首页> 中文期刊>计算机集成制造系统 >基于数据驱动的大型化工装置健康评估与故障源诊断

基于数据驱动的大型化工装置健康评估与故障源诊断

     

摘要

Aiming at the subhealthy of large in-service chemical process equipment,a method of healthy quantitative evaluation and fault source diagnosis for large process equipment systems was proposed based on practical operation process parameters and their inherent relationship analysis.The performances of control loop were evaluated by using minimum variance control,and performance of control valves were assessed by using bicoherence-based nonlinearity measures so as to realize healthy quantitative evaluation for equipment.Fault separation and oscillatory source localization were evaluated by special independent component analysis,and oscillation propagation analysis was defined based on Granger causality test.The theories and methods were introduced and used to evaluate a large-scale vinyl chloride production plant.The results of quantitative evaluation for the plant loop were obtained,and the oscillation disturbance and propagation were determined.The field survey demonstrated that the results could reflect practical situations of the plant.%针对大型在役化工过程装置亚健康问题,提出一套基于实际操作过程参数及其内在联系分析,进行大型在役化工过程装置定量健康评估与故障源诊断的方法.该方法依据最小方差控制方法评估控制回路性能,基于双相干频谱非线性分析评估控制阀的性能,实现装置健康状态定量评估;采用功率谱独立源成分进行系统波动源分析,进而通过格兰杰因果检验确定波动故障传播路径和根源.给出有关理论和方法,并据此对大型氯乙烯装置进行健康评估和故障源定位,获得装置性能定量评估结果,确定异常控制回路、故障源及其传播路径.现场验证结果表明,评估结果能够较好地反映装置的实际情况.

著录项

相似文献

  • 中文文献
  • 外文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号