一种新的氧化铝质量分数建模与控制策略

阎纲^{1, 2}，梁昔明^{2, 3}

(1. 湖南财政经济学院 信息管理系，湖南 长沙，410205；
2. 中南大学 信息科学与工程学院，湖南 长沙，410083；
3. 北京建筑工程学院 理学院，北京，100044)

摘 要：针对氧化铝质量分数的建模与控制问题，提出一种新的基于最小二乘支持向量机(LS-SVM)和预测控制的建模与控制策略。首先，针对LS-SVM建模时的参数选取问题，提出一种基于混沌优化的CHAOS LS-SVM算法获得最优氧化铝质量分数预测模型。然后，提出一种基于LS-SVM的氧化铝质量分数预测控制算法，采用混沌优化在线求解最优控制律。仿真结果表明：CHAOS LS-SVM算法建立的氧化铝质量分数预测模型，其泛化能力要比基于神经网络(NN)的氧化铝质量分数预测模型的强；基于LS-SVM的氧化铝质量分数预测控制算法，其控制精度要比基于NN的氧化铝质量分数预测控制算法的高。

关键词：氧化铝质量分数；最小二乘支持向量机；预测控制；混沌优化

A novel model and control strategy for alumina concentration

YAN Gang^{1, 2}, LIANG Xi-ming^{2, 3}

(1. Department of Information Management, Hunan University of Finance and Economics, Changsha 410205, China;
2. School of Information Science and Engineering, Central South University, Changsha 410083, China;
3. School of Science，Beijing University of Civil Engineering and Architecture, Beijing 100044, China)

Abstract:Considering the problem of alumina concentration modeling and control, a novel modeling and control strategy based on least squares support vector machine (LS-SVM) and predictive control were proposed. First, aiming at the problem of parameter selection of LS-SVM, a CHAOS LS-SVM algorithm based on chaos optimization was presented to obtain optimal alumina concentration prediction model. Then, an alumina concentration predictive control algorithm based on LS-SVM was developed, which uses chaos optimization to solve optimal control law online. The simulation results show that the generalization ability of alumina concentration prediction model established by CHAOS LS-SVM algorithm is stronger than that of alumina concentration prediction model based on neural network (NN), and the control precision of alumina concentration predictive control algorithm based on LS-SVM is higher than that of alumina concentration predictive control algorithm based on NN.

Key words:alumina concentration; least squares support vector machine (LS-SVM); predictive control; chaos optimization