Structured Scenario System Model for Unconventional Emergency Based on Random Forest
In order to reduce the losses caused by the occurrence of unconventional emergencies,a structured scenario system model of unconventional emergencies based on random forest is constructed.According to the theoretical knowledge of random forest,python is used to conduct experimental simulation,and the collapse accident of traffic tunnel engineering is treated as an unconventional emergency.First of all,aiming at the problems of the traditional structural scenario model of unconventional emergencies,the structured description space of unconventional emergencies is constructed,the random forest model is used to classify the unconventional emergencies,and the structured description of the unconventional emergencies is carried out according to the classification results.Secondly,a comparative experiment based on the Naive Bayes model and the support vector machine model is constructed,in the same experimental environment,different data sets are used for multiple experimental analysis and comparison.Finally,the accident type is predicted based on the random forest regression model,the decision tree model is used to visualize the data,and the time efficiency and accuracy of predicting accident type using important features are calculated.The practicability of the random forest regression model is verified by the actual traffic tunnel collapse accident.The result shows that(1)the average accuracy of the random forest model is improved by 8.77%compared with the other two models;(2)the simulation results are close to the real values,and the simulation accuracy value is 1.The study result provides emergency decision-making basis for national emergency management departments,and helps decision-makers to put forward reasonable emergency response and emergency early-warning plans in the shortest time,it can effectively reduce the economic losses caused by non-conventional emergencies,therefore,the random forest model has high value.
tunnel engineeringstructured scenario systemrandom forestunconventional emergencyemergency type
NETL
NSTL
万方数据
