摘要
基于共因失效理论的隐蔽性故障率模型,修正了RCMⅡ(Reliability Centered MaintenanceⅡ)模型假设条件的缺陷。在新模型计算中,进行参数估计所建立的一、二阶原点距方程组存在复杂的非线性关系,用一般的数学方法求解较困难。通过参数变换,应用BP神经网络模型和自适应线性神经网络模型,得到较为精确的解值,并确定其分布函数。
With the model of hidden fault probability based on the theory of CCF （common cause failure）, the errors in the assumed conditions of the model of RCM Ⅱ （Reliability Centered Maintenance Ⅱ） are corrected. In the calculation of the new model, there is a complicated non - linear connection existing in establishment of the ranks 1 and 2 origin distance equations for parameter assessment and it is difficult to obtain a solution with the common mathematical methods. The problem is easily solved with the aid of BP and Adaline neural networks via parameter transformation, As a result, an accurate solution is obtained and its distribution function determined.
出处
《机械制造》
2007年第6期4-6,共3页
Machinery
基金
863项目"复杂机械装备可靠性设计与评价"(编号:2006AA04Z408)
国家973重大基础研究发展计划(编号:2006CB605005)
关键词
隐蔽性故障
模型
神经网络
参数估计
Hidden Fault Probability Model Neural Network Parameter Assessment
