基于支持向量机的蒙脱土悬浮液电动电位预测

Prediction of montmorillonite suspensions zeta-potential based on support vector machine

应用多元线性回归、人工神经网络、支持向量机3种方法,对加入聚乙二醇、十二烷基苯磺酸钠、石油磺酸盐和部分水解聚丙烯酰胺四种处理剂的蒙脱土悬浮液的电动电位进行预测。在模型训练中,分别采用了神经网络集成和非启发式参数优化来提高人工神经网络和支持向量机模型的泛化能力。检验结果表明,参数优化的支持向量机模型预测精度最高,其平均误差率为3.88%,最大误差率为7.55%。

Three models were used to research four kinds of agents（PEG,SDBS,PS and HPAM） effecting on zeta-potential of montmorillonite suspensions,which were based on multiple linear regression（MLR）,artificial neural network（ANN） and support vector machine（SVM）.The methods of neural network ensemble and parameter optimization were used to improve the generalization ability of models of ANN and SVM in the training. The results indicate that the parameter optimization SVM model is the most accurate model.Its average error rate is 3.88%and maximum error rate is 7.55%.