吴良,陈铮

• 1:东华大学机械工程学院
• 2:东华大学机械工程学院 上海201620
• 3:上海201620

摘要(Abstract)：

介绍了支持向量机(SVM)技术中的支持向量回归模型,并结合实例运用SVM技术构建了42CrMo钢热处理力学性能预测的数学模型。研究表明,在小样本条件下,应用SVM技术构建数学模型的最大预测相对误差为4.78%;而且随着检验精度的提高,模型的预测精度保持基本不变,泛化能力明显优于用人工神经网络的BP模型。认为在材料热处理领域应用SVM技术构建预测力学性能的数学模型,能较好地解决小样本和模型预测精度间的矛盾。

关键词(KeyWords)： 支持向量机(SVM);预测模型;热处理;泛化能力

Abstract:

Keywords:

基金项目(Foundation):

作者(Author): 吴良,陈铮

参考文献(References)：

• [1]潘健生,张伟民,陈乃录,等.热处理信息化若干问题的思考[J].金属热处理,2004,29(1):13-16.PAN Jian-sheng,ZHANG Wei-min,CHEN Nai-lu,et al.The issues of information technique of heat treatment[J].Heat Treatment of Metals,2004,29(1):13-16.
• [2]Dobrzanski L A,Sitek W.Application of a neural network in modelling of hardenability of constructional steels[J].Journal of Materials ProcessingTechnology,1998,78(1-3):59-66.
• [3]Vermeulen W G,Van der Wolk P J,Weijer A P de,et al.Prediction of jominy hardness profiles of steels using artificial neural networks[J].Journal ofMaterials Engineering and Performance,1996,5(1):30-36.
• [4]Wang J,Van der Wolk P J,Dorrepaal C,et al.Effects of carbon and cooling rate on continuous cooling transformations predicted by artificial network[J].ISIJ International,1999,39(10):1038.
• [5]吴良.提高BP人工神经网络热处理工艺及性能预测模型泛化能力的研究[J].金属热处理,2003,28(5):42-45.WU Liang.How to improve the generalized ability of ANN-based BP model used to predict technologies of heat treatment and properties of materials[J].HeatTreatment of Metals,2003,28(1):42-45.
• [6]Vapnik V.The nature of statistical learning theory[M].New York:Springer Verlag,1995.
• [7]Nello Cristianini,John Shawe-Taylor.支持向量机导论[M].北京:电子工业出版社,2004.
• [8]Ming Ge,Du R,Zhang Guicai,et al.Fault diagnosis using support vector machine with an application in sheet metal stamping operations[J].MechanicalSystems and Signal Processing,2004,18:143-159.
• [9]陈念贻,陆文聪,叶晨洲,等.支持向量机及其他核函数算法在化学计量中的应用[J].计算机与应用化学,2002,19(6):691-696.CHEN Nian-yi,LU Wen-cong,YE Chen-zhou,et al.Application of support vector machine and kernel function in chemometrics[J].Computers and AppliedChemistry,2002,19(6):691-696.
• [10]邓乃扬,田英杰.数据挖掘中的新方法——支持向量机[M].北京:科学出版社,2005.
• [11]王泽华,鲍国栋,等.热处理工艺参数对42CrMo钢电渣熔铸件力学性能的影响[J].机械工程材料,2005,29(2):13-16.WANG Ze-hua,BAO Guo-dong,et al.Influence of heat treatment processes on mechanical properties of 42CrMo electro-slag castings[J].Materials forMechanical Engineering,2005,29(2):13-16.

文章评论(Comment)：