支持向量机在酒精发酵过程建模中的应用  

Application of Support Vector Machine in Alcohol Fermentation Process Modeling

作　　者：成娟娟[1] 谭永红[1] 吴宏栋[1] 

机构地区：[1]桂林电子科技大学智能系统与工业控制研究室

出　　处：《酿酒科技》2008年第2期55-57,共3页Liquor-Making Science & Technology

基　　金：国家自然科学基金资助项目(60572055)

摘　　要：支持向量机(SVM)是近年来发展起来的一种基于统计学习理论,采用结构风险最小化的新型学习机器。将支持向量机应用到酒精发酵生产过程建模中,仿真结果表明,支持向量机模型是有效的,SVM模型比BP网络模型有更好的预测能力。Due to the complexity and non-linearity of microbial fermentation process, most mathematical models could not describe the behavior of biochemistry systems very well. Support vector machine （SVM） is a novel machine learning algorithm based on the statistical learning theory, which employs the principle of structural minimization. In this paper, a modeling scheme based on SVM was proposed for the alcohol fermentation process. Simulation results showed that the use of SVM method in alcohol fermentation process modeling was effective, and SVM model had better generalization ability than BP network model.

关 键 词：酒精发酵 支持向量机 模型 仿真 

分 类 号：TS262.2[轻工技术与工程—发酵工程] TS261.4[轻工技术与工程—食品科学与工程]