论文部分内容阅读
本文以散装生铁块的随机样本分析结果为基础 ,应用 Monte Carlo模拟法考察了组合样本中 Si、Mn、C、S、P含量 (或份样平均值 )的分布情况 ,研究了组分含量的置信区间 ,对参数法和非参数法的结果进行了比较。结果表明 ,对近似符合正态分布 (或 t-分布 )的原始总体或当组合样中的份样数目较多时 ,两种方法所估计的置信区间相近。但对偏离正态分布 (或 t-分布 )的总体且组合样中的份样数目较少时 ,非参数法较参数法的结果更符合实际。
Based on the results of random sample analysis of bulk pig iron, the distribution of Si, Mn, C, S, P contents (or sample average) in the combined samples was investigated by Monte Carlo simulation. The effects of component content Confidence intervals, the results of the parametric and non-parametric methods were compared. The results show that the confidence intervals estimated by the two methods are similar for the original population that approximate the normal distribution (or t-distribution) or when the number of aliquots in the composite samples is large. However, the non-parametric method is more practical than the parametric method for the total deviations from the normal distribution (or t-distribution) and the fewer samples in the combined samples.