论文部分内容阅读
最小二乘支持向量机(least squares support vector machine,LSSVM)为一种遵循结构风险最小化原则的核函数学习机器,其训练仅需求解一个线性方程组,且超参数较标准支持向量机更少。由于其实现简单且预测效果良好,近年来在化学、化工领域的应用日益广泛。本文研究了基于LSSVM的软测量建模过程中的数据预处理和优选超参数等问题。并将其应用于常压塔塔顶汽油干点的软测量建模。计算结果表明,其预测精度能够满足生产实际要求,是一种简单有效的非线性软测量建模工具。
The least squares support vector machine (LSSVM) is a kernel function learning machine that follows the principle of structural risk minimization. Its training only needs to solve a system of linear equations and has fewer superelements than standard support vector machines . Due to its simple implementation and good predictive effect, it has been widely used in chemical and chemical fields in recent years. In this paper, the problems of data preprocessing and optimal hyperparameters in soft sensor modeling based on LSSVM are studied. And it is applied to the soft sensor modeling of the dry point of the top of the atmospheric tower overhead gas. The calculation results show that the prediction accuracy can meet the actual production requirements and is a simple and effective nonlinear soft sensor modeling tool.