Parallel GPU-Acceleration of Metaphorless Optimization Algorithms: Application for Solving Large-Scale Nonlinear Equation Systems DOI Creative Commons
Bruno Silva, Luiz Guerreiro Lopes, Fábio Mendonça

et al.

Applied Sciences, Journal Year: 2024, Volume and Issue: 14(12), P. 5349 - 5349

Published: June 20, 2024

Traditional population-based metaheuristic algorithms are effective in solving complex real-world problems but require careful strategy selection and parameter tuning. Metaphorless optimization have gained importance due to their simplicity efficiency. However, research on applicability for large systems of nonlinear equations is still incipient. This paper presents a review detailed description the main metaphorless algorithms, including Jaya enhanced (EJAYA) three Rao best-worst-play (BWP) algorithm, new max–min greedy interaction (MaGI) algorithm. article improved GPU-based massively parallel versions these using more efficient parallelization strategy. In particular, novel GPU-accelerated implementation MaGI algorithm proposed. The developed were implemented Julia programming language. Both high-end professional-grade GPUs powerful consumer-oriented GPU used testing, along with set hard, large-scale equation system gauge speedup gains from parallelizations. computational experiments produced substantial gains, ranging 33.9× 561.8×, depending test parameters testing. highlights efficiency proposed considered.

Language: Английский

Supercell thunderstorm algorithm (STA): a nature-inspired metaheuristic algorithm for engineering optimization DOI Creative Commons
Mohamed H. Hassan,

Salah Kamel

Neural Computing and Applications, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 1, 2025

Abstract In this paper, an optimization algorithm called supercell thunderstorm (STA) is proposed. STA draws inspiration from the strategies employed by storms, such as spiral motion, tornado formation, and jet stream. It a computational specifically designed to simulate model behavior of thunderstorms. These storms are known for their rotating updrafts, strong wind shear, potential generating tornadoes. The procedures based on three distinct approaches: exploring divergent search space using exploiting convergent through navigating with aid To evaluate effectiveness proposed in achieving optimal solutions various problems, series test sequences were conducted. Initially, was tested set 23 well-established functions. Subsequently, algorithm’s performance assessed more complex including ten CEC2019 functions, second experimental sequence. Finally, applied five real-world engineering problems validate its effectiveness. results compared those contemporary metaheuristic methods. analysis clearly demonstrates that developed outperforms other methods terms performance.

Language: Английский

Citations

1
Improving teaching-learning-based optimization algorithm with golden-sine and multi-population for global optimization DOI

Aosheng Xing,

Yong Chen,

Jinyi Suo

et al.

Mathematics and Computers in Simulation, Journal Year: 2024, Volume and Issue: 221, P. 94 - 134

Published: Feb. 17, 2024

Language: Английский

Citations

7
Swarm Bipolar Algorithm: A Metaheuristic Based on Polarization of Two Equal Size Sub Swarms DOI Open Access
Purba Daru Kusuma, Ashri Dinimaharawati

International journal of intelligent engineering and systems, Journal Year: 2024, Volume and Issue: 17(2), P. 377 - 389

Published: Feb. 28, 2024

This paper presents a new metaphor-free metaheuristic search called the swarm bipolar algorithm (SBA).SBA is developed mainly based on non-free-lunch (NFL) doctrine, which mentions non-existence of any general optimizer appropriate to answer all varieties problems.The construction SBA splitting into two equal-sized swarms diversify searching process while performing intensification within subswarms.There are types finest members: member for whole and in every sub-swarm.There four directed searches performed iteration: (1) toward member, (2) sub-swarm (3) middle between members, (4) relative randomly picked from another sub-swarm.The performance assessed through assessments with set 23 functions representing optimization problem.In benchmark assessment, contended five metaheuristics: northern goshawk (NGO), language education (LEO), coati (COA), fully informed (FISA), total interaction (TIA).The result superiority among its contenders by being better than NGO, LEO, COA, FISA, TIA 21, 16, 16,21,and 18 functions.The single assessment evaluate each strategy involved SBA.The shows that members best SBA.

Language: Английский

Citations

4
Fully Informed Search Algorithm for Estimating the Parameters of Li-Ion Battery Model under UDDS Drive Cycle Profile DOI Open Access
Walid Merrouche, Badis Lekouaghet,

Islam abd elsammed Boughiout

et al.

Transportation research procedia, Journal Year: 2025, Volume and Issue: 84, P. 275 - 282

Published: Jan. 1, 2025

Language: Английский

Citations

0
A Self-Adaptive Population-Based Hybrid Optimisation Technique for Multireservoir Benchmark Problems DOI

K. B. Baladaniya,

P. L. Patel, P. V. Timbadiya

et al.

