Show simple item record

dc.contributor.author王舒阳
dc.contributor.author姚斌
dc.contributor.author姚博世
dc.contributor.author冯伟
dc.contributor.author何昱超
dc.contributor.author曹新城
dc.date.accessioned2018-11-26T08:20:32Z
dc.date.available2018-11-26T08:20:32Z
dc.date.issued2017
dc.identifier.citation工具技术,2017,(5)
dc.identifier.issn1000-7008
dc.identifier.other10.3969/j.issn.1000-7008.2017.05.012
dc.identifier.urihttps://dspace.xmu.edu.cn/handle/2288/165027
dc.description.abstract主轴热变形是影响数控机床加工精度的主要因素。为提高主轴热误差的预测精度,提出了基于信息粒化支持向量机(SVM)的主轴热误差综合预测模型。使用信息粒化方法对采样温度数据与主轴热误差数据进行预处理,分别建立基于SVM的主轴热误差的回归预测模型和时间序列模型,通过计算两个模型权重系数,最终建立主轴热误差综合预测模型。以2MZK7150五轴数控可转位刀片工具磨床为研究对象,实验表明,较之于单一模型该模型具有良好的泛化能力和较高建模精度。
dc.description.abstractSpindle thermal deformation has an extraordinary influence on the NC machining precise. Aiming to the im- provement of the spindle thermal deformation predict ability, the spindle thermal error comprehensive predict model is pro- posed on the basis of the information granulation and support vector machine(SVM). Information granulation method is em- ployed to pretreat the sampling temperature and spindle thermal error. The regression prediction model and time series mod- el of spindle thermal error are also carried out based on the SVM. Finally, the spindle thermal error comprehensive predict model is established using the weight coefficients of the two models. The 2MZK7150 five-axis NC indexable insert tool grinding lathe machining experiment shows that the established model has better generalization ability and higher modeling precision compared with the unitary model.
dc.description.sponsorship陕西省科技统筹创新工程计划(2014KTD201-04-03)
dc.language.isozh_CN
dc.subject主轴热误差
dc.subject信息粒化
dc.subject支持向量机
dc.subject预测模型
dc.subjectprediction model
dc.title基于信息粒化支持向量机的主轴热误差综合预测模型
dc.title.alternativeSpindle Thermal Error Comprehensive Prediction Model Based on Information Granulation and SVM
dc.typeArticle


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record