Study on alarm optimization method based on historical data of chemical production process
In order to effectively reduce the missing alarm rate and false alarm rate of alarm device, two methods of alarm op-timization were studied based on the historical data of chemical production process. Taking the missing alarm rate and false a-larm rate as optimization indexes, an objective function of alarm threshold was established. The optimization design of alarm threshold was conducted by the method of numerical optimization from the view of minimized objective function. The number of alarm was reduced using the method of adaptive alarm delay base on the variation coefficient of alarm duration. The case study showed that the method based on the historical data of chemical production process is of great significance for improving the safety of petrochemical production process, which can effectively reduce the number of invalid alarm, the missing alarm rate and false alarm rate, and improve the quality of alarming.
alarm thresholdmissing alarm ratefalse alarm rateprobability density function
NETL
NSTL
万方数据
维普
