0
Research Papers: Gas Turbines: Turbomachinery

Adjoint Method for Shape Optimization in Real-Gas Flow Applications

[+] Author and Article Information
M. Pini

Assistant Professor
Propulsion and Power,
Aerospace Engineering Faculty,
Delft University of Technology,
Kluyverweg 1,
Delft 2629 HS, The Netherlands
e-mail: m.pini@tudelft.nl

G. Persico

Assistant Professor
Laboratorio di Fluidodinamica delle Macchine,
Dipartimento di Energia,
Politecnico di Milano,
via Lambruschini, 4,
Milano 20156, Italy
e-mail: giacomo.persico@polimi.it

D. Pasquale

Dipartimento di Ingegneria
Meccanica e Industriale,
Universitá degli studi di Brescia,
via Branze, 38,
Brescia 25123, Italy
e-mail: david.pasquale@ing.unibs.it

S. Rebay

Associate Professor
Dipartimento di Ingegneria
Meccanica e Industriale,
Universitá degli studi di Brescia,
via Branze, 38,
Brescia 25123, Italy
e-mail: stefano.rebay@ing.unibs.it

Contributed by the Turbomachinery Committee of ASME for publication in the JOURNAL OF ENGINEERING FOR GAS TURBINES AND POWER. Manuscript received May 24, 2014; final manuscript received June 26, 2014; published online October 14, 2014. Editor: David Wisler.

J. Eng. Gas Turbines Power 137(3), 032604 (Oct 14, 2014) (13 pages) Paper No: GTP-14-1245; doi: 10.1115/1.4028495 History: Received May 24, 2014; Revised June 26, 2014

An adjoint-based shape optimization approach for supersonic turbine cascades is proposed for application to organic Rankine cycle (ORC) turbines. The algorithm is based on an inviscid discrete adjoint method and encompasses a fast look-up table (LuT) approach to accurately deal with real-gas flows. The turbine geometry is defined by adopting state-of-the-art parameterization techniques (NURBS), enabling to handle both global and local control of the shape of interest. A preconditioned steepest descent method has been chosen as gradient-based optimization algorithm to efficiently search for the nearest minimum. The potential of the optimization approach is first verified by application on the redesign of an existing converging–diverging turbine nozzle operating in thermodynamic regions characterized by relevant real-gas effects. A significant efficiency improvement and a more uniform flow at the blade outlet section are achieved, with expected beneficial effects on the aerodynamics of the downstream rotor. The optimized configuration is also assessed by means of high-fidelity turbulent simulations, which point out the capability of the present inviscid approach in optimizing supersonic turbine cascades with very limited computational burdens. Finally, the newly developed real-gas adjoint method is compared against adjoints based on ideal equations of state on the same design problem. Results show that the performance gain obtained by a fully real-gas optimization strategy is by far higher than that achieved with simplified approaches in case of ORC turbines. This proves the relevance of including accurate thermodynamic models in all steps of ORC turbine design.

FIGURES IN THIS ARTICLE
<>
Copyright © 2015 by ASME
Your Session has timed out. Please sign back in to continue.

