Improved Multi-Objective Particle Swarm Optimization Applied to Design Problem

Authors: Kapse Swapnil, K. Shankar

Abstract:

Aiming at optimizing the weight and deflection of cantilever beam subjected to maximum stress and maximum deflection, Multi-objective Particle Swarm Optimization (MOPSO) with Utopia Point based local search is implemented. Utopia point is used to govern the search towards the Pareto Optimal set. The elite candidates obtained during the iterations are stored in an archive according to non-dominated sorting and also the archive is truncated based on least crowding distance. Local search is also performed on elite candidates and the most diverse particle is selected as the global best. This method is implemented on standard test functions and it is observed that the improved algorithm gives better convergence and diversity as compared to NSGA-II in fewer iterations. Implementation on practical structural problem shows that in 5 to 6 iterations, the improved algorithm converges with better diversity as evident by the improvement of cantilever beam on an average of 0.78% and 9.28% in the weight and deflection respectively compared to NSGA-II.

Keywords: Utopia point, multi-objective particle swarm optimization, local search, cantilever beam.

Digital Object Identifier (DOI): doi.org/10.5281/zenodo.1128823

References:

[1] J. Kennedy & R. C. Eberhart, Swarm Intelligence. San Mateo, CA: Morgan Kaufmann, 2001.
[2] Y. Shi and R. Eberhart, Parameter selection in particle swarm optimization. In Evolutionary Programming VIZ: Proc. EP98, New York: Springer Verlag, 1998, pp. 591-600.
[3] E. Elbeltagi, T. Hegazy and D. Grierson. Comparison among five evolutionary-based optimization algorithms, Advanced Engineering Informatics Volume 19 Issue 1, Pages 43-53
[4] K. Deb. Multi-Objective Optimization Using Evolutionary Algorithms. John Wiley and Sons, Chichester, UK. ISBN: 047187339X, 2001
[5] J. Moore & R. Chapman. Application of Particle Swarm to Multi- objective Optimization. Department of Computer Science and Software Engineering, Auburn University, 1999.
[6] Q. Lin, J. Li, Z. Du, J. Chen, Z. Ming. A novel multi-objective particle swarm optimization with multiple search strategies. European Journal of Operational Research 247, 2015, 732–744
[7] K. Harada, K. Ikeda & S. Kobayashi. Hybridization of Genetic Algorithm and Local Search in Multi- objective Function Optimization: Recommendation of GA then LS GECCO’06, July 8–12, 2006, Seattle, Washington, USA
[8] K. Deb, A. Pratap, S. Agarwal, T. Meyarivan. A fast and elitist multi- objective genetic algorithm: NSGA-II. IEEE Transactions on Evolutionary Computation 6(2):182–197. 2002, Doi: 10.1109/4235.996017
[9] M. L. Zheng, Q. Wang, S. N. Zhang & S. Y. Zheng. Population recombination strategies for multi-objective particle swarm optimization. Soft Computing, 2016. DOI 10.1007/s00500-016-2078
[10] E. Zitzler, K. Deb, L. Thiele. Comparison of multi- objective evolutionary algorithms: empirical results, Evolutionary Computation. 2000, 8 (2) 173–195.
[11] J.A. Adeyemo and F.A. O. Otieno. Multi-Objective Differential Evolution Algorithm for Solving Engineering Problems. Journal of Applied Sciences, 2009, 9: 3652-3661.
[12] M. R. Sierra & C. A. Coello Coello. Improving PSO-based multi-objective optimization using crowding, mutation and epsilon-dominance. Evolutionary Multi-criterion Optimization, Lecture Notes in Computer Science,2005, 3410, 505–519
[13] K. P. Tripathi., S. Bandyopadhyay, S. Pal. Multi-Objective Particle Swarm Optimization with time variant inertia and acceleration coefficients. Information Sciences 177 (2007) 5033–5049
[14] Kiranyaz, Serkan, Ince, Turker, Gabbouj & Moncef, Multidimensional Particle Swarm Optimization for Machine Learning and Pattern Recognition‖, Springer-Verlag Berlin Heidelberg, 2014, pp. 48-49