水准点(测量)
计算机科学
解算器
集合(抽象数据类型)
软件
微分方程
常微分方程
数学优化
离散化
计算科学
数学
大地测量学
数学分析
程序设计语言
地理
作者
Bernhard Manhartsgruber
标识
DOI:10.1115/fpmc2022-90185
摘要
Abstract Mathematical modelling and simulation have become widespread tools in the fluid power industry. The sheer number of available software packages of commercial and academic nature as well as the abundance of published modeling strategies for certain components gives multiple possibilities for modeling one and the same fluid power system. The mathematical model is required to capture the behavior of the real system with a certain degree of accuracy while using a minimum of computational resources for reaching this goal. This trade-off cannot be assessed by looking on model complexity on the level of differential equations only. The computational efficiency is defined by the combination of mathematical modelling on the high level of differential equations, the choice of a solver, i.e. the time discretization, and the details of software implementation like the optimized compilation of right hand sides for the solvers. In order to compare different modeling strategies on various software platforms, well defined benchmark problems are needed. This paper presents a first attempt towards such a set of benchmark problems targeted at the specific problems of fluid power simulation. In a first benchmark case, a system with reversing flow at an orifice under the additional difficulty of a stiff differential equation system is presented. A second benchmark case uses the coupling between dry friction in a hydraulic cylinder and wave propagation in a transmission line which has been shown to give notorious modelling problems in the literature. The goal of the present publication is to present a proposal for such benchmark systems in order to make the accuracy and computational performance of simulation approaches more comparable.
科研通智能强力驱动
Strongly Powered by AbleSci AI