Non–Geometric Sensitivities Using the Adjoint Method

Authors: Marcelo Hayashi, João Lima, Bruno Chieregatti, Ernani Volpe

Abstract:

The adjoint method has been used as a successful tool to obtain sensitivity gradients in aerodynamic design and optimisation for many years. This work presents an alternative approach to the continuous adjoint formulation that enables one to compute gradients of a given measure of merit with respect to control parameters other than those pertaining to geometry. The procedure is then applied to the steady 2–D compressible Euler and incompressible Navier–Stokes flow equations. Finally, the results are compared with sensitivities obtained by finite differences and theoretical values for validation.

Keywords: Adjoint method, optimisation, non–geometric sensitivities, boundary conditions.

Digital Object Identifier (DOI): doi.org/10.5281/zenodo.1338772

References:

[1] O. Pirroneau, “On optimal profiles in stokes flow,” Journal of Fluid Mechanics, vol. 59, no. 1, pp. 117–128, 1973.
[2] A. Jameson, “Aerodynamic design via control theory,” in 12th IMACS World Congress on Scientific Computation, ser. MAE Report 1824, Paris, July 1988.
[3] H. Cabuk, C.-H. Sung, and V. Modi, “Adjoint operator approach to shape design for incompressible flows,” in 3rd International Conference on Inverse Design Concepts and Optimization in Engineering Sciences (ICIDES), College Park, PA, 1991, pp. 391–404.
[4] S. Taasan, G. Kuruvila, and M. D. Salas, “Aerodynamic design and optimization in one shot,” in 30th Aerospace Sciences Meeting and Exhibit, Reno, NV, January 1992, aIAA 92–0025.
[5] G. Kuruvila, S. Taasan, and M. D. Salas, “Airfoil optimization by the one–shot method, optimum design methods in aerodynamics,” AGARD–FDP–VKI Special Course, 1994.
[6] L. C. C. Santos, “A study on aerodynamic design optimization using an adjoint method,” Institut f¨ur Entwurfsaerodynamik, Braunschweig, Tech. Rep. IB 129 - 95/12, July 1995.
[7] A. Jameson, L. Martinelli, and N. A. Pierce, “Optimum aerodynamic design using the Navier-Stokes equations,” in 35th Aerospace Sciences Meeting & Exhibit, American Institute of Aeronautics and Astronautics. Reno, NV: AIAA, January 1997, pp. 1–20.
[8] S. Kim, “Design optimization of high–lift configurations using a viscous adjoint–based method,” Ph.D. dissertation, Stanford University, 2001.
[9] B. Mohammadi and O. Pirroneau, Applied Shape Optimization for Fluids, 1st ed. Oxford University Press, 2001.
[10] J. J. Alonso and I. M. Kroo, “Advanced algorithms for design and optimization of quiet supersonic platforms,” in AIAA Computational Fluid Dynamics Conference, Reno, NV, January 2002, aIAA–2002–0144.
[11] D. G. Cacuci, Weber, C. F., O. E. M., and J. H. Marable, “Sensitivity theory for general systems of non–linear equations,” Nuclear Science and Engineering, vol. 75, pp. 88–110, 1980.
[12] M. Hall and D. Cacuci, “Physical interpretation of the adjoint functions for sensitivity analysis of atmospheric models,” Journal of Atmospheric Sciences, vol. 40, pp. 2537–2546, October 1983.
[13] S. K. Nadarajah, “The discrete adjoint approach to aerodynamic shape optimization,” Ph.D. dissertation, Stanford University, 2003.
[14] H. Kim and K. Nakahashi, “Unstructured adjoint method for Navier-Stokes equations,” JSME International Journal, vol. 48, no. 2, 2005.
[15] S. Kim, J. J. Alonso, and A. Jameson, “Multi–element high–lift configuration design optimization using viscous continuous adjoint method,” Journal of Aircraft, vol. 41, no. 5, September–October 2004.
[16] H. J. Kim, D. Sasaki, S. Obayashi, and K. Nakahashi, “Aerodynamic optimization of supersonic transport wing using unstructured adjoint method,” AIAA Journal, vol. 39, no. 6, pp. 1011–1020, June 2001.
[17] J. P. Thomas, K. C. Hall, and E. H. Dowell, “Discrete adjoint approach for modeling unsteady aerodynamic design sensitivities,” AIAA Journal, vol. 43, no. 9, pp. 1931–1936, September 2005.
