Abstract
A novel mesh-free heuristic method for solving differential equations is proposed. The new approach can cope with linear, nonlinear, and partial differential equations (DE), and systems of DEs. Candidate solutions are expressed using a linear combination of kernel functions. Thus, the original problem is transformed into an optimization problem that consists in finding the parameters that define each kernel. The new optimization problem is solved applying a Covariance Matrix Adaptation Evolution Strategy. To increase the accuracy of the results, a Downhill Simplex local search is applied to the best solution found by the mentioned evolutionary algorithm. Our method is applied to 32 differential equations extracted from the literature. All problems are successfully solved, achieving competitive accuracy levels when compared to other heuristic methods. A simple comparison with numerical methods is performed using two partial differential equations to show the pros and cons of the proposed algorithm. To verify the potential of this approach with a more practical problem, an electric circuit is analyzed in depth. The method can obtain the dynamic behavior of the circuit in a parametric way, taking into account different component values.
Similar content being viewed by others
References
Allahviranloo T, Salahshour S, Abbasbandy S (2012) Explicit solutions of fractional differential equations with uncertainty. Soft Comput 16:297–302
Babaei M (2013) A general approach to approximate solutions of nonlinear differential equations using particle swarm optimization. Appl Soft Comput 13(7):3354–3365
Balasubramaniam P, Kumar AVA (2009) Solution of matrix Riccati differential equation for nonlinear singular system using genetic programming. Genet Program Evolvable Mach 10:71–89
Bronson JR, Sastry SP, Levine JA, Whitaker RT (2014) Adaptive and unstructured mesh cleaving. In: 23rd International meshing roundtable (IMR23), vol 82, pp 266–278
Bryden KM, Ashlock DA, Corns S, Willson SJ (2006) Graph-based evolutionary algorithms. IEEE Trans Evol Comput 10(5):550–567
Chaquet JM, Carmona EJ (2012) Solving differential equations with Fourier series and evolution strategies. Appl Soft Comput 12:3051–3062
Chen H, Kong L, Leng WJ (2011) Numerical solution of PDEs via integrated radial basis function networks with adaptive training algorithm. Appl Soft Comput 11:855–860
Choi B, Lee JH (2009) Comparison of generalization ability on solving differential equations using backpropagation and reformulated radial basis function networks. Neurocomputing 73:115–118
Colutto S, Fruhauf F, Fuchs M, Scherzer O (2010) The CMA-ES on Riemannian manifolds to reconstruct shapes in 3-D voxel images. IEEE Trans Evol Comput 14(2):227–245
El-Emam NN, Al-Rabeh RH (2011) An intelligent computing technique for fluid flow problems using hybrid adaptive neural network and genetic algorithm. Appl Soft Comput 11:3283–3296
Guo R, Guo D (2009) Random fuzzy variable foundation for grey differential equation modeling. Soft Comput 13:185–201
Hangelbroek T, Ron A (2010) Nonlinear approximation using Gaussian kernels. J Funct Anal 259(1):203–219
Hansen N (2006) The CMA evolution strategy: a comparing review. In: Lozano JA, Larrañaga P, Inza I, Bengoetxea E (eds) Towards a new evolutionary computation, vol 192. Studies in fuzziness and soft computing. Springer, Berlin, pp 75–102
Hansen N (June 2011) The CMA evolution strategy: A tutorial. https://www.lri.fr/~hansen/cmatutorial.pdf
Hansen N, Kern S (2004) Evaluating the CMA evolution strategy on multimodal test functions. In: Eighth international conference on parallel problem solving from nature PPSN VIII, pp 282–291. Springer
He J, Xu J, Yao X (2000a) Solving equations by hybrid evolutionary computation techniques. IEEE Trans Evol Comput 4(3):295–304
He S, Reif K, Unbehauen R (2000b) Multilayer neural networks for solving a class of partial differential equations. Neural Netw 13:385–396
Howard D, Roberts SC (2001) Genetic programming solution of the Convection-diffusion equation. In: Proceedings of the 3rd annual conference on genetic and evolutionary computation, GECCO’01, pp 34–41. Morgan Kaufmann Publishers Inc
Howard D, Brezulianu A, Kolibal J (2011) Genetic programming of the stochastic interpolation framework: convection-diffusion equation. Soft Comput 15:71–78
Khan JA, Zahoor RMA, Qureshi IM (2009) Swarm intelligence for the solution of problems in differential equations. In: Second international conference on environmental and computer science, 2009. ICECS ’09, pp 141–147
Kirstukas SJ, Bryden KM, Ashlock DA (2005) A hybrid genetic programming approach for the analytical solution of differential equations. Int J Gen Syst 34:279–299
