Solving the Turbine Balancing Problem Using a Metropolis Algorithm Hybridized with the Hooke-Jeeves Method
Abstract
Combinatorial optimization problems have been a great challenge for metaheuristics. One of them, the turbine balancing problem, which is NP-hard, is solved here. In order to do so, we use a Metropolis Algorithm, the Particle Collision Algorithm (PCA), hybridized with the well-known Hooke-Jeeves pattern search method. The aim of this algorithm, called Hooke-Jeeves PCA, is to perform a wide search in the solution space using a stochastic optimization method (the PCA) and then scan the promising areas with a local search technique (Hooke-Jeeves). This algorithm is favorably compared against a state-of-the-art metaheuristic, differential evolution. Our results show that Hooke-Jeeves PCA has the potential to be applied to other combinatorial optimization problems.
Downloads
Published
How to Cite
Issue
Section
License
DECLARAÇÃO DE TRANSFERÊNCIA DE DIREITOS AUTORAIS
Os autores do manuscrito submetido declaram ter conhecimento que em caso de aceitação do artigo, a Revista Cereus, passa a ter todos os direitos autorais sobre o mesmo. O Artigo será de propriedade exclusiva da Revista, sendo vedada qualquer reprodução, em qualquer outra parte ou meio de divulgação, impressa ou eletrônica.