CIESC Journal ›› 2021, Vol. 72 ›› Issue (11): 5696-5706.DOI: 10.11949/0438-1157.20210712

• Process system engineering • Previous Articles     Next Articles

Reliability monitoring of fluorochemical process operation unit based on GLSAFIS

Feng XUE1,2(),Xintong LI1,Kun ZHOU1,Zhiqiang WEI3,Xiaoxia GE3,Zhiqiang GE4,Kai SONG1()   

  1. 1.Tianjin Key Laboratory of Chemical Safety and Equipment, School of Chemical Engineering and Technology, Tianjin University, Tianjin 300350, China
    2.Suzhou Chuangyuan Industrial Investment Co. , Ltd. , Suzhou 215000, Jiangsu, China
    3.Health, Safety and Environmental Protection Department, Juhua Group Co. , Ltd. , Quzhou 324004, Zhejiang, China
    4.State Key Laboratory of Industrial Control Technology, College of Control Science and Engineering, Zhejiang University, Hangzhou 310027, Zhejiang, China
  • Received:2021-05-26 Revised:2021-08-16 Online:2021-11-12 Published:2021-11-05
  • Contact: Kai SONG

基于GLSAFIS的氟化工过程操作单元可靠性监测

薛峰1,2(),李欣铜1,周琨1,魏志强3,葛晓霞3,葛志强4,宋凯1()   

  1. 1.天津大学化工学院,天津化工安全与装备重点实验室,天津 300350
    2.苏州创元产业投资有限公司,江苏 苏州 215000
    3.巨化集团有限公司健康安全环保部,浙江 衢州 324004
    4.浙江大学控制科学与工程学院,工业控制技术国家重点实验室,浙江 杭州 310027
  • 通讯作者: 宋凯
  • 作者简介:薛峰(1995—),男,硕士,3014207238@tju.edu.cn
  • 基金资助:
    国家重点研发计划项目(2018YFC0808600)

Abstract:

The highly toxic characteristics of fluorochemical products make it extremely important to monitor the operational reliability of fluorochemical process equipment. For this reason, this paper proposes a fuzzy inference system based on global-local structural analysis (GLSAFIS) to evaluate the operational reliability of fluorochemical process operation units online. After selecting the process variables of the operating unit according to the fluorochemical process flow, the global-local feature extraction of the operating unit process variables is carried out through the global-local structure analysis algorithm (GLSA). More importantly, GLSA transferred process data into a much lower feature space, which made it much more efficient to design fuzzy logics for a fuzzy inference system (FIS) and less dependent on expert knowledge. The effectiveness of the proposed method is evaluated through the fluorochemical R-22 production process located in East China and the Tennessee Eastman benchmark process. The results show that this method can accurately reflect the operating conditions of the actual chemical process operation unit, and plays an important role in monitoring the chemical process operation.

Key words: reliability estimation, global-local structure analysis, fuzzy inference system, fluorochemical process, Tennessee Eastman Process

摘要:

氟化工产物的剧毒特性,使得对于氟化工过程设备的运行可靠性监控异常重要。为此,提出了基于全局-局部结构分析的模糊推理系统(GLSAFIS)在线评估氟化工过程操作单元运行可靠性。依据氟化工工艺流程选取操作单元过程变量后,通过全局-局部结构分析算法(GLSA)对操作单元过程变量进行全局-局部特征提取。这些低维的全局-局部特征代替原始变量作为模糊推理系统(FIS)的输入,不仅可以克服噪声的影响,降低对专家知识的依赖,同时可以通过特征空间的降维压缩,大大加速后续模糊推理系统的逻辑设计。最后模糊推理系统的实施,使得本方法可以对化工过程操作单元可靠性进行在线评估。国内某氟化工厂二氟一氯甲烷(R-22)生产过程的实际应用以及田纳西伊斯曼模拟过程的仿真结果均证实了所提方法可以准确地反映实际化工过程操作单元的运行状况,大大提高了对实际化工过程的安全监控力度。

关键词: 可靠性评估, 全局-局部结构分析, 模糊推理系统, 氟化工过程, 田纳西伊斯曼过程

CLC Number: