王飞, 姜文宇, 刘彬彬, 郑晓翠, 房龄航. 利用灾害链规则的灾害模型服务链编制方法[J]. 武汉大学学报 ( 信息科学版), 2020, 45(8): 1168-1178. DOI: 10.13203/j.whugis20200130
引用本文: 王飞, 姜文宇, 刘彬彬, 郑晓翠, 房龄航. 利用灾害链规则的灾害模型服务链编制方法[J]. 武汉大学学报 ( 信息科学版), 2020, 45(8): 1168-1178. DOI: 10.13203/j.whugis20200130
WANG Fei, JIANG Wenyu, LIU Binbin, ZHENG Xiaocui, FANG Linghang. Disaster Model Service Chain Orchestration Method Using Disaster Chain Rules[J]. Geomatics and Information Science of Wuhan University, 2020, 45(8): 1168-1178. DOI: 10.13203/j.whugis20200130
Citation: WANG Fei, JIANG Wenyu, LIU Binbin, ZHENG Xiaocui, FANG Linghang. Disaster Model Service Chain Orchestration Method Using Disaster Chain Rules[J]. Geomatics and Information Science of Wuhan University, 2020, 45(8): 1168-1178. DOI: 10.13203/j.whugis20200130

利用灾害链规则的灾害模型服务链编制方法

Disaster Model Service Chain Orchestration Method Using Disaster Chain Rules

  • 摘要: 灾害模型能够提供灾情模拟、灾损评估等灾情分析功能,在一体化综合减灾的风险评估、预测预警、灾害处置、灾损评估等环节发挥着重要作用。传统的灾害模型以单灾种为主,部署于各类孤立的业务系统中,较少考虑不同灾害之间的复杂链式关系,难以满足多灾种耦合场景下模型群交互运算的灾情辅助分析需求。针对上述问题,提出了一种利用灾害链规则的灾害模型服务链编制方法。首先,根据灾害场景分析需求和灾害链规则构建灾害模型服务链;其次,基于优化的服务质量(quality of service, QoS)指标评价体系从单灾种灾害模型群中选择最优的组合链路来实现复杂灾害场景的模拟分析功能;然后,为了保证最优组合链路的稳定性,提出了动态调整机制以解决链路模型节点失效断开的问题;最后,为了支持分布式异构灾害模型群的互操作处理,提出标准化的网络服务链发布方法。相较于传统的孤立的单灾种灾害模型,使用所提方法能够提供灵活性更高、扩展性更好、稳定性更强的多灾种复合灾情模拟分析功能。以地震为原生灾害构成复合灾害场景为例,从灾害模型服务链构建、物理模型服务链优化选择与动态调整、灾害模型服务链网络发布3个方面对所提方法进行验证。

     

    Abstract:
      Objectives   Traditional disaster models focus on single disaster, deployed in various types of isolated business systems. For lack of considering complex chain relationships between different disasters, it's difficult to meet the requirements of disaster assistance analysis with the disaster cluster in multi-disaster coupling scenarios. Given the above problems, we propose a disaster model service chain orchestration method using disaster chain rules. In order to elaborate the proposed method, we take the earthquake as a primary disaster to compose a composite disaster scenario as an example, and prove the proposed method from three aspects: The construction of the logical disaster model service chain, the optimization and dynamic adjustment of the physical model service chain, and the web publishing of the disaster model service chain.
      Methods   The disaster chain theory is adopted to orchestrate the disaster model service chain. Firstly, the logic disaster model service chain is constructed based on the analysis requirements of disaster scenario and disaster chain rules. Secondly, the optimized quality of service(QoS) index evaluation system is adopted to select the optimal combined service link from the disaster model cluster to realize the simulation analysis function for complex disaster scenarios. In order to ensure the stability of the chain, a dynamic adjustment mechanism is proposed to solve the failure of model node in the chain. Finally, to support the interoperable processing of distributed heterogeneous disaster model cluster, a standardized web service chain publishing method is proposed.
      Results   Compared with the traditional isolated single disaster model, our method considers the complex chain relationship between different disasters based on the disaster chain theory and have the ability to provide multi‑disaster composite disaster simulation analysis function with higher flexibility, better scalability and stronger stability.
      Conclusions   The disaster chain rules provide a theoretical guarantee for the service chain orchestration method with strong flexibility and high intelligence. Besides, the main chain and the auxiliary chain are adopted to settle the problem of logic conflict caused by the circular structure of the disaster chain. Meanwhile, based on the QoS index evaluation system, the disaster model service chain optimization method and dynamic adjustment method are proposed to ensure efficient and stable response of distributed heterogeneous disaster model. In general, the proposed method can help realize the multi-hazard composite disaster situation analysis function and provide flexible, intelligent, and stable disaster analysis information to integrated emergency management.

     

/

返回文章
返回