摘要
针对复杂系统可靠性分配中的关键问题和难点,提出了一种基于神经网络的采用逆向思维进行复杂系统可靠性分配的方法。利用误差反向传播的改进算法,以系统可靠度和各子系统自身约束条件为网络输入,以各子系统可靠度相互比例为网络输出,对样本数据进行训练。借用神经网络中的权值和阈值,反映了不同系统可靠度及对应各子系统自身约束条件下,各个子系统可靠性之间的相互关系,从而得到子系统可靠性分配权重,实现对复杂系统可靠性进行精确分配。
For solving the key problems and puzzles in complex system reliability allocation, a reliability allocation method based on the neural network and converse thinking is proposed. Using the improved error backward propagation, the system reliability and the self-restrain conditions of every sub-system are taken as the input of neural network; the mutual proportion of every sub-system＇s reliability is taken as the output of neural network. The neural network is trained by sample data. The interrelations of every sub-system＇s reliability in the conditions of certain system reliability and self-restrain conditions are reflected by the meanings of the neural network weight and threshold. Then the reliability allocation weight of every sub-system can be achieved, and the complex system reliability can be allocated accurately.
出处
《机械》
2009年第5期5-8,18,共5页
Machinery
基金
国家自然科学基金项目(50875021)
关键词
神经网络
可靠性分配
复杂系统
算法
逆向思维
neural network
reliability allocation
complex system
algorithm
converse thinking
