考虑模型误差的浸出过程优化方法  

Optimization method of leaching process considering model error

作　　者：袁青云[1] 王福利[1,2] 何大阔[1,2] 张淑宁[1] 

机构地区：[1]东北大学信息科学与工程学院,沈阳110004 [2]东北大学流程工业综合自动化国家重点实验室,沈阳110004

出　　处：《控制与决策》2014年第8期1408-1412,共5页Control and Decision

基　　金：国家自然科学基金项目(61074074;61004083);国家863计划项目(2011AA060204);国家973计划子课题项目(2009CB320601)

摘　　要：由于浸出过程较为复杂,其过程模型难以准确地反映实际过程,导致基于该模型的过程优化结果不是实际最优值.基于此,提出一种考虑模型误差的浸出过程优化方法,利用高斯混合模型对浸出过程混合模型的误差均值和方差进行描述,并将其引入优化目标中.构建考虑模型误差的浸出过程优化模型,并以二阶振荡粒子群优化算法完成对优化模型的求解.最后通过仿真实验表明了所提出方法的有效性.Due to the leaching process is relatively complex, the process model is difficulty to accurately reflect the actual process, the process optimization results based on the model is not the actual optimal value. Based on that, an optimization method of leaching process considering model error is proposed. The method applies Gaussian mixture model to describe the error mean and the error variance of leaching process hybrid model which are brought into the optimization objectives to establish the optimization model of leaching process with taking the process model error into account. Then the second order oscillation particle swarm optimization algorithm is used to complete the solution of the optimization model. Finally, simulation results show the effectiveness of the proposed method.

关 键 词：浸出过程 高斯混合模型 优化模型 粒子群优化算法 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]