| 基于通用Dombi算子的测量不确定度传递方法研究* |
| |
| 引用本文： | 蒋薇,张玘. 基于通用Dombi算子的测量不确定度传递方法研究*[J]. 仪器仪表学报, 2016, 37(2): 270-276 |
| |
| 作者姓名： | 蒋薇 张玘 |
| |
| 作者单位： | 国防科学技术大学机电工程与自动化学院,国防科学技术大学机电工程与自动化学院 |
| |
| 基金项目： | 国家自然科学基金(51275523)项目资助 |
| |
| 摘 要： | 随机模糊变量RFVs法是近年来提出的一种基于可能性理论的测量不确定度评定和表示方法。RFVs法应用t-范数传递不确定度随机分量,是传统概率论方法的近似。针对Frank t-范数传递多个随机分量时近似误差累积变大的问题,提出使用二参数通用Dombi算子传递随机分量。首先简单介绍了RFVs的含义;然后说明了最优t-范数及其参数的选择方法,得到了GDO的最优参数,并用于获得联合可能性分布,结果表明GDO对于合成随机分量有很大改善;最后将GDO用于有功功率测量不确定度评定,并与Frank t-范数、传统GUM法和实验数据比较,使用GDO得到的置信区间能够很好地近似传统GUM法和实验数据分析结果。具有2个参数的GDO具有更大的灵活性,可以在传递多个不确定度随机分量的情况下获得满意的结果。
|
| 关 键 词： | 测量不确定度 通用Dombi算子 随机模糊变量 可能性分布 t-范数 |
Study on the propagation method of measurement uncertaintybased on general Dombi operator |
| |
| Jiang Wei and Zhang Qi. Study on the propagation method of measurement uncertaintybased on general Dombi operator[J]. Chinese Journal of Scientific Instrument, 2016, 37(2): 270-276 |
| |
| Authors: | Jiang Wei and Zhang Qi |
| |
| Affiliation: | College of Mechatronic Engineering and Automation, National University of Defense Technology and College of Mechatronic Engineering and Automation, National University of Defense Technology |
| |
| Abstract: | The random-fuzzy variables (RFVs) method based on the theory of possibility has been developed as a promising method for the evaluation and expression of measurement uncertainty in recent years. The RFVs method uses t-norms to propagate random contributions of the uncertainty, which is the approximation of the traditional method based on probability theory. Aiming at the problem that the error would cumulate and become bigger when Frank t-norm is applied to propagate multiple random contributions, a general Dombi operator (GDO) with two parameters is proposed and used for the propagation of random contributions. First, the meaning of the RFVs is briefly presented; second, the method for choosing optimal t-norm and its optimal parameters is illustrated, and the optimal parameters of GDO are obtained and used to get the joint possibility distribution. The result presents that an important improvement in the combination of the random contributions is achieved with GDO. Finally, the GDO was applied to the measurement uncertainty evaluation for active power; and the result was compared with that of the Frank t-norm, traditional GUM methods and the experiment data. The confidence interval issued with GDO is thus a good approximation of those provided by traditional GUM method and the experiment data. The GDO is a very flexible operator due to its two parameters, and can achieve a satisfied result in the propagation of multiple random contributions of uncertainty. |
| |
| Keywords: | |
| 本文献已被 CNKI 等数据库收录！ |
| 点击此处可从《仪器仪表学报》浏览原始摘要信息 |
|
点击此处可从《仪器仪表学报》下载免费的PDF全文 |
|
