INVESTIGATION OF THE TIME EFFICIENCY OF THE SEVENTH-ORDER WENO-S SCHEME
-
摘要: WENO-S格式是一类适合于含间断问题数值模拟的加权本质无振荡格式. 这类格式的光滑因子满足对单频波为常数, 这使得其近似色散关系与线性基底格式一致, 并且具有良好的小尺度波动模拟能力.计算效率是数值方法性能指标的一个重要方面. 由于WENO-S格式的光滑因子在各子模板上的计算公式除下标不同外形式一致, 在计算线性对流方程相邻数值通量时, 部分光滑因子完全相同.为此提出一种消除WENO-S格式冗余光滑因子计算的方法. 该方法要求一条网格线上用于重构或插值的量可以表示为一个序列. 基于此要求分析其对于几种不同物理问题的可行性和使用方法. 以7阶WENO-S格式为例介绍了格式性质和去除冗余光滑因子计算的方法. 该方法中预先计算和存储一条网格线上的所有光滑因子, 在网格点较多的情况下, 光滑因子计算次数约为原7阶WENO-S格式的1/4. 对一维对流问题、球面波传播问题、二维旋转问题、二维小扰动传播问题及一维和二维无黏流动问题进行了数值模拟. 结果表明该格式对多种流动结构具有良好的捕捉能力, 并且同时具有良好的计算效率, 去除冗余计算后又降低了约20%的计算时间.Abstract: The WENO-S scheme is a class of weighted essentially non-oscillatory schemes suitable for numerical simulations of problems with discontinuities. The smoothness indicator of this kind of scheme is constant for single-frequency waves, which makes this kind of scheme have exactly the same approximate dispersion relationship with its linear base scheme, and thus has an excellent ability to simulate small-scale waves. Time efficiency is crucial for numerical methods. For a WENO-S scheme, the formula of the smoothness indicator on each sub-stencil has the same formula except for different subscripts. Then some smoothness indicators are the same when calculating adjacent numerical fluxes of linear convection equations. So, a method is proposed to remove redundant computations of smoothness indicators. The premise of this approach is that the quantity used for reconstruction or interpolation on a grid line can be represented as a sequence. According to this requirement, the feasibility and application requirements for several different physical problems are analyzed. The seventh-order WENO-S scheme is employed to illustrate the advantages of the WENO-S schemes, including good properties near extreme points, good stability near discontinuities, and outstanding spectral properties. Then the method of eliminating the computation of the redundant smoothness indicators is introduced. In numerical computation, all smoothness indicators in a grid line are calculated and stored in advance. With this approach, the count of the smoothness indicator calculation is about 1/4 of the original one for the seventh-order WENO-S scheme when there are many grid points. Numerical examples include one-dimensional advection, spherical wave propagation, two-dimensional rotation, small disturbance propagation, and one- and two- dimensional inviscid flow problems. The numerical results show that this scheme can capture a variety of flow structures well and have good time efficiency. Furthermore, the proposed method reduces the computational time by about 20%.
-
Key words:
- WENO scheme /
- smoothness indicator /
- shock-capturing scheme /
- high order scheme /
- time efficiency
-
表 1 几种WENO格式计算N + 1个数值通量所需浮点运算量
Table 1. The number of floating point operations required to calculate N + 1 numerical fluxes for several WENO schemes
Scheme +, − *, / |·| Total WENO7-JS0 59(N + 1) 87(N + 1) 0 146(N + 1) WENO7-JS1 55(N + 1) 75(N + 1) 0 130(N + 1) WENO7-Z 62(N + 1) 75(N + 1) N + 1 138(N + 1) WENO7-S0 77(N + 1) 51(N + 1) 5(N + 1) 133(N + 1) WENO7-S1 47N + 77 39N + 51 2N + 5 88N + 133 表 2 权系数偏离值的平均值与最大值
Table 2. The average values and maximum values of the deviations of weights
