基于大数据的特种设备宏观安全风险预警方法研究

特种设备监管机构在监督检验过程中积累了大量数据资料，为充分挖掘其中的潜在价值，增强监管针对性，提高检验效率，应用并优化关联规则挖掘、社区发现、可视化等数据挖掘技术，构建1种宏观安全风险预警方法。该方法以特种设备安全监督检验大数据为基础，通过挖掘单台设备微观因素间的关联关系，实现整类特种设备宏观安全风险的识别与预警；以2008—2016年全国长管拖车检验数据为例，进行应用实践分析。研究结果表明:该方法可以对区域范围、缺陷类别等多种宏观指标进行预警，指导针对性监管与检验。

Research on method of macro safety risk early warning of special equipment based on big data

The regulators of special equipment have accumulated a lot of data in the process of supervision and inspection. In order to fully mine the potential value of the data, enhance the supervision pertinence and improve the inspection efficiency, a method of macro safety risk early warning was constructed by applying and optimizing the data mining technologies such as the association rule mining, community discovery, visualization and so on. Based on the big data of the safety supervision and inspection on the special equipment, this method could realize the identification and early warning on the macro safety risk of the whole special equipment through mining the association between the micro factors of a single equipment. The application practical analysis was carried out by taking the inspection data of the national long tube trailers from 2008 to 2016 as example, and the results showed that this method can conduct the early warning on multiple macro indexes such as the regional scope, defect category and others, so as to guide the targeted regulation and inspection.