Water Resources Management, Journal Year: 2025, Volume and Issue: unknown

Published: April 14, 2025

Language: Английский

Citations

0
Multi-strategy synthetized equilibrium optimizer and application DOI Creative Commons

Quandang Sun,

Xinyu Zhang, Ruixia Jin

et al.

PeerJ Computer Science, Journal Year: 2024, Volume and Issue: 10, P. e1760 - e1760

Published: Jan. 12, 2024

Background Improvement on the updating equation of an algorithm is among most improving techniques. Due to lack search ability, high computational complexity and poor operability equilibrium optimizer (EO) in solving complex optimization problems, improved EO proposed this article, namely multi-strategy synthetized (MS-EO). Method Firstly, a simplified strategy adopted improve reduce complexity. Secondly, information sharing updates concentrations early iterative stage using dynamic tuning form (SS-EO) enhance exploration ability. Thirdly, migration golden section are used for particle construct Golden SS-EO (GS-EO) Finally, elite learning implemented worst late MS-EO strengthen exploitation The strategies embedded into balance between by giving full play their respective advantages. Result Finding Experimental results functions from CEC2013 CEC2017 test sets demonstrate that outperforms quite few state-of-the-art algorithms running speed operability. experimental feature selection several datasets show also provides more

Language: Английский

Citations

2
Stochastic Shaking Algorithm: A New Swarm-Based Metaheuristic and Its Implementation in Economic Load Dispatch Problem DOI Open Access
Purba Daru Kusuma, Anggunmeka Luhur Prasasti

International journal of intelligent engineering and systems, Journal Year: 2024, Volume and Issue: 17(3), P. 276 - 289

Published: May 3, 2024

This paper introduces a novel metaheuristic named the stochastic shaking algorithm (SSA), which is rooted in swarm intelligence principles.The innovation lies its unique utilization of iteration for selecting references during guided searches through approach.The optimization process involves two sequential steps: primary reference first step finest member, while second step, it mean all finer members plus one.This then combined with randomly chosen solution within space, serving as secondary reference.SSA undergoes evaluation contexts.The assessing performance using set 23 classic functions theoretical use case.The tackling economic load dispatch problem (ELD), practical case featuring system 13 generators various energy resources.The study compares SSA against five other metaheuristics-One to One Based Optimization (OOBO), Kookaburra Algorithm (KOA), Language Education (LEO), Total Interaction (TIA), and Walrus (WaOA).Results indicate SSA's superiority over OOBO, KOA, LEO, TIA, WaOA 21, 13, 11, 16, 14 out functions, respectively.Additionally, reveals intense competition among six metaheuristics.

Language: Английский

Citations

2
Solution of optimal reactive power dispatch by Lévy-flight phasor particle swarm optimization DOI Creative Commons
Milad Gil, Ebrahim Akbari, Abolfazl Rahimnejad

et al.

Intelligent Systems with Applications, Journal Year: 2024, Volume and Issue: 23, P. 200398 - 200398

Published: June 15, 2024

Optimal reactive power dispatch (ORPD) problems are important tools for the sake of security and economics systems. The ORPD nonlinear optimization to minimize real losses voltage profile enhancement by optimizing several discrete continuous control variables. This paper proposes a Lévy-flight phasor particle swarm (LPPSO) solving while considering in two standard simulation results demonstrate that LPPSO algorithm proves itself as an acceptable method reaching more optimal solution problems.

Language: Английский

Citations

2
A novel community development algorithm and its application to optimize main steam temperature of supercritical units DOI
WU Ming-liang, Dongsheng Yang, Yingchun Wang

et al.

Expert Systems with Applications, Journal Year: 2024, Volume and Issue: 254, P. 124190 - 124190

Published: June 10, 2024

Language: Английский

Citations

1
A Novel Quantum Algorithm for Solving Optimization Problems in Electrical Engineering DOI
Ajay Sharma, Kuldeep Sharma, Kanchan Yadav

et al.

Published: Feb. 21, 2024

The ability of theoretical and monadic quantum models against number comparison to conventional Quantum Genetic Algorithm (QGA), quantitative particle swarm optimization, ant colony groups with simulated annealing types is analyzed in the context electrical engineering. Several experiments were done, involving a collection multiple data sets standard for proposed circuit layout within power distribution signal processing applications. In regard comparative analysis, each algorithm presents its particular strengths QGA-competitive convergence speed; ͟QPSO – quick conversion individuals into global optimum during evolution's course robust solutions quality an unstable environment without tuning. related works algorithms include evaluating it metaheuristics systems, nature-inspired hybrid heuristic approaches as well physics motivated optimization schemes. paper emphasizes superiority over their classical counterparts, which major innovation space. This detailed analysis contributes further comprehending potentials computing overcoming tough challenges faced by engineers.

Language: Английский

Citations

0