Function WENO7-JS WENO7-Z WENO7-S ${f_1}\left( x \right)$ average 1.56 × 10−3 2.40 × 10−3 2.58 × 10−4 minimum 3.61 × 10−2 1.00 × 10−2 7.58 × 10−4 ${f_2}\left( x \right)$ average 8.49 × 10−4 9.31 × 10−6 4.38 × 10−8 minimum 2.40 × 10−3 4.42 × 10−5 7.27 × 10−8 ${f_3}\left( x \right)$ average 9.03 × 10−2 4.11 × 10−2 2.07 × 10−2 minimum 1.87 × 10−1 1.14 × 10−1 5.43 × 10−2 表 3 组合波远距传输问题的计算时间
Table 3. Computational time of long-distance advection of combined waves
Scheme WENO7-JS0 WENO7-JS1 WENO7-Z WENO7-S0 WENO7-S1 time/s 636.39 580.32 645.67 546.09 416.99 表 4 球面波传播问题的计算时间, dt = 0.01
Table 4. Computational time of spherical wave propagation problem with dt = 0.01
Scheme WENO7-JS0 WENO7-JS1 WENO7-Z WENO7-S0 WENO7-S1 time/s 1.422 1.336 1.465 1.219 0.953 表 5 几种格式不同网格下计算结果的L1误差和上冲/下冲幅度
Table 5. The L1 error and up/down overshooting amplitude of the computational results with different schemes and grids
Grid WENO7-JS WENO7-Z WENO7-S L1 error 200 × 200 1.759 1.428 1.600 400 × 400 0.959 0.788 0.886 800 × 800 0.524 0.430 0.484 up/down overshooting amplitude $\delta$ 200 × 200 1.11 × 10−4 1.45 × 10−2 5.02 × 10−3 400 × 400 9.86 × 10−5 9.13 × 10−3 2.72 × 10−4 800 × 800 1.27 × 10−4 2.58 × 10−2 2.87 × 10−8 表 6 二维旋转问题的计算时间
Table 6. Computational time for the two-dimensional rotation problem
Scheme WENO7-JS0 WENO7-JS1 WENO7-Z WENO7-S0 WENO7-S1 time/s 114.731 107.400 117.493 99.744 76.327 表 7 二维小扰动传播问题的计算时间
Table 7. Computational time of two-dimensional small disturbance propagation problem
Scheme WENO7-JS0 WENO7-JS1 WENO7-Z WENO7-S0 WENO7-S1 time/s 8.842 7.780 8.377 7.485 6.150 表 8 Shu-Osher问题计算时间, N = 2000
Table 8. Computational time of Shu-Osher problem, N = 2000
Scheme WENO7-JS0 WENO7-JS1 WENO7-Z WENO7-S0 WENO7-S1 time/s 3.330 3.037 3.317 3.015 2.255 表 9 激波旋涡相互作用问题的计算时间
Table 9. Computational time of shock vortex interaction problem
Scheme WENO7-JS0 WENO7-JS1 WENO7-Z WENO7-S0 WENO7-S1 time/s 14.350 13.195 14.520 13.007 10.159 -
[1] Harten A. High resolution schemes for hyperbolic conservation laws. Journal of Computational Physics, 1983, 49(3): 357-393 doi: 10.1016/0021-9991(83)90136-5 [2] 张涵信. 无波动、无自由参数的耗散差分格式. 空气动力学学报, 1988, 6(2): 143-165 (Zhang Hanxin. Non-oscillatory and non-free-parameter dissipation difference scheme. Acta Aerodynamica Sinica, 1988, 6(2): 143-165 (in Chinese) [3] Liu XD, Osher S, Chan T. Weighted essentially non-oscillatory shock-capturing schemes. Journal of Computational Physics, 1994, 115(1): 200-212 doi: 10.1006/jcph.1994.1187 [4] Jiang GS, Shu CW. Efficient implementation of weighted ENO schemes. Journal of Computational Physics, 1996, 126(1): 202-228 doi: 10.1006/jcph.1996.0130 [5] Henrick AK, Aslam TD, Powers JM. Mapped weighted essentially non-oscillatory schemes: achieving optimal order near critical points. Journal of Computational Physics, 2005, 207(2): 542-567 doi: 10.1016/j.jcp.2005.01.023 [6] Borges R, Carmona M, Costa B, et al. An improved weighted essentially non-oscillatory scheme for hyperbolic conservation laws. Journal of Computational Physics, 2008, 227(6): 3192-3211 [7] 骆信, 吴颂平. 改进的5阶WENO-Z + 格式. 