Abstract
In this paper, we present a new model-based trust-region derivative-free optimization algorithm which can handle nonlinear equality constraints by applying a sequential quadratic programming (SQP) approach. The SQP methodology is one of the best known and most efficient frameworks to solve equality-constrained optimization problems in gradient-based optimization [see e.g. Lalee et al. (SIAM J Optim 8:682–706, 1998), Schittkowski (Optim Lett 5:283–296, 2011), Schittkowski and Yuan (Wiley encyclopedia of operations research and management science, Wiley, New York, 2010)]. Our derivative-free optimization (DFO) algorithm constructs local polynomial interpolation-based models of the objective and constraint functions and computes steps by solving QP sub-problems inside a region using the standard trust-region methodology. As it is crucial for such model-based methods to maintain a good geometry of the set of interpolation points, our algorithm exploits a self-correcting property of the interpolation set geometry. To deal with the trust-region constraint which is intrinsic to the approach of self-correcting geometry, the method of Byrd and Omojokun is applied. Moreover, we will show how the implementation of such a method can be enhanced to outperform well-known DFO packages on smooth equality-constrained optimization problems. Numerical experiments are carried out on a set of test problems from the CUTEst library and on a simulation-based engineering design problem.
Similar content being viewed by others
References
Adams, B.M., Bauman, L.E., Bohnhoff, W.J., Dalbey, K.R., Ebeida, M.S., Eddy, J.P., Eldred, M.S., Hough, P.D., Hu, K.T., Jakeman, J.D., Swiler, L.P., Vigil, D.M.: Dakota, a multilevel parallel object-oriented framework for design optimization, parameter estimation, uncertainty quantification, and sensitivity analysis: Version 5.4 user’s manual. Sandia Technical Report SAND2010-2183 (2009, updated 2013) (2013)
Ashley, H., Landahl, M.: Aerodynamics of Wings and Bodies. Dover Books on Aeronautical Engineering Series. Addison-Wesley Publishing Company, New York (1965)
Audet, C., Dennis, J.J.E.: Mesh adaptive direct search algorithms for constrained optimization. SIAM J. Optim. 17, 188–217 (2006)
Audet, C., Dennis, J.J.E.: A progressive barrier for derivative-free nonlinear programming. SIAM J. Optim. 20, 445–472 (2009)
Bueno, L., Friedlander, A., Martinez, J., Sobral, F.: Inexact restoration method for derivative-free optimization with smooth constraints. SIAM J. Optim. 23(2), 1189–1213 (2013)
Byrd, R.H.: Robust trust region methods for constrained optimization. In: Third SIAM Conference on Optimization, Houston, Texas (1987)
Colson, B.: Trust-region algorithms for derivative-free optimization and nonlinear bilevel programming. Ph.D Thesis, The University of Namur, Rempart de la Vierge 8, 5000 Namur, Belgium (2003)
Conn, A.R., Gould, N.I.M., Toint, P.L.: Trust-Region Methods. MPS-SIAM Series on Optimization. Society for Industrial and Applied Mathematics, Philadelphia (2000)
Conn, A.R., Scheinberg, K., Toint, P.L.: On the convergence of derivative-free methods for unconstrained optimization. In: Iserles, A., Buhmann, M. (eds.) Approximation Theory and Optimization: Tributes to M. J. D. Powell, pp. 83–108. Cambridge, England (1997)
Conn, A.R., Scheinberg, K., Toint, P.L.: A derivative free optimization algorithm in practice. In: Proceedings of the 7th AIAA/USAF/NASA/ISSMO Symposium on Multidisciplinary Analysis and Optimization, September 2–4 (1998)
Conn, A.R., Scheinberg, K., Vicente, L.N.: Geometry of interpolation sets in derivative free optimization. Mathe. Program. Ser. A, B 111, 141–172 (2008)
Conn, A.R., Scheinberg, K., Vicente, L.N.: Introduction to Derivative-Free Optimization. MPS-SIAM Series on Optimization. SIAM, Philadelphia (2008)
Conn, A.R., Scheinberg, K., Vicente, L.N.: Global convergence of general derivative-free trust-region algorithms to first- and second-order critical points. SIAM J. Optim. 20(1), 387–415 (2009)
Currie, J., Wilson, D.: OPTI: Lowering the barrier between open source optimizers and the industrial MATLAB user. In: Sahinidis, N., Pinto, J. (eds.) Foundations of Computer-Aided Process Operations. Savannah (2012)
Diniz-Ehrhardt, M.A., Martinez, J.M., Pedroso, L.G.: Derivative-free methods for nonlinear programming with general lower-level constraints. Comput. Appl. Math. 30, 19–52 (2011)
Dolan, E.D., Moré, J.J.: Benchmarking optimization software with performance profiles. Math. Program. 91, 201–203 (2002)
