We compare the performance of six recent global optimization algorithms: Imperialist Competitive Algorithm (ICA), Firefly Algorithm (FA), Water Cycle Algorithm (WCA), Whale Swarm Optimization (WSO), Fireworks Algorithm (FWA) and Quantum Particle Swarm Optimization (QPSO). These methods have been introduced in the last few years and have found very limited or no applications in geophysical exploration problems thus far. The methods are first tested on two multi-minima analytic objective functions often used to test optimization algorithms: The Rastrigin and the Schwefel functions. Then, they are compared on the residual statics corrections, which is a highly non-linear geophysical optimization problem. In particular, we are interested in testing the convergence capabilities of these methods as the number of unknown model parameters increases. The different approaches are compared against a standard implementation of the Particle Swarm optimization (PSO), that is a popular global search method. The tests on the analytical functions and on the residual statics corrections demonstrate that FA, WCA, WSO and FWA outperform the other approaches in solving multi-minima and high-dimensional optimization problems. Conversely, PSO and ICA show limited exploration capabilities and lower convergence rates with respect to the other approaches.

Comparisons of recent global optimization algorithms: tests on analytic objective functions and residual statics corrections

Mattia Aleardi
;
Silvio Pierini;Angelo Sajeva
2019-01-01

Abstract

We compare the performance of six recent global optimization algorithms: Imperialist Competitive Algorithm (ICA), Firefly Algorithm (FA), Water Cycle Algorithm (WCA), Whale Swarm Optimization (WSO), Fireworks Algorithm (FWA) and Quantum Particle Swarm Optimization (QPSO). These methods have been introduced in the last few years and have found very limited or no applications in geophysical exploration problems thus far. The methods are first tested on two multi-minima analytic objective functions often used to test optimization algorithms: The Rastrigin and the Schwefel functions. Then, they are compared on the residual statics corrections, which is a highly non-linear geophysical optimization problem. In particular, we are interested in testing the convergence capabilities of these methods as the number of unknown model parameters increases. The different approaches are compared against a standard implementation of the Particle Swarm optimization (PSO), that is a popular global search method. The tests on the analytical functions and on the residual statics corrections demonstrate that FA, WCA, WSO and FWA outperform the other approaches in solving multi-minima and high-dimensional optimization problems. Conversely, PSO and ICA show limited exploration capabilities and lower convergence rates with respect to the other approaches.
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11568/993965
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 0
  • ???jsp.display-item.citation.isi??? ND
social impact