References

Pini, M., Persico, G., Casati, E., and Dossena, V., 2013, “Preliminary Design of a Centrifugal Turbine for ORC Applications,” ASME J. Eng. Gas Turbines Power, 135(4), p. 042312. [CrossRef]
Petrovic, M. V., Dulikravich, G. S., and Martin, T. J., 2001, “Optimization of Multistage Turbines Using a Through-Flow Code,” J. Power Energy, 215(5), pp. 559–569. [CrossRef]
Larocca, F., 2008, “Multiple Objective Optimization and Inverse Design of Axial Turbomachinery Blade,” J. Propul. Power, 24(5), pp. 1093–1099. [CrossRef]
Pasquale, D., Persico, G., and Rebay, S., 2014, “Optimization of Turbomachinery Flow Surfaces Applying a CFD-Based Troughflow Method,” ASME J. Turbomach., 136(3), p. 031013. [CrossRef]
Pierret, S., and Van Den Braembussche, R., 1999, “Turbomachinery Blade Design Using a Navier–Stokes Solver and Artificial Neural Network,” ASME J. Turbomach., 121(2), pp. 326–332. [CrossRef]
Rai, M., 2000, “Aerodynamic Design Using Neural Networks,” AIAA J., 38(1), pp. 173–182. [CrossRef]
Leonard, O., and Van Den Braembussche, R., 1992, “Design Method for Subsonic and Transonic Cascade With Prescribed Mach Number Distribution,” ASME J. Turbomach., 114(3), pp. 553–560. [CrossRef]
Demeulenaere, A., Leonard, O., and Van Den Braembussche, R., 1997, “A Two-Dimensional Navier–Stokes Inverse Solver for Compressor and Turbine Blade Design,” J. Power Energy, 211(4), pp. 299–307. [CrossRef]
Coello, C., 2000, “An Updated Survey of GA-Based Multiobjective Optimization Techniques,” ACM Comput. Surv., 32(2), pp. 109–143. [CrossRef]
Verstraete, T., Alsalihi, Z., and Van Den Braembussche, R., 2010, “Multidisciplinary Optimization of a Radial Compressor for Microgas Turbine Applications,” ASME J. Turbomach., 132(2), p. 031004. [CrossRef]
Pasquale, D., Ghidoni, A., and Rebay, S., 2013, “Shape Optimization of an Organic Rankine Cycle Radial Turbine Nozzle,” ASME J. Eng. Gas Turbines Power, 135(4), p. 042308. [CrossRef]
Lee, S. Y., and Kim, K. Y., 2000, “Design Optimization of Axial Flow Compressor Blades With Three-Dimensional Navier–Stokes Solver,” KSME Int. J., 14(9), pp. 1005–1012.
Oyama, A., 2004, “Transonic Axial-Flow Blade Optimization: Evolutionary Algorithms/Three-Dimensional Navier–Stokes Solver,” J. Propul. Power, 20(4), pp. 612–619. [CrossRef]
Chen, N., Zhang, H., Huang, W., and Xu, Y., 2005, “Study on Aerodynamic Design Optimization of Turbomachinery Blades,” J. Therm. Sci., 14(4), pp. 298–304. [CrossRef]
Pierret, S., Coelho, R., and Kato, H., 2006, “Multidisciplinary and Multiple Operating Points Shape Optimization of Three-Dimensional Compressor Blades,” Struct. Multidiscip. Optim., 33(1), pp. 61–70. [CrossRef]
Peter, J., and Dwight, R., 2010, “Numerical Sensitivity Analysis for Aerodynamic Optimization: A Survey of Approaches,” Comput. Fluids, 39(3), pp. 373–391. [CrossRef]
Quoilin, S., Broek, M. V. D., Declaye, S., Dewallef, P., and Lemort, V., 2013, “Techno-Economic Survey of Organic Rankine Cycle (ORC) Systems,” Renewable Sustainable Energy Rev., 22(0), pp. 168–186. [CrossRef]
Lang, W., Almbauer, R., and Colonna, P., 2013, “Assessment of Waste Heat Recovery for a Heavy-Duty Truck Engine Using an ORC Turbogenerator,” ASME J. Eng. Gas Turbines Power, 135(4), p. 042313. [CrossRef]
Pasquale, D., Ghidoni, A., and Rebay, S., 2013, “Shape Optimization of an Organic Rankine Cycle Radial Turbine Nozzle,” ASME J. Eng. Gas Turbines Power, 135(4), p. 042305. [CrossRef]
Cinnella, P., and Congedo, P., 2006, “GA-Hardness of Dense-Gas Flow Optimization Problems,” Applied Simulation and Modelling, Rhodes, Greece, June 26–28.
Giles, M., and Pierce, N., 2000, “An Introduction to the Adjoint Approach to Design,” Oxford Computing Laboratory, Oxford, UK.
Giles, M., Ghate, D., and Duta, M., 2005, “Using Automatic Differentiation for Adjoint CFD Code Development,” Oxford Computing Laboratory, Oxford, UK.
Carpentieri, G., Koren, B., and van Tooren, M., 2007, “Adjoint-Based Aerodynamic Shape Optimization on Unstructured Meshes,” J. Comput. Phys.224(1), pp. 267–287. [CrossRef]
Hoschek, J., Lasser, D., and Schumaker, L., 1996, Fundamentals of Computer Aided Geometric Design, A.K. Peters, Natick, MA.
Farin, G., 2002, Curves and Surfaces for CAGD: A Practical Guide, 5th ed., Academic Press, Waltham, MA.
de Boer, A., van der Schoot, M., and Bijl, H., 2007, “Mesh Deformation Based on Radial Basis Function Interpolation,” Comput. Struct., 85(11), pp. 784–795. [CrossRef]
Colonna, P., and Rebay, S., 2004, “Numerical Simulation of Dense Gas Flows on Unstructured Grids With an Implicit High Resolution Upwind Euler Solver,” Int. J. Numer. Methods Fluids, 46(7), pp. 735–765. [CrossRef]
Harinck, J., Colonna, P., Guardone, A., and Rebay, S., 2010, “Influence of Thermodynamic Models in 2D Flow Simulations of Turboexpanders,” ASME J. Turbomach., 132(1), p. 011001. [CrossRef]
Pecnik, R., Rinaldi, E., and Colonna, P., 2012, “Computational Fluid Dynamics of a Radial Compressor Operating With Supercritical CO2,” ASME J. Eng. Gas Turbines Power, 134(12), p. 122301. [CrossRef]
Boger, A., 2001, “Efficient Method for Calculating Wall Proximity,” AIAA J., 39(12), pp. 2404–2406. [CrossRef]
Pini, M., Spinelli, A., Persico, G., and Rebay, S., 2014, “Consistent Look-Up Table Interpolation Method for Real-Gas Flow Simulations,” Comput. Fluids (in press).
Saad, Y., and Schultz, M., 1986, “GMRES: A Generalized Minimal Residual Algorithm for Solving Nonsymmetric Linear Systems,” SIAM J. Sci. Stat. Comput., 7(3), pp. 856–869 [CrossRef].
Hascoet, L., and Pascual, V., 2004, Tapenade 2.1 User's Guide, Unité de recherche INRIA Sophia Antipolis, Cedex, France.
Rinaldi, E., Pecnik, R., and Colonna, P., 2014, “Implicit Schemes for Compressible Flows of Dense Gases,” J. Comput. Phys. (in press).
Ghidoni, A., Pelizzari, E., Rebay, S., and Selmin, V., 2006, “3D Anisotropic Unstructured Grid Generation,” Int. J. Numer. Methods Fluids, 51(9–10), pp. 1097–1115. [CrossRef]
Jameson, A., and Kim, S., 2003, “Reduction of the Adjoint Gradient Formula in the Continuous Limit,” AIAA Paper No. 2003-0040. [CrossRef]
Kim, S., Hosseini, K., Leoviriyakit, K., and Jameson, A., 2005, “Enhancement of the Adjoint Design Methods Via Optimization of Adjoint Parameters,” AIAA Paper No. 2005-0448. [CrossRef]
Colonna, P., Harinck, J., Rebay, S., and Guardone, A., 2008, “Real-Gas Effects in Organic Rankine Cycle Turbine Nozzles,” J. Propul. Power, 24(2), pp. 282–294. [CrossRef]
Rebay, S., Colonna, P., Pasquale, D., and Ghidoni, A., 2009, “Simulation of the Turbulent Dense Gas Flow Through The Nozzle of an Organic Rankine Cycle Turbine,” 8th European Conference on Turbomachinery, Fluid Dynamics and Thermodynamics, Graz, Austria, Mar. 23–27, pp. 1137–1148.

