Abstract: To assess the uncertainty for small sample measurement in the Automatic Test System （ATS）, a novel method based on grey system theory and Bayesian information fusion theory is adopted. The new method is advanced on the typical grey evaluation model of measurement uncertainty, and it introduces the grey relational theory to gain a new grey evaluation model of measurement uncertainty further, and employees the idea of Bayesian information fusion which combines the evaluation results of historical data and current data to get the final uncertainty at last. According to a special example analysis of a transfer unit in the ATS measurement chain, the experimental results indicate that compared with other common methods, the uncertainty results are in good agreements with each other, and the results got from small sample ATS static measurement have a less variation, and the results got from small sample ATS dynamic measurement whose amount of computations is smaller are closer to its dynamic measurement characteristics.

Key words: Automatic Test System（ATS）, small sample measurement, grey system theory, Bayesian information fusion theory, measurement uncertainty evaluation

摘要： 提出一种基于灰色系统理论和贝叶斯信息融合理论评定小样本自动测试系统（Automatic Test System，ATS）测量不确定度的新方法。该方法在已有测量不确定度灰色评定模型的基础上，引入灰色关联理论，建立新灰色评定模型，并引入贝叶斯信息融合思想，融合历史测量数据和当前测量数据的评定结果，得到最终测量不确定度。以某ATS测量链中具体的传递单元作为实例，按该方法计算测量不确定度，并将评定结果与其他常用方法评定结果进行比较，达到很好的一致性，且该方法评定小样本ATS静态测量得到的结果准确度高，评定小样本ATS动态测量，更符合其动态测量特性且计算量小。

关键词: 自动测试系统, 小样本测量, 灰色系统理论, 贝叶斯信息融合理论, 测量不确定度评定