piecewise constant coefficients and impulse actions specified at prescribed times. The method is based on guaranteed-accuracy polyhedral approximations of reachable sets at researcher-specified times. Every approximation is constructed using the preceding one. A procedure is described for choosing parameters of the method that ensure the required accuracy with close-to-minimal time costs.     

A comparison of double-focusing sector field ICP-MS, ICP-OES and octopole collision cell ICP-MS for the high-accuracy determination of calcium in human serum

Human serum is routinely measured for total calcium content in clinical studies. A definitive high-accuracy and low-uncertainty method is required for reference measurements to underpin medical diagnoses. This study presents a novel octopole collision cell ICP-MS, high-accuracy, methodology and comparison of that technique with double-focusing sector field ICP-MS and an ICP-OES method. Double-matched isotope dilution mass spectrometry (IDMS) was employed for ICP-MS techniques and an exact matching bracketing technique using scandium as an internal standard was used for ICP-OES analysis. Medium resolution mode was utilised for double-focusing sector field ICP-MS analysis to resolve the dominant interferences on the ⁴⁴Ca/⁴²Ca isotope pair. Hydrogen reaction gas was employed to chemically resolve a number of polyatomic interferences predominantly through charge transfer reactions in the octopole collision cell. Comparison data presented for NIST CRM 909b human serum analysis from all three techniques demonstrates highest accuracy (99.6%) and lowest uncertainty (1.1%) for octopole collision cell ICP-MS. Data from ICP-OES using a non-IDMS technique produces comparably accurate data and low-uncertainties. The much higher total expanded uncertainties for double-focusing sector field ICP-MS compared with octopole collision cell data are explained by lower precision on the measurement of the ⁴⁴Ca/⁴²Ca isotope ratio. Data for octopole collision cell ICP-MS submitted for an international blind trial comparison (CCQM K-14) demonstrated excellent agreement with the mean of all participants with a low expanded uncertainty.     

Identification of micro-scale calorimetric devices

Micro-calorimetric devices using Si-based sensors are very useful for the study of gas–solid reactions, in which very low mass of reactants are necessary. But in fact the consequence of using flat detectors is an increase of the uncertainty in the measured energy. In this work a calorimetric gas sensor based on Xensor chip is analysed studying the local x–y contributions of dissipation to the sensitivity related to the value in the centre. We study also the effects of the gas-flow on the sensitivity, comparing the results obtained with two Xensor type chips. Finally we carry out a deeper analysis of the x–y effects on the calorimetric detector for dissipations in the reactant shell extremely close to the detector surface to visualize the link between the power density distribution and the output signal.     

The predictive accuracy for estimating infinite dilution activity coefficients by γ∞-based UNIFAC

The predictive accuracy for estimating infinite dilution activity coefficients by a modification of the UNIFAC method wherein the group interaction parameters were based on only data (referred to as -based UNIFAC) has been studied. Estimates and measured values were compared for six prototypical solutes in a series of homologous n-alkanes, l-alcohols and alkanenitrile solvents. Despite the fact that the interaction parameters were derived using only data, this approach still gave serious errors due to several inherent problems in the original UNIFAC model. Its performance is sometimes even poorer than that of the original UNIFAC method. For example for nitromethane in alcohols and p-dioxane in nitriles values predicted by the -based UNIFAC are essentially zero. The large errors for these systems are most likely due to inaccurate interaction parameters in the -based UNIFAC method.     

Determination of Low Contents of Fenpiverine Bromide by Extraction Spectrophotometry

An ion-pair extraction spectrophotometry method was developed for the determination of fenpiverine bromide in tablets. To determine this substance it was necessary to find experimental conditions that would allow eliminating the influence of other components of the tablets. Attention was paid to the fact that a suitable pH of water phase was necessary to achieve high selectivity.Received December 11, 2002; accepted April 24, 2003 Published online July 28, 2003     

对在建标技术报告中正确贯彻《测量不确定度表示指南》的理解

针对计量人员在贯彻《测量不确定表示指南》,尤其是在撰写“建标技术报告”时遇到的一些体问题,进行了探讨。     

Some aspects of quality assessment of the k0-based method of NAA

Neutron activation analysis is one of the analytical techniques often used for certification of reference materials. The k₀-based method of instrumental neutron activation analysis can also be applied in intercomparison runs in the certification process and therefore it is desirable to know its accuracy in advance. Possible systematic errors related to the application of nuclear data at given neutron flux rate parameters, that can affect the uncertainties of the results obtained by this specific method, are elucidated and error propagation factors calculated for a typical irradiation position in the TRIGA Mark II reactor of the Jozef Stefan Institute. It was found that these uncertainties are at the level of 1–2% on the average.     

Improving (Q)SAR predictions by examining bias in the selection of compounds for experimental testing

ABSTRACT

Existing data on structures and biological activities are limited and distributed unevenly across distinct molecular targets and chemical compounds. The question arises if these data represent an unbiased sample of the general population of chemical-biological interactions. To answer this question, we analyzed ChEMBL data for 87,583 molecules tested against 919 protein targets using supervised and unsupervised approaches. Hierarchical clustering of the Murcko frameworks generated using Chemistry Development Toolkit showed that the available data form a big diffuse cloud without apparent structure. In contrast hereto, PASS-based classifiers allowed prediction whether the compound had been tested against the particular molecular target, despite whether it was active or not. Thus, one may conclude that the selection of chemical compounds for testing against specific targets is biased, probably due to the influence of prior knowledge. We assessed the possibility to improve (Q)SAR predictions using this fact: PASS prediction of the interaction with the particular target for compounds predicted as tested against the target has significantly higher accuracy than for those predicted as untested (average ROC AUC are about 0.87 and 0.75, respectively). Thus, considering the existing bias in the data of the training set may increase the performance of virtual screening.     

Linear stability of WENO schemes coupled with explicit Runge–Kutta schemes

This paper focuses on the results of the linear stability analysis of the finite‐difference weighted essentially non‐oscillatory (WENO) schemes with optimal weights. The standard WENO schemes between the third and 11th order, the order‐optimised WENO schemes of the sixth and eighth order and the bandwidth‐optimised WENO schemes of the third and fourth order are considered. Several explicit Runge–Kutta schemes including the recently published strong stability‐preserving explicit Runge–Kutta schemes are considered for time discretisation. The stability limits as well as dissipation and dispersion properties dependent on the Courant–Friedrichs–Lewy number are presented for a hyperbolic model equation. The different combinations of space and time discretisation schemes are compared in terms of their accuracy and efficiency. For a parabolic model equation, the viscous term is discretised with high‐order central differences. The stability limits for the parabolic problem are presented as well. Numerical results of linear test cases are shown. Copyright © 2011 John Wiley & Sons, Ltd.