数据统计处理中关于相关系数R的错误理解

本文旨在纠正一种在化学家中流传甚久的对统计数学中相关系数的错误理解.这种理解认为在数据拟合中单凭相关系数R大于某个人为指定值就可以判定数据拟合的优劣.     

A method for analysis of DMFC performance curves

Our recent analytical model of direct methanol fuel cell is used to fit experimental performance curves. Based on evolutionary genetic algorithm, a new fitting procedure is developed. The idea is to fit simultaneously a set of performance curves, which have several common fitting parameters. On the first stage of evolution all parameters are determined independently. Starting from certain step a mean values of common parameters are calculated and further evolution corrects the mean values. For different sets of experimental curves the method gives well reproducible results. The physical parameters resulted from fitting and the effect of crossover on cell performance are discussed.     

On Subgroups Generated by Small Classes in Finite Groups

Let G be a finite group and M(G) be the subgroup of G generated by all noncentral elements of G that lie in the conjugacy classes of the smallest size. Recently several results have been proved regarding the nilpotency class of M(G) and F(M(G)), where F(M(G)) denotes the Fitting subgroup of M(G). We prove some conditional results regarding the nilpotency class of M(G).     

Dynamic Modelling and Identification of a Waterjet Cutting System

This paper describes a new dynamic model for a waterjet cutting system that includes a double-acting reciprocating intensifier pump. Since the system operates at high pressures the fluid flow is assumed to be compressible. The dynamic model includes the characteristics of the intensifier pump, the check valves, the accumulator, the system piping and compressible jet flow through the nozzle. The system model is presented as a set of differential-algebraic equations. Experimental results for an actual system are used to identify the discharge coefficient of the nozzle, certain unknown parameters associated with the check valve, and to determine the velocity profile of the piston in the intensifier pump. This is accomplished by formulating and solving a parameter optimization problem. The paper also includes numerical simulation results that validate the dynamic model.     

基于函数拟合的玉米叶片反射光谱时间变化规律分析

为了探索玉米(Zea mays L.)叶片反射光谱时间变化规律,对21个叶位的玉米叶片进行了每天一次的反射光谱测量,获得了玉米活体叶片整个生命周期反射光谱数据1 261条。在此基础上采用光谱相关图方法,对400～960 nm的可见光、近红外波段的玉米叶片反射光谱进行分段拟合,获得了7个表征玉米叶片反射光谱时间变化规律的拟合参数,并对这7个参数的时间变化趋势采用二元二次多项式进行拟合。结果表明该方法对单片叶片反射光谱的拟合效果非常理想,其中98.7%的叶片反射光谱拟合复相关系数r大于0.99,80.9%的叶片反射光谱拟合均方根误差RMSE小于0.001 5。将所有数据的拟合结果与原始数据进行比较,复相关系数r为0.9978,均方根误差RMSE为0.010 5,拟合结果表明该方法较好地保持了反射光谱时间变化的趋势,能够充分利用叶片反射光谱的相关性,有效地表征玉米叶片反射光谱随时间变化的规律。     

Retention mechanism for ion-pair chromatography with chaotropic reagents

During the last decade, the extensive use of ion-pair chromatography (IPC) in protein, peptides and basic drugs applications prompted chromatographers to evaluate new additives, since traditional ion-pairing reagents (IPRs) are not usually compatible with LC-MS hyphenation and tend to stick very strongly to the stationary phase, thereby impairing the initial column properties. Chaotropic salts received a great share of credit as tentative IPRs, since they proved to be able to mimic the role of classical IPRs, thereby increasing the retention of oppositely charged analytes. Very few quantitative theoretical studies faced the retention modelling when chaotropic additives are made use of in a chromatographic system and, unfortunately, they used a stoichiometric approach. We hereby debate the present state of the theory and illustrate the first attempt to explain the retention mechanism in the presence of chaotropic reagents in RP-HPLC at a thermodynamic level. We quantitatively validate this model for typical positively and negatively charged analytes as well as for ionic liquid, zwitterionic and neutral analytes.     

基于最大长度序列的绝对位移精密检测方法

为了检测精密机构的微小位移,根据位移编码与机器视觉技术,提出一种基于最大长度序列（Maximum Length Sequence—MLS）的绝对位移检测方法。该方法设计了基于MLS的绝对位移检测的视觉靶标方案,采用基于灰度叠加及二值化方法进行图像处理和靶标绝对位置识别;研究了基于拟合相位的精密定位方法,实现一维绝对位移的精密检测,并进行实验验证。实验结果表明,该方法可有效实现一维绝对位移检测,定位精度在±2μm范围内,具有良好的重复性和鲁棒性,且无须视觉标定。     

基于遗传算法的曲线拟合方法用于重叠荧光光谱的定量解析

将修正高斯模型为遗传算法的适应性函数用于拟合荧光光谱图,有效地解决了分子荧光光谱法多组分检测过程中荧光光谱相互干扰的问题。考察了不同尿液中内源性荧光物质对加替沙星(GFLX)荧光的干扰。应用拟合荧光光谱图可有效地消除内源性荧光物质的干扰。在优化条件下,GFLX的浓度在0.06～3.5 μg·mL-1范围内, 与其荧光强度之间具有良好的线性,相关系数为0.999 4。检出限为0.02 μg·mL-1,回收率为99.2%～109.4%,相对标准偏差为1.3%～2.7%。结果表明,拟合荧光光谱法无需分离即可实现对尿液中GFLX准确的定量检测,适用于常规定量检测和药物代谢研究。     

采用曲线拟合方法进行EPR波谱基线校正

提供了一种以曲线拟合方式进行的EPR波谱基线校正的简单方法.程序置于486微机上,采用C⁺⁺语言编写,在Windws环境下运行.使用菜单、模块化方式.操作简单,快捷.在联机进行数据采集时尤为方便.本程序对需要进行基线校正的谱线,只需使用鼠标器在实验谱上选择一定的数据点,通过程序运算,比较,验算即可得到较好的基线校正.