基于Agent的危险化学品安全生产信息采集系统研究
|
摘要
安全生产监管是政府的重要职责,直接关系到国家财产和人民生命安全。安全生产监管中,以危险化学品的监管为重中之重。提高安全生产监管的信息化程度是降低监管强度、提高监管效率行之有效的方法。而危险化学品的生产分布地域广,种类多,涉及行业多,如何获取和分析危险化学品生产信息是政府部门在面对众多危险化学品生产企业、大量危险化学品生产信息的难题,也是进行监管监察的首要任务和前提条件。
本文总结了危险化学品监管及生产过程信息采集现状,对危险化学品安全生产监管中的信息采集进行了需求分析,研究了Agent及多Agent系统的基础理论,将Agent技术应用于危险化学品安全生产过程中的信息采集。文章研究了危险化学品生产过程的监控策略、信息采集方案,基于Agent技术,本文构建了由管理Agent、采集Agent、故障诊断Agent、通信Agent组成的分布式信息采集多Agent系统,运行在各企业危险源参数的采集点上,将省安监局监管中心对全省危险化学品生产过程监管的信息采集任务下移到企业端,利用多Agent系统实现分散于各企业的危险源生产状态参数的信息采集、监控及分析；设计了该信息采集系统的数据库以及各类Agent的结构和功能；利用模糊数学理论及故障诊断模型实现Agent对异常情况的诊断及推理；研究Agent之间的通信及协调机制,从而通过Agent之间的合作、并行求解使全省危险化学品安全生产信息采集的问题得到较好地解决。论文最后以焚硫转化工艺的信息采集为例介绍了该系统的工作模式,模拟了温度、气体浓度等参数的信息采集过程,实现了Agent之间的通信与协作。
Work-safety Monitoring is a critical governmental responsibility, directly affecting human safety as well as protecting capital assets. In this regard, hazardous chemical monitoring is of paramount importance. But the collection and analysis of hazardous chemical databases can be a difficult information technology problem, since manufacturers are diversely distributed and are responsible for a wide variety of products. A well-developed safety monitoring information system has been proven to be a feasible solution to this problem, lowering the amount of required work while increasing the effectiveness of the monitoring process.
This thesis first summarizes the methods of requisition and collection of hazardous chemical information that are currently employed in industry, study the theory of Agent and Multi-Agent system.The author then demonstrates the deployment of the Agent Technology System to collect and analyze hazardous chemical information. The author studies strategies for monitoring the production of hazardous chemicals, builds the Agent information-collection model, the system is used to connect industrial enterprises located in Yunnan Province through to the Hazardous Chemical Information Center at the Yunnan Provincial Bureau, so as to monitor the work-safety of their manufacturing processes, designs the database and the function and structure of the information-collection system across different Agent technologies.Further, the author customizes the Agent technological implementations using both fuzzy math and problem diagnosis modeling, to predict or diagnose abnormal conditions in the manufacturing process and studies the communication and coordination mechanisms between Agents. Thereby, through the cooperation and parallel solution, the problem of information-collection for hazardous chemicals work safety can be solved. At last, taking the example of the sulfur burning and conversion process, the author introduces a working model of the system that simulates the information collecting process (e.g., data of temperature and gas concentration parameters), and instantiates the communication and coordination between different types of Agents.
本文总结了危险化学品监管及生产过程信息采集现状,对危险化学品安全生产监管中的信息采集进行了需求分析,研究了Agent及多Agent系统的基础理论,将Agent技术应用于危险化学品安全生产过程中的信息采集。文章研究了危险化学品生产过程的监控策略、信息采集方案,基于Agent技术,本文构建了由管理Agent、采集Agent、故障诊断Agent、通信Agent组成的分布式信息采集多Agent系统,运行在各企业危险源参数的采集点上,将省安监局监管中心对全省危险化学品生产过程监管的信息采集任务下移到企业端,利用多Agent系统实现分散于各企业的危险源生产状态参数的信息采集、监控及分析；设计了该信息采集系统的数据库以及各类Agent的结构和功能；利用模糊数学理论及故障诊断模型实现Agent对异常情况的诊断及推理；研究Agent之间的通信及协调机制,从而通过Agent之间的合作、并行求解使全省危险化学品安全生产信息采集的问题得到较好地解决。论文最后以焚硫转化工艺的信息采集为例介绍了该系统的工作模式,模拟了温度、气体浓度等参数的信息采集过程,实现了Agent之间的通信与协作。
Work-safety Monitoring is a critical governmental responsibility, directly affecting human safety as well as protecting capital assets. In this regard, hazardous chemical monitoring is of paramount importance. But the collection and analysis of hazardous chemical databases can be a difficult information technology problem, since manufacturers are diversely distributed and are responsible for a wide variety of products. A well-developed safety monitoring information system has been proven to be a feasible solution to this problem, lowering the amount of required work while increasing the effectiveness of the monitoring process.
