Lecture notes in computer science, Год журнала: 2024, Номер unknown, С. 223 - 234
Опубликована: Янв. 1, 2024
Язык: Английский
Journal of Computational Design and Engineering, Год журнала: 2023, Номер 10(6), С. 2223 - 2250
Опубликована: Окт. 26, 2023
Abstract The coati optimization algorithm (COA) is a meta-heuristic proposed in 2022. It creates mathematical models according to the habits and social behaviors of coatis: (i) In group organization coatis, half coatis climb trees chase their prey away, while other wait beneath catch it (ii) Coatis avoidance predators behavior, which gives strong global exploration ability. However, over course our experiment, we uncovered opportunities for enhancing algorithm’s performance. When confronted with intricate problems, certain limitations surfaced. Much like long-nosed raccoon gradually narrowing its search range as approaches optimal solution, COA exhibited tendencies that could result reduced convergence speed risk becoming trapped local optima. this paper, propose an improved (ICOA) enhance efficiency. Through sound-based envelopment strategy, can capture more quickly accurately, allowing converge rapidly. By employing physical exertion have greater variety escape options when being chased, thereby exploratory capabilities ability Finally, lens opposition-based learning strategy added improve To validate performance ICOA, conducted tests using IEEE CEC2014 CEC2017 benchmark functions, well six engineering problems.
Язык: Английский
Процитировано
13
Mathematics, Год журнала: 2022, Номер 10(20), С. 3765 - 3765
Опубликована: Окт. 12, 2022
The group teaching optimization algorithm (GTOA) is a meta heuristic simulating the mechanism. inspiration of GTOA comes from Each student will learn knowledge obtained in teacher phase, but each student’s autonomy weak. This paper considers that has different learning motivations. Elite students have strong self-learning ability, while ordinary general motivation. To solve this problem, proposes motivation strategy and adds random opposition-based restart to enhance global performance (MGTOA). In order verify effect MGTOA, 23 standard benchmark functions 30 test IEEE Evolutionary Computation 2014 (CEC2014) are adopted proposed MGTOA. addition, MGTOA also applied six engineering problems for practical testing achieved good results.
Язык: Английский
Процитировано
22
Applied Sciences, Год журнала: 2022, Номер 12(19), С. 10144 - 10144
Опубликована: Окт. 9, 2022
The Gorilla Troops Optimizer (GTO) is a novel Metaheuristic Algorithm that was proposed in 2021. Its design inspired by the lifestyle characteristics of gorillas, including migration to known position, an undiscovered moving toward other following silverback gorillas and competing with for females. However, like Algorithms, GTO still suffers from local optimum, low diversity, imbalanced utilization, etc. In order improve performance GTO, this paper proposes modified (MGTO). improvement strategies include three parts: Beetle-Antennae Search Based on Quadratic Interpolation (QIBAS), Teaching–Learning-Based Optimization (TLBO) Quasi-Reflection-Based Learning (QRBL). Firstly, QIBAS utilized enhance diversity position silverback. Secondly, teacher phase TLBO introduced update behavior 50% probability. Finally, quasi-reflection generated QRBL. optimal solution can be updated comparing these fitness values. MGTO comprehensively evaluated 23 classical benchmark functions, 30 CEC2014 10 CEC2020 functions 7 engineering problems. experimental results show has competitive promising prospects real-world optimization tasks.
