南京理工大学学报（自然科学版）Issue(6)：922-925,4.
基于径向基神经网络的义齿材料磨损量预测模型
Wear loss prediction model of denture material based on radial basis function neural network
摘要
Abstract
To research the wear matching of denture materials and teeth,low speed reciprocating wear tests between teeth and TC4 alloys are performed in artificial saliva with different normal loads, sliding frequencies and cycles. Taking 11 groups of test results as training samples, a wear loss prediction model for denture material is proposed based on the radial basis function neural network (RBFNN). The mean absolute error of this model is 0. 649 2 by using the 10-fold cross validation method,which verifies the correctness and rationality of the model. Dependency degree of each factor is calculated. The results show that the influences of the tooth normal load, sliding frequency and cycle on the mean absolute error are 0. 626 2,0. 628 8 and 0. 488 6 respectively.关键词
径向基神经网络/义齿/磨损量/TC4钛合金/十折交叉验证法Key words
radial basis function neural network/dentures/wear loss/TC4 alloys/ten-fold cross-vali-dation分类
机械制造引用本文复制引用
郑侃,贾修一,廖文和..基于径向基神经网络的义齿材料磨损量预测模型[J].南京理工大学学报（自然科学版）,2013,(6):922-925,4.基金项目
中央高校基本科研业务费专项资金(2012XQTR001) （2012XQTR001）
江苏省自然科学基金(BK2012402) （BK2012402）
