A high-order cell-centered Lagrangian scheme for one-dimensional elastic–plastic problems. (20th November 2015)

• Record Type: Journal Article
• Authors: Cheng, Jun-Bo; Toro, Eleuterio F.; Jiang, Song; Yu, Ming; Tang, Weijun
• Is Part Of: Computers & fluids. Volume 122(2015)
• Page Start: 136
• Page End: 152
• Publication Date: 2015-11-20
• Subjects: Two-rarefaction Riemann solver -- Elastic–plastic flows -- Cell-centered Lagrangian scheme -- High-order scheme -- Hypo-elastic constitutive model; Fluid dynamics -- Data processing -- Periodicals; 532.050285
• DOI: 10.1016/j.compfluid.2015.08.029
• Languages: English
• ISSNs: 0045-7930

Abstract

Highlights: Develop a two-rarefaction Riemann solver (TRRSE) for 1D elastic–plastic flows. Propose high-order cell-centered Lagrangian schemes for 1D elastic–plastic problems. Show how to limit the time step to keep our high-order scheme positivity-preserving. Our 3rd-order scheme seems to be convergent, stable and essentially non-oscillatory. TRRSE performs better than the node Riemann solvers in resolving rarefaction waves. Abstract: We construct the 2nd-order and 3rd-order cell-centered Lagrangian schemes for 1D elastic–plastic problems with the hypo-elastic constitutive model and the von Mises yield criterion. The basic procedure of the construction is the following: first, we carefully analyze the wave structure of the Riemann problem for elastic–plastic materials and develop a two-rarefaction Riemann solver with elastic waves (TRRSE). Then, based on the developed TRRSE, we propose the 2nd-order and 3rd-order cell-centered Lagrangian schemes for 1D elastic–plastic solid problems. Moreover, we show that our scheme is positivity-preserving, provided the time step is suitably small. A number of numerical experiments are carried out, and the numerical results show that our 3rd-order scheme achieves the desired order of accuracy. Finally, we apply our 2nd-order and 3rd-order schemes to the numerical solution of the problems with elastic shock waves and elastic rarefaction waves, and the numerical results are compared with the reference solution and with the results obtained by other authors. The comparison shows that the current high-order scheme appears to be convergent, stable and essentially non-oscillatory. Moreover, for shock waves the numerical results of our 2nd-order scheme agree very well with those computed by the 2nd-order scheme developed by Maire et al. (2013), while for rarefaction waves the current second-order scheme performs better than Maire et al.'s (2013) second-order scheme. Besides, our third-order scheme performs better than the 2nd-order scheme developed by Maire et al. (2013).