Abstract
Keywords - Artificial Neural Network, Energy Equivalent Speed, model, vehicle, accident
Abstract During a crash involving two vehicles, two phases can be distinguished: the compression phase and the restitution phase. The compression phase lasts from the contact of the vehicles to the point of maximum compression. During this phase, the energy is stocked until the point of maximum deformation. The restitution phase begins when maximum deformation has been reached and ends when the vehicles separate. During this phase, the deformation energy is released. Deformation energy can be written as follows:
Ed= ½ m x EES², with m, the mass of the vehicle (kg), and EES, the Energy Equivalent Speed (.m.s-¹).
The EES is a parameter which is used to estimate the deformation energy absorbed by the vehicles during the collision. EES is currently estimated by purely empirical methods.
The objective of this paper is to propose a model for EES estimation in the case of frontal crashes. In order to do this, an Artificial Neural Network (ANN) is employed. The building of an ANN is composed of two steps: learning and generalization. To estimate the EES with our model, we used the LAB (Laboratory of Accidentology and Biomecanics) database. The methodology and the results are presented in the paper: our model estimates the EES with an error of 4,59 km/h.