管强, 汤银才. 基于线性退化轨道的区间型建模分析及应用[J]. 应用概率统计, 2018, 34(4): 427-440. DOI: 10.3969/j.issn.1001-4268.2018.04.007
引用本文: 管强, 汤银才. 基于线性退化轨道的区间型建模分析及应用[J]. 应用概率统计, 2018, 34(4): 427-440. DOI: 10.3969/j.issn.1001-4268.2018.04.007
GUAN Qiang, TANG Yincai. Analysis and Application of Interval Modeling for Linear Degradation Path[J]. Chinese Journal of Applied Probability and Statistics, 2018, 34(4): 427-440. DOI: 10.3969/j.issn.1001-4268.2018.04.007
Citation: GUAN Qiang, TANG Yincai. Analysis and Application of Interval Modeling for Linear Degradation Path[J]. Chinese Journal of Applied Probability and Statistics, 2018, 34(4): 427-440. DOI: 10.3969/j.issn.1001-4268.2018.04.007

基于线性退化轨道的区间型建模分析及应用

Analysis and Application of Interval Modeling for Linear Degradation Path

  • 摘要: 目前在退化试验中,产品失效一般定义为其性能退化至低于或高于一个指定临界值(单点型退化),尽管这种定义简单实用, 但却不尽合理, 不能完全描述产品退化失效问题.本文改进单点型退化模型, 提出区间型退化模型,并讨论了退化轨道是线性时的区间退化模型问题, 获得了各种线性区间退化寿命分布.本文利用数值积分和蒙特卡罗法模拟分析对比区间退化模型和单点退化模型的寿命分布,揭示区间退化和单点退化的关系. 最后通过实例分析体现区间退化在实际中更为合理有效.

     

    Abstract: At present, in degradation tests, product failure is generally defined as degradation of performance below or above a specified critical value (that is called single point degradation). Although this definition is simple and practical, it is not reasonable enough and degradation failure of the product can not be completely described. In this paper, a single point degradation model is improved, and an interval degeneration model is proposed. We discuss the interval degradation model when the degradation path is liner, and obtain life distribution functions for all kinds of linear interval degradation model. Numerical integration and Monte Carlo simulation methods are used to analyze and compare the life distribution of the interval degradation model and the single point degradation model, and the relationship between the interval degradation and the single point degradation is revealed. Finally, an real data example is analysis to show that interval egradation is more reasonable and effective in practice.

     

/

返回文章
返回