摘要
采用人工神经网络技术解决公差与配合选择问题，对其中相关的数据预处理等技术也作了较为深入的研究，最后以国家标准推荐的13种优先配合为例，训练得到了网络的结构参数。
It's long time to develop machinery industry,there is a lot of successful experiences come from each fields.It is an important problem to help designers selecting fits and tolerances in current design according to successful samples.This paper discusses how to select fits and tolerances by means of ANN,and profoundly strdies some related datd pretreatment etc.Lastly,the parameters of the BP network are acquired by training based on 13 GB recommendatory precedence fittings.
出处
《机械制造》
北大核心
2001年第5期13-15,共3页
Machinery
