Research On Rapid Prediction Of Driver Injury In Frontal Collision Of Minivan

In recent years,the harm caused by vehicle traffic accidents can’t be ignored,and China is still one of the countries with a high incidence of traffic accidents in the world.How to reduce the severity of passenger injury in traffic accidents and how to quickly take rescue measures is a research hotspot in the field of automobile safety.The occupant injury prediction algorithm is an effective post-accident safety technology that can quickly predict occupant injury after an accident occurs.It can provide technical support for medical personnel to carry out rapid triage,thus effectively improving the efficiency of accident rescue and reducing the degree of occupant injury.Therefore,this paper is of great significance for the study of driver injury prediction.Based on the above reasons,this paper takes minivan as the research object to carry out the research work of driver injury prediction under the frontal collision.Firstly,the statistical analysis of accident data is carried out for the type of minivan,and the factors affecting the severity of driver injury are determined according to the results of correlation analysis.Secondly,the finite element model and driver restraint system model of the minivan are built and verified.Based on the established model,combined with the experimental design strategy,and according to the injury evaluation criteria,the construction of the driver injury data sets are completed.Finally,the BP neural network is used to build the driver injury prediction model,and the improved whale algorithm is used to optimize the BP prediction model.The real accidents are applied to the driver damage prediction model,and evaluate the prediction performance of the model.The results of this paper show that:(1)The severity of driver injury is significantly correlated with longitudinal maximum delta velocity,seat belt wear,and airbag deployment.(2)The prediction accuracy of MWOA-BP prediction model is higher than that of BP prediction model,and the average absolute error value of MWOA-BP prediction model for driver head injury prediction MAE is 0.1,mean square error value MSE is 0.1;MAE value for chest injury prediction is 0.125,mean square error value is 0.15.(3)The MWOA-BP prediction model is verified by using 62 accidents,and the prediction results of 54 accidents are consistent with the actual results,3 accidents have a predicted value greater than the actual value,and 5 accidents have a predicted value less than the actual value;The ROC analysis result is AUC=0.79,and the established injury prediction model has good prediction performance.Finally,according to the research results,driver-side safety suggestions and damage protection suggestions are proposed.