用户名: 密码: 验证码:
面向城市应急指挥的智能决策系统的研究与实现
详细信息    本馆镜像全文|  推荐本文 |  |   获取CNKI官网全文
摘要
随着社会的不断发展,人们对自身的安全越来越重视。对各种突发事件的快速响应和应急处理,将能够有力地保障人们的生命财产安全。维护社会的持续稳定的发展,建立起高效、科学的城市应急指挥系统正在成为各级政府日益关注的建设任务。
     本文在对国内外城市应急指挥系统的发展和应用现状总结的基础上,通过对城市应急指挥系统的工作特征进行分析,提出并建立了一个基于智能决策过程的城市应急指挥系统的总体结构。由于应急指挥系统强调对事件的快速决策和处置,需要建立预案库。预案是对各种应急响应事件的处理模板,是决策指挥的重要参考和依据。但是,常规的人工编制预案的模式,不能适应不断变化的环境,人工预案更新不及时。智能决策就是一个面向预案的智能生成和智能决策指挥的过程。本文建立了一个基于机器学习理念的智能决策模型,它包括了三个逻辑相关的循环工作步骤:预案管理、推理决策和学习评估。在预案管理阶段,建立了预案特征模型,提出了预案管理方法,是智能决策的重要基础。推理决策阶段主要是根据事件的特征属性,运用基于案例的推理方法(CBR),并结合适当的推理规则,对预案库进行推理匹配,定位和选取相应的应急预案,支持指挥调度。学习评估则是现实智能决策的关键一环,它对已完成的案件而形成的案例进行学习评价,确定该案例的代表性,抽取案例特征,运用关联规则方法,进行数据挖掘,生成基于特征属性的决策树。学习评估实现了在案例评价基础上的预案模型的自学习过程,使得基于智能决策的应急指挥系统具备了智能性和学习能力,从而可以提高应急指挥的快速响应能力和准确处置能力。在应急指挥系统中采用了面向服务的架构(SOA)和组件模型,提高了系统的适应性,又不失系统效率,提高了软件的复用度和可维护性。
     本文提出的基于智能决策的应急指挥系统已经在广州市城市建设管理监控指挥系统中投入了实际的运行,并取得了一定的成果。
With the development of society, people concerns their safety than ever before. Rapid emergency response and process to all kinds of accidents can protect people' s lives and assets effectively. Maintaining continued and steady development of society and building high efficient city emergency conducting system (CECS) are becoming the task which is concerned by all levels governments day by day.
     This paper analyses the work characters of CECS based on the summary of development and application actuality of CECS in China and abroad. The author puts forward a new system structure of CECS based on intelligent decision-making process. A planning case base would be created due to the CECS emphasizes rapid decision-making and treatment. Planning case is a model treating all kinds of accidents; it is an important reference of Decision-making and conducting system, but general mode of planning case making by people manually can not adapt the continued changes of circumstance. Further more the handmade planning cases can not be updated in time. Intelligent decision-making is a process of intelligent conducting based on planning case. This paper builds an intelligent decision-making model based on machine learning concept, which includes three circular work steps: planning case management, reasoning and decision-making, learning and evaluation. In the phase of planning case management, a character model is created, and a method to manage the planning case is developed. These two phases are important to intelligent decision-making. In the phase of reasoning and decision making, the system runs the program to reference and to match an appropriate case in the case base using CBR technique according to the main characters in the accident. The system locates and selects corresponding case to support man' s conducting and scheduler. The learning and evaluation is a critical phase in realization of intelligent decision-making, it finishs the process of learn and evaluate the case, confirms the case, and extract its character, performs data mining applying the association rule analysis method. Finally a decision-making tree will be created based on character attributes. The learning and evaluation realizes self learning process based on planning case mode at the basis of case evaluation. It makes CECS has the ability of rapid response and emergency process. SOA technique and module mode will be applied in the realization of software, which makes the system has more applicability and higher efficiency and maintainability.
     TheⅠ-CECS has been developed in the project of Guang-Zhou city monitoring conducting system and the effect is acceptable.
