基于可能性理论的测量数据处理  

Processing of Measurement Data Based on Possibility Theory

作　　者：肖明珠[1] 陈光(礻禹)[2] 

机构地区：[1]中国工程物理研究院电子工程研究所,四川绵阳621900 [2]电子科技大学自动化工程学院,成都610054

出　　处：《信息与电子工程》2006年第2期98-102,共5页information and electronic engineering

摘　　要：测量数据的可能性分布充分地反映测量值的不完备程度,不仅表达了测量值与真值的相容程度,还表示了不同置信水平下的置信区间。利用概率分布到可能性分布的最优转换以及简单三角形转换,可以建立3种常见的测量数据的可能性分布模型,并据此提出了3种测量不确定度的评估方法。The measurement data includes many types of uncertainty such as vagueness, nonspecificity and stochasticity, and can not always be handed by probability theory. Based on possibility theory, how the measurement data is represented by possible distribution, and how the measurement uncertainty is evaluated by three methods are discussed firstly. Then, methods of probability-possibility transformation are described. According to these transformations, possible distribution model of measurement data can be build. Finally, three examples of main situations are demonstrated.

关 键 词：数理统计学 数据处理 可能性理论 测量误差 不确定度 

分 类 号：O212.1[理学—概率论与数理统计]