A penalty function-based differential evolution algorithm for constrained global optimization | Computational Optimization and Applications Skip to main content
Springer Nature Link
Log in

A penalty function-based differential evolution algorithm for constrained global optimization

  • Published:
Computational Optimization and Applications Aims and scope Submit manuscript

Abstract

We propose a differential evolution-based algorithm for constrained global optimization. Although differential evolution has been used as the underlying global solver, central to our approach is the penalty function that we introduce. The adaptive nature of the penalty function makes the results of the algorithm mostly insensitive to low values of the penalty parameter. We have also demonstrated both empirically and theoretically that the high value of the penalty parameter is detrimental to convergence, specially for functions with multiple local minimizers. Hence, the penalty function can dispense with the penalty parameter. We have extensively tested our penalty function-based DE algorithm on a set of 24 benchmark test problems. Results obtained are compared with those of some recent algorithms.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Subscribe and save

Springer+ Basic
¥17,985 /Month
  • Get 10 units per month
  • Download Article/Chapter or eBook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Price includes VAT (Japan)

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2

Similar content being viewed by others

Notes

  1. Further discussions on the initial choice of U ∗ is presented in the discussion section later in the paper.

  2. If x is the first feasible point generated by CDE then U ∗=f(x). Subsequently, the feasibility of every point x generated by CDE is checked. If the point x is feasible and f(x)<U ∗ holds then U ∗ is updated as U ∗=f(x). The penalty function value L(x) is then calculated.

  3. This is done by calculating ψ(x i,1) which is then used to evaluate L(x i,1) at x i,1 via (8).

  4. The selection rule of CDE is similar to (6). The only difference is that CDE uses the penalty function L(x) in (6) instead of f(x).

  5. We have made this choice based on our observation on results using numerical testing.

  6. Notice that since i=1,2,…,N, we can consider a upper bound \(U_{*}^{i,k}\) corresponding to y i,k . Hence, prior to the creation of T k , the best known upper bound \(U_{*}^{N,k-1} ({=}U_{*})\) is the overall upper bound generated at the end of T k−1. However, all upper bounds, \(U_{*}^{i,k}\), at the k-th iteration are not necessarily distinct. Consider any two consecutive updates at the k-th iteration, say \(U_{*}^{i,k}\) (the update occurred corresponding to y i,k ) and \(U_{*}^{j,k}\) (the update occurred corresponding to y j,k , j>i, j=i+p) then \(U_{*}^{i,k}=U_{*}^{i+1,k}=\cdots=U_{*}^{i+p-1} ({>}U_{*}^{j,k})\). The case when all upper bounds are distinct is explained in the next subsection.

  7. In the example below (17) the two updates \(U_{*}^{1,k}\) and \(U_{*}^{\frac{N}{2},k}\) can be treated as \(\{U_{*}^{1,k},U_{*}^{\frac{N}{2},k}\}=\{U_{*}^{1,k},U_{*}^{2,k}\}\).

  8. The upper bound \(U_{*}^{i,k}\) in the relation corresponds to y i,k ∈T i,k .

  9. The feasible x i,k may also be obtained during the generation of the initial set, i.e., x i,k =x i,1∈S 1. Hence, we can use \(\bar{k}\ge0\) by assuming the members of S 1 as the members of the 0-th trial population where all members are accepted.

  10. Results for CDE are based the decimal points accuracies of f ∗ presented in Table 1.

  11. A slightly high mfe incurs for problems G6, G11 and G24 when we use the population size N=50.

  12. G17 is a non-smooth function, and it is not clear as to how SQP can be implemented.

References

  1. Miettinen, K., Makela, M., Toivanen, J.: Numerical comparison of some penalty based constraint handling techniques in genetic algorithm. J. Glob. Optim. 27, 427–446 (2003)

    Article  MathSciNet  MATH  Google Scholar 

  2. Runarsson, T.P., Yao, X.: Stochastic ranking for constrained evolutionary optimization. IEEE Trans. Evol. Comput. 4(3), 284–294 (2000)

    Article  Google Scholar 

  3. Datta, R., Deb, K.: A bi-objective based hybrid evolutionary-classical algorithm for handling equality constraints. In: Evolutionary Multicriterion Optimization Conference, EMO’2011. LNCS, vol. 6576, pp. 313–327. Springer, Berlin (2011)

    Chapter  Google Scholar 

  4. Storn, R., Price, K.: Differential evolution—a simple and efficient heuristic for global optimization over continuous spaces. J. Glob. Optim. 11, 341–359 (1997)

