论文部分内容阅读
针对湿法炼锌净化过程中杂质离子浓度检测的大滞后特性和模型失效问题,提出了基于在线支持向量回归的离子浓度预测模型.该模型对每个新样本进行增量学习,并能删除数据集中的一个旧样本.进而提出用分块矩阵的方法解决更新算法计算复杂的问题.将该建模方法应用于离子浓度的预测,结果表明预测模型具有较好的泛化性能,模型更新时间明显缩短,有效地提高了适应工况的实时性.
Aiming at the large lag characteristic and model failure of impurity ion concentration detection in the process of purification of zinc hydrometallurgy, an ion concentration prediction model based on online support vector regression is proposed. The model performs incremental learning for each new sample and can delete the data And then a block-matrix method is proposed to solve the problem of computational complexity of the update algorithm. The proposed method is applied to the prediction of ion concentration. The results show that the prediction model has good generalization performance and the model update time is obvious Shorten and effectively improve the real-time adaptability of working conditions.