测量误差分析及数据处理若干要点系列论文(四)——统计学习理论及支持向量机方法统用于形位误差评定  

Some Key Points of Measurement Error Analysis and Data Processing Series Papers(4)--Statistical Learning Theory and Support Vector Machine Method Used for the Evaluation of the Shape and Position Errors

作　　者：林洪桦[1] Lin Honghua(Beijing Institute of Technology,Beijing 100081,China)

机构地区：[1]北京理工大学,北京100081

出　　处：《自动化与信息工程》2020年第4期1-5,共5页Automation & Information Engineering

摘　　要：阐述统计学习理论(SLT)及支持向量机(SVM)方法统用于形位误差评定。首先,简要复述SLT-SVM方法及特点;然后,详细描述多测点形位误差评定要点与难点、统用SVM方法基本依据及评定算法等;最后,验证现代数据处理对策四要诀:实、佳、智、验。This paper expounds the statistical learning theory and the support vector machine method used for the evaluation of the shape and position errors.First of all,the SLT-SVM method and its characteristics are sketched;then,the key points and difficulties of the shape and position errors evaluation for multiple measurement points are detailed described,as well as the basis and the evaluation algorithms of the unified SVM method;finally,the four key points of modern data processing countermeasures are verified:real,good,intelligent and empirical.

关 键 词：数据处理 数学模型 误差 统示法 

分 类 号：TP274[自动化与计算机技术—检测技术与自动化装置]