This thesis first summarizes the methods of requisition and collection of hazardous chemical information that are currently employed in industry, study the theory of Agent and Multi-Agent system.The author then demonstrates the deployment of the Agent Technology System to collect and analyze hazardous chemical information. The author studies strategies for monitoring the production of hazardous chemicals, builds the Agent information-collection model, the system is used to connect industrial enterprises located in Yunnan Province through to the Hazardous Chemical Information Center at the Yunnan Provincial Bureau, so as to monitor the work-safety of their manufacturing processes, designs the database and the function and structure of the information-collection system across different Agent technologies.Further, the author customizes the Agent technological implementations using both fuzzy math and problem diagnosis modeling, to predict or diagnose abnormal conditions in the manufacturing process and studies the communication and coordination mechanisms between Agents. Thereby, through the cooperation and parallel solution, the problem of information-collection for hazardous chemicals work safety can be solved. At last, taking the example of the sulfur burning and conversion process, the author introduces a working model of the system that simulates the information collecting process (e.g., data of temperature and gas concentration parameters), and instantiates the communication and coordination between different types of Agents.
引文
[1]赵庆闲,邵辉.危险化学品安全管理[M].北京：中国石化出版社,2005
[2]周剑峰,陈国华,万木生等.基于多Agent的危险源分布式监控系统[J].华南理工
大学学报(自然科学版).2007,35(9)：107-112
[3]陆汝铃.知识科学与计算科学[M].北京：清华大学出版社,2003,1
[4]赵来军.安全管理中的几个管理决策问题研究.
[5]周琦.危险化学品安全监管信息化的研究[D].硕士学位论文,江苏,江苏大学,2005
[6]国家安全生产信息化“十一五”专项规划
[7]王晓宁,王明贤,陈万金.现代信息技术对我国危险化学品监管体系的影响[J].2005,25(7)：29-31,39
[8]康荣学,刘骥,吴宗之等.重大危险源监控系统发展历程[J].2006.2(6)：78-82
[9]杨妮妮.移动代理在煤矿安全生产监控联网中的应用[J].矿山机械,2007,35(8)：120-122
[10]刘大有,杨鲲等.Agent研究现状与发展趋势[J].软件学报.2000,11(3)：315～321
[11]Wooldridge MJ, Jennings NR. Intelligent agents:Theory and practice[J].Knowledge Engineering Review,1995,10 (2):115-152.
[12]Shoham Y. Agent oriented programming. Artificial Intelligence,1993,60 (1):51～ 92
[13]王汝传,徐小龙,黄海平.智能Agent及其在信息网络中的应用[M].北京：北京邮电大学出版社,2006
[14]王勇,黄国兴.多Agent信息系统知识发现研究[J].计算机科学,2008,35(11)：184-186
[15]石纯一,张伟.基于Agent的计算[M].北京：清华大学出版社,2007
[16]赵丽平.基于Mulit-agent的电能质量监控网络的研究[D].博士学位论文,成都,西南交通大学,2006
[17]谭现虎.基于多Agent的水质监测系统研究与实现[D].硕士学位论文,北京,中国科学技术大学,2007
[18]Scheibel, John R.Iman, Imada Ebben, Thomas G, Blomgren Rick.Expert system-based online rotor crack monitor for utility steam turbines. Proceeding of the
American Power Conference,1989,51:369-380
[19]彭强.复杂系统远程智能故障诊断技术研究[D].硕士学位论文,南京,南京理工大学,2004
[20]S.J.Russell. Provably Bounded Optimal Agents. Proceedings of the 13th International Joint Conference on Artificial Intelligence.1993:40-48
[21]H Wang,C Wang. APACS.A Multi-Agent System with Repository Support.Knowledge-Based Systems,1996,9(3):329-337
[22]http://www.soft6.com/product/3016
[23]王道平,张义忠.故障智能诊断系统的理论与方法[M].北京：冶金工业出版社,2001
[24]蒋伟进,许宇胜.基于多Agent的复杂系统故障诊断研究[J].机械科学与技术,2004,23(10)：242-1249
[25]张曾科.模糊数学在自动化技术中的应用[M].北京：清华大学大学出版社,1997.
