Performance assessment of meta-heuristics for composite layup optimisation

Shahin Jalili Dargalusani; Reza Khani; Alireza Maheri; Yousef Hosseinzadeh

doi:10.1007/s00521-021-06519-0

Performance assessment of meta-heuristics for composite layup optimisation

Shahin Jalili Dargalusani^* (Corresponding Author), Reza Khani, Alireza Maheri, Yousef Hosseinzadeh

^*Corresponding author for this work

University of Tabriz

Research output: Contribution to journal › Article › peer-review

2 Citations (Scopus)

1 Downloads (Pure)

Abstract

This paper investigates the performance of several meta-heuristic algorithms, including particle swarm optimisation (PSO), different variants of differential evolution (DE), biogeography-based optimisation (BBO), cultural algorithm (CA), optics-inspired optimisation (OIO), and league championship algorithm (LCA), for optimum layup of laminated composite plates. The study provides detailed Pseudo codes for different algorithms. The buckling capacity maximisation of a 64-layer laminated composite plate under various load scenarios has been considered as the benchmark problem, in which the design variables are the stacking sequences of layers. A deep statistical comparison (DSC) method is employed to rank the performance of different algorithms. The DSC uses a non-parametric two-sample Kolmogorov–Smirnov (KS) test to conduct the performance comparisons between the algorithms. The overall performance
rankings obtained from the DSC suggest that the LCA, OIO, and PSO algorithms
perform remarkably better in comparison to other algorithms. The comparisons provide some interesting conclusions on the performance of different algorithms

Original language	English
Journal	Neural Computing and Applications
DOIs	https://doi.org/10.1007/s00521-021-06519-0
Publication status	Published - 21 Sept 2021

Keywords

Composite structure
Meta-heuristics
Deep statistical comparison
Kolmogorov– Smirnov test

Access to Document

10.1007/s00521-021-06519-0Licence: CC BY

Jalili_etal_NCA_performance_assessment_of_AAM
This is the peer-reviewed accepted version of [cite]. Version of record available at [doi]
Accepted author manuscript, 1.07 MBLicence: Unspecified

Cite this

@article{21bb9831a020474ba6d688f0b6ad1aa9,

title = "Performance assessment of meta-heuristics for composite layup optimisation",

abstract = "This paper investigates the performance of several meta-heuristic algorithms, including particle swarm optimisation (PSO), different variants of differential evolution (DE), biogeography-based optimisation (BBO), cultural algorithm (CA), optics-inspired optimisation (OIO), and league championship algorithm (LCA), for optimum layup of laminated composite plates. The study provides detailed Pseudo codes for different algorithms. The buckling capacity maximisation of a 64-layer laminated composite plate under various load scenarios has been considered as the benchmark problem, in which the design variables are the stacking sequences of layers. A deep statistical comparison (DSC) method is employed to rank the performance of different algorithms. The DSC uses a non-parametric two-sample Kolmogorov–Smirnov (KS) test to conduct the performance comparisons between the algorithms. The overall performancerankings obtained from the DSC suggest that the LCA, OIO, and PSO algorithmsperform remarkably better in comparison to other algorithms. The comparisons provide some interesting conclusions on the performance of different algorithms",

keywords = "Composite structure, Meta-heuristics, Deep statistical comparison, Kolmogorov– Smirnov test",

author = "{Jalili Dargalusani}, Shahin and Reza Khani and Alireza Maheri and Yousef Hosseinzadeh",

year = "2021",

month = sep,

day = "21",

doi = "10.1007/s00521-021-06519-0",

language = "English",

journal = "Neural Computing and Applications",

issn = "0941-0643",

publisher = "Springer London",

}

TY - JOUR

T1 - Performance assessment of meta-heuristics for composite layup optimisation

AU - Jalili Dargalusani, Shahin

AU - Khani, Reza

AU - Maheri, Alireza

AU - Hosseinzadeh, Yousef

PY - 2021/9/21

Y1 - 2021/9/21

N2 - This paper investigates the performance of several meta-heuristic algorithms, including particle swarm optimisation (PSO), different variants of differential evolution (DE), biogeography-based optimisation (BBO), cultural algorithm (CA), optics-inspired optimisation (OIO), and league championship algorithm (LCA), for optimum layup of laminated composite plates. The study provides detailed Pseudo codes for different algorithms. The buckling capacity maximisation of a 64-layer laminated composite plate under various load scenarios has been considered as the benchmark problem, in which the design variables are the stacking sequences of layers. A deep statistical comparison (DSC) method is employed to rank the performance of different algorithms. The DSC uses a non-parametric two-sample Kolmogorov–Smirnov (KS) test to conduct the performance comparisons between the algorithms. The overall performancerankings obtained from the DSC suggest that the LCA, OIO, and PSO algorithmsperform remarkably better in comparison to other algorithms. The comparisons provide some interesting conclusions on the performance of different algorithms

AB - This paper investigates the performance of several meta-heuristic algorithms, including particle swarm optimisation (PSO), different variants of differential evolution (DE), biogeography-based optimisation (BBO), cultural algorithm (CA), optics-inspired optimisation (OIO), and league championship algorithm (LCA), for optimum layup of laminated composite plates. The study provides detailed Pseudo codes for different algorithms. The buckling capacity maximisation of a 64-layer laminated composite plate under various load scenarios has been considered as the benchmark problem, in which the design variables are the stacking sequences of layers. A deep statistical comparison (DSC) method is employed to rank the performance of different algorithms. The DSC uses a non-parametric two-sample Kolmogorov–Smirnov (KS) test to conduct the performance comparisons between the algorithms. The overall performancerankings obtained from the DSC suggest that the LCA, OIO, and PSO algorithmsperform remarkably better in comparison to other algorithms. The comparisons provide some interesting conclusions on the performance of different algorithms

KW - Composite structure

KW - Meta-heuristics

KW - Deep statistical comparison

KW - Kolmogorov– Smirnov test

U2 - 10.1007/s00521-021-06519-0

DO - 10.1007/s00521-021-06519-0

M3 - Article

SN - 0941-0643

JO - Neural Computing and Applications

JF - Neural Computing and Applications

ER -

Performance assessment of meta-heuristics for composite layup optimisation

Abstract

Keywords

Access to Document

Fingerprint

Cite this