    Article  MathSciNet  MATH  Google Scholar 

  5. Mallipeddi, R., Suganthan, P.N.: Ensemble of constraint handling techniques. IEEE Trans. Evol. Comput. 14(4), 561–579 (2010)

    Article  Google Scholar 

  6. Mezura-Montes, E., Miranda-Varela, M.E., Gómez-Ramón, R.C.: Differential evolution in constrained numerical optimization: an empirical study. Inf. Sci. 180(22), 4223–4262 (2010)

    Article  MATH  Google Scholar 

  7. Landa Becerra, R., Coello Coello, C.A.: Cultured differential evolution for constrained optimization. Comput. Methods Appl. Mech. Eng. 195(33–36), 4303–4322 (2006)

    Article  MathSciNet  MATH  Google Scholar 

  8. Takahama, T., Sakai, S.: Constrained optimization by the ε constrained differential evolution with gradient-based mutation and feasible elites. In: Proceedings of the IEEE Congress on Evolutionary Computation, Vancouver, BC, Canada, pp. 303–315 (2006)

    Google Scholar 

  9. Mezura-Montes, E., Velázquez-Reyes, J., Coello Coello, C.A.: Modified differential evolution for constrained optimization. In: Proceedings of the IEEE Congress on Evolutionary Computation, Vancouver, BC, Canada, pp. 324–331 (2006)

    Google Scholar 

  10. Huang, V.L., Qin, A.K., Suganthan, P.N.: Self-adaptive differential evolution algorithm for constrained real-parameter optimization. In: Proceedings of the IEEE Congress on Evolutionary Computation, Vancouver, BC, Canada, pp. 17–24 (2006)

    Google Scholar 

  11. Huang, F., Wang, L., He, Q.: An effective co-evolutionary differential evolution for constrained optimization. Appl. Math. Comput. 186(1), 340–356 (2007)

    Article  MathSciNet  MATH  Google Scholar 

  12. Deb, K.: An efficient constraint handling method for genetic algorithms. Comput. Methods Appl. Mech. Eng. 186, 311–338 (2000)

    Article  MATH  Google Scholar 

  13. Runarsson, T.P., Yao, X.: Search biases in constrained evolutionary optimization. IEEE Trans. Syst. Man Cybern., Part C 35(2), 223–243 (2005)

    Article  Google Scholar 

  14. Kaelo, P., Ali, M.M.: A numerical comparison of some modified differential evolution algorithms. Eur. J. Oper. Res. 169(3), 1176–1184 (2006)

    Article  MathSciNet  MATH  Google Scholar 

  15. Homaifar, A., Lai, S.H.Y., Qi, X.: Constrained optimization via genetic algorithm. Simulation 62(4), 242–254 (1994)

    Article  Google Scholar 

  16. Joines, J., Houck, C.: On the use of non-stationary penalty functions to solve non-linearly constrained problems with GAs. In: Fogel, D. (ed.) Proceedings of the First IEEE Conference on Evolutionary Computation, pp. 579–584. IEEE Press, New York (1994)

    Google Scholar 

  17. Michalewicz, Z., Attia, N.: Evolutionary optimization of constrained problems. In: Proceedings of the Third Annual Conference on Evolutionary Programming, pp. 98–108 (1994)

    Google Scholar 

  18. Mezura-Montes, E., Coello Coello, C.A.: Constraint-handling in nature-inspired numerical optimization: past, present and future. Swarm Evol. Comput. 1(4), 173–194 (2011)

    Article  Google Scholar 

  19. Powell, D., Skolnick, M.M.: Using genetic algorithms in engineering design optimization with non-linear constraints. In: Proceedings of the Fifth International Conference on Genetic Algorithms, pp. 424–430 (1993)

    Google Scholar 

  20. Farmani, R., Wright, J.A.: Self-adaptive fitness formulation for constrained optimization. IEEE Trans. Evol. Comput. 7(5), 445–455 (2003)

    Article  Google Scholar 

  21. Tessema, B., Yen, G.G.: An adaptive penalty formulation for constrained evolutionary optimization. IEEE Trans. Syst. Man Cybern., Part A 39(3), 565–578 (2009)

    Article  Google Scholar 

  22. Koziel, S., Michalewicz, Z.: Evolutionary algorithms, homomorphous mappings, and constrained parameter optimization. Evol. Comput. 7(1), 19–44 (1999)

