大数据驱动的极端气象灾害态势要素提取模型与优化

The Construction and Optimization of Extraction Model on Key Situation Elements in Extreme Meteorological Disaster Via Big Data Driven

大数据驱动的灾害精准监测与预警是新时代提升公共服务质量、实现精准防灾减灾的重要手段。针对目前关于灾害态势要素提取缺乏可操作性的方法,以极端气象灾害中的台风灾害为例,基于“致灾体-承灾体-救灾体”理论设计了灾害态势要素提取框架模型,采用大数据分析中常用的主成分分析法对影响极端气象灾害态势的诸多要素进行降维处理并提取关键态势要素;引入博弈论的思想,构建基于合作博弈的灾害态势要素权重优化模型,并通过案例分析验证了该方法的合理性和可行性。通过推动自然灾害类突发事件监测与预警理论的创新,推动应急管理由注重灾后救助向注重灾前预防转变,更好地从根本上减轻灾害风险。

Accurate monitoring and early warning of disaster based on big data is an important mean to improve the quality of public services and achieve accurate disaster prevention and mitigation in the new era.In view of the lack of operational methods for extracting disaster situation elements at present,the framework model of disaster situation element extraction was proposed,based on the triangle theory of public security by taking typhoon disasters in extreme meteorological disaster as an example.Then the Principal Component Analysis(PCA)method of big data analysis was used to extract the key factors of extreme meteorological disasters through data dimensionality reduction.The element’s weight optimization model of disaster situation based on cooperative game was constructed by introducing game theory.And the rationality and feasibility of the method were verified through case analysis.By promoting the innovation of the monitoring and early warning theory of natural disaster emergencies,this research promoted the transformation of emergency management from relief after disaster to prevention before disaster,which fundamentally reduced disaster risk.