案例驱动下韧性城市应对级联灾害风险的情景设计方法

Taking cascading disaster risks as the research object, in order to overcome the dilemma of “the existing expertise cannot meet the requirements of event” in “scenario-response” decision-making, this project proposes a research on the case-driven scenario design method for resilient city to respond to cascading disaster risks. The research mode of this project is “scenario modelling-scenario arrangement-scenario reuse-capability evaluation”, and the main content contains: (1) Based on the multi-source heterogeneous mass data, the key scenario elements are identified, then the “context-scenario” model of cascading disaster risks is built. From top to bottom and from the surface to the center, a multi-layer nested full space-time three-dimensional gene model for cascading disaster risks response case. (2) Based on the theory of biological evolution, scenario evolutionary forest of cascading disaster risks is planned. Its structure is designed as far as the breadth and depth, and the substitutive and derivative expansion method are illustrated. (3) The deduction of cascading path of disaster risk scenarios of the target case and the copy and adaptation of response plan of each phase can be achieved based on the generated similar case set with filtration. The “filter-deduce-copy-adapt” scenario reuse method of cascading disaster risks is proposed. (4) The case-driven capability of responding to cascading disaster risks is evaluated for resilient city with the application verification in the sample city. This project would be conductive to build and perfect the preparedness system of cascading disaster risks response, and provide support for the corresponding emergency response decision-making.

以级联灾害风险为研究对象，为克服“情景-应对”决策“事件超越经验”的困境，本项目在案例驱动下研究韧性城市应对级联灾害风险的情景设计方法，以“情景建模-情景布局-情景重用-能力评价”为研究模式，主要研究内容包括：（1）基于多源异构的海量数据识别情景关键要素，构建级联灾害风险“情境-情景”模型，自上而下由表及里形成多层嵌套的全时空立体的级联灾害风险应对案例基因模型。（2）基于生物进化理论规划级联灾害风险情景演化森林，从广度与深度对其进行结构设计，并阐述相应的案例库替代与派生扩容方法。（3）基于过滤生成的相似案例集实现对目标案例灾害风险情景级联路径的推演和各阶段应对方案的复制与适配，提出“过滤-推演-复制-适配”的级联灾害风险情景重用方法。（4）通过在样本城市应用验证，面向韧性城市评价应对级联灾害风险的案例驱动能力。本项目将有利于构建与完善级联灾害风险应对准备体系，为相应的应对决策提供支持。

在韧性城市建设的进程中，提升临灾阶段级联灾害风险的应对能力是强化复杂灾害风险治理的重要举措。历史案例作为应对经验的重要载体，能有效缓解“事件超越经验”的困境。因此，基于“情景-应对”型范式，本项目提出案例驱动的级联灾害风险应对情景设计方法，旨在辅助应急决策。.本项目主要研究内容包括级联灾害风险情景建模、级联灾害风险应对案例库结构构建、案例驱动的级联灾害风险应对情景推理、级联灾害应急准备控错机制设计。首先，阐述了级联灾害风险链机理并构建了灾害风险情景的三棱锥模型，以及基于本体建模法构建了级联灾害风险三层本体系统。其次，设计了一个嵌套交叉的灾害风险应对案例库架构，主要分为“案例-情境-情景-后果”的多层级纵向分类体系与“网络-链-对”的多尺度横向分化体系。然后，构建了案例驱动的级联灾害风险情景画像生成模型，主要包含本体重用源选择、风险本体模板生成、风险本体重用等环节。最后，作为内容的扩展延伸，基于案例推理设计临灾阶段级联灾害应急准备控错机制，通过多阶段情景检索与混合情景重用获取相似情景的控错措施，促进管理层面的风险应对。此外，基于上述研究内容，本项目主要选取了台风级联灾害风险作为阐述方法流程的具体场景。.本项目研究进展主要贡献有以下四点：①构建了相对完备的级联灾害风险三层本体系统，有利于标准化与规范化风险建模，以及促进风险情景信息沟通与共享；②构建了适合存储级联灾害风险应对案例的应急案例库结构，既有利于灾害风险情景的快速准确查询，又有利于支持灾害风险推演；③构建了案例驱动的级联灾害风险情景画像生成模型，有利于认知灾害风险演化与获取可用应对任务；④构建了基于案例推理的级联灾害应急准备控错决策支持模型，有利于在厘清控错情景的基础上为决策者提供合理可行的控错方案。综上所述，本项目充分利用案例，并结合情景分析法与案例推理服务级联灾害风险应对，为智能应急决策提供重要支持。