基于数据驱动的电力安全生产事故风险预警研究

Research on the risk early warning of power production safety accident based on data drive

智能电网的快速发展和广泛应用为电力企业提供了来源复杂、结构多样的海量数据,使得智能电网成为大数据最重要的应用领域之一。如何应用大数据技术实现对电力安全生产数据的采集、存储和挖掘,进而提高电力企业的安全生产水平成为当前重要的研究课题。本文首先阐述了电力大数据和大数据技术的基本概念,然后借鉴传统的"海因里希法则"的思想,从隐患的角度出发,运用大数据技术对我国某省电力安全生产数据进行处理,通过这些数据建立某省电力企业的安全事故比例模型。再通过回归分析,对未来可能存在的隐患数量进行预测。最后根据安全事故比例模型对未来可能发生事故、事件数量做出预测,确定隐患数量的控制目标,形成一套安全生产预警模型,从而达到消除隐患、减少事故发生目的。

The rapid development and wide application of smart grid provide massive data with complex sources and diverse structures for power enterprises,making smart grid become one of the most important application fields of big data. How to apply big data technology to achieve data collection,storage and mining,and then improve the level of safe production of power enterprises has become an important research topic. This paper first expounds the basic concepts of big data and big data technology of electric power,then draws lessons from the traditional " Heinrich rule" and proceeds from the perspective of hidden trouble,uses big data technology to process the massive data generated in the safe production of power in a province of China,through which a safety accident proportion model of electric power enterprises in a province is established. Then,through regression analysis,the number of potential dangers in the future is predicted. Finally,according to the proportion model of safety accidents,the number of possible accidents and accidents in the future is predicted,and determine the control target of the number of hidden dangers,and in the end a set of production safety early warning model is formed in order to eliminate hidden dangers and reduce accidents.