[26]彭祖赠,孙韫玉.模糊(Fuzzy)数学及其应用[M].武汉：武汉大学出版社,2002
[27]刘普寅,吴孟达.模糊理论及其应用[M].长沙：国防科技大学出版社,1998.
[28]Zadeh L A. Fuzzy sets. Information and Control.1965,8:330-35
[29]邵辉,王凯全.危险化学品生产安全[M].北京：中国石化出版社,2005
[30]关磊,刘骥,魏利军等.危险化学品重大危险源安全监控通用技术规范研究[J].中国安全生产科学技术,2008,4(6)：15-19
[31]孙红岩.大型旋转机械智能诊断多Agent系统的研究[D].博士学位论文,重庆,重庆大学,2007
[32]Labrou Y.Finin T.A proposal for a new KQML specification.Draft:University of Maryland Baltimore Country,http://www.cs.umbc.edu,1996
[33]黄晓宇,李磊,李东.基于多Agent系统的DCS模型[J].计算机工程与设计,2004,25(8)：1384-1387
[34]历小军,俞欢军,王建飞等.用智能体实施化工企业多层优化的新策略[J].信息与控制,2003,32(4)：324-330
[35]王海东.水泥生产过程多Agent控制系统的研究[D].博士学位论文,长沙,中南大学,2007
[36]刘必雄,魏连,许榕生.基于Agent技术的多源日志采集系统的设计与实现[J].计算机系统应用,2008,2：71-74
[37]赵新宇,林作铨.具有监控能力的Agent模型[J].计算机科学,2006,33(3)：11-17
[38]王仲生.智能检测与控制技术[M].西安：西北工业大学出版社,2002.9
[39]Zadeh L A.Fuzzy algorithm.Information and Control,1968, (12):94-102
[40]褚蕾蕾,陈绥阳等.计算智能的数学基础[M].北京：科学出版社,2002
[41]Muszynska A. Vibrational Diagnostics of Rotating Machinery Malfunctions. Vki.Vibration and Rotor Dynamics.CI1992.pp.40
[42]崔彦平,付其凤,葛杏卫等.机械设备故障诊断发展历程及展望[J].河北工业科技,2004,21(4)：59-62
[43]G Hartvigsen, D Johansen. Cooperation in a distributed artificial intelligence environment-The storm-cast application[J].Engineering Application of Artificial Intelligence,1990,3(3):229-23
[44]刘占生,须根法,夏松波等.旋转机械故障诊断的自学习方法研究[J].振动工程学报,1996,9(4)：405-410
[45]Foundation for Intelligent Physical Agents, FIPA 97 Specification:Agent Communication Language[M].http://www.fipa.org
[46]Bratman M.Intentions,Plans, and Practical Reason. Cambriage:Harvard University Press,1987
[47]Cohen P R.Levesque H J.Intention is choice with commitment.Artificial Intelligence,1990,42(2-3):213-261
[48]Rao A S,Georgeff M P. BDI Agents:From Theory to Practice. In:Proceedings of the 1st International Conference on Multi-Agent Systems (ICMAS'95). San Francisco. ACM Press,1995.312-319
[49]Rao A S,Georgeff M P. Modeling rational agents within a BDI-architecture.Allen J, Fikes R,Sandewall E editors. In:Proceedings of the 2nd International Conference on Principles of Knowledge Representation and Reasoning(KR'91),1991.473～484
[50]胡晓辉,周兴社,党建武.一种面向Agent的分布监控系统行为建模和设计方法[J].计算机应用,2005,25(12)：284-286
