基于QGA-LSSVM的醋酸乙烯聚合率软测量建模

Research on Soft Sensor Modeling of Vinyl Acetate Polymerization Rate Based on Hybrid QGA-LSSVM

针对最小二乘支持向量机(LS-SVM)在建立醋酸乙烯(VAC)聚合率软测量模型过程中最优模型参数的选择问题,提出了利用一种量子遗传算法来自动选取LS-SVM模型正则化参数和核函数参数的方法;把LS-SVM模型参数的选择问题转化为优化问题,利用全局搜索能力强的量子遗传算法优化LS-SVM建模过程的重要参数,建立了基于QGA-LSSVM方法的VAC聚合率软测量模型;仿真结果表明:与已有的神经网络和支持向量机软测量方法相比,该模型泛化能力强,精度高,更有利于醋酸乙烯聚合率测量工程实际运用。

An quantum genetic algorithm （QGA） was proposed to ow：rcome the disadvantage that it＇ s difficult to get better parameter values of least squares support vector machine （LS--SVM） and the mixed kernel function in the processing of establish the soft sensing of vinyl acetate （VAC） polymerization rate. The method can convert the LS--SVM model parameters of selection into optimization problem, the best parameters of LS--SVM would be selected by QGA which has the ability of better search, and the QGA--LSSVM mode about soft sensing of VAC polymerization rate was constructed. The simulation result indicated that compared with the methods based on neural network and support vector machine, the QGA--LSSVM model has more effective generation performance and high precision, and it is more conducive to the practical application of engineering measurements.