多重关联情境下基于数据驱动的城市关键基础设施运行风险动态评估与防范策略研究

The operation of urban critical infrastructures is influenced by many risk factors, such as natural factors, man-made factors and so on. It has been a hot topic to scientifically assess and effectively prevent the operation risk of urban critical infrastructures in the field of public safety. However, the existence of multiple interdependencies, data driven and dynamic changes increases the difficulty of this research. Therefore, it is necessary to pay much attention to how to solve the problems on dynamic assessment and prevention of urban critical infrastructures based on data driven in the multiple interdependent scenarios. But unfortunately, there still lacks of systematic theoretical analysis and proper method studies so far. The main line of this project is designed as “theoretical analysis – method exploration – strategy formulation – empirical verification”. Firstly, the analysis of multiple interdependent scenarios is conducted to clarity the mechanisms, features and influences of different interdependencies. Secondly, the processing and fusion methods for multi-source heterogeneous data are given mentioned feature extraction, type classification, format conversion and integrated fusion. Thirdly, the dynamic assessment on operation risk of urban critical infrastructures is conducted to present a series of methods for dynamic assessment on risk probability / consequence, classification of risk grades and mapping of risk tendency. Then, a study on dynamic prevention strategies for operation risk is conducted from top-level design, strategy set generation, strategy formulation and dynamic adjustment. Finally, the effectiveness of the proposed methods and strategies are verified through an empirical research in typical areas. The research results will provide a systematic resolution to ensure the operation safety of urban critical infrastructures.

城市关键基础设施运行受到自然、人为等诸多风险诱因的影响，对其科学评估和有效防范是公共安全领域的研究热点。然而，多重关联性、数据驱动性和动态变化性的存在增加了研究难度。如何解决多重关联情境下基于数据驱动的城市关键基础设施运行风险动态评估与防范问题亟待关注，但目前尚缺乏系统的理论剖析和适用的方法研究。本项目以“理论分析—方法探索—策略制定—实证检验”为主线，首先解析多重关联情境，明确各类关联的机理与特征及其影响；其次，从特征提取、类型划分、格式转换、集成融合入手给出多源异构数据的处理与融合方法；再次，开展城市关键基础设施运行风险动态评估，提出风险概率/后果动态评估、风险等级划分、风险态势制图等一系列方法；然后，从顶层设计、策略集生成、策略制定、动态调整等方面研究运行风险动态防范策略；最后，通过典型样区的实证研究验证方法与策略的有效性。研究成果将为保障城市关键基础设施运行安全提供系统性解决方案。

城市关键基础设施运行受到自然、人为等诸多风险诱因的影响，对其科学评估和有效防范是公共安全领域的研究热点。然而，多重关联性、数据驱动性和动态变化性的存在增加了研究难度。如何解决多重关联情境下基于数据驱动的城市关键基础设施运行风险动态评估与防范问题亟待关注，但目前尚缺乏系统的理论剖析和适用的方法研究。本项目以“多重关联情境解析→多源异构数据处理与融合→风险动态评估方法探索→风险防范策略制定与优化→典型样区实证检验”为主要研究内容，为城市关键基础设施运行风险动态评估与防范提供了集理论分析、数据佐证、方法支撑、策略保障为一体的系统性解决方案。通过积极开拓数据获取渠道，已形成了国别和城市等不同尺度下跨年份、多维度的城市关键基础设施相关指标数据以及北京市热力系统、电力系统和燃气系统为典型对象的跨年份风险事件数据。依托相关研究工作，发表（含录用）论文11篇，其中SCI检索英文论文4篇、国家自然科学基金委管理科学部认定A级重要期刊论文6篇、中文核心期刊论文1篇，中文专著2部（已与科学出版社签订出版合同），协助培养博士后1名、博士生3名、硕士生1名，累计5人次出席重要国际学术会议并做主题报告，组织/协助组织3次学术交流报告/内部研讨会，受邀至国内知名高校、国内学术会议做主题报告5次，获得学术年会优秀论文奖2次，参加20余次国内重要学术会议，团队成员3人次晋升高级职称，形成的政策建议被《中国科学院专报信息》、《国家高端智库建设工作简报》等采用或收录，相关研究成果也得到了新疆生产建设兵团公路科学技术研究所、首都机场集团等企事业单位的应用证明和好评。