详细信息

Multi-loop process fluctuating source locating and identifying method, involves drawing granger causality relationship map, and simplifying serial control loop points according to process knowledge    

文献类型:专利

英文题名:Multi-loop process fluctuating source locating and identifying method, involves drawing granger causality relationship map, and simplifying serial control loop points according to process knowledge

作者:WU Q;ZHENG J;XIA C;ZHOU F

机构:[1]UNIV EAST CHINA SCI&TECHNOLOGY

申请号:CN101887247-A

申请日:2009-05-12

公开日:2010-11-17

语种:英文

收录:DERWENT

摘要:NOVELTY - The method involves performing spectral independent component analysis of a frequency domain for data to obtain component-related ratio and mixing matrix. Two parameters are combined to select data for post analysis. A Granger causality test and analysis is realized by using Granger causality relationship testing method to analyze causality relationship between variations to obtain causality relationship magnitudes in pairs between the variations. A Granger causality relationship map is drawn. Serial control loop points are simplified according to process knowledge. USE - Method for locating and identifying a multi-loop process fluctuating source. ADVANTAGE - The method enables utilizing the causality map to directly express causality influence relationship between loop variations, so that propagation path of the fluctuating interruption can be expressed. The method enables utilizing processing knowledge to simplify the causality relationship map and automatically searching the threshold to filtrate secondary causality relationship branches to obtain the main propagation path of fluctuation, so that the fluctuating source can be located and identified. The method enables the exact location of the fluctuating source to realize post maintenance and repair of the fault loop, so that the device-grade multi-loop performance is reduced. DESCRIPTION OF DRAWING(S) - The drawing shows a schematic illustration of a multi-loop process fluctuating source locating and identifying method.

参考文献:

正在载入数据...

版权所有©华东理工大学 重庆维普资讯有限公司 渝B2-20050021-7 
渝公网安备 50019002500408号 违法和不良信息举报中心