摘要：【中文】本发明公开了一种规则几何形状零件尺寸波动的估计方法，通过数理统计理论建立样本零件尺寸标准差与该批次零件尺寸标准差的关系，可精确计算出尺寸标准差置信区间，进而结合序贯抽样理论不断缩短置信区间，最终实现对尺寸标准差的精确估计。本发明结果可直接用于规则几何形状零件的尺寸波动估计，解决了现有零件尺寸波动估计方法在抽样不确定性条件下存在的检测样本量难确定、估计精度难衡量以及不易实施的问题。 【EN】The invention discloses a method for estimating the size fluctuation of a regular geometric shape part, which establishes the relation between the standard deviation of the size of a sample part and the standard deviation of the size of the batch of parts through a mathematical statistics theory, can accurately calculate the confidence interval of the standard deviation of the size, further continuously shortens the confidence interval by combining a sequential sampling theory, and finally realizes the accurate estimation of the standard deviation of the size. The result of the invention can be directly used for the size fluctuation estimation of the parts with regular geometric shapes, and solves the problems of difficult determination of the detection sample amount, difficult estimation precision and difficult implementation of the existing part size fluctuation estimation method under the sampling uncertainty condition.

摘要：【中文】一种规则几何形状零件尺寸分布中心的估计方法，包括以下步骤：步骤1：零件抽样；步骤2：计算样本零件尺寸均值及标准差；步骤3：计算整批零件尺寸分布中心置信区间；步骤4：采用序贯抽样的方法，以样本零件尺寸均值估计该批次零件的尺寸分布中心；本发明基于小样本计算出零件尺寸分布中心在给定置信度及估计误差下的估计值；根据已获取的样本信息，灵活判断抽样是否应该继续，基于估计误差约束确定抽样终止条件，在保证精度的基础上，大大减少检测样本量，节省检测成本，提高效率。 【EN】A method for estimating the size distribution center of a regular geometric part comprises the following steps: step 1: sampling parts; step 2: calculating the average value and the standard deviation of the sizes of the sample parts; and step 3: calculating the central confidence interval of the size distribution of the whole batch of parts; and 4, step 4: estimating the size distribution center of the batch of parts by a sequential sampling method according to the average value of the sizes of the sample parts; the method calculates the estimated value of the size distribution center of the part under the given confidence coefficient and estimation error based on the small sample; according to the acquired sample information, whether sampling should continue or not is flexibly judged, and sampling termination conditions are determined based on estimation error constraints, so that the detection sample amount is greatly reduced, the detection cost is saved, and the efficiency is improved on the basis of ensuring the precision.