Applications of exponential-approximate method with multi-resolution in spare requirement determination
-
摘要: 针对Weibull分布的更新函数较难确定的问题,研究了在平均寿命相等情况下指数分布与Weibull分布之间的贴近性.在此基础上,提出利用指数分布的更新函数模型计算Weibull分布的更新函数,能够比较方便有效地得到近似解.通过实例计算,分别比较了指数方法(直接利用指数分布的更新函数)、线性加权模型以及几何加权模型等三种方法的精度.结果表明:当时间较短时,线性加权和几何加权模型比指数方法精确度有所提高;当时间较长时,几何加权模型的精度较高.利用该结论能够为工程应用提供方便.Abstract: As to the fact that it is difficult to determine the renewal function (RF) of the Weibull distribution, the closeness between the two distributions with the same average life was studied. On this condition, the RF of exponential distribution was applied to compute that of the Weibull distribution, and the approximate solution would be gained expediently. Furthermore, the accuracy of the three methods was compared. That is, exponential method (The RF of the exponential), linear weight model and geometrical weight model respectively. The result shows that compared with the exponential method, the precision of the two models are highly enhanced when the time is short; and the precision of the geometrical weight model is still high when time is long. This conclusion would provide convenience for the engineering.
-
Key words:
- renewal function /
- exponential approximation /
- linearly weighted /
- geometrically weighted
-
[1] Eric Smeitink,Rommert Dekker.A simple approximation to the renewal function[J].IEEE Transactions on Reliability,1990,39(1):71-75 [2] McConalogue D J.Numerical treatment of convolution integrals involving distributions with densities having singularities at the origin[J].Communications in Statistics,1981,B10:265-280 [3] Jiang R.A Gamma-normal series truncation approximation for computing the Weibull renewal function[J].Reliability Engineering & System Safety,2008,93:616-626 [4] Jiang R.A simple approximation for the renewal function with an increasing failure rate[J].Reliability Engineering & System Safety,2010,95:963-969 [5] Jiang R.A simple approximation for the Weibull renewal function[C]//Proceedings of the 2009 IEEE IEEM.Hong Kong:IEEE,2009:1146-1149
点击查看大图
计量
- 文章访问数: 2549
- HTML全文浏览量: 162
- PDF下载量: 588
- 被引次数: 0