基于分布式混合贝叶斯网络的煤泥浮选过程安全运行与产品质量一体化控制方法  

An integrated safe operation and product quality control method for coal slurry flotation process based on distributed hybrid Bayesian network

作　　者：毛腾 褚菲 王建文[1] 王福利 MAO Teng;CHU Fei;WANG Jian-wen;WANG Fu-li(School of Information and Control Engineering,China University of Mining and Technology,Xuzhou 221116,China;Artificial Intelligence Research Institute,China University of Mining and Technology,Xuzhou 221116,China;College of Information Science and Engineering,Northeastern University,Shenyang 110819,China)

机构地区：[1]中国矿业大学信息与控制工程学院,江苏徐州221116 [2]中国矿业大学人工智能研究院,江苏徐州221116 [3]东北大学信息科学与工程学院,辽宁沈阳110819

出　　处：《控制与决策》2025年第2期497-506,共10页Control and Decision

基　　金：国家自然科学基金项目(61973304,61873049,62073060);江苏省六大人才高峰项目(DZXX-045);中央高校基础研究基金项目(2022ZZCX01K01)。

摘　　要：针对煤泥浮选过程中因原煤属性变化、关键变量参数波动导致的产品质量下降甚至异常工况问题,提出一种基于分布式混合贝叶斯网络的煤泥浮选过程安全运行与产品质量一体化控制方法.该方法将煤泥浮选过程划分为若干相互关联的局部模块,并建立相应的局部混合贝叶斯网络模型;然后利用关联变量确定煤泥浮选过程的全局混合贝叶斯网络模型,有效提升建模的效率和精度.该模型在离散贝叶斯网络的基础上通过引入连续节点提升控制决策的推理精度.当煤泥浮选过程发生异常工况时,首先利用贡献图算法识别导致异常工况的局部模块,通过贝叶斯推理获取安全运行控制决策,消除异常工况;在此基础上结合模拟退火算法获取产品质量控制决策,进一步提升产品煤质量.最后,通过煤泥浮选过程仿真实验验证所提方法的有效性.To address the issues of decreased product quality and abnormal conditions caused by variations in raw coal properties and parameter fluctuations in coal slurry flotation process,this paper proposes an integrated safe operation and product quality control method for the coal slurry flotation process based on a distributed hybrid Bayesian network(BN).The method divides the coal slurry flotation process into interrelated local modules and establishes corresponding local hybrid BN models.Then,by utilizing the correlated variables,the global hybrid BN model of the coal slurry flotation process is determined,effectively improving the efficiency and accuracy of modeling.The model improves the inference accuracy of control decisions by introducing continuous nodes based on the discrete BN.When abnormal conditions occur in the coal slurry flotation process,the contribution graph algorithm is used to identify the local modules causing the abnormal conditions.The BN inference is then employed to obtain safe operation control decisions,eliminating the abnormal conditions.Furthermore,by combining the simulated annealing algorithm on this basis,product quality control decisions are obtained to improve the coal product quality.Finally,the effectiveness of the proposed method is validated through the simulation experiments of the coal slurry flotation process.

关 键 词：分布式建模 混合贝叶斯网络 煤泥浮选过程 安全运行与产品质量一体化控制 贡献图算法 模拟退火算法 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]