论文部分内容阅读
基于乏信息失效数据,提出了机械产品可靠性的最大熵评估模型.根据可靠性经验值公式,获得失效数据的可靠性经验值向量,并逆推出离散失效频率向量即获得统计直方图;基于区间映射的牛顿迭代方法获得具有最大熵的概率密度函数,对其积分获得失效概率分布函数,进而得到可靠性估计真值函数.仿真案例和试验案例研究证明该方法可以很好地评估已知分布的可靠性并有效地解决只有失效数据而没有概率分布任何先验信息的可靠性评估问题.在寿命给定时,最大熵方法获得的可靠性取值与已知分布获得的可靠性取值之间的差值非常小仅为3.40%.
Based on the failure information of lacking information, a maximum entropy evaluation model of mechanical product reliability is proposed.According to reliability empirical formula, the reliability empirical value vector of failure data is obtained, and the discrete failure frequency vector is inversely derived to obtain a statistical histogram. Based on the interval And the probability density function with the maximum entropy is obtained by the Newton iterative method of mapping, and then the failure probability distribution function is obtained by integrating the probability function, and then the reliability estimation truth function is obtained. Simulation results and experimental case studies show that this method can well evaluate the distribution of the known distribution Reliability and effective solution to the problem of reliability assessment of a data set without any priori probability distribution only with failure data.When the life is given, the reliability value obtained by the maximum entropy method and the reliability value obtained by the known distribution The difference is very small at only 3.40%.