基于模糊LS-SVM的净化过程钴离子浓度软测量

Soft cobalt-ion concentration measurement for purification process based on fuzzy LS-SVM

针对锌湿法冶炼净化过程的复杂性,提出一种基于模糊最小二乘支持向量机方法的净化过程钴离子浓度软测量模型。考虑到不同时期以及不同区域样本数据所起的作用不一样,提出一种综合模糊加权函数,有效提高了最小二乘支持向量机的性能。以净化过程生产数据进行实验验证分析,计算结果表明:模糊最小二乘方法性能优于传统最小二乘方法,模型具有精度高、泛化性能好等特点,钴离子浓度软测量结果满足现场工艺参数要求,可以作为过程信息用于净化过程的优化控制。

Aiming at the complexity of the purification process in zinc hydrometallurgy, a cobaltion concentration soft measurement model is proposed based on fuzzy least square support vector machine （LS-SVM） . Owing to the differ- ent effect of the sample in different periods and in different sample space, a comprehensive weight function is presen- ted, the performance of LS-SVM to deal with the noisy data is improved effectively. The experimental verification a- nalysis is performed using the industrial production data from purification process. The experimental results show that the weighted FLS-SVM model performs better than the traditional LS-SVM model and has high accuracy and good generalization property. The soft measurement results satisfy the technology requirements of production and can be used as the operation optimization information for the purification process.