Design of early warning model based on time series data for production safety

摘要: Abstract Chemical equipment failure, toxic leaks and other abnormal conditions often have significant impact on the production line moreover there is a risk of life safety. Therefore, detection early warning for states stifling risks are prime importance. In this context, time series prediction method commonly used one building models warning. The forecasting problem in petrochemical industry could be very complicated because large number multi-nature processes takes place industry. This work presents methodology to predict alarms catalytic reforming unit Based refinery data obtained by monitoring status, complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) decompose complex nonlinear series. best parameters classification model were explored effective prediction. results show that, safety based AI7005 unit, reasonable correlation accuracy under proper preprocess, features extraction selection. experimental validation shows 75.9%, which acceptable valuable practice enterprise’s management. study not only provides feasible path practical application, but also gives program reference similar studies analysis.