共载冗余系统可靠性分析与优化设计

发布时间：2018-05-11 11:19

本文选题：共载冗余 + 可靠性建模　；参考：《西北工业大学》2015年博士论文

【摘要】：冗余技术是工程领域中为达到系统的可靠性目标而经常采用的主要手段之一，如何准确的预计、评估、分析冗余系统的可靠性便成为冗余设计的重中之重。传统的冗余系统可靠性理论忽略了系统各单元寿命之间的相互影响，认为其寿命是相互独立的，这种理论在分析共载冗余系统时必然会带来比较大的偏差，无法满足工程需要。目前共载冗余系统可靠性理论尚处于发展阶段，对于工程中实际存在的共载冗余系统，尚有许多问题亟待解决。为此，基于共载冗余系统可靠性分析与优化设计的工程需求，本文对共载冗余系统包含控制装置、系统中单元退化失效服从随机过程、系统包含共因失效等情况下的可靠性理论进行研究，并分析了包含共载冗余分系统的串-并联系统的优化设计问题。论文主要内容与结论包括如下几个方面： 1.建立了考虑控制装置的共载冗余系统的可靠性模型 (1)针对不可修系统各单元寿命服从指数分布的情况，对于控制装置具有恒定失效率和恒定可靠度的情况，根据齐次马尔可夫过程特性分别给出了可靠性模型表达式，，对于控制装置服从一般寿命分布的情形，建立了系统可靠性的Monte Carlo仿真算法； (2)对于不可修系统各单元寿命服从一般分布的情形，运用事件空间法，根据全概率理论、条件概率理论和等效累积失效概率理论，建立了系统可靠性模型，并给出了Monte Carlo仿真算法；通过在某排水系统算例的应用验证了上述两个模型的合理性和仿真算法的有效性； (3)对于各单元寿命和维修时间均服从指数分布的可修系统，绘出系统状态转移图，根据马尔可夫链数值计算方法计算系统的瞬态可靠性特征量，通过在某电压源换流阀系统算例的应用验证了模型的合理性。 2.建立了单元退化失效为随机过程的共载冗余系统的可靠性模型分别针对单元退化失效服从Gamma过程或Wiener过程的系统，根据Gamma过程或Wiener过程的独立增量特性，运用事件空间法，分别建立了系统可靠性模型，并给出了系统可靠性的Monte Carlo仿真算法。最后，分别以三角带传动系统和风机通风系统为例，验证了所提出模型的合理性和仿真算法的有效性。 3.建立了考虑共因失效的共载冗余系统的可靠性模型 (1)针对系统各单元寿命服从指数分布的情况，基于β因子模型，根据齐次马尔可夫过程特性给出了可靠性模型的表达式；基于二项失效率(BFR)模型，绘出系统状态转移图，通过马尔可夫链数值计算方法求解系统可靠性； (2)针对各单元寿命服从一般分布的情形，基于β因子模型，根据事件空间法给出了可靠性模型的表达式，并给出Monte Carlo仿真算法；基于二项失效率(BFR)模型，给出了系统可靠性的Monte Carlo仿真算法；最后以某锂离子电池组成的电源系统为例，分别验证了所提模型的合理性和仿真算法的有效性。 4.对包含共载冗余分系统的系统进行可靠性优化设计 (1)根据对多个时刻可靠度要求建立了串-并联系统冗余分配优化模型，包括分系统中部件同类型的串-并联系统和分系统中部件不同类型的串-并联系统； (2)根据共载冗余分系统可靠性模型的特点，对1阶邻域启发式算法和遗传算法进行了研究，然后通过算例分析表明了优化模型的全面性及优化算法的有效性。
[Abstract]:Redundancy technology is one of the main methods used in the field of engineering to achieve the reliability goal of the system. How to accurately predict, evaluate, and analyze the reliability of redundant systems is the most important part of redundant design. The traditional redundancy system reliability theory neglects the interaction between the life of each unit of the system and considers the life span of the system. It is independent of each other. This theory will inevitably bring large deviations in the analysis of CO loaded redundant systems and can not meet the needs of engineering. At present, the reliability theory of the common redundant system is still at the stage of development. There are still many questions to be solved for the actual redundant systems in the project. For the engineering requirements of the analysis and optimization design, this paper studies the reliability theory of the co loaded redundancy system including the control device, the unit degradation failure obeys the random process, the system includes the common cause failure and so on, and analyzes the optimization design of the series parallel system including the co loaded redundancy system. The main contents of the paper are the main contents of the paper. And the conclusion includes the following aspects:
1. a reliability model of a co loading redundant system considering control devices is established.
(1) for the condition that the life of each unit of the inpairable system obeys the exponential distribution, the reliability model expression is given according to the homogeneous Markov process characteristic for the control device with constant failure rate and constant reliability. The Monte Carl of the system reliability is established for the control device obeying the general life distribution. O simulation algorithm;
(2) the system reliability model is established by using the event space method, according to the full probability theory, the conditional probability theory and the equivalent cumulative failure probability theory by using the event space method, and the Monte Carlo simulation algorithm is given by using the event space method, and the above two models are verified by the application of a drainage system example. The rationality of the algorithm and the effectiveness of the simulation algorithm.
(3) for the repairable system with exponential distribution of the life and maintenance time of each unit, the state transfer diagram of the system is drawn, and the transient reliability characteristic of the system is calculated according to the Markov chain numerical calculation method, and the rationality of the model is verified by the application of a voltage source converter valve system.
2. the reliability model of a co loading redundant system with unit degradation failure as a stochastic process is established.
According to the system of Gamma process or Wiener process, according to the independent increment characteristic of the Gamma process or the Wiener process, the system reliability model is established by using the event space method, and the Monte Carlo simulation algorithm of the system reliability is given. Finally, the triangle belt transmission system and the fan ventilation system are respectively used. As an example, the rationality of the proposed model and the effectiveness of the simulation algorithm are verified.
3. a reliability model of a co loading redundant system considering common cause failure is established.
(1) in view of the exponential distribution of the life of each unit of the system, based on the beta factor model, the expression of the reliability model is given according to the properties of the homogeneous Markov process. Based on the two term BFR model, the system state transfer graph is drawn, and the system reliability is solved by the Markov chain number calculation method.
(2) on the basis of the beta factor model, the expression of the reliability model is given based on the event space method, and the Monte Carlo simulation algorithm is given based on the beta factor model, and the Monte Carlo simulation algorithm for the system reliability is given based on the two term loss of efficiency (BFR) model.
Finally, a power system consisting of a lithium ion battery is taken as an example to verify the rationality of the proposed model and the effectiveness of the simulation algorithm.
4. reliability optimization design of the system including the redundant redundant subsystems.
(1) the redundant allocation optimization model of series parallel system is established according to the reliability requirements for multiple time, including the series parallel system in the same type of components in the subsystem and the series parallel system of different types of components in the subsystem.
(2) according to the characteristics of the reliability model of the shared redundant subsystem, the 1 order neighborhood heuristic algorithm and the genetic algorithm are studied. Then the optimization model is proved to be comprehensive and the optimization algorithm is effective by example analysis.

【学位授予单位】：西北工业大学
【学位级别】：博士
【学位授予年份】：2015
【分类号】：TB114.3

【参考文献】