摘要
为了实现资源的深度利用和确保生产过程的平稳运行,有色冶金过程常采用多个大型级联反应槽/器进行生产,给全流程的优化与控制带来了巨大挑战.本文针对具有级联反应器的有色冶金过程,提出一种分层协调优化与控制框架,包括生产规划层、全流程多目标协调优化层和局部优化层的分层协调优化方法,通过协调优化层与局部优化层的递阶协调,实现全流程各反应槽输出设定值的优化;针对不同采样频率的被控量,分别设计了模型预测控制器和基于T-S模糊逻辑的模糊控制,构建误差补偿数据驱动模型对模型进行校正,以实现设定点的跟踪控制;为提高优化控制方法性能,提出了自适应前馈校正方法,根据检测反馈值,优化后续反应槽的输出设定,使生产流程在运行工况迁移后仍能优化运行.通过锌矿物浮选过程的案例研究,验证了所提协调优化与控制方法的有效性,提高了精矿品位预测精度,实现了流程的优化运行.
In order to achieve full utilization of resources and ensure the smooth operation of the production process,non-ferrous metallurgical processes often employ multiple cascaded large-scale tanks/reactors for production.This brings significant challenges to optimizing and controlling the plant-wide process.This article proposes a hierarchical coordinated optimization and control framework for nonferrous metallurgical processes with cascaded reactors.For the optimization of set-points for various reactors,a hierarchical coordinated optimization method is designed.This method comprises a global multi-objective coordinated optimization layer and local optimization layers.By employing a hierarchical coordination mechanism between the global coordinated layer and the local optimization layers,it enables the achievement of optimal operation for the plant-wide process.For the different controlled variables with different sampling frequencies,model predictive controllers and fuzzy control based on the T-S fuzzy logic are designed.A datadriven error compensation model is created to calibrate the models for achieving better set-point tracking control performance.To enhance the performance of the optimization control,an adaptive feedforward correction method is proposed.It optimizes the setpoints of subsequent reactors based on measured feedback values,enabling the production process to continue optimal operation even after operation condition changes.Through a case study of the zinc flotation process,the effectiveness of the proposed coordinated optimization and control methods is validated.It improves the accuracy of predicting concentrate grade and accomplishes the optimal operation of the process.
作者
谢世文
谢永芳
黄廷文
XIE ShiWen;XIE YongFang;HUANG TingWen(School of Automation,Central South University,Changsha 410083,China;Texas A&M University at Qatar,Doha 23874,Qatar)
出处
《中国科学：技术科学》
EI
CSCD
北大核心
2024年第9期1778-1788,共11页
Scientia Sinica(Technologica)
基金
国家自然科学基金(批准号:62233018,62373381)
湖南省自然科学基金杰出青年基金(编号:2023JJ10079)资助项目。
关键词
分层协调优化
模型预测控制
T-S模糊控制
自适应前馈校正
级联有色冶金过程
hierarchical coordinated optimization
model predictive control
T-S fuzzy control
adaptive feedback correction,cascaded non-ferrous metallurgical processes