引文
[1] 王文俊.应急联动系统建设中的若干问题探讨[Z].http://iems911.ncase.net/
    [2] 陈立.尽早建立我市城市应急指挥系统的几点建议[Z].http://www.hz.gov.cn/
    [3] 刘铁民.重大事故应急体系建设[J].劳动保护,2004(4):6-10
    [4] 马品仲.国内外应急指挥联动技术研究及发展趋势[J].高科技与产业化,2004(11)
    [5] Weiliang Zeng. Coalition Emergency Response System, City Security and City Sustainable Development. Chinese BusinessReview, 2005, (10): 78-81
    [6] 陈秀万,杨凯欣,方裕等.基于空间信息技术的城市应急救援联动系统研究[J].地理与地理信息科学,2003,19(4):49-52
    [7] 曾元武,万宝林,钟远军等.广东省国土资源应急指挥系统建设[J].测绘与空间地理信息,2005;28(6):64-66
    [8] 马品仲.城市突发事件应急系统的开发和实施[J].中国信息导报,2004,(10):37-39
    [9] 王波,魏克俊,费树岷.城市应急救援指挥系统信息平台研究[J].中国公共安全(学术版),2006,5(2):33-36
    [10] 曾元武,万宝林,钟远军等.广东省国土资源应急指挥系统建设[J].测绘与空间地理信息,2005,28(6):64-66
    [11] 程效军,徐文昌.公安应急指挥系统中关键技术的探讨[J].同济大学学报,2003,31(1):52-55
    [12] 陈锋,林才经,林祥娟.120指挥调度系统在突发公共卫生事件应急体系中的地位[J].中国急救医学,2005,25(8):593-594
    [13] 陈红,冯启民,陈天恩.城市地震应急指挥决策支持软件数据库的建立[J].世界地震工程,2005,21(4):77-81
    [14] 姜立新,帅向华,张建福等.地震应急指挥管理信息系统的探讨[J].地震,2003,23(2):115-120
    [15] 王建鹏,马吉苹.基于GIS的城市应急系统研究[J].地理空间信息,2004,2(3):25-27
    [16] 徐志胜,冯凯,徐亮等.基于GIS的城市公共安全应急决策支持系统的研究[J].安 全与环境学报,2004,4(6):82-85
    [17] 米红,杨帆,曾东海.GIS技术和数学模型在城市应急系统中的应用[J].测绘科学,2005,30(1):69-72
    [18] 迟文学,孙刚,武峥.基于GIS民航应急救援与辅助决策支持系统研究[J].交通与计算机,2005,23(127):46-48
    [19] 张荣梅,涂序彦.基于CBR的交通事故处理智能决策支持系统[J].计算机工程与应用,2002,(2):247-249
    [20] 曹立,吴明晖,应晶.基于网格的城市应急联动中心方案设计研究[J].计算机应用研究,2004,(10):61-63
    [21] 陈建军,袁玉平.应急指挥系统建设方案设计与研究[J].武汉理工大学学报(信息与管理工程版),2005,27(2):122-127
    [22] 张建华,刘仲英.案例推理和规则推理结合的紧急预案信息系统[J].同济大学学报(自然科学版),2002,30(7):891-894
    [23] 汪季玉,王金桃.基于案例推理的应急决策支持系统研究[J].管理科学,2003,16(6):46-50
    [24] 廖光煊,翁韬,朱霁平等.城市重大事故应急辅助决策支持系统研究[J].中国工程科学,2005,7(7):7-13
    [25] 孙鉴坤,陈娟.应急指挥决策系统的设计与分析[J].科技广场,2006,(10):32-35
    [26] Gan, Jie-Fu, Li QiQu, Yi-Zhen. Public health emergency management system framework based on digital city. 2005 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2005
    [27] Ying Liu, Nan Zou, Gang-len Chang. An integrated emergency evacuation system for real-time operations - a case study of Ocean City, Maryland under hurricane attacks. 2005 IEEE Intelligent Transportation Systems Conference (ITSC)
    [28] 于卫红,贾传荧.基于XML的海上搜救智能决策支持系统知识库.大连海事大学学报,2005,31(4):30-32
    [29] 李高和,张艳莉,石龙等.基于XML的智能决策支持系统研究.计算机工程与应用,2005,(26):177-180
    [30] 王文俊,盂凡问,王月龙等.基于本体的应急预案研究[J].计算机工程,2006,32(19):170-172
    [31] 张春平,关志超,杨东援等.深圳应急指挥中心的最优路径规划技术研究[J].中山大学学报(自然科学版),2003,42(2):160-164
