摘要
为了解决磨矿过程这一典型的复杂工业过程的优化控制问题,基于某选矿厂的两段全闭路磨矿流程展开研究,提出了基于模糊规则和案例推理相结合的磨矿专家系统控制方法,解决了实际应用中案例检索相似度较低,无法求解的问题。同时,为了进一步提高模型的准确性,解决由于检索到的案例与当前工况之间的差异导致的案例解之间存在差别的问题,采用RBF神经网络方法建立了增量补偿模型对检索到的案例解进行补偿。并最终将这一方法应用到实际生产中,实现了实际磨矿过程中生产关键控制参数的实时优化。
In order to solve the optimization control problem of the grinding process which is a typical complex industrial process,a research is proceeded in a concentrator with two-stage fully closed-circuit grinding process.A grinding expert system based on the combination of fuzzy rules and case-based reasoning is proposed,which solves the problem of low similarity and unsolvable in case retrieval.In order to improve the accuracy of the model,which restricted by the differences of the case solutions caused by the differences between the retrieved cases and the current working conditions,the RBF neural network method is used to establish an incremental compensation model for case solution compensation.Finally,the real-time optimization of key production control parameters in the actual grinding process is achieved by applying the model in the concentrator.
作者
杨佳伟
余刚
刘道喜
YANG Jia-Wei;YU Gang;LIU Dao-Xi(School of Resources&Civil Engineering,Northeastern University,Shenyang 110819,China;Beijing General Research Institute of Mining and Metallurgy,Beijing 100160,China;State Key Laboratory of Process Automation in Mining&Metallurgy,Beijing 102628,China;Beijing Key Laboratory of Process Automation in Mining&Metallurgy,Beijing 102628,China)
出处
《控制工程》
CSCD
北大核心
2021年第1期42-48,共7页
Control Engineering of China
关键词
磨矿优化
案例推理
产生式规则
专家系统
Grinding optimization
case-based reasoning
production rule
expert system