Process Alarm Thresholds Optimizations Combing Operators' Comprehensive Alarm Processing Capabilities

The alarm management system is used to detect abnormal conditions in the industrial production process in real time and generate prompt information.The operator then intervenes and processes abnormal conditions to reduce the accident rate in the production process and improve the production efficiency and safety of the plant.In the actual production process,the frequent occurrence of false alarms reduces the efficiency of the operator's handling of the alarms,and brings a great negative impact on the production process.The appropriate alarm threshold setting can effectively reduce the false alarm rate.In the study of alarm threshold setting and optimization,the operator's judgment on the alarm and the influence of processing capability are rarely considered.Therefore,this thesis combines human engineering theory and data mining technology with alarm threshold optimization,and proposes an alarm threshold optimization method that combines the operator's comprehensive alarm processing capabilities.Firstly,the traditional APRIORI algorithm for association rules mining is improved,and the lift is introduced to reduce the number of rules and improve the quality of rules.The improved APRIORI algorithm is used to mine association rules of process alarm variables and related alarm processing variables.On this basis,the operator's judgment ability and processing ability index are defined.Secondly,the nuisance alarm rate and alarm processing rate are used as fuzzy inputs to create the operator's comprehensive alarm processing capabilities.Using the kernel density estimator to estimate the probability density of the process variables in the positive and negative state,the false alarm rate and the missed alarm rate were obtained.Combined with the results of human data mining,the weights of false alarm rate and missed alarm rate were obtained,then establishing an alarm threshold optimization objective function based on the operator's comprehensive capabilities.Finally,taking the TE process as an example,the thresholds are calculated by the traditional threshold setting method and the method proposed in this thesis.Through the comparison and analysis,the new alarm threshold method proposed in this thesis can better adapt to the operator's operating habits and capabilities.The effectiveness of the method was verified.