湿法冶金草酸钴粒度分布混合建模方法

On the Hybrid Modeling Method of Cobalt Oxalate Grain Size Distribution in Hydrometallurgy Synthesis Process

由于缺乏有效的检验手段,无法实现湿法冶金合成过程草酸钴粒度分布的实时检测.提出了一种支持向量机(SVM)与合成过程动态机理模型相结合的混合建模方法,用以预报草酸钴的粒度分布.通过分析合成过程机理,利用粒数衡算、物料衡算关系建立动态机理模型;利用支持向量机来对不能用机理模型表现的合成过程的动态特性进行误差补偿.仿真分析验证了该方法相比于机理模型的有效性,将其应用到实际生产过程中,取得了满意的效果.

Due to lack of efficient measuring means,the real-time measurement is unattainable to the grain size distribution of cobalt oxalate in its hydrometallurgical synthesis process. A hybrid modeling method of the distribution was therefore proposed combining SVM with the dynamic mechanism model of the synthesis process to predict the grain size distribution. Analyzing the mechanism of synthesis process,the population balance equation and material balance were used to develop a dynamic model for the distribution,with SVM used to compensate for the error resulting from the dynamic characteristic in synthesis process,which could not be revealed by the dynamic mechanism model. Simulation analysis was verified that hybrid method is more effective in comparison with the dynamic mechanism model. The application of the proposed method to a practical industrial process showed its satisfactory results.