湿法冶金中预测金产量的混合建模方法

由于缺乏有效的检测手段，无法实现湿法冶金全过程金产量的精确在线检测.提出了一种最小二乘支持向量机(LS-SVM)与湿法冶金全过程静态机理模型相结合的混合建模方法，用以预测金的产量.通过对湿法冶金全过程的机理分析，利用物料衡算关系建立金的产量静态机理模型，利用LS-SVM对机理模型不能描述的过程特性进行误差补偿.仿真结果表明，该方法的预测性能优于机理模型和单独的LS-SVM构建的模型，验证了该方法的有效性.

Hybrid Modeling Method of Forecasting Gold Production in the Hydrometallurgy Process

Due to lack of efficient measuring means, it is unable to realize the accurate on-line measurement of gold production in the whole hydrometallurgy process. Therefore a hybrid modeling method was proposed, which combined LS-SVM with the static mechanism model of the whole hydrometallurgy process to predict the gold production. By analyzing the mechanism of the whole hydrometallurgy process, the material balance was used to establish the static mechanism model of the gold production, and LS-SVM was applied to compensate for the error resulting from the process characteristic, which could not be described by the mechanism model. Simulation results showed that the forecasting performance of the proposed method is better than the mechanism model and the model established only by LS-SVM, and verified that the proposed method is effective.