Constructive cooperative coevolution for large-scale global optimisation

Abstract

This paper presents the Constructive Cooperative Coevolutionary (C 3 ) algorithm, applied to continuous large-scale global optimisation problems. The novelty of C 3 is that it utilises a multi-start architecture and incorporates the Cooperative Coevolutionary algorithm. The considered optimisation problem is decomposed into subproblems. An embedded optimisation algorithm optimises the subproblems separately while exchanging information to co-adapt the solutions for the subproblems. Further, C 3 includes a novel constructive heuristic that generates different feasible solutions for the entire problem and thereby expedites the search. In this work, two different versions of C 3 are evaluated on high-dimensional benchmark problems, including the CEC'2013 test suite for large-scale global optimisation. C 3 is compared with several state-of-the-art algorithms, which shows that C 3 is among the most competitive algorithms. C 3 outperforms the other algorithms for most partially separable functions and overlapping functions. This shows that C 3 is an effective algorithm for large-scale global optimisation. This paper demonstrates the enhanced performance by using constructive heuristics for generating initial feasible solutions for Cooperative Coevolutionary algorithms in a multi-start framework.

Dynamical Sphere Regrouping Particle Swarm Optimization: A Proposed Algorithm for Dealing with PSO Premature Convergence in Large-Scale Global Optimization

“…According to this new range obtained, the algorithm sets a new maximum velocity with each regrouping, i.e., there are j upper limits determined depending on the search space dimension, as in Equation (7).…”

“…The IEEE Congress on Evolutionary Computation (CEC) tool is a widely used benchmark for LSGO problems. It includes optimization functions that simulate real-world problems and has been used to evaluate algorithms in [5][6][7][8][9][10][11][12][13][14][15][16][17][18][19]. Several metaheuristics have been applied to address the challenges associated with solving Large-Scale Global Optimization (LSGO) problems.…”

Energy Model for Motion Planning of 2D-Belt Press Line Tending Robots

“…The results will provide data for comparing robots with mechanical brakes against robots without mechanical brakes. Note that some of these results were obtained from the resulting Pareto-front by multi-objective optimisation with the mo C 3 algorithm (for cycle-time of 3.5 s and below), others (for cycle-time above 3.5 s) that are not on the Paretofront by single objective optimisation with the C 3i DE algorithm proposed by Glorieux et al [17,18]. It is also important to note that, since these optimisation algorithms are stochastic, each motion planning optimisation was repeated 10 times in order to get statistically reliable results.…”