论文部分内容阅读
设系统A由K个独立的子系统B_1,B_2,…,B_K并(串)联而成,设第i个子系统B_i又由m_i个相互独立的成败型元件C_(i1),C_(i2),…C_(imi)串(并)联而成,设有多层试验数据: 元件C_(ij)试验N_(ij)次,成功S_(ij)次,失败F_(ij)次(i=1,2,…,K,j=1,2,…,m_i) 子系统B_i有成败型试验数据:试验N_i次,成功S_i次,失败F_i次(i=1,2,…,K) 系统A有成败型试验数据:试验N次,成功S次,失败F次。 本文给出利用此多层成败型试验数据,求系统A的可靠性置信下限的近似解的方法,本文利用一、二阶矩拟合的原则将上述数据折合为原系统A的伪成败型数据:伪试验数N~*,伪成功数S~*,然后从N~*,S~*出发利用单个成败型元件之可靠性的经典精确方法求出原系统A的可靠性置信下限的近似值。本文推导了伪试验数N~*,伪成功数S~*的计算公式,并给出了计算实例。
Suppose that the system A is composed of K independent subsystems B_1, B_2, ..., B_K and connected in series, and the i th subsystem B_i is composed of m_i success independent components C_ (i1), C_ (i2) (Ij) times, successful S_ (ij) times and failed F_ (ij) times (i = 1) , ..., K, j = 1,2, ..., m_i) Subsystem B_i has test data of success or failure: test N_i times, succeed S_i times, fail F_i times (i = 1,2, ..., K) System A Success or failure type test data: test N times, successfully S times, failed F times. In this paper, a method of using this multi-layer test data to find the approximate solution to the confidence lower bound of system A is given. In this paper, we use the principle of first and second order moments to convert the above data into pseudo-failure data of the original system A : Pseudo-test number N *, pseudo-successful number S ~ *, and then from N *, S * *, the approximation of the lower confidence limit of the reliability of the original system A is obtained by the classical and exact method using the reliability of a single success / failure component. This paper deduces the pseudo-test number N ~ *, the number of pseudo-success S *, and gives a calculation example.