[51]C Hewitt.Open information system semantics for distributed artificial intelligence,1991,47(1):79-106
[52]W Shen,N Douglas,J P Barthes.Multi-Agent Systems for concurrent Intelligent Manufacturing[M].New York,NY:Taylor&Francis Inc,2001
[53]刘书,潘成胜,张德育.Mobile Agent在网络系统监控中数据采集的设计与应用[J].武汉工程大学学报,2009,31(3)：77-80
[54]夏娜,蒋建国,胡社教等.基于Multi-agent的分布式智能视觉监控系统研究[J].计算机工程与应用,2003,31：41-42,51
[55]倪峰,左丽娟,倪天倪.一种基于Multi-agent的实时监控系统的设计和实现[J].计算机应用研究,2002,5：71-72,75
[56]赵毅德.基于Agent的分布式安全数据采集框架的研究与实现[D].硕士学位论文,北京,北京邮电大学,2007
[57]张军,张荣涛,孙宇.分布式监测监控系统中Agent模型及实现研究[J].计算机工程,2001,27(2)：93-95
[58]朱全博,张士杰.多Agent技术在生产过程监控与管理系统中的应用[J].微计算机信息,2006,33(10-3)：68-71
[59]Labrou Y,Finin T,Yun Peng.Agent communication languages:the current landscape[J].Intelligent Systems,1999,14(5):45-52
[60]刘彦斌,朱小冬.基于Multi-agent的实时系统运行故障监控研究[J].微计算机信息,2006,22(10-1)：224-226
[61]肖诗松,陈涛.基于插件技术的日志采集Agent系统的设计与实现[J].东南大学学报(自然科学版),2008,38(9)：90-93
[62]Georgeff M.Communication and Interaction in Multi-agent Planning[C].In:Proc of the International Joint Conference on Artifical Intelligence,1983:125-129
[63]Davies W H,Edwards P.Agent-K:an integration of AOP and KQML[C].In:Proc 3rd International Conference on Information and Knowledge Management,Gaithersburg,Maryland,1994:68-75
[64]严梅娟,李建华.多代理技术在校园信息采集系统中的应用[J].计算机应用研究,2004,3：194-196
[65]李娟.高校数字图书馆中基于Agent的智能信息采集与服务原型系统的研究[D].硕士学位论文,武汉,武汉理工大学,2005
[66]张晖.基于移动Agrnt信息采集系统与安全性研究[D].硕士学位论文,南京,南京理工大学,2007
[67]庄旭东.高炉本体温度监控与故障诊断中的模糊数学模型研究[D].硕士学位论文,浙江,浙江大学理学院,2006
[68]侯健,苑飞.模糊推理方法及模糊系统的逼近性能[M].天津：天津科学技术出版社,2009
[69]井新宇.温湿度数据采集与监控系统[D].硕士学位论文,无锡,江南大学,2008
[2]周剑峰,陈国华,万木生等.基于多Agent的危险源分布式监控系统[J].华南理工
大学学报(自然科学版).2007,35(9)：107-112
[3]陆汝铃.知识科学与计算科学[M].北京：清华大学出版社,2003,1
[4]赵来军.安全管理中的几个管理决策问题研究.
[5]周琦.危险化学品安全监管信息化的研究[D].硕士学位论文,江苏,江苏大学,2005
[6]国家安全生产信息化“十一五”专项规划
[7]王晓宁,王明贤,陈万金.现代信息技术对我国危险化学品监管体系的影响[J].2005,25(7)：29-31,39
[8]康荣学,刘骥,吴宗之等.重大危险源监控系统发展历程[J].2006.2(6)：78-82
[9]杨妮妮.移动代理在煤矿安全生产监控联网中的应用[J].矿山机械,2007,35(8)：120-122
[10]刘大有,杨鲲等.Agent研究现状与发展趋势[J].软件学报.2000,11(3)：315～321
[11]Wooldridge MJ, Jennings NR. Intelligent agents:Theory and practice[J].Knowledge Engineering Review,1995,10 (2):115-152.
