多状态系统的动态可靠性度量及重要度分析

发布时间：2018-12-26 13:18

【摘要】：为有效表征性能退化条件下多状态系统的可靠性,确定各组件(状态)对系统的影响,文中对性能状态进行分析,提出了新的动态可靠性度量指标和性能水平限制下的重要度计算方法.首先,针对多状态系统马尔可夫过程的组合爆炸问题,应用通用生成函数法求解系统(组件)各状态的概率和可靠度、性能状态期望值、性能状态平均期望值和性能状态驻留时间长度等4种动态可靠性度量指标.然后,为满足系统(组件)的不同性能需求,综合评估各组件(状态)对系统的影响,用性能期望值定义了给定性能水平下的Birnbaum重要度、Fussell-Vesely重要度、风险业绩值和风险降低值.四缸柴油机燃油供给系统的实例结果表明,文中方法既可准确地求解系统及其组件(油泵和喷油阀)的性能状态概率,以及性能退化条件下各性能状态的可靠度、期望值、平均期望值和驻留时间长度,又能根据性能水平限制下的4种重要度综合评估各性能状态及组件对系统的影响,有助于识别出系统最薄弱的环节,提高系统可靠性.
[Abstract]:In order to effectively characterize the reliability of the multi-state system under the condition of performance degradation, the influence of each component (state) on the system is determined, and the performance state is analyzed in this paper. A new dynamic reliability metric and a method for calculating importance under performance level constraints are proposed. Firstly, aiming at the combinatorial explosion problem of Markov process in multi-state system, the general generating function method is applied to solve the probability and reliability of each state of the system (component), and the expected value of the performance state. The performance state average expectation value and the performance state dwell time length are four dynamic reliability measures. Then, in order to meet the different performance requirements of the system (components), the impact of each group of components (states) on the system is evaluated synthetically, and the Birnbaum importance, Fussell-Vesely importance at a given performance level is defined with the performance expectation value. Risk performance value and risk reduction value. The results of a fuel supply system for a four-cylinder diesel engine show that the proposed method can not only accurately calculate the performance state probability of the system and its components (oil pump and fuel injection valve), but also the reliability and expected value of each performance state under the condition of performance degradation. The average expectation value and the length of residence time can be used to comprehensively evaluate the effects of each performance state and components on the system according to the four kinds of important degree limited by the performance level, which is helpful to identify the weakest links of the system and improve the system reliability.
【作者单位】：华南理工大学机械与汽车工程学院;江西理工大学机电工程学院;
【基金】：国家自然科学基金资助项目(61164009,61463021)~~
【分类号】：TB114.3

【参考文献】