论文部分内容阅读
磨矿分级过程的控制目标是将一、二级溢流浓度和细度稳定控制在质量指标区间内.磨矿分级过程是胖系统,完成动态优化目标后控制器仍有剩余自由度,因此需考虑局部稳态经济优化.针对这一目标,提出一种考虑局部稳态经济目标的多模型预测控制方案.首先,建立了基于现场数据库的球磨机和分级机传函矩阵模型;然后考虑局部经济性能,将稳态经济目标以罚函数形式嵌入动态优化目标函数;为消除球磨机换球引起的模型失配的影响,建立了一种基于换球规律的多模型切换策略.仿真结果表明了所提出控制方案的有效性.
The control objective of the grinding and classification process is to control the concentration and fineness of the primary and secondary overflows within the quality index interval. The grinding and classification process is a fat system, and the controller still has remaining degree of freedom after the dynamic optimization goal is completed Considering the local steady-state economic optimization, aiming at this goal, a multi-model predictive control scheme considering local steady-state economic goals is proposed.Firstly, a ball mill and classifier transfer matrix model based on the field database is established, and then the local economic performance , The steady-state economic objective is embedded into the dynamic optimization objective function in the form of penalty function. A multi-model switching strategy based on the ball-changing rule is established to eliminate the influence of model mismatch caused by the ball change in the ball mill. The simulation results show that the proposed control The effectiveness of the program.