[12]Shoham Y. Agent oriented programming. Artificial Intelligence,1993,60 (1):51～ 92
[13]王汝传,徐小龙,黄海平.智能Agent及其在信息网络中的应用[M].北京：北京邮电大学出版社,2006
[14]王勇,黄国兴.多Agent信息系统知识发现研究[J].计算机科学,2008,35(11)：184-186
[15]石纯一,张伟.基于Agent的计算[M].北京：清华大学出版社,2007
[16]赵丽平.基于Mulit-agent的电能质量监控网络的研究[D].博士学位论文,成都,西南交通大学,2006
[17]谭现虎.基于多Agent的水质监测系统研究与实现[D].硕士学位论文,北京,中国科学技术大学,2007
[18]Scheibel, John R.Iman, Imada Ebben, Thomas G, Blomgren Rick.Expert system-based online rotor crack monitor for utility steam turbines. Proceeding of the
American Power Conference,1989,51:369-380
[19]彭强.复杂系统远程智能故障诊断技术研究[D].硕士学位论文,南京,南京理工大学,2004
[20]S.J.Russell. Provably Bounded Optimal Agents. Proceedings of the 13th International Joint Conference on Artificial Intelligence.1993:40-48
[21]H Wang,C Wang. APACS.A Multi-Agent System with Repository Support.Knowledge-Based Systems,1996,9(3):329-337
[22]http://www.soft6.com/product/3016
[23]王道平,张义忠.故障智能诊断系统的理论与方法[M].北京：冶金工业出版社,2001
[24]蒋伟进,许宇胜.基于多Agent的复杂系统故障诊断研究[J].机械科学与技术,2004,23(10)：242-1249
[25]张曾科.模糊数学在自动化技术中的应用[M].北京：清华大学大学出版社,1997.
[26]彭祖赠,孙韫玉.模糊(Fuzzy)数学及其应用[M].武汉：武汉大学出版社,2002
[27]刘普寅,吴孟达.模糊理论及其应用[M].长沙：国防科技大学出版社,1998.
[28]Zadeh L A. Fuzzy sets. Information and Control.1965,8:330-35
[29]邵辉,王凯全.危险化学品生产安全[M].北京：中国石化出版社,2005
[30]关磊,刘骥,魏利军等.危险化学品重大危险源安全监控通用技术规范研究[J].中国安全生产科学技术,2008,4(6)：15-19
[31]孙红岩.大型旋转机械智能诊断多Agent系统的研究[D].博士学位论文,重庆,重庆大学,2007
[32]Labrou Y.Finin T.A proposal for a new KQML specification.Draft:University of Maryland Baltimore Country,http://www.cs.umbc.edu,1996
[33]黄晓宇,李磊,李东.基于多Agent系统的DCS模型[J].计算机工程与设计,2004,25(8)：1384-1387
[34]历小军,俞欢军,王建飞等.用智能体实施化工企业多层优化的新策略[J].信息与控制,2003,32(4)：324-330
[35]王海东.水泥生产过程多Agent控制系统的研究[D].博士学位论文,长沙,中南大学,2007
[36]刘必雄,魏连,许榕生.基于Agent技术的多源日志采集系统的设计与实现[J].计算机系统应用,2008,2：71-74
[37]赵新宇,林作铨.具有监控能力的Agent模型[J].计算机科学,2006,33(3)：11-17
[38]王仲生.智能检测与控制技术[M].西安：西北工业大学出版社,2002.9
[39]Zadeh L A.Fuzzy algorithm.Information and Control,1968, (12):94-102
[40]褚蕾蕾,陈绥阳等.计算智能的数学基础[M].北京：科学出版社,2002
[41]Muszynska A. Vibrational Diagnostics of Rotating Machinery Malfunctions. Vki.Vibration and Rotor Dynamics.CI1992.pp.40
[42]崔彦平,付其凤,葛杏卫等.机械设备故障诊断发展历程及展望[J].河北工业科技,2004,21(4)：59-62
[43]G Hartvigsen, D Johansen. Cooperation in a distributed artificial intelligence environment-The storm-cast application[J].Engineering Application of Artificial Intelligence,1990,3(3):229-23
[44]刘占生,须根法,夏松波等.旋转机械故障诊断的自学习方法研究[J].振动工程学报,1996,9(4)：405-410
