基于支持向量机的数据建模在软测量建模中的应用

支持向量机是一种新的机器学习算法 ,它的基础是 Vapnik创建的统计学习理论。与传统学习方法相比 ,该理论采用了结构风险最小化准则 ,在最小化样本点误差的同时缩小模型泛化误差的上界 ,提高了模型的泛化能力。同时该理论把机器学习问题转化为一个二次规划问题 ,可以得到唯一的全局最优解。本文应用支持向量机技术进行数据建模研究 ,并用数据建模技术建立了加氢裂化装置分馏塔的航煤干点的软测量模型。

Data Modeling Based on Support Vector Machine and Its Application to Soft Sensing

Support vector machine is a new type of machine learning algorithm based on statistic learning theory presented by Vapnik. Compared with conventional learning algorithms, SLT enhances the generalization ability of the models by employing structural risk minimization criterion to minimize the errors at the samples and decrease simultaneously the upper bound of the predict error of the models. Furthermore, the global optimal solution can be uniquely obtained owing to that SLT converts machine learning into quadratic programming. The data modeling technique based on SVM is presented in this paper, and it has been successfully applied to model jet fuel endpoint estimator for a fractionator of hydrocracking unit.