Multicriteria Optimization of Injury Prevention Systems to Impact

Abstract

A sensitivity analysis and multicriteria control optimization formulation is derived for mechanical systems. This formulation is implemented into an interactive optimum design code and it is applied to optimize protection systems for the prevention of injuries. The limiting isolation capabilities of the systems are determined. The effect of pre-acting control is investigated. Control forces as well as the time at which the control should act before the instant of impact are considered as design variables. The same idea used by the authors in previous articles for minimum time control problems is applied here to find the preview time. Dynamic response index, maximum acceleration, rattlespace, or maximum power of the resisting force among others can be used as performance criteria. In order to handle the multicriteria problem, both the reduced feasible region method and a min-max upper bound method are utilized. The adjoint system approach is used to calculate the sensitivities. The dynamic response of the systems and its sensitivity are discretized via space-time finite elements. The equations of motion and the sensitivity equations are integrated at-once as it is typical for the static response. This way, displacement, velocity or acceleration control conditions can be imposed easily at any point in time. Also, adjoint system response is obtained without needing primary response memorization. Mathematical programming is used for the optimal control process.

Optimal multiobjective design of a vehicle restraint system using pre-acting control

“…The model shown in Figure 1 represents the human thorax (Cardoso et al , 2008, 2010). Mass m 1 represents the sternum, rib structure and thoracic contents and m 2 represents the remaining portion of the thorax and the part of the total body mass that is coupled to the thorax by the vertebral column.…”

“…A survey of problems related to the limiting performance analysis of injury prevention systems is given in Balandin et al (2003), namely the optimal design of a crashworthy helicopter seat, automobile seatbelts, and sporting helmets. In Cardoso et al (2010), the reduced feasible region (or ε -constraint) is used and a min-max upper bound multicriteria strategy to optimize injury prevention systems with preview control is applied. To find the preview time, it used the same idea as in Moita et al (2008) and Cardoso et al (2008) for minimum time problems.…”