New Particle Swarm Optimizer Algorithm with Chaotic Maps for Combinatorial Global Optimization Problems. An Application to the Deconvolution of Mössbauer Spectra
Journal
Data-Driven Innovation for Intelligent Technology : Perspectives and Applications in ICT
ISSN
2197-6503
Publisher
Springer
Date Issued
2024-01-01
Author(s)
González Morfín, Juan
Type
Resource Types::text::book::book part
Abstract
In this chapter, we present a novel method for addressing global optimization problems inspired by evolutionary algorithms found in nature. We integrate the Comprehensive Learning Particle Swarm Optimization (CLPSO) algorithm with random value generation based on chaotic maps. The resulting algorithm is applied to the computationally complex task of deconvoluting Mossbauer spectra. We implement ten chaotic maps to generate random values and compare their performance with traditional random number generators. Through experiments, we demonstrate that the developed algorithm excels in exploring the search space and exhibits fast intensification in finding the global minimum. In addition, we perform a comprehensive review of existing solutions to the Mossbauer spectrum deconvolution problem, highlighting the scarce availability of developments in this area. We also present a user-friendly program designed with an intuitive interface to facilitate the deconvolution process by Spector Mossbauer. This program will be freely distributed without operational restrictions. Experimental validation is performed on Mossbauer spectra generated using the developed program and those obtained by experimental means, affirming the efficiency of the new algorithm conceived. ©Springer.
License
Acceso Restringido
How to cite
Martinez-Rios, F., Jiménez-López, O., Alvarez Guillen, L.A. (2024). New Particle Swarm Optimizer Algorithm with Chaotic Maps for Combinatorial Global Optimization Problems. An Application to the Deconvolution of Mössbauer Spectra. In: Ponce, H., Brieva, J., Lozada-Flores, O., Martínez-Villaseñor, L., Moya-Albor, E. (eds) Data-Driven Innovation for Intelligent Technology. Studies in Big Data, vol 148. Springer, Cham. https://doi.org/10.1007/978-3-031-54277-0_7