基于支持向量机的加工误差预测建模方法研究  

Support Vector Machines-Based Method for Machining Error Prediction Modeling

作　　者：江亮[1] 刘健[1] 潘双夏[1] 

机构地区：[1]浙江大学机械设计研究所,杭州310027

出　　处：《组合机床与自动化加工技术》2005年第8期13-15,共3页Modular Machine Tool & Automatic Manufacturing Technique

基　　金：国家863计划项目(2002AA421110);浙江省重大科技项目(021101566)

摘　　要：误差补偿是提高机械加工精度的有效途径,其中误差的建模是关键。在分析现有加工误差预测技术不足的基础上,提出基于支持向量机的加工误差回归建模和预测方法,并对实际应用中的问题进行了分析和总结。通过实例验证及与其它建模方法的对比,表明该方法具有优良的预测性能,为加工误差预测提供了一种新的可行方法。Error compensation is an effective approach to enhance machining precision, and error modeling is the key factor for the compensation. Firstly, the present machining error prediction technologies and their disadvantages were analyzed and compared. Then support vector machines（ SVM）-based method of machining error modeling and prediction was put forward. Also some problems in real application of the method were analyzed and summarized, Finally, a real application example and comparison with other modeling methods was given,which indicates that the proposed method has an excellent prediction performance and is a new feasible approach for machining error prediction,

关 键 词：支持向量机 误差建模 误差预测 参数选择 

分 类 号：TH161.5[机械工程—机械制造及自动化]