[45]Foundation for Intelligent Physical Agents, FIPA 97 Specification:Agent Communication Language[M].http://www.fipa.org
[46]Bratman M.Intentions,Plans, and Practical Reason. Cambriage:Harvard University Press,1987
[47]Cohen P R.Levesque H J.Intention is choice with commitment.Artificial Intelligence,1990,42(2-3):213-261
[48]Rao A S,Georgeff M P. BDI Agents:From Theory to Practice. In:Proceedings of the 1st International Conference on Multi-Agent Systems (ICMAS'95). San Francisco. ACM Press,1995.312-319
[49]Rao A S,Georgeff M P. Modeling rational agents within a BDI-architecture.Allen J, Fikes R,Sandewall E editors. In:Proceedings of the 2nd International Conference on Principles of Knowledge Representation and Reasoning(KR'91),1991.473～484
[50]胡晓辉,周兴社,党建武.一种面向Agent的分布监控系统行为建模和设计方法[J].计算机应用,2005,25(12)：284-286
[51]C Hewitt.Open information system semantics for distributed artificial intelligence,1991,47(1):79-106
[52]W Shen,N Douglas,J P Barthes.Multi-Agent Systems for concurrent Intelligent Manufacturing[M].New York,NY:Taylor&Francis Inc,2001
[53]刘书,潘成胜,张德育.Mobile Agent在网络系统监控中数据采集的设计与应用[J].武汉工程大学学报,2009,31(3)：77-80
[54]夏娜,蒋建国,胡社教等.基于Multi-agent的分布式智能视觉监控系统研究[J].计算机工程与应用,2003,31：41-42,51
[55]倪峰,左丽娟,倪天倪.一种基于Multi-agent的实时监控系统的设计和实现[J].计算机应用研究,2002,5：71-72,75
[56]赵毅德.基于Agent的分布式安全数据采集框架的研究与实现[D].硕士学位论文,北京,北京邮电大学,2007
[57]张军,张荣涛,孙宇.分布式监测监控系统中Agent模型及实现研究[J].计算机工程,2001,27(2)：93-95
[58]朱全博,张士杰.多Agent技术在生产过程监控与管理系统中的应用[J].微计算机信息,2006,33(10-3)：68-71
[59]Labrou Y,Finin T,Yun Peng.Agent communication languages:the current landscape[J].Intelligent Systems,1999,14(5):45-52
[60]刘彦斌,朱小冬.基于Multi-agent的实时系统运行故障监控研究[J].微计算机信息,2006,22(10-1)：224-226
[61]肖诗松,陈涛.基于插件技术的日志采集Agent系统的设计与实现[J].东南大学学报(自然科学版),2008,38(9)：90-93
[62]Georgeff M.Communication and Interaction in Multi-agent Planning[C].In:Proc of the International Joint Conference on Artifical Intelligence,1983:125-129
[63]Davies W H,Edwards P.Agent-K:an integration of AOP and KQML[C].In:Proc 3rd International Conference on Information and Knowledge Management,Gaithersburg,Maryland,1994:68-75
[64]严梅娟,李建华.多代理技术在校园信息采集系统中的应用[J].计算机应用研究,2004,3：194-196
[65]李娟.高校数字图书馆中基于Agent的智能信息采集与服务原型系统的研究[D].硕士学位论文,武汉,武汉理工大学,2005
[66]张晖.基于移动Agrnt信息采集系统与安全性研究[D].硕士学位论文,南京,南京理工大学,2007
[67]庄旭东.高炉本体温度监控与故障诊断中的模糊数学模型研究[D].硕士学位论文,浙江,浙江大学理学院,2006
[68]侯健,苑飞.模糊推理方法及模糊系统的逼近性能[M].天津：天津科学技术出版社,2009
[69]井新宇.温湿度数据采集与监控系统[D].硕士学位论文,无锡,江南大学,2008