Chemical Engineering Science, Год журнала: 2025, Номер unknown, С. 121456 - 121456
Опубликована: Март 1, 2025
Язык: Английский
Journal of Computational Design and Engineering, Год журнала: 2024, Номер 11(4), С. 151 - 183
Опубликована: Июнь 12, 2024
Abstract The slime mould algorithm (SMA), as an emerging and promising swarm intelligence algorithm, has been studied in various fields. However, SMA suffers from issues such easily getting trapped local optima slow convergence, which pose challenges when applied to practical problems. Therefore, this study proposes improved SMA, named HESMA, by incorporating the covariance matrix adaptation evolution strategy (CMA-ES) storing best position of each individual (SBP). On one hand, CMA-ES enhances algorithm’s exploration capability, addressing issue being unable explore vicinity optimal solution. other SBP convergence speed prevents it diverging inferior solutions. Finally, validate effectiveness our proposed conducted experiments on 30 IEEE CEC 2017 benchmark functions compared HESMA with 12 conventional metaheuristic algorithms. results demonstrated that indeed achieved improvements over SMA. Furthermore, highlight performance further, 13 advanced algorithms, showed outperformed these algorithms significantly. Next, five engineering optimization problems, experimental revealed exhibited significant advantages solving real-world These findings further support practicality complex design challenges.
Язык: Английский
Процитировано
6
International Journal of Hydrogen Energy, Год журнала: 2024, Номер 83, С. 1003 - 1023
Опубликована: Авг. 14, 2024
Язык: Английский
Процитировано
6
Scientific Reports, Год журнала: 2024, Номер 14(1)
Опубликована: Окт. 19, 2024
The Whale Optimization Algorithm (WOA) is regarded as a classic metaheuristic algorithm, yet it suffers from limited population diversity, imbalance between exploitation and exploration, low solution accuracy. In this paper, we propose the Spiral-Enhanced (SEWOA), which incorporates nonlinear time-varying self-adaptive perturbation strategy an Archimedean spiral structure into original WOA. enhances diversity of space, aiding algorithm in escaping local optima. optimization dynamic improves algorithm's search capability effectiveness proposed validated multiple perspectives using CEC2014 test functions, CEC2017 23 benchmark functions. experimental results demonstrate that enhanced significantly balances global search, Additionally, SEWOA exhibits excellent performance solving three engineering design problems, showcasing its value wide range potential applications.
Язык: Английский
Процитировано
6
Energy, Год журнала: 2024, Номер 302, С. 131798 - 131798
Опубликована: Май 27, 2024
Язык: Английский
Процитировано
5
Knowledge-Based Systems, Год журнала: 2024, Номер 304, С. 112409 - 112409
Опубликована: Авг. 30, 2024
Язык: Английский
Процитировано
5
Journal of Computational Design and Engineering, Год журнала: 2022, Номер 9(6), С. 2196 - 2234
Опубликована: Сен. 13, 2022
Abstract Salp swarm algorithm (SSA) is a well-established population-based optimizer that exhibits strong exploration ability, but slow convergence and poor exploitation capability. In this paper, an endeavour made to enhance the performance of basic SSA. The new upgraded version SSA named as ‘adaptive strategy-based (ABSSA) algorithm’ proposed in paper. First, exploratory scope food source navigating commands are enriched using inertia weight boosted global best-guided mechanism. Next, novel velocity clamping strategy designed efficiently stabilize balance between operations. addition, adaptive conversion parameter tactic modify position update equation effectively intensify local competency solution accuracy. effectiveness ABSSA verified by series problems, including 23 classical benchmark functions, 29 complex optimization problems from CEC 2017, five engineering design tasks. experimental results show developed approach performs significantly better than standard other competitors. Moreover, implemented handle path planning obstacle avoidance (PPOA) tasks autonomous mobile robots compared with some intelligent approach-based planners. indicate ABSSA-based PPOA method reliable algorithm.
Язык: Английский
Процитировано
20
Entropy, Год журнала: 2023, Номер 25(1), С. 178 - 178
Опубликована: Янв. 16, 2023
Multi-level thresholding image segmentation divides an into multiple regions of interest and is a key step in processing analysis. Aiming toward the problems low accuracy slow convergence speed traditional multi-level threshold methods, this paper, we present based on improved slime mould algorithm (ISMA) symmetric cross-entropy for global optimization tasks. First, elite opposition-based learning (EOBL) was used to improve quality diversity initial population accelerate speed. The adaptive probability adjust selection enhance ability jump out local optimum. historical leader strategy, which selects optimal information as position update, found accuracy. Subsequently, 14 benchmark functions were evaluate performance ISMA, comparing it with other well-known algorithms terms accuracy, speed, significant differences. tested method proposed paper eight grayscale images compared criteria algorithms. experimental metrics include average fitness (mean), standard deviation (std), peak signal noise ratio (PSNR), structure similarity index (SSIM), feature (FSIM), utilized segmentation. results demonstrated that superior algorithms, can be effectively applied task
Язык: Английский
Процитировано
13
Quality and Reliability Engineering International, Год журнала: 2024, Номер 40(4), С. 1502 - 1525
Опубликована: Янв. 23, 2024
Abstract The pivotal problem in reliability analysis is how to use as few actual assessments possible obtain an accurate failure probability. Although adaptive Kriging provides a viable method address this problem, unsatisfied surrogate accuracy and modeling samples often lead unacceptable computing burden. In paper, optimized combining efficient sampling (AOK‐ES) proposed: first, enhance the approximation ability, high‐fidelity model (OKM) established; further, ensure quality of OKM calculation, improved Latin hypercube importance approach are developed correspondingly. Six different types case studies demonstrate superiority proposed AOK‐ES. results that AOK‐ES holds potential reduce cost while ensuring accuracy.
Язык: Английский
Процитировано
4
Artificial Intelligence Review, Год журнала: 2024, Номер 57(6)
Опубликована: Май 9, 2024
Abstract In this study, the Learning Search Algorithm (LSA) is introduced as an innovative optimization algorithm that draws inspiration from swarm intelligence principles and mimics social learning behavior observed in humans. The LSA optimizes search process by integrating historical experience real-time information, enabling it to effectively navigate complex problem spaces. By doing so, enhances its global development capability provides efficient solutions challenging tasks. Additionally, improves collective capacity incorporating teaching active behaviors within population, leading improved local capabilities. Furthermore, a dynamic adaptive control factor utilized regulate algorithm’s exploration abilities. proposed rigorously evaluated using 40 benchmark test functions IEEE CEC 2014 2020, compared against nine established evolutionary algorithms well 11 recently algorithms. experimental results demonstrate superiority of algorithm, achieves top rank Friedman rank-sum test, highlighting power competitiveness. Moreover, successfully applied solve six real-world engineering problems 15 UCI datasets feature selection problems, showcasing significant advantages potential for practical applications problems.
Язык: Английский
Процитировано
4
Journal of Energy Storage, Год журнала: 2024, Номер 94, С. 112412 - 112412
Опубликована: Июнь 13, 2024
Язык: Английский
Процитировано
4