Dynamic fluid-solid interactions are widely found in chemical engineering, such as in particle-laden flows, which usually contain complex moving boundaries. The immersed boundary method （IBM） is a convenient approac...Dynamic fluid-solid interactions are widely found in chemical engineering, such as in particle-laden flows, which usually contain complex moving boundaries. The immersed boundary method （IBM） is a convenient approach to handle fluid-solid interactions with complex geometries. In this work, Uhlmann＇s direct-forcing IBM is improved and implemented on a supercomputer with CPU-GPU hybrid architec- ture. The direct-forcing IBM is modified as follows： the Poisson＇s equation for pressure is solved before evaluation of the body force, and the force is only distributed to the Cartesian grids inside the immersed boundary. A multidirect forcing scheme is used to evaluate the body force. These modifications result in a divergence-free flow field in the fluid domain and the no-slip boundary condition at the immersed boundary simultaneously. This method is implemented in an explicit finite-difference fractional-step scheme, and validated by 2D simulations of lid-driven cavity flow, Couette flow between two concentric cylinders and flow over a circular cylinder. Finally, the method is used to simulate the sedimentation of two circular particles in a channel. The results agree very well with previous experimental and numerical data, and are more accurate than the conventional direct-forcing method, especially in the vicinity of a moving boundary.展开更多
This paper aims to solve large-scale and complex isogeometric topology optimization problems that consumesignificant computational resources. A novel isogeometric topology optimization method with a hybrid parallelstr...This paper aims to solve large-scale and complex isogeometric topology optimization problems that consumesignificant computational resources. A novel isogeometric topology optimization method with a hybrid parallelstrategy of CPU/GPU is proposed, while the hybrid parallel strategies for stiffness matrix assembly, equationsolving, sensitivity analysis, and design variable update are discussed in detail. To ensure the high efficiency ofCPU/GPU computing, a workload balancing strategy is presented for optimally distributing the workload betweenCPU and GPU. To illustrate the advantages of the proposedmethod, three benchmark examples are tested to verifythe hybrid parallel strategy in this paper. The results show that the efficiency of the hybrid method is faster thanserial CPU and parallel GPU, while the speedups can be up to two orders of magnitude.展开更多
基金supported by the National Natural Science Foundation of China(NSFC) under Grant Nos.21225628,51106168 and 11272312the "Strategic Priority Research Program" of Chinese Academy of Sciences(CAS) under Grant No.XDA07080102
文摘Dynamic fluid-solid interactions are widely found in chemical engineering, such as in particle-laden flows, which usually contain complex moving boundaries. The immersed boundary method （IBM） is a convenient approach to handle fluid-solid interactions with complex geometries. In this work, Uhlmann＇s direct-forcing IBM is improved and implemented on a supercomputer with CPU-GPU hybrid architec- ture. The direct-forcing IBM is modified as follows： the Poisson＇s equation for pressure is solved before evaluation of the body force, and the force is only distributed to the Cartesian grids inside the immersed boundary. A multidirect forcing scheme is used to evaluate the body force. These modifications result in a divergence-free flow field in the fluid domain and the no-slip boundary condition at the immersed boundary simultaneously. This method is implemented in an explicit finite-difference fractional-step scheme, and validated by 2D simulations of lid-driven cavity flow, Couette flow between two concentric cylinders and flow over a circular cylinder. Finally, the method is used to simulate the sedimentation of two circular particles in a channel. The results agree very well with previous experimental and numerical data, and are more accurate than the conventional direct-forcing method, especially in the vicinity of a moving boundary.
基金the National Key R&D Program of China(2020YFB1708300)the National Natural Science Foundation of China(52005192)the Project of Ministry of Industry and Information Technology(TC210804R-3).
文摘This paper aims to solve large-scale and complex isogeometric topology optimization problems that consumesignificant computational resources. A novel isogeometric topology optimization method with a hybrid parallelstrategy of CPU/GPU is proposed, while the hybrid parallel strategies for stiffness matrix assembly, equationsolving, sensitivity analysis, and design variable update are discussed in detail. To ensure the high efficiency ofCPU/GPU computing, a workload balancing strategy is presented for optimally distributing the workload betweenCPU and GPU. To illustrate the advantages of the proposedmethod, three benchmark examples are tested to verifythe hybrid parallel strategy in this paper. The results show that the efficiency of the hybrid method is faster thanserial CPU and parallel GPU, while the speedups can be up to two orders of magnitude.
文摘系统地探讨使用CPU/GPU协同处理理论对高分辨率卫星影像进行MTF补偿的方法。首先在GPU上对方法进行基本实现,并通过3种性能优化策略(执行配置优化、存储访问优化和指令优化)进一步提高了方法的执行效率。在Intel Xeon E5650CPU和NVIDIA Tesla C2050GPU组成的CPU/GPU系统中对高分一号卫星全色影像进行MTF补偿,加速比达到42.80倍。在此基础上,为充分利用CPU的计算性能,使用CPU/GPU负载分配策略将部分负载分配给CPU进行处理,使用该策略后,方法加速比达到47.82倍,相应的处理时间压缩至1.62s,可满足对高分辨率卫星影像进行近实时MTF补偿的需求。
