组合取样的误差理论研究及其Monte Carlo模拟

Theoretical Study on Composite Sampling and Monte Carlo Simulation

基于多项分布理论 ,建立了组合取样方差与取样总体中各子区的大小及其组分含量之间的关系和计算公式 ,探讨了组合取样常数的物理意义 .以颗粒物质为例 ,探讨了组合取样的逻辑质量单元的概念及其意义 ,为确定组合取样中的份样质量提供了理论依据 .本文对于完善组合取样误差理论 。

Based on the multinomial distribution theory, an equation for estimating the composite sampling errors was derived in this paper, by which the composite sampling variance is quantitatively related with the concentrations of the sought for component, the sizes of the subgroups of the population and the number of increments in the sample. Monte Carlo technique was employed to simulate the composite sampling and verify the above equation with satisfactory results. This equation is generally applicable in composite sampling so as to enrich the theory of sampling in analytical chemistry.