基于支持向量机的锅炉水位研究毕业论文.docVIP

支持向量机在发电厂锅炉建模的预测研究 摘要 支持向量机(Support Vector Machine ,SVM)是Corinna Cortes和Vapnik等于1995年首先提出，建立在统计学习理论的VC 维理论和结构风险最小原理基础上，根据有限的样本信息在模型的复杂性（即对特定训练样本的学习精度）和学习能力（即无错误地识别任意样本的能力）之间寻求最佳折衷，以期获得最好的推广能力。在解决小样本、非线性及模式识别中表现出许多特有的优势，并能够推广应用到拟合等其他机器学习问题中。Support vector machine in the power plant boiler modeling Abstract For the first time the Support Vector Machine (Support Vector Machine, SVM) was proposed by Corinna Cortes and Vapnik in 1995. The Support Vector Machine is an algorithm based on the theory of VC of statistical learning theory and the theory on the basis of structure of minimum risk. According to the limited sample information it can choose the best parameters between the complexity of the model that is study accuracy of the Specific training sample and learning ability (i.e. the ability of correct to recognition any sample),so as to get the best promotion ability. The purpose of this study: the Support Vector Machine algorithm to introduce into the prediction of power plant reheat steam temperature, and by the support vector machine algorithm construct the temperature of the power plant reheat steam system to control in the extent permitted ,and to predict the temperature of the power plant reheat steam system. Support vector machine (SVM) that to solve the problems of the small sample, nonlinear and high dimensional pattern recognition have many unique advantages, and it can promote the application of the function fitting in other machine learning problems. Thus the power plant boiler reheat steam problem can also be seen as an approximation problem of nonlinear function that it was influence by many influence factors, and it make greatly simplified for power plant boiler modeling problem. Keywords: Support vector machine (SVM); boiler drum of water; machine learning; statistical learning theory. 目录 绪论 5 背景及意义 5 论文的主要内容 5 第一章 统计学习理论原理及凸最优解的基本理论 7 1.1统计学习理论原理 7 1.1.1基于数据的机器学习问题建模 7 1.1.2 经验风险最小化 8 1.1.3学习过程的复杂性 8 1.1.4推广性界的问题 9 1.1.5 VC维理论 10 1.1.6结构风险最小化 11 1.2凸