在基于实例推理设计系统中,为了能够灵活、快速、准确地进行实例调整,提出了基于回归分析和规则推理相结合的调整机制。以推力轴承设计为研究对象,阐述了该方法的实施过程。确定实例属性和润滑性能为自变量和因变量,利用热弹性流体动力润滑程序进行数学试验得到样本数据,并绘制曲线图。根据曲线变化趋势提取设计规则,剖析润滑机理,从而指导调整的方向;采用回归分析建立样本数据的回归模型,指导实例做定量调整。最后,将提出的调整机制在某推力轴承方案设计中进行应用,得到了预期的设计效果,从而验证了该方法的可行性和有效性。
The adaptation mechanism of combining regression analysis with rule-based reasoning was proposed in order to adapt case flexibly, effectively and accurately in case-based reasoning (CBR) design system. The employment of the method was demonstrated through its application to the thrust bearing design. Taking case attributes and lubricating properties as independent and dependent variables respectively, sample data were obtained through numerical experiment using the thermo-elasto-hydrodynamic (TEHD) program and curve graphs were drawn accordingly. To guide the adaptation direction, design rules were extracted and the lubricating mechanism was analyzed according to the drawings' tendency. And the regression models of sample data were constructed by regression analysis, which could guide the quantitative adaptation. Fi- nally, this adaptation mechanism was applied to a certain thrust bearing design. The expected result was obtained and it shows that the method is feasible and effective.