    [32] XUE Jingsheng, SUN Jizhou, LIU Muxing etal. Data mining and visualization of grid-based city emergency system. Transactions of Tianjin University, 2005, 12(4) 125-128
    [33] 罗忠良,王克运,康仁科等.基于案例推理系统中案例检索算法的探索[J].计算机工程与应用,2005.25:230-232
    [34] Zhang Yan, Zhao Zheng. Application of clustering algorithm in emergency response system. Computer Engineering and Applications, 2006, 42(31): 191-193
    [35] 冯兴杰,周谆.Aprori算法的改进[J].计算机工程,2005,7(31):172-173
    [36] 陈耿,朱玉全,杨鹤标等.关联规则挖掘中若干关键技术的研究[J].计算机研究与发展,2005,42(10):1785-1789
    [37] 许中华,王君,李世祥.一种优化的Apriori算法[J].软件导刊,2005,(13):24-26
    [38] 黄进,尹治本.关联规则挖掘的Apriori算法的改进[J].电子科技大学学报,2003,32(1):76-79
    [39] 马盈仓,陈建文,程代杰.关联规则挖掘的研究及对Apriori算法的改进[J].计算机应用与软件,2002,24(6):65-68
    [40] NI Min, XU Xiao-fei, DENG Sheng-chun etal. NIA2: A fast indirect association m ining algorithm. Journal of Harbin Institute of Technology, 2005, 12(5): 511-516
    [41] 陶灵姣,孙继银等.远程教育考试成绩分析决策树的构造方法[J].计算机工程与设计,2006,27(6):976-978
    [42] 段玉春,朱晓艳等.一种改进的ID3算法[J].南阳师范学院学报,2006,5(9):63-64
    [43] 马瑜,王有刚.ID3算法应用研究[J].信息技术,2006,(12)84-86
    [44] 唐华松,姚耀文.数据挖掘中决策树算法的探讨[J].计算机应用研究,2001,(8):18-22
    [45] Barletta R. An introduction to case based reasoning[J]. AI Expert, 1991, (8): 43-49.
    [46] Schaaf J W. Fish and shrink: A next step towards efficient case retriexral in large scaled case bases[A]. Advances in Case--based Reasoning[C]. Lausanne: Third European Workshop, 1996, 362-376.
    [47] Zhang Shuyu, Zhu Zhongyin. Parallelism of spatial data mining based on autocorrelation decision tree. Journal of Systems Engineering and Electronics, 2005, 16(4): 947-956
    [48] NIU Xiao-tai, MENG Bo, PENG Ding-zhi. Research on Computing Models of CBR Similarity. Wuhan University Journal of Natural Sciences, 2004, 9(4): 407-410
    [49] 王树林,袁志宏.专家系统设计原理.北京,科学出版社,1991
    [50] 陈文伟.智能决策技术.北京,电子工业出版社,1998
    [51] Margaret H.Dunham著.数据挖掘教程.北京,清华大学出版社,2005
    [52] LI Kan, LIU Yu-shu. Agent-Based Data Mining Framework for the High-Dimensional Environment. Journal of Beijing Institute of Technology, 2005, 14(2): 113-116
    [53] WANGYuan-zhen, ZHANGZhi-bing, YIBao-lin etal. Data Mining Based on Computational Intelligence. Wuhan University Journal of Natural Sciences, 2005, 10(2): 371-374

© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号

地址:北京市海淀区学院路29号 邮编:100083

电话:办公室:(+86 10)66554848;文献借阅、咨询服务、科技查新:66554700