论文部分内容阅读
根据轴承－转子系统摩擦学设计的需要，研究了从大量仿真计算数据中获取设计知识的方法，并采用自组织特征映射神经网络，对输入样本进行了概念聚类，获得了概念性的设计知识，且利用计算机自动搜索算法获得产生式规则知识．
According to the needs of tribological design of bearing-rotor system, the method of acquiring design knowledge from a large amount of simulation data is studied. The self-organizing feature mapping neural network is used to carry out the concept clustering of the input samples and get the conceptual design knowledge , And using computer automatic search algorithm to obtain the rules of production rules.