基于时空大数据的石油安全智能分析方法研究  

Research on an intelligent analysis method for oil security using spatio-temporal big data

作　　者：赵黎冬 王德利[1] 王永志 代涛[3] 王安建[3] 孟祥卉 曹亚琴 徐可 霍雨佳 ZHAO LiDong;WANG DeLi;WANG YongZhi;DAI Tao;WANG AnJian;MENG XiangHui;CAO YaQin;XU Ke;HUO YuJia(College of Geo-exploration Science and Technology,Jilin University,Changchun 130061,China;Institute of Integrated Information for Mineral Resources Prediction,Jilin University,Changchun 130061,China;Institue of Mineral Resources,Chinese Academy of Geological Sciences,Beijing 100037,China)

机构地区：[1]吉林大学地球探测科学与技术学院,长春130061 [2]吉林大学综合信息矿产预测研究所,长春130061 [3]中国地质科学院矿产资源研究所,北京100037

出　　处：《地球物理学进展》2023年第2期958-966,共9页Progress in Geophysics

基　　金：国家重点研发计划(2021YFC2901801);中国地质调查局“石油安全智能推演系统”联合资助。

摘　　要：传统石油安全战略分析方法存在数据更新不及时、处理复杂数据困难、效率较低等不足,难以适应大数据时代快速分析的要求.为了满足石油安全供应对多元异构大数据高效分析的需求,本文设计了涵盖石油全产业链与安全密切相关指标体系,即综合考虑生产、运输、储备、加工与消费等环节的多层次影响因素;提出了事件驱动的石油安全智能分析模式,当突发战争、公共卫生、天气、海盗、灾害等重大事件时,可对石油全产业链中一个或多个环节进行多角度的智能分析与挖掘.采用大数据、人工智能、云计算、物联网等技术手段,构建石油安全智能分析平台,通过时空大数据驱动自动计算石油安全相关环节造成的影响,从而实现一种面向石油安全的新型“事件-计算-响应-应对措施”工作模式.以马六甲海峡因多种事件造成通道阻塞为例,采用2021年中国从全球石油进口数据为基数,可“一键式”高效测算并以多种可视化形式实时展现2023年前6个月对全国及单一港口的海上石油进口影响.实验证明本方法是一种高效的石油安全战略分析方法,为新时代石油资源战略研究从定性分析向智能分析范式转变提供参考.Traditional analysis methods of oil safety strategy are difficult to adapt to the requirements of rapid analysis in the era of big data because it has some disadvantages such as data update is not timely, hard to process complex data, less efficient, etc. In order to meet the demands for rapid analysis of multi-heterogeneous big data in oil safety supply, this paper designs a comprehensive index system closely related to oil safety covering oil whole industry chain. The system considers the multi-level influencing factors involving oil's production, storage and transportation, processing and consumption. This paper also proposes an event-driven intelligent analysis pattern for oil safety supply. One or more links of the entire oil safety industry chain will be analyzed intelligently from multi-angle and multi-level when wars, public health, weather, piracy, disasters and other major events happened suddenly. Using big data, artificial intelligence, cloud computing, Internet of Things and other techniques to build an intelligent analysis platform for oil safety, and the platform can automatically calculate the impact of oil safety related links driven by spatio-temporal big data. So as to realize a new “event-computing-response-measure” working mode oriented to oil safety. Taking the channel blockage caused by various events in the Strait of Malacca as an example and China's global oil import data in 2021 is used basis to calculate oil import fall influences of the first half of 2023. The client sends computation request and the platform can efficiently calculate under “One Click ” and real-time display the impact of offshore oil imports on the whole country and a single port in a variety of visual forms. Experiments show that this method is an efficient oil safety strategy analysis method,which provides a reference for the transformation of oil resource strategy research from qualitative analysis to an intelligent analysis paradigm in the coming new era.

关 键 词：大数据 时空大数据 石油安全 人工智能 全产业链 事件驱动 

分 类 号：P631[天文地球—地质矿产勘探]