基于互信息和MPCA的单基药塑化过程监测方法  

Process monitoring for plasticizing process of single-base gun propellant based on mutual information and MPCA

作　　者：杨明毅[1,2,3] 王军义 白鑫林[1,2] 徐志刚 余廷江[4] 陈舒渤 YANG Ming-yi;WANG Jun-yi;BAI Xin-lin;XU Zhi-gang;YU Ting-jiang;CHEN Shu-bo(State Key Laboratory of Robotics,Shenyang Institute of Automation,Chinese Academy of Sciences,Shenyang 110016,China;Institutes for Robotics and Intelligent Manufacturing,Chinese Academy of Sciences,Shenyang 110169,China;University of Chinese Academy of Sciences,Beijing 100049,China;Luzhou North Chemical Industries Co.,Ltd.,Luzhou 646003,China)

机构地区：[1]中国科学院沈阳自动化研究所机器人学国家重点实验室,辽宁沈阳110016 [2]中国科学院机器人与智能制造创新研究院,辽宁沈阳110169 [3]中国科学院大学,北京100049 [4]泸州北方化学工业有限公司,四川泸州646003

出　　处：《应用化工》2022年第1期290-296,共7页Applied Chemical Industry

基　　金：装备发展部预先研究项目;国防火炸药科研专项项目;中国科学院青年创新促进会资助项目(2021203)。

摘　　要：针对现阶段单基药塑化过程缺乏可靠的渐变式故障检测和异常工况的报警及评价能力等问题,同时传统数据驱动方法难以处理过程中包含的多种线性与非线性关系混合特征,又无法突出变量间耦合相关性的差异,提出一种基于归一化互信息的多向主成分分析故障检测方法。该方法通过归一化互信息刻画过程中多维变量间的复杂耦合关系,并以此对不同维度变量之间的相关性特征进行权值修正,得到体现每个变量与其他维度变量之间耦合作用关系的数据集,并分别建立与其相应的监测模型。最后再利用贝叶斯推理将不同模型的监测结果融合成一组概率指标实现过程监测。实验结果表明了所提方法的有效性,可实现塑化过程多种故障的快速检测和异常工况的监测预警。The plasticizing process of single-base gun propellant is lack of reliable gradual fault detection ability and alarm evaluation under abnormal working conditions.Meanwhile, traditional data-driven method is difficult to deal with the mixed characteristics of linear and nonlinear relations contained in the process, and it is unable to highlight the difference of coupling correlation between variables.A multiway principal component analysis method based on normalized mutual information was proposed.In this method, the complex coupling relationship between multidimensional variables is characterized by normalized mutual information, and the correlation between variables of different dimensions is modified by weight.Then the data sets that reflects the coupling relationship between each variable and other dimension variables are obtained, and the corresponding monitoring models are established respectively.Finally, according to the Bayesian inference, the monitoring results of different models are fused into a set of global monitoring statistics to realize process monitoring.The experimental results show that the proposed method is effective, which can realize the rapid detection of multiple faults and the monitoring and warning of abnormal conditions in plasticizing process.

关 键 词：单基发射药 塑化过程 多向主成分分析 故障检测 过程监测预警 

分 类 号：TQ562[化学工程—炸药化工] TJ55[兵器科学与技术—军事化学与烟火技术]