力学学报, 2019, 51(6): 1927-1939 (Luo Xin, Wu Songping. An improved fifth-order WENO-Z + scheme. Chinese Journal of Theoretical and Applied Mechanics, 2019, 51(6): 1927-1939 (in Chinese) doi: 10.6052/0459-1879-19-249 [8] Ha Y, Kim CH, Lee YJ, et al. An improved weighted essentially non-oscillatory scheme with a new smoothness indicator. Journal of Computational Physics, 2013, 232(1): 68-86 doi: 10.1016/j.jcp.2012.06.016 [9] Hu XY, Wang Q, Adams NA. An adaptive central-upwind weighted essentially non-oscillatory scheme. Journal of Computational Physics, 2010, 229(23): 8952-8965 doi: 10.1016/j.jcp.2010.08.019 [10] Zhang SH, Shu CW. A new smoothness indicator for the WENO Schemes and its effect on the convergence to steady state. Journal of Scientific Computing, 2007, 31: 273-305 doi: 10.1007/s10915-006-9111-y [11] Yamaleev NK, Carpenter MH. A systematic methodology for constructing high-order energy stable WENO schemes. Journal of Computational Physics, 2009, 228(11): 4248-4272 doi: 10.1016/j.jcp.2009.03.002 [12] Deng XG, Zhang HX. Developing high-order weighted compact nonlinear schemes. Journal of Computational Physics, 2000, 165(1): 22-44 doi: 10.1006/jcph.2000.6594 [13] 邓小刚, 刘昕, 毛枚良等. 高精度加权紧致非线性格式的研究进展. 力学进展, 2007, 37(3): 417-427 (Deng Xiaogang, Liu Xin, Mao Meiliang, et al. Advances in high-order accurate weighted compact nonlinear schemes. Advances in Mechanics, 2007, 37(3): 417-427 (in Chinese) doi: 10.3321/j.issn:1000-0992.2007.03.009 [14] 张树海. 加权型紧致格式与加权本质无波动格式的比较. 力学学报, 2016, 48(2): 336-347 (Zhang Shuhai. The comparison of weighted compact schemes and weno scheme. Chinese Journal of Theoretical and Applied Mechanics, 2016, 48(2): 336-347 (in Chinese) doi: 10.6052/0459-1879-15-190 [15] Friedrichs O. Weighted essentially non-oscillatory schemes for the interpolation of mean values on unstructured grids. Journal of Computational Physics, 1998, 144(1): 194-212 doi: 10.1006/jcph.1998.5988 [16] Hu C, Shu CW. Weighted essentially non-oscillatory schemes on triangular meshes. Journal of Computational Physics, 1999, 150(1): 97-127 doi: 10.1006/jcph.1998.6165 [17] Wu CH, Wu L, Zhang SH. A smoothness indicator constant for sine functions. Journal of Computational Physics, 2020, 419: 109661 doi: 10.1016/j.jcp.2020.109661 [18] Wu CH, Wu L, Li H, et al. Very high order WENO schemes using efficient smoothness indicators. Journal of Computational Physics, 2021, 432: 110158 doi: 10.1016/j.jcp.2021.110158 [19] Pirozzoli S. On the spectral properties of shock-capturing schemes. Journal of Computational Physics, 2006, 219(2): 489-497 doi: 10.1016/j.jcp.2006.07.009 [20] Don WS, Borges R. Accuracy of the weighted essentially non-oscillatory conservative finite difference schemes. Journal of Computational Physics, 2013, 250: 347-372 doi: 10.1016/j.jcp.2013.05.018 [21] Lele SK. Compact difference schemes with spectral-like resolution. Journal of Computational Physics, 1992, 103(1): 16-42 doi: 10.1016/0021-9991(92)90324-R [22] Li XL, Leng Y, He ZW. Optimized sixth-order monotonicity-preserving scheme by nonlinear spectral analysis. International Journal for Numerical Methods in Fluids, 2013, 73(6): 560-577 doi: 10.1002/fld.3812 [23] 毛枚良, 燕振国, 刘化勇等. 高阶加权非线性格式的拟线性频谱分析方法研究. 