欢迎访问《兵工学报》官方网站,今天是 分享到:

兵工学报 ›› 2019, Vol. 40 ›› Issue (4): 874-879.doi: 10.3969/j.issn.1000-1093.2019.04.023

• 论文 • 上一篇    下一篇

基于灰自助和未确知有理数的小样本试验数据估计

柯肇捷1, 周文雅2   

  1. (1.大连理工大学 石油与化学工程学院, 辽宁 盘锦 124221; 2.大连理工大学 航空航天学院, 辽宁 大连 116024)
  • 收稿日期:2018-05-14 修回日期:2018-05-14 上线日期:2019-06-10
  • 通讯作者: 周文雅(1981—), 男, 副教授, 硕士生导师 E-mail:zwy@dlut.edu.cn
  • 作者简介:柯肇捷(1998—), 男, 本科生。 E-mail: 18310811599@163.com
  • 基金资助:
    国家自然科学基金项目(11502041)

Parameter Estimation Model of Small Test Samples Based on Grey Bootstrap Method and Unascertained Rational Number

KE Zhaojie1, ZHOU Wenya2   

  1. (1.School of Petroleum and Chemical Engineering, Dalian University of Technology, Panjin 124221, Liaoning, China; 2.School of Astronautics and Aeronautics, Dalian University of Technology, Dalian 116024, Liaoning, China)
  • Received:2018-05-14 Revised:2018-05-14 Online:2019-06-10

摘要: 针对小样本试验数据的参数估计问题,提出了一种基于灰自助和未确知有理数的新方法。使用灰自助方法对小样本试验数据进行自助再抽样,通过灰色模型GM(1,1)挖掘生成较多的样本数据;构建虚拟样本的未确知有理数表达模型,提出了未确知有理数阶数的优化算法;分别建立了基于未确知有理数的小样本试验数据点估计和区间估计模型,并提出了相应的估计置信度模型;对某型装备干扰功率测试数据进行了参数估计。示例结果表明所提方法合理可行,不涉及原始数据的概率分布特征,能在较高的可靠度下解决小样本数据的参数估计问题。

关键词: 参数估计, 小样本, 灰自助, 未确知有理数

Abstract: A method based on grey bootstrap method and unascertained rational number is proposed for the parameter estimation of small samples. The experimental data of small samples are resampled by using the grey bootstrap method, and more samples are generated by using the grey model GM(1,1). An expression model based on unascertained rational number of virtual samples is developed, and an optimization algorithm of the order of unascertained rational number is proposed. The point estimation and interval estimation models based on unascertained rational number are introduced, respectively, and the correlative estimation confidence models are given. The point estimation and interval of the interference power test data are estimated by the proposed method. The examples show that the proposed method has no demand for the probability distribution of small samples, and has high estimation reliability. Key

Key words: parameterestimation, smalltestsample, greybootstrapmethod, unascertainedrationalnumber

中图分类号: