zbMATH — the first resource for mathematics

A methodology for the development of discrete adjoint solvers using automatic differentiation tools. (English) Zbl 1184.76832
Summary: A methodology for the rapid development of adjoint solvers for computational fluid dynamics (CFD) models is presented. The approach relies on the use of automatic differentiation (AD) tools to almost completely automate the process of development of discrete adjoint solvers. This methodology is used to produce the adjoint code for two distinct 3D CFD solvers: a cell-centred Euler solver running in single-block, single-processor mode and a multi-block, multi-processor, vertex-centred, magneto-hydrodynamics (MHD) solver. Instead of differentiating the entire source code of the CFD solvers using AD, we have applied it selectively to produce code that computes the transpose of the flux Jacobian matrix and the other partial derivatives that are necessary to compute sensitivities using an adjoint method. The discrete adjoint equations are then solved using the Portable, Extensible Toolkit for Scientific Computation (PETSc) library. The selective application of AD is the principal idea of this new methodology, which we call the AD adjoint (ADjoint). The ADjoint approach has the advantages that it is applicable to any set of governing equations and objective functions and that it is completely consistent with the gradients that would be computed by exact numerical differentiation of the original discrete solver. Furthermore, the approach does not require hand differentiation, thus avoiding the long development times typically required to develop discrete adjoint solvers for partial differential equations, as well as the errors that result from the necessary approximations used during the differentiation of complex systems of conservation laws. These advantages come at the cost of increased memory requirements for the discrete adjoint solver. However, given the amount of memory that is typically available in parallel computers and the trends toward larger numbers of multi-core processors, this disadvantage is rather small when compared with the very significant advantages that are demonstrated. The sensitivities of drag and lift coefficients with respect to different parameters obtained using the discrete adjoint solvers show excellent agreement with the benchmark results produced by the complex-step and finite-difference methods. Furthermore, the overall performance of the method is shown to be better than most conventional adjoint approaches for both CFD solvers used.

76N25 Flow control and optimization for compressible fluids and gas dynamics
76W05 Magnetohydrodynamics and electrohydrodynamics
76M99 Basic methods in fluid mechanics
Full Text: DOI
[1] Balay S., Proceedings of the Modern Software Tools in Scientific Computing pp 163– (1997) · doi:10.1007/978-1-4612-1986-6_8
[2] Balay, S., Buschelman, K., Eijkhout, V., Gropp, W.D., Kaushik, D., Knepley, M.G., McInnes, L.C., Smith, B.F. and Zhang, H., PETSc user manual, Technical report ANL-95/11–Revision 2.3.0, Argonne National Laboratory. 2004
[3] Carle, A. and Fagan, M., ADIFOR 3.0 overview, Technical report CAAM-TR-00-02, Rice University 2000
[4] Carle A., AIAA Paper 1998–4807 pp 1998– (1998)
[5] DOI: 10.1006/jcph.1994.1057 · Zbl 0832.65098 · doi:10.1006/jcph.1994.1057
[6] DOI: 10.1006/jcph.1998.6114 · Zbl 0921.65059 · doi:10.1006/jcph.1998.6114
[7] DOI: 10.1002/fld.885 · Zbl 1134.76431 · doi:10.1002/fld.885
[8] Driver J., Proceedings of the 44th AIAA Aerospace Sciences Meeting and Exhibit (2006)
[9] Dwight R.P., Proceedings of the 44th AIAA Aerospace Sciences Meeting and Exhibit (2006)
[10] DOI: 10.1016/j.future.2004.11.005 · doi:10.1016/j.future.2004.11.005
[11] Faure, C. and Papegay, Y., Chapter title, Odyssée Version 1.6: The Language Reference Manual, INRIA Rapport Technique 211
[12] Forth S.A., Automatic Differentiation: From Simulation to Optimization pp 153– (2002)
[13] Gaitonde D.V., Proceedings of the 43rd AIAA Aerospace Sciences Meeting and Exhibit (2005)
[14] Giering R., Proceedings of the GAMM 2002 (2002)
[15] DOI: 10.1016/j.future.2004.11.003 · doi:10.1016/j.future.2004.11.003
[16] DOI: 10.1023/A:1011430410075 · Zbl 0996.76023 · doi:10.1023/A:1011430410075
[17] Gockenbach M.S., Understanding Code Generated by TAMC (2000)
[18] Griewank A., Evaluating Derivatives (2000)
[19] Hascoët, L. and Pascual, V., TAPENADE 2.1 user’s guide, Technical report 300, INRIA 2004
[20] Hascoët L., Automatic Differentiation: Applications, Theory, and Tools (2005)
[21] DOI: 10.1016/j.future.2004.11.010 · doi:10.1016/j.future.2004.11.010
[22] Horwedel J.E., Automatic Differentiation of Algorithms: Theory, Implementation and Application pp 243– (1991)
[23] DOI: 10.1007/BF01061285 · Zbl 0676.76055 · doi:10.1007/BF01061285
[24] Lee K.H., Proceedings of the 44th AIAA Aerospace Sciences Meeting and Exhibit (2006)
[25] Marta A.C., Proceedings of the 3rd AIAA Flow Control Conference (2006)
[26] Marta A.C., Proceedings of the 14th AIAA/AHI International Space Planes and Hypersonic Systems and Technologies Conference (2006)
[27] Martins J.R.R.A., Proceedings of the 39th Aerospace Sciences Meeting (2001)
[28] DOI: 10.1145/838250.838251 · Zbl 1072.65027 · doi:10.1145/838250.838251
[29] DOI: 10.2514/1.11478 · doi:10.2514/1.11478
[30] DOI: 10.1023/B:OPTE.0000048536.47956.62 · Zbl 1145.76418 · doi:10.1023/B:OPTE.0000048536.47956.62
[31] Martins J.R.R.A., Proceedings of the 2nd AIAA Multidisciplinary Design Optimization Specialist Conference (2006)
[32] Martins J.R.R.A., Proceedings of the 11th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference (2006)
[33] DOI: 10.1016/j.jcp.2004.03.001 · Zbl 1071.65025 · doi:10.1016/j.jcp.2004.03.001
[34] DOI: 10.1023/B:JOMP.0000027955.75872.3f · Zbl 1085.76050 · doi:10.1023/B:JOMP.0000027955.75872.3f
[35] Mohammadi B., Computational Differentiation: Techniques, Applications, and Tools pp 309– (1996)
[36] Int. J. Supercomputer Appl. High Perform. Comput. 8 pp 159– (1994)
[37] DOI: 10.1145/1114268.1114270 · Zbl 1136.65310 · doi:10.1145/1114268.1114270
[38] DOI: 10.2514/1.10415 · doi:10.2514/1.10415
[39] DOI: 10.1017/S0022112074002023 · Zbl 0281.76020 · doi:10.1017/S0022112074002023
[40] DOI: 10.2514/2.2413 · doi:10.2514/2.2413
[41] DOI: 10.2514/2.2414 · doi:10.2514/2.2414
[42] DOI: 10.1137/0907058 · Zbl 0599.65018 · doi:10.1137/0907058
[43] DOI: 10.1006/jcph.1996.0252 · Zbl 0933.76070 · doi:10.1006/jcph.1996.0252
[44] Van der Weide E., Proceedings of the 44th AIAA Aerospace Sciences Meeting and Exhibit (2006)
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.