船舶与海洋工程流固耦合数值方法研究进展

Research developments in numerical methods of fluid-structure interactions in naval architecture and ocean engineering

  • 摘要: 流固耦合问题较为复杂,通常难以通过理论推导求得,而数值模拟则能提供一种有效的解决方案,并被广泛用于船舶与海洋工程领域。流固耦合数值方法根据其网格离散方式,可以分为贴体网格方法、非贴体网格方法、重叠网格方法和粒子类方法4类,对这4类方法的特点及研究进展进行概述并总结得出:贴体网格方法和重叠网格方法均能精确捕捉界面的变形和演化,适合高雷诺数流动问题,在考虑结构变形时一般采用贴体网格方法,而考虑复杂几何形状的刚体运动时则常采用重叠网格方法;非贴体网格方法能够避免网格的更新操作,使计算较为简单,目前多用于模拟流动控制、水下柔性仿生航行器的研发以及多体运动干扰等问题;粒子类方法因其固有的拉格朗日属性,在模拟涉及自由液面剧烈变形、砰击、爆炸等强非线性流固耦合问题中发挥着重要作用。不同的流固耦合问题属性决定了不同方法的适用性,如何选取适合的数值方法,同时结合各类方法的优势开发新的计算方法以应对更为复杂的问题,是流固耦合算法开发的重要发展方向。

     

    Abstract: It is a challenge to solve complex fluid-structure interaction (FSI) problems through theoretical derivations, whereas numerical simulation provides an effective solution and is widely applied in naval architecture and marine engineering. Based on grid treatment, FSI methods are classified into the body-fitted grid method, non-body-fitted grid method, overset grid method and particle-based method. The research development of these four types of methods is then reviewed. Both the body-fitted grid method and overset grid method can accurately capture the interface and are suitable for high Reynolds number flow problems, and the former is generally employed when structural deformation is considered, while the latter often works well when considering rigid body motion with complex geometric shapes. The non-body-fitted grid method can avoid the mesh update operation to make calculations simpler, and is widely used in the simulation of flow control, development of underwater flexible bionic vehicles and interference of multi-body motion. The particle-based method plays an increasingly important role in simulating strong nonlinear fluid-structure interaction problems involving severe free surface deformation, slamming, explosion, etc. The properties of different FSI problems determine the applicability of different methods. How to select a suitable numerical method and combine the advantages of various methods to develop novel numerical methods that can handle more challenging problems are important development directions for FSI algorithms.

     

/

返回文章
返回