基于历史过渡特性的新过渡模态建模方法  

Modeling algorithm for new transitional mode of multi-mode process based on the characteristics of history data

作　　者：谭帅[1] 王福利[1,2] 彭俊[1] 

机构地区：[1]东北大学信息科学与工程学院,沈阳110004 [2]东北大学流程工业综合自动化教育部重点实验室,沈阳110004

出　　处：《仪器仪表学报》2012年第7期1533-1540,共8页Chinese Journal of Scientific Instrument

基　　金：国家973计划(2009CB320601);国家自然科学基金(61074074)资助项目

摘　　要：传统的统计过程质量控制是通过检测和采集大量的过程质量特征数据以实现过程质量控制,过渡过程在短期内无法获得充足的建模数据,导致统计错判概率增加。研究了多模态生产过程中的过渡模态建模特征,提出了一种基于历史过渡模型相关特性的新过渡模态建模方法,充分利用已有过渡模型中所蕴含的过程特性,弥补新过程数据不足无法准确描述新过渡过程特性的缺点,可准确描述新出现的未建模过渡模态。通过在田纳西-伊斯曼过程中的仿真结果证明了所提出方法的可行性和有效性。Traditional statistical process quality control is realized through monitoring the process characteristics extracted from a large number of process data. However, the production efficiency is decreased as the inaccurate fault detections increase due to lacking the modeling data that are used to estimate the process characteristics and control limits. In this paper, we propose a modeling algorithm for new transitional mode of multi-mode process. The unmod- eled mode does not have enough modeling data to describe the process characteristics completely. The characteristics contained in history models are fully utilized to remedy the weakness. A large number of simulations in Tennessee- Eastman process show the feasibility and effectiveness of the proposed method.

关 键 词：多模态过程 新过渡模态建模 动态时间弯曲 

分 类 号：TH164[机械工程—机械制造及自动化]