Driessen, L.: Simulation-based optimization for product and process design. Ph.D Thesis, Tilburg University, Tilburg, Netherlands (2006)
Fasano, G., Morales, J.L., Nocedal, J.: On the geometry phase in model-based algorithms for derivative-free optimization. Optim. Methods Softw. 24(1), 145–154 (2009)
Gould, N.I.M., Orban, D., Toint, P.L.: CUTEst: a constrained and unconstrained testing environment with safe threads. Comput. Optimiz. Appl. (to appear)
Lalee, M., Nocedal, J., Plantenga, T.: On the implementation of an algorithm for large-scale equality constrained optimization. SIAM J. Optim. 8, 682–706 (1998)
Liuzzi, G., Lucidi, S., Sciandrone, M.: Sequential penalty derivative-free methods for nonlinear constrained optimization. SIAM J. Optim. 20(5), 2614–2635 (2010)
Neeb, D., Schwanekamp, T., Gülhan, A.: Preliminary aerodynamic shape optimization of the spaceliner by means of engineering methods. In: 17th AIAA International Space Planes and Hypersonic Systems and Technologies Conference, American Institute of Aeronautics and Astronautics (2011)
Omojokun, E.O.: Trust region algorithms for optimization with nonlinear equality and inequality constraints. Ph.D. Thesis, University of Colorado, Boulder (1989)
Perez, R., Jansen, P., Martins, J.: pyopt: A python-based object-oriented framework for nonlinear constrained optimization. Struct. Multidiscip. Optim. 45(1), 101–118 (2012)
Powell, M.: A direct search optimization method that models the objective and constraint functions by linear interpolation. In: Gomez, S., Hennart, J.P. (eds.) Advances in Optimization and Numerical Analysis, Mathematics and Its Applications, vol. 275, pp. 51–67. Springer, Netherlands (1994)
Powell, M.: Direct search algorithms for optimization calculations. Acta Numerica 7, 287–336 (1998)
Powell, M.J.D.: Developments of NEWUOA for minimization without derivatives. IMA J. Numer. Anal. 28(4), 649–664 (2008)
Rios, L.M., Sahinidis, N.V.: Derivative-free optimization: a review of algorithms and comparison of software implementations. J. Glob. Optim. 56(3), 1247–1293 (2013)
Scheinberg, K., Toint, P.L.: Self-correcting geometry in model-based algorithms for derivative-free unconstrained optimization. SIAM J. Optim. 20(6), 3512–3532 (2010)
Schittkowski, K.: A robust implementation of a sequential quadratic programming algorithm with successive error restoration. Optim. Lett. 5(2), 283–296 (2011). doi:10.1007/s11590-010-0207-9
Schittkowski, K., Yuan, Y.X.: Sequential quadratic programming methods. In: Cochran, J.J., Cox, L.A., Keskinocak, P., Kharoufeh, J.P., Smith, J.C. (eds.) Wiley Encyclopedia of Operations Research and Management Science. Wiley, New York (2010)
Streit, T., Martin, S., Eggers, T.: Approximate heat transfer methods for hypersonic flow in comparison with results provided by numerical Navier–Stokes solution. DLR-Forschungsbericht, 94–36 (1994)
Wächter, A., Biegler, L.T.: On the implementation of an interior-point filter line-search algorithm for large-scale nonlinear programming. Math. Program. 106(1), 25–57 (2006)
Author information
Authors and Affiliations
Corresponding author
Appendix A: Test problem statistics and detailed results
Appendix A: Test problem statistics and detailed results
Table 1 depicts the equality-constrained test problems taken from the CUTEst testing environment. It shows the name of the problem and gives specific details on the number of variables, the number of constraints and the type of the objective and constraint functions. As used in the CUTEst library, ’N’ stands for ’no objective function’ where a feasibility problem has to be solved, ’L’ stands for ’linear function’, ’Q’ means ’quadratic function’, ’S’ stands for ’sum of squares’ and ’O’ stand for any ’other’ type of function which could for instance involve exponential terms. If problems involve constraints of different type, the most non-linear type is given in the table. Furthermore, the table shows the number of function evaluations needed by each solver to attain four significant figures of the reference objective function value \(f^*\) from IPOPT, and to satisfy \(\Vert h\Vert _\infty \le 10^{-4}\).
Rights and permissions
About this article
Cite this article
Tröltzsch, A. A sequential quadratic programming algorithm for equality-constrained optimization without derivatives. Optim Lett 10, 383–399 (2016). https://doi.org/10.1007/s11590-014-0830-y
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11590-014-0830-y