ABSTRACT
To improve numerical calculation efficiency, the relationship between the time step and relative error presented in this study addresses errors caused by the time step during dynamic structural analysis. As a result, not only is the automatic time-step adjustment of the boundary on the target error proven to be effective, but studies on time-adaptive theory have also been enriched, and the method proposed in this article successfully realizes the prediction of the time step. The feasibility of the optimization method was further validated. Specifically, compared with the traditional time-discretization method, Newmark-, the new time-step optimization method could be adopted to save 60.22% of the computing time provided that the corresponding computational accuracy is ensured. Therefore, computational efficiency was substantially improved in an effective manner. Differently from the a posteriori error estimation time-adaptive method, the new method is a bold attempt at an a priori method.
Disclosure statement
No potential conflict of interest was reported by the authors.