Active and passive security generalization algorithm and typical case analysis

In order to be able to effectively predict the safety performance of vehicles and reduce the cost of enterprise safety tests, a generalized model for automotive active and passive safety simulation is proposed. The frontal driver's side collision model under AEB intervention is created by MADYMO software, and the collision acceleration obtained from the bench test is used as the model input to simulate the human body in different seating postures, and the damage values of each part of the Hybrid III 50th dummy are read, and the active-passive simulation model is established by BP neural network according to the correlation relationship between the two, and the model input is the inclination angle centered at the waist of the dummy, and the output is the damage values of each part of the dummy. The input of the model is the tilt angle centered on the waist of the dummy, and the output is the damage value of each part of the dummy. The results surface: the prediction accuracy is more than 80%. Therefore, the design of the model algorithm can predict the degree of injuries suffered by the driver when receiving a proof collision, which is conducive to the rapid evaluation of vehicle safety performance.