Язык: Английский
Процитировано
21
Soft Computing, Год журнала: 2025, Номер unknown
Опубликована: Фев. 7, 2025
Язык: Английский
Процитировано
0
Renewable Energy, Год журнала: 2025, Номер unknown, С. 123032 - 123032
Опубликована: Апрель 1, 2025
Язык: Английский
Процитировано
0
IEEE Access, Год журнала: 2022, Номер 10, С. 75040 - 75062
Опубликована: Янв. 1, 2022
The arithmetic optimization algorithm (AOA) is based on the distribution character of dominant operators and imitates addition (A), subtraction (S), multiplication (M) division (D) to find global optimal solution in entire search space. However, basic AOA has some drawbacks premature convergence, easily falls into a local value, slow convergence rate, low calculation precision. To improve overall ability overcome AOA, an enhanced (EAOA) Lévy variation differential sorting proposed solve function project optimization. increases population diversity, broadens space, enhances improves filters out agent, avoids stagnation, accelerates rate. EAOA realizes complementary advantages avoid falling optimum convergence. sixteen benchmark functions five engineering design projects are applied verify effectiveness feasibility EAOA. compared with other algorithms by minimizing fitness such as artificial bee colony, ant line optimizer, cuckoo search, dragonfly algorithm, moth-flame optimization, sine cosine water wave algorithm. experimental results show that superior algorithms, can effectively balance exploration exploitation obtain best solution. In addition, faster higher precision stronger stability.
Язык: Английский
Процитировано
16
Опубликована: Фев. 6, 2024
This paper introduces a novel multi-strategy enhanced dung beetle optimization (MSDBO) algorithm that is designed to address several issues identified in the standard algorithm. Specifically, MSDBO aims enhance convergence speed, reduce susceptibility local optima, and increase search accuracy. By incorporating three strategies: tent chaotic mapping for population initialization, golden sinusoidal strategy position updating, Lévy flight balancing exploration exploitation, enhanced. The evaluated using twelve benchmark test functions compared against five state-of-the-art algorithms. results consistently show exhibits faster speeds more accurate solutions than other algorithms across most of functions. In addition, also applied optimize parameters valve plate, including close angle, cross triangle groove sizes, wrap angle. outcomes reveal effectively minimizes pressure ripples piston chamber, resulting reduced flow rate fluctuations noise emission initial design. study highlights potential tackling complex nonlinear engineering problems.
Язык: Английский
Процитировано
3
Journal of Computational Science, Год журнала: 2022, Номер 64, С. 101867 - 101867
Опубликована: Сен. 16, 2022
Язык: Английский
Процитировано
15
Current Journal of Applied Science and Technology, Год журнала: 2024, Номер 43(4), С. 12 - 20
Опубликована: Март 9, 2024
The dung beetle optimization (DBO) algorithm is a newly swarm intelligence inspired by the biological behaviors of beetles while it still has disadvantages easy convergence to local optimal, slow speed, and poor global search capability. This paper proposes an adaptive with golden sine (Gold-SA), denoted as Gold-SA-based DBO (GSDBO) algorithm. Firstly, PWLCM chaotic mapping introduced generate population individuals increase diversity explore more space. Secondly, position update formula for mathematical model ball-rolling behavior without obstacle replaced that Gold-SA, which can accelerate speed improve accuracy. Finally, weight coefficients are used stage thief beetles. strategy boost balance exploration vs exploitation, simultaneously. Furthermore, GSDBO proved be effective comparing some algorithms on benchmark functions different characteristics. results demonstrate accuracy stability.
Язык: Английский
Процитировано
2
International Journal of Applied Metaheuristic Computing, Год журнала: 2023, Номер 14(1), С. 1 - 27
Опубликована: Фев. 24, 2023
Arithmetic optimization algorithm (AOA) is a recent population-based metaheuristic widely used for solving problems. However, the emerging large-scale problems pose great challenge AOA due to its prohibitive computational cost traverse huge solution space effectively. This article proposes parallel Spark-AOA using Scala on Apache Spark computing platform. leverages intrinsic nature of and native iterative in-memory computation support through resilient distributed datasets (RDD) accelerate process. divides solutions population into several subpopulations that are multiple RDD partitions manipulated concurrently. Simulation experiments different benchmark functions with up 1,000-dimension three engineering design demonstrate outperforms considerably standard Spark-based implementations two metaheuristics both in terms run-time quality.
Язык: Английский
Процитировано
5