论文部分内容阅读
为避免海量报警影响正常操作,对报警阈值进行了优化.考虑了变量相关性,利用主元分析法(PCA)分析变量的重要程度,并求出变量权重;结合国际标准,根据变量权重分配变量报警数,优化报警阈值;在平行坐标图上将变量数据及报警阈值可视化,查看正常和异常工况区域,考察各变量之间的变化趋势,提取有效报警值;对某工业原油常减压操作数据进行验证,结果表明优化方法能有效减少报警数.
In order to avoid the influence of the mass alarm on the normal operation, the alarm threshold is optimized.Considering the correlation of the variables, PCA is used to analyze the importance of the variables and to find the weight of the variables.According to the international standard, Alarm number, and optimize the alarm threshold; visualize the variable data and alarm thresholds on the parallel coordinate map, check the normal and abnormal working conditions, examine the changing trend among variables, and extract the effective alarm value; Data validation, the results show that the optimization method can effectively reduce the number of alarms.