[1]蔡爽,胡瑾秋,张来斌.基于传递熵与KELM的炼油化工过程风险传播路径分析方法[J].中国安全生产科学技术,2018,14(3):19-26.[doi:10.11731/j.issn.1673-193x.2018.03.003]
CAI Shuang,HU Jinqiu,ZHANG Laibin.Analysis method of risk propagation paths in refining petrochemical process based on transfer entropy and KELM[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2018,14(3):19-26.[doi:10.11731/j.issn.1673-193x.2018.03.003]
点击复制
基于传递熵与KELM的炼油化工过程风险传播路径分析方法
CAI Shuang,HU Jinqiu,ZHANG Laibin.Analysis method of risk propagation paths in refining petrochemical process based on transfer entropy and KELM[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2018,14(3):19-26.[doi:10.11731/j.issn.1673-193x.2018.03.003]
《中国安全生产科学技术》[ISSN:1673-193X/CN:11-5335/TB]
- 卷:
- 14
- 期数:
- 2018年3期
- 页码:
- 19-26
- 栏目:
- 学术论著
- 出版日期:
- 2018-03-31
文章信息/Info
- Title:
- Analysis method of risk propagation paths in refining petrochemical process based on transfer entropy and KELM
- 文章编号:
- 1673-193X(2018)-03-0018-08
- Keywords:
- refining petrochemical process; reasoning model; transfer entropy; kernel extreme learning machine (KELM); propagation path
- 分类号:
- X937
- 文献标志码:
- A
- 摘要:
- 基于炼油化工过程复杂，设备众多，某一设备的监测变量发生扰动可能会传播至其相邻设备引发出一系列故障链。现有方法多是针对某一设备进行监测与诊断，以期降低事故后果，而忽视了对过程风险传播路径的预测以防止事故的发生。因此，提出一种基于传递熵与核极限学习机的炼油化工过程风险传播路径分析方法，该方法针对某一工艺扰动，分析其在风险发展过程中的扰动传播过程，基于传递熵分析法建立炼油化工过程风险传播推绎模型；并提出一种基于KELM的风险传播搜索方法，预测风险传播路径；将该方法应用于分馏塔冲塔过程。研究结果表明：该方法可辨识出未来一段时间内风险的可能传播路径，以便操作人员及时采取预防措施，保证过程安全及产品质量。
- Abstract:
- The refining petrochemical process is complex and has a large amount of equipments, so the disturbance in the monitored variables of a certain equipment may propagate to its adjacent equipments and cause a series of fault chain. The existing methods are mostly monitoring and diagnosing a certain equipment to reduce the accident consequence, but neglect the prediction on the risk propagation paths of the process to prevent the occurrence of accidents. An analysis method on the risk propagation paths of the refining petrochemical process based on the transfer entropy and the kernel extreme learning machine (KELM) was put forward. Aiming at the disturbance of a certain technology, the disturbance propagation process in the risk development process was analyzed, and the reasoning model of risk propagation in the refining petrochemical process was established based on the analysis method of transfer entropy. A searching method of risk propagation based on KELM was proposed to predict the risk propagation paths. The method was applied in the flooding process of the fractionator. The results showed that the proposed method can identify the probable propagation paths of risk during a period of time in the future, so that the operators can adopt the prevention measures in time to ensure the process safety and the product quality.
参考文献/References:
［1］DAHLSTRAND F. Consequence analysis theory for alarm analysis［J］. Knowledge-Based Systems, 2002, 15(1): 27-36.
［2］WAN Yiming, YANG Fan, LYU Ning, et al. Statistical root cause analysis of novel faults based on digraph models［J］. Chemical Engineering Research & Design, 2013, 91(1): 87-99.
［3］YANG Fan, SHAH S L, XIAO De-yun. Signed directed graph modeling of industrial processes and their validation by Data-Based methods［C］//2010 CONFERENCE ON CONTROL AND FAULT-TOLERANT SYSTEMS (SYSTOL'10)．Nice,France, 2010: 387-392.
［4］MAURYA M R, RENGASWAMY R, VENKATASUBRAMANIAN V. A systematic framework for the development and analysis of signed digraphs for chemical processes. 1. Algorithms and analysis［J］. Industrial & Engineering Chemistry Research, 2003, 42(20): 4789-4810.
［5］MAURYA M R, RENGASWAMY R, VENKATASUBRAMANIAN V. A systematic framework for the development and analysis of signed digraphs for chemical processes.2.Control loops and flowsheet analysis［J］. Industrial & Engineering Chemistry Research, 2003, 42(20): 4811-4827.
［6］HE Bo, CHEN Tao, YANG Xianhui. Root cause analysis in multivariate statistical process monitoring: Integrating reconstruction-based multivariate contribution analysis with fuzzy-signed directed graphs［J］. Computers & Chemical Engineering, 2014, 64: 167-177.
［7］MAURYA M R, RENGASWAMY R, VENKATASUBRAMANIAN V. A signed directed graph-based systematic framework for steady-state malfunction diagnosis inside control loops［J］. Chemical Engineering Science, 2006, 61(6): 1790-1810.
［8］CHANG C T, CHEN C Y. Fault diagnosis with automata generated languages［J］. Computers & Chemical Engineering, 2011, 35(2): 329-341.
［9］BAUER M, THORNHILL N F. A practical method for identifying the propagation path of plant-wide disturbances［J］. Journal of Process Control, 2008, 18(7): 707-719.
［10］HAN Liu, GAO Huihui, XU Yuan, et al. Combining FAP, MAP and correlation analysis for multivariate alarm thresholds optimization in industrial process［J］. Journal of Loss Prevention in the Process Industries, 2016, 40: 471-478.
［11］GAO Huihui, XU Yuan, GU Xiangbai, et al. Systematic rationalization approach for multivariate correlated alarms based on interpretive structural modeling and Likert scale［J］. Chinese Journal of Chemical Engineering, 2015, 23(12): 1987-1996.
［12］ABELE L, ANIC M, GUTMANN T, et al. Combining Knowledge Modeling and Machine Learning for Alarm Root Cause Analysis［J］. IFAC Proceedings Volumes, 2013, 46(9):1843-1848.
［13］GAO J, TULSYAN A, YANG F, et al. A transfer entropy method to quantify causality in stochastic nonlinear systems［J］. IFAC-PapersOnLine, 2016, 49(7): 454-459.
［14］ZHAO Xiaojun, SHANG Pengjian, LIN Aijing. Transfer mutual information: A new method for measuring information transfer to the interactions of time series［J］. Physica a: Statistical Mechanics and Its Applications, 2017, 467: 517-526.
［15］SCHREIBER T. Measuring information transfer［J］. Physical Review Letters, 2000, 85(2): 461-464.
［16］HUANG Guangbin, ZHU Qinyu, SIEW C K. Extreme learning machine: Theory and applications［J］. Neurocomputing, 2006, 70(1/3): 489-501.
［17］ HAJIHOSSEINI P, SALAHSHOOR K, MOSHIRI B. Process fault isolation based on transfer entropy algorithm［J］. ISA Transactions, 2014, 53(2): 230-240.
相似文献/References:
备注/Memo
- 备注/Memo:
- 国家自然科学基金项目（51574263）；中国石油大学(北京)青年创新团队C计划项目(C201602)