Figures

Grahic Jump Location
Fig. 1

Left: thermodynamic region covered by a LuT for the siloxane MDM in the T−s diagram. Blue dots located on the vapor saturation line represent the support points of the grid. Right: thermodynamic grid of 10,000 elements built within the tabulated region. Notice the high distortion of the mesh, as it is originally generated using different independent variables.

Grahic Jump Location
Fig. 2

Fully automated adjoint-based optimization algorithm. The aerodynamic gradient refers to the sensitivities with respect to all grid points, the active gradient is the vector of derivatives against surface grid points, whereas the surface gradient indicates the sensitivities with respect to NURBS control points. Red labels denote the names of the numerical solvers.

Grahic Jump Location
Fig. 3

Profile contour and control points distribution of the converging–diverging blade. Red circles indicate the design variables, while blue ones are kept fixed during optimization.

Grahic Jump Location
Fig. 4

Left: predicted Mach flow-field using the baseline and optimized cascade. Right: comparison of blade loading between the baseline and optimized cascade.

Grahic Jump Location
Fig. 5

Sketch of the nonsmoothed (a) and smoothed and normally projected (b) gradient vectors with respect to blade surface mesh nodes. The gradient components are properly scaled to be better visualized. Notice that flow uniformity, i.e., the cost functional, highly depends on the deflection imparted by the rear suction profile.

Grahic Jump Location
Fig. 6

Convergence history of the design process

Grahic Jump Location
Fig. 7

Left: predicted Mach number distribution of the baseline cascade. Right: predicted Mach number distribution of the optimized cascade.

Grahic Jump Location
Fig. 8

Left: isentropic Mach number distribution along the blade surface of the baseline and optimized configurations. Right: predicted spanwise Mach number distribution at outflow boundary.

Grahic Jump Location
Fig. 9

Left: predicted isentropic Mach number of the baseline blade surface for inviscid and turbulent flows. Right: predicted isentropic Mach number of the optimized blade surface for inviscid and turbulent flows.

Grahic Jump Location
Fig. 10

Left: predicted inviscid fishtail wave pattern of the baseline configuration. Right: predicted turbulent fishtail shock pattern of the baseline configuration.

Grahic Jump Location
Fig. 11

Left: predicted inviscid fishtail wave pattern of the optimized configuration. Right: predicted turbulent fishtail shock pattern of the optimized configuration.

Grahic Jump Location
Fig. 12

Mach flowfield obtained through the three optimized blades (PIG, PVdW, and LuT)

Grahic Jump Location
Fig. 13

Left: predicted isentropic Mach number distribution along the blade profile computed through the Span–Wagner EoS for the three optimized blades (PIG, PVdW, and LuT). Notice that curvilinear abscissa s/smax is used to plot the trends in place of the nondimensional streamwise coordinate. Right: area ratio (Ar = a/o) for the optimized PIG and LuT cascades. Notice that Aropt = ArLuT>ArPIG and correspondingly results that ArLuT>ArPVdW leading to the formation of an over-expanded freejet outside the bladed channel.

Grahic Jump Location
Fig. 14

Outlet Mach number pitchwise distribution obtained with the three optimized blades (PIG, PVdW, and LuT)

Tables

Errata

Discussions

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In