神经网络求解隐蔽性故障率  

Solution to Hidden Fault Probability with Neural Network

作　　者：倪爱伟[1] 谢里阳[1] 翁刚[1] 吴克勤[1] 闫骏[1] 

机构地区：[1]东北大学机械工程与自动化学院,沈阳110004

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)

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.

关 键 词：隐蔽性故障 模型 神经网络 参数估计 

分 类 号：TP391.75[自动化与计算机技术—计算机应用技术]