煤炭科技2025，Vol.46Issue(4)：29-34,6.DOI:10.19896/j.cnki.mtkj.2025.04.006
强化学习驱动的重介分选密度优化控制研究与应用
Research and application of reinforcement learning driven density optimization control in dense-medium sorting
宋万军 1白龙1
作者信息
- 1. 国家能源集团国神公司 上榆泉煤矿选煤厂,山西 河曲 036500
- 折叠
摘要
Abstract
In-depth research and analysis on density circuit process and related characteristics of dense-medium sorting were conduc-ted.An optimization control method based on online model free reinforcement learning was proposed,making the density control system of dense-medium separation suspension asymptotically stable and tracked the set value of suspension density online,and improved effi-ciency and accuracy of dense-medium sorting.Meanwhile,the optimization control method was simulated and validated using MATLAB simulation experiments.The results indicate that this method has precise control effect.关键词
重介分选过程/强化学习/策略迭代/优化控制/最优性能指标Key words
dense-medium sorting process/reinforcement learning/strategy iteration/optimization control/optimal performance indicators分类
矿业与冶金引用本文复制引用
宋万军,白龙..强化学习驱动的重介分选密度优化控制研究与应用[J].煤炭科技,2025,46(4):29-34,6.