[18] S. Nadarajah and A. Jameson, “Optimum shape design for unsteady three-dimensional viscous flows using a non-linear frequency domain method,” in 24th Applied Aerodynamics Conference, American Institute of Aeronautics and Astronautics. San Francisco, CA: AIAA, June 2006.
[19] S. K. Nadarajah and A. Jameson, “Optimum shape design for unsteady flows with time–accurate continuous and discrete adjoint methods,” AIAA Journal, vol. 45, no. 7, pp. 1478–1491, July 2007.
[20] K. Mani and D. Mavriplis, “Unsteady discrete adjoint formulation for two–dimensional flow problems with deforming meshes,” AIAA Journal, vol. 46, no. 6, pp. 1351–1364, June 2008.
[21] M. Giles, N. Pierce, and E. S¨uli, “Progress in adjoint error correction for integral functionals,” Comput Visual Sci, vol. 6, pp. 113–121, 2004.
[22] M. Giles and E. S¨uli, “Adjoint methods for pdes: a posteriori error analysis and postprocessing by duality,” Acta Numerica, vol. 11, pp. 145–236, 2002.
[23] D. A. Venditti and D. L. Darmofal, “A multilevel error estimation and grid adaptive strategy for improving the accuracy of integral outputs,” AIAA Paper 99–3292, 1999.
[24] D. A. Venditti and D. L. Darmofal, “Adjoint error estimation and grid adaptation for functional outputs: Application to quasi–one–dimensional flow,” Journal of Computational Physics, vol. 164, pp. 204–227, 2000.
[25] D. A. Venditti and D. L. Darmofal, “Anisotropic grid adaptation for functional outputs: application to two-dimensional viscous flows,” Journal of Computational Physics, vol. 187, pp. 22–46, 2003.
[26] M. B. Giles and N. A. Pierce, “Superconvergent lift estimates through adjoint error analysis,” 1998, http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.27.152.
[27] M. B. Giles and N. A. Pierce, “Adjoint recovery of superconvergent functionals from approximate solutions of partial differential equations,” Oxford University Computing Laboratory, Oxford, Report 98/18, August 1999.
[28] M. B. Giles and N. A. Pierce, “Improved lift and drag estimates using adjoint Euler equations,” AIAA Paper 99–3293, 1999.
[29] N. Pierce and M. Giles, “Adjoint and defect error bounding and correction for functional estimates,” Journal of Computational Physics, vol. 200, pp. 769–794, 2004.
[30] A. Jameson, L. Martinelli, and N. A. Pierce, “Optimum aerodynamic design using the Navier-Stokes equations,” Theoretical and Computational Fluid Dynamics, vol. 1, no. 10, pp. 213–237, 1998.
[31] A. Jameson, A. Sriram, and L. Martinelli, “A continuous adjoint method for unstructured grids,” in AIAA Computational Fluid Dynamics Conference, Orlando, FL, June 2003, aIAA 2003–3955.
[32] L. A. Lusternick and V. J. Sobolev, Elements of Functional Analysis, 1st ed., ser. Russian Monographs and Texts on Advanced Mathematics and Physics. Delhi: Hindustan Pub. Co., 1961, vol. V.
[33] P. M. Morse and H. Feshbach, Methods of Theoretical Physics, 1st ed. NY: MacGraw–Hill, 1953, vol. 1.
[34] M. Hayashi, M. Ceze, and E. Volpe, “Characteristics-based boundary conditions for the Euler adjoint problem,” Int. J. Numer. Meth. Fluids, vol. 71, no. 10, pp. 1297–1321, April 2013.
[35] A. Jameson and S. Kim, “Reduction of the adjoint gradient formula for aerodynamic shape optimization problems,” AIAA Journal, vol. 41, no. 11, pp. 2114–2129, November 2003.
[36] E. V. Volpe, “Continuous formulation of the adjoint problem for unsteady incompressible Navier–Stokes flows,” ANP–NDF–EPUSP, SP, Technical Report 5, June 2013.
[37] A. Jameson, W. Schimidt, and E. Turkel, “Numerical solution of the euler equations by finite volume methods using runge–kutta time–stepping schemes,” AIAA Journal, 1981.
[38] D. Barkley, H. M. Blackburn, and S. J. Sherwin, “Direct optimal growth for timesteppers,” International Journal for Numerical Methods in Fluids, vol. 1, no. 231, pp. 1–21, September 2002.