Kumar M, Yadav N (2011) Multilayer perceptrons and radial basis function neural network methods for the solution of differential equations. A survey. Comput Math Appl 62:3796–3811
Lagaris IE, Likas A, Fotiadis DI (1998) Artificial neural networks for solving ordinary and partial differential equations. IEEE Trans Neural Netw 5:987–1000
Leveque RJ (2002) Finite volume methods for hyperbolic problems. Cambridge University Press, Cambridge
Liu GR (2010) Meshfree methods: moving beyond the finite element method. CRC Press, Inc., Boca Raton
MacNeil PE (2012) A technique for generating approximate solutions and its application to coulomb interactions. In: Southeastcon, 2012 Proceedings of IEEE, pp 1–5
Mehrkanoon S, Suykens JAK (2015) Learning solutions to partial differential equations using LS-SVM. Neurocomputing 159:105–116
Mosleh M (2013) Fuzzy neural network for solving a system of fuzzy differential equations. Appl Soft Comput 13(8):3597–3607
Nelder JA, Mead R (1965) A simplex method for function minimization. Comput J 7(4):308–313
Ozisik MN (1994) Finite difference methods in heat transfer. CRC Press, Inc., Boca Raton
Panagant N, Bureerat S (2014) Solving partial differential equations using a new differential evolution algorithm. Math Probl Eng 2014(747490):1–10
Parisi DR, Mariani MC, Laborde MA (2003) Solving differential equations with unsupervised neural networks. Chem Eng Process 42:715–721
Peterson LE (2011) Covariance matrix self-adaptation evolution strategies and other metaheuristic techniques for neural adaptive learning. Soft Comput 15:598
Press WH, Vetterling WT, Teukolsky SA, Flannery BP (2002) Numerical Recipes in C++: the art of scientific computing, 2nd edn. Cambridge University Press, New York
Puffer F, Tetzlaff R, Wolf D (1995) A learning algorithm for cellular neural networks (CNN) solving nonlinear partial differential equations. In: International symposium on signals, systems, and electronics (ISSSE ’95), pp 501–504
Rudd K, Ferrari S (2015) A constrained integration (cint) approach to solving partial differential equations using artificial neural networks. Neurocomputing 155:277–285
Seaton T, Brown G, Miller JF (2010) Analytic solutions to differential equations under graph-based genetic programming. In: Esparcia-Alcázar Anna Isabel, Ekárt Anikó, Sara Silva, Dignum Stephen, Şima Uyar A (eds) Genetic Programming: 13th European Conference, EuroGP 2010, pp 232–243. Springer Berlin Heidelberg
Shirvany Y, Hayati M, Moradian R (2009) Multilayer perceptron neural networks with novel unsupervised training method for numerical solution of the partial differential equations. Appl Soft Comput 9:20–29
Sobester A, Nair PB, Keane AJ (2008) Genetic programming approaches for solving elliptic partial differential equations. IEEE Trans Evol Comput 12:469–478
Spivak M (1980) Calculus. Publish or Perish, 4 edition
Suli E, Mayers DF (2003) An introduction to numerical analysis. Cambridge University Press, Cambridge
Sun M, Yan X, Sclabassi RJ (2003) Solving partial differential equations in real-time using artificial neural network signal processing as an alternative to finite-element analysis. In: International conference on neural networks and signal processing, 2003, vol 1, pp 381–384
Thompson JF, Soni BK, Weatherill NP (1999) Handbook of grid generation. CRC Press, Boca Raton
Tsoulos IG, Lagaris IE (2006) Solving differential equations with genetic programming. Genet Program Evolvable Mach 7:33–54
Tsoulos IG, Gavrilis D, Glavas E (2009) Solving differential equations with constructed neural networks. Neurocomputing 72:2385–2391
Yao K (2015) Uncertain differential equation with jumps. Soft Comput 19:2063–2069
Yazdi HS, Pourreza R (2010) Unsupervised adaptive neural-fuzzy inference system for solving differential equations. Appl Soft Comput 10:267–275
Yazdi HS, Pakdaman M, Modaghegh H (2011) Unsupervised kernel least mean square algorithm for solving ordinary differential equations. Neurocomputing 74:2062–2071
Acknowledgements
The authors wish to thank Nikolaus Hansen for sharing with the research community CMA-ES implementations in several programming languages. In particular, the ANSI C implementation (https://www.lri.fr/~hansen/cmaes_inmatlab.html) has been used in the present work.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that they have no conflict of interest.
Additional information
Communicated by V. Loia.
Rights and permissions
About this article
Cite this article
Chaquet, J.M., Carmona, E.J. Using Covariance Matrix Adaptation Evolution Strategies for solving different types of differential equations. Soft Comput 23, 1643–1666 (2019). https://doi.org/10.1007/s00500-017-2888-9
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00500-017-2888-9