Multi-objective optimization of a magnetorheological seat belt system for increasing passenger safety

Abstract

Seat belt effectiveness plays an important role in reducing occupant injuries during severe braking. Although, conventional three-point seat belt system is widely used, but it has certain limitations, particularly regards to adaptation to different occupants and collision scenarios. To address this challenge, in this paper, the effect of adding a magnetorheological (MR) damper on the performance of the seat belt mechanism is investigated. For this purpose, the multi-body dynamic model of a half-vehicle, including the seat, passenger, and belt mechanism including a MR damper, is developed using the bond graph method. In order to model the seat belt system including the MR damper, the nonlinear modified Bouc–Wen model with hysteresis behavior is used. Using the developed bond graph model, the state differential equations of the system are derived. Then, using the particle swarm optimization (PSO) algorithm, the optimal belt parameters are determined to reduce head acceleration, neck forces and torques, and chest forces and accelerations during vehicle severe braking. The optimization results show that using MR damper instead of simple viscous damper reduces thoracic injury criterion (TIC) and weighted injury criterion (WIC) by 50%. By varying input voltage, the MR damper is continuously adjusting the seat belt rotational damping and force for optimized occupant motion.