首页 | 本学科首页   官方微博 | 高级检索  
     

基于GMM的多模态过程模态识别与过程监测
引用本文:谭帅,常玉清,王福利,王姝.基于GMM的多模态过程模态识别与过程监测[J].控制与决策,2015,30(1):53-58.
作者姓名:谭帅  常玉清  王福利  王姝
作者单位:1. 东北大学流程工业综合自动化国家重点实验室,沈阳110004;
2. 华东理工大学化工过程先进控制和优化技术教育部重点实验室,上海200237.
基金项目:国家自然科学基金项目(61403072,61374146,61174130);国家863计划项目(2011AA060204);中央高校基本科研专项资金项目(N120304004);中国博士后科学基金项目(2013M530937)
摘    要:多模态复杂过程的多变量、多工序、变量时变性以及模态转换时间不确定等多种原因, 导致面向多模态生产过程的监测问题十分复杂. 对此, 基于高斯混合模型的监测方法, 结合定性知识和定量知识, 解决了多模态过程监测中离线数据模态划分、稳定模态和过渡模态的监测模型建立以及在线数据的模态识别等关键问题, 最终实现了对多模态过程的监测.

关 键 词:多模态过程  过程监测  模态识别  连续退火机组
收稿时间:2013-11-14
修稿时间:2014/2/26 0:00:00

Mode identification and process monitoring for multiple mode processes based on GMM
TAN Shuai CHANG Yu-qing WANG Fu-li WANG Shu.Mode identification and process monitoring for multiple mode processes based on GMM[J].Control and Decision,2015,30(1):53-58.
Authors:TAN Shuai CHANG Yu-qing WANG Fu-li WANG Shu
Affiliation:1. Stat Key Laboratory of Integrated Automation of Process Industry Technology,Northeastern University,Shenyang 110004,China;
2. Key Laboratory of Advanced Control and Optimization for Chemical Processes of Ministry of Education,East China University of Science and Technology,Shanghai 200237,China.
Abstract:Considering the process high dimensionality, multi-operation, time-variant characteristics, and unknown mode duration, it is challenging to conduct the statistical analysis and online monitoring for multi-mode processes. The process monitoring model based on the Gaussian mixture model(GMM) combining quantitative with qualitative information solves several key points, such as mode classification of offline data, model building for stable modes and transitional modes, and mode identification of online data. Process monitoring for multi-mode processes is realized by using the proposed method.
Keywords:multiple mode processes  process monitoring  mode identification  continuous annealing
本文献已被 CNKI 万方数据 等数据库收录!
点击此处可从《控制与决策》浏览原始摘要信息
点击此处可从《控制与决策》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号