摘要
针对现场安全监管数据库风险识别错误率较高、识别效率较低的问题,提出面向典型作业场景的现场安全监管数据库风险识别方法。针对典型作业场景现场安全监管数据库中存在的异常数据,根据异常数据类型,分别采用移动平均线法和AR模型法对数据进行预处理;通过灰色关联聚类算法提取数据风险特征;引入树突状细胞算法,将MAP作为抗原综合评价指标实现现场安全监管数据库风险识别。实验结果表明,所提方法风险识别错误率更低、识别效率更高,具有较好的应用价值。
In response to the high error rate and low identification efficiency of on-site safety supervision database risk identification,a risk identification method of on-site safety supervision database for typical operation scenarios is proposed.Aiming at the abnormal data in the on-site safety supervision database of typical operation scenarios,the moving average method and AR model are used to preprocess the data according to the type of abnormal data.The grey correlation clustering algorithm is used to extract data risk features.The dendritic cell algorithm is introduced,and MAP is used as an antigen comprehensive evaluation index to achieve risk identification in the on-site safety supervision database.The experimental results show that the proposed method has a lower risk identification error rate and higher recognition efficiency,and has good application value.
作者
王天师
WANG Tianshi(Zhongshan Power Supply Bureau of Guangdong Power Grid Co.,Ltd.,Zhongshan 528400,China)
出处
《微型电脑应用》
2024年第2期66-69,共4页
Microcomputer Applications
基金
广东省中国南方电网有限责任公司科技项目(GDKJXM20190028)。
关键词
典型作业场景
现场安全监管
数据库
风险识别
树突状细胞算法
typical operation scenario
on-site safety supervision
database
risk identification
dendritic cell algorithm
作者简介
王天师(1983-),男,本科,高级工程师,研究方向为安全管理。
