论文部分内容阅读
系统关键故障的发生,会导致系统处于各种离散性能降级状态。针对传统的基于马尔可夫链蒙特卡罗(Markovchain Monte Carlo,MCMC)的自适应重要抽样法只适用于连续变量系统的不足,提出考虑混合变量的基于MCMC的自适应重要抽样法,以支持系统性能可靠性的高效仿真。该方法首先将由关键故障产生的不同失效域组成失效空间,并通过初始样本点在失效空间中随机游走构造马尔可夫链模拟样本;然后综合考虑连续变量与离散变量,利用核密度估计构建混合核抽样密度函数;再根据该密度函数进行重要抽样仿真并计算系统的性能可靠性;最后对该方法的仿真效率进行理论分析。通过电液舵机(Electro-Hydrostatic Actuator,EHA)案例对方法的正确性和仿真效率进行验证。
The occurrence of critical system failures will result in the system being in a state of discrete performance degradation. Aiming at the shortcomings of the traditional Markovchain Monte Carlo (MCMC) -based adaptive significant sampling method only for continuous variable systems, an adaptive MCMC-based adaptive significant sampling method considering mixed variables is proposed to support the system Efficient simulation of performance and reliability. In this method, different failure domains generated by the key failure are firstly constructed into the failure space, and the Markov chain simulation samples are constructed by random walk of the initial sample points in the failure space. Then, considering the continuous variables and the discrete variables, the kernel density estimation is used to construct a mixture Nuclear sampling density function; then according to the density function of important sampling and simulation system performance and reliability; and finally the theoretical analysis of the simulation efficiency of the method. The correctness of the method and the simulation efficiency were verified by the Electro-Hydrostatic Actuator (EHA) case.