Abstract

A benchmark problem is defined to test potential solutions for re-usable submodels representing static (‘stick’) and dynamic dry (‘Coulomb’) friction in dynamic system simulation that are based on relevant physical port behavior. Next an alternative submodel is proposed with several advantages over other possible implementations. It is shown that dynamic causality and dynamic model structure can always be prevented and that some solutions for numerical computation can be completely embedded in a one-port resistor, if necessary externally modulated by the normal force between the contact surfaces. This makes these solutions highly re-usable, although they create a trade-off between accuracy and numerical stiffness. The key issue is that the generic expression for the contact force derived originally to solve the case with dynamic structure, results in a numerically robust and efficient submodel with static port causality and static model structure that does not create this numerical stiffness.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call