空气动力学学报, 2015, 33(1): 1-9 (Mao Meiliang, Yan Zhenguo, Liu Huayong, et al. Study of quasi-linear spectral analysis method of high-order weighted nonlinear schemes. Acta Aerodynamica Sinica, 2015, 33(1): 1-9 (in Chinese) [24] Zhao S, Lardjane N, Fedioun I. Comparison of improved finite-difference WENO schemes for the implicit large eddy simulation of turbulent non-reacting and reacting high-speed shear flows. Computers and Fluids, 2014, 95: 74-87 doi: 10.1016/j.compfluid.2014.02.017 [25] Li H, Luo Y, Zhang SH. Assessment of upwind/symmetric weno schemes for direct numerical simulation of screech tone in supersonic jet. Journal of Scientific Computing, 2021, 87: 3 doi: 10.1007/s10915-020-01407-6 [26] 李虎, 罗勇, 刘旭亮等. 一种面向激波噪声计算的加权优化紧致格式及非线性效应分析. 力学学报, 2022, 54(10): 2747-2759 (Li Hu, Luo Yong, Liu Xuliang, et al. A weighted-optimization compact scheme for shock-associated noise computation and its nonlinear effect analysis. Chinese Journal of Theoretical and Applied Mechanics., 2022, 54(10): 2747-2759 (in Chinese) doi: 10.6052/0459-1879-22-254 [27] Cravero I, Puppo G, Semplice M, et al. Cool WENO schemes. Computers and Fluids, 2018, 169: 71-86 doi: 10.1016/j.compfluid.2017.07.022 [28] Castro M, Costa B, Don WS. High order weighted essentially non-oscillatory WENO-Z schemes for hyperbolic conservation laws. Journal of Computational Physics, 2011, 230(5): 1766-1792 doi: 10.1016/j.jcp.2010.11.028 [29] Balsara DS, Shu CW. Monotonicity preserving weighted essentially non-oscillatory schemes with increasingly high-order of accuracy. Journal of Computational Physics, 2000, 160(2): 405-452 doi: 10.1006/jcph.2000.6443 [30] Baeza A, Burger R, Mulet P, et al. On the efficient computation of smoothness indicators for a class of a WENO reconstructions. Journal of Scientific Computing, 2019, 80: 1240-1263 doi: 10.1007/s10915-019-00974-7 [31] 刘旭亮, 武从海, 李虎等. WENO格式的一种新型光滑因子及其应用. 空气动力学学报, 2022, 40: 1-14 (Liu Xuliang, Wu Conghai, Li Hu, et al. A new type of smoothness indicator for weighted essentially non-oscillatory schemes and its application. Acta Aerodynamica Sinica, 2022, 40: 1-14 (in Chinese) [32] Rathan S, Raju GN, Bhise AA. Simple smoothness indicator WENO-Z scheme for hyperbolic conservation laws. Applied Numerical Mathematics, 2020, 157: 255-275 [33] 柳军, 刘伟, 曾明等. 高超声速三维热化学非平衡流场的数值模拟. 力学学报, 2003, 35(6): 730-734 (Liu Jun, Liu Wei, Zeng ming, et al. Numerical simulation of 3 D hypersonic thermochemical nonequilibrium flow. Chinese Journal of Theoretical and Applied Mechanics, 2003, 35(6): 730-734 (in Chinese) doi: 10.3321/j.issn:0459-1879.2003.06.011 [34] 王东红. 多介质流体界面追踪方法研究及误差分析. [博士论文]. 南京: 南京航空航天大学, 2014: 1-7Wang Donghong. Investigation of front tracking method for multi-component flow and error analysis. [PhD Thesis]. Nanjing: Nanjing University of Aeronautics and Astronautics, 2014: 1-7 (in Chinese) [35] Hennessy JL, Patterson DA. 计算机体系结构量化研究方法(英文版第5版). 北京: 人民邮电出版社出版, 2013: 193-255(Hennessy JL, Patterson DA. Computer Architecture: A Quantitative Approach, Fifth Edition. Beijing: Morgan Kaufmann Publishers Inc, 2013: 193-255 (in Chinese)) [36] Hardin JC, Ristorcelli JR, Tam CKW. Benchmark problems and solutions. ICASE/LaRC Workshop on Benchmark Problems in Computational Aeroacoustics, NASA CP-3300, 1995: 1-14 [37] Zalesak ST. Fully multidimensional flux-corrected transport algorithms for fluids. Journal of Computational Physics, 1979, 31(3): 335-362 doi: 10.1016/0021-9991(79)90051-2 [38] Liu XL, Zhang SH, Zhang HX, et al. A new class of central compact schemes with spectral-like resolution I: Linear schemes. Journal of Computational Physics, 2013, 248: 235-256 -