    Article  Google Scholar 

  23. Hamida, S.B., Schoenauer ASCHEA, M.: New results using adaptive segregational constraint handling. In: Proceedings of the Congress on Evolutionary Computation (CEC2002), pp. 884–889. IEEE Service Center, Piscataway (2002)

    Google Scholar 

  24. Montes, E.M., Coello Coello, A.C.: A simple multi-membered evolution strategy to solve constrained optimization problems. IEEE Trans. Evol. Comput. 9(1), 1–17 (2005)

    Article  Google Scholar 

  25. Hedar, A.R., Fukushima, M.: Derivative-free filter simulated annealing method for constrained continuous global optimization. J. Glob. Optim. 35, 521–549 (2006)

    Article  MathSciNet  MATH  Google Scholar 

  26. Liang, J.J., Runarsson, T.P., Mezura-Montes, E., Clerc, M., Suganthan, P.N., Coello Coello, C.A., Deb, C.: In: Problem Definition and Evolution Criteria for the CEC 2006 Special Session on Constrained Real-Parameter Optimization, IEEE Congress on Evolutionary Computation, Vancouver, Canada, 17–21 July (2006)

    Google Scholar 

  27. Derrac, J., García, S., Molina, D., Herrera, F.: A practical tutorial on the use of nonparametric statistical tests as a methodology for comparing evolutionary and swarm intelligence algorithms. Swarm Evol. Comput. 1, 3–18 (2011)

    Article  Google Scholar 

  28. Wang, Y., Cai, Z., Zhou, Y.: Accelerating adaptive trade-off model using shrinking space technique for constrained evolutionary optimization. Int. J. Numer. Methods Eng. 77, 1501–1534 (2009)

    Article  MathSciNet  MATH  Google Scholar 

  29. Michalewicz, Z., Schoenauer, M.: Evolutionary algorithms for constrained parameter optimization problems. Evol. Comput. 4(1), 1–32 (1996)

    Article  Google Scholar 

  30. Floudas, C.A., Pardalos, P.M.: A Collection of Test Problems for Constrained Global Optimization Algorithms. Springer, Berlin (1987)

    Google Scholar 

  31. Michalewicz, Z., Nazhiyath, G., Michalewicz, M.: A note on usefulness of geometrical crossover for numerical optimization problems. In: Fogel, L.J., Angeline, P.J., Bäck, T. (eds.) Proceedings of the Fifth Annual Conference on Evolutionary Programming, pp. 305–312. MIT Press, Cambridge (1997)

    Google Scholar 

  32. Himmelblau, D.: Applied Nonlinear Programming. McGraw-Hill, New York (1972)

    MATH  Google Scholar 

  33. Hock, W., Schittkowski, K.: Test Examples for Nonlinear Programming Codes. Lecture Notes in Economics and Mathematical Systems. Springer, Berlin (1981)

    Book  MATH  Google Scholar 

  34. Kim, J.H., Myung, H.: Evolutionary programming techniques for constrained optimization problems. IEEE Trans. Evol. Comput. 1(2), 129–140 (1997)

    Article  Google Scholar 

  35. Michalewicz, Z., Logan, T.D., Swaminathan, S.: Evolutionary operators for continuous convex parameter spaces. In: Sebald, A.V., Fogel, L.J. (eds.) Proceedings of the Fourth Annual Conference on Evolutionary Computation, pp. 84–97. World Scientific, Singapore (1994)

    Google Scholar 

  36. Epperly, T.: Global optimization test problems with solutions. http://citeseer.nj.nec.com/147308.html

  37. Xia, Q.: Global optimization test problems. http://www.mat.univie.ac.at/neum/glopt/xia.txt

Download references

Author information

Authors and Affiliations

  1. School of Computational and Applied Mathematics, Witwatersrand University, Wits 2050, Johannesburg, South Africa

    M. M. Ali

  2. Center for Discrete Mathematics and Theoretical Computer Science, Fuzhou University, Fuzhou, 350002, China

    W. X. Zhu

Authors
  1. M. M. Ali
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. W. X. Zhu
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to M. M. Ali.

Additional information

M.M. Ali was supported by the National Research Foundation of South Africa under Grant FA2006042300001.

W.X. Zhu was supported by the National Natural Science Foundation of China under Grant 61170308.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Ali, M.M., Zhu, W.X. A penalty function-based differential evolution algorithm for constrained global optimization. Comput Optim Appl 54, 707–739 (2013). https://doi.org/10.1007/s10589-012-9498-3

Download citation

  • Received:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10589-012-9498-3

Keywords

Search

Navigation