Swarm and Evolutionary Computation最新文献_第7页

An adaptive differential evolution algorithm with exponential crossover based on a learning strategy within the difference vector 基于差分向量内学习策略的指数交叉自适应差分进化算法

IF 8.5 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Swarm and Evolutionary Computation

Pub Date : 2025-12-04 DOI: 10.1016/j.swevo.2025.102247

Junpeng Chen , Zhenyu Meng

Most Differential Evolution (DE) researchers tend to adopt the binomial crossover operation in tackling optimization problems. However, we find that the DE variants using exponential crossover can also achieve superior performance to those using binomial crossover, as long as appropriate parameter control strategies are applied. Therefore, this paper proposes a new DE algorithm, an adaptive Differential Evolution algorithm with exponential crossover based on Learning Strategy within Difference vector (DLS-DE), to fill the gap in this field. The main contributions of this work are summarized as follows: First, a two-phase parameter control strategy is designed to regulate the scale factor

F

for balancing exploration and exploitation. In addition, considering the dispersion of effective parameter values, an adaptive

σ_{F}

strategy is proposed to adjust the sampling distribution and enhance parameter adaptability. Second, a differential vector learning strategy is developed to identify and incorporate promising difference vector information during an individual’s stagnation, enabling the search direction to adapt based on its past performance. Finally, the algorithm employs exponential crossover, where the crossover rate

C R

is automatically generated, and a fitness-independent parameter weight update mechanism is adopted to mitigate premature convergence. The performance of DLS-DE is evaluated on 88 benchmark functions from the CEC2013, CEC2014, and CEC2017 test suites. Statistical analyses, including the Friedman test and the Wilcoxon rank-sum test, demonstrate its effectiveness and competitiveness compared with ten state-of-the-art algorithms. In addition, DLS-DE is applied to an Economic Load Dispatch (ELD) problem in a power system with 40 generating units, achieving satisfactory results.

大多数差分进化研究者倾向于采用二项交叉操作来解决优化问题。然而，我们发现，只要采用适当的参数控制策略，使用指数交叉的DE变体也可以取得比使用二项交叉的DE变体更好的性能。为此，本文提出了一种新的差分进化算法——基于差分向量内学习策略的指数交叉自适应差分进化算法（DLS-DE）来填补这一领域的空白。本文的主要贡献如下：首先，设计了一种两阶段参数控制策略来调节规模因子F，以实现勘探与开采的平衡。此外，考虑到有效参数值的离散性，提出了一种自适应σF策略来调整采样分布，增强参数自适应能力。其次，开发了一种差分向量学习策略，用于在个体停滞期间识别和合并有希望的差分向量信息，使搜索方向能够根据其过去的表现进行调整。最后，该算法采用指数交叉，自动生成交叉率CR，并采用与适应度无关的参数权重更新机制来缓解过早收敛。DLS-DE的性能在CEC2013、CEC2014和CEC2017测试套件中的88个基准函数上进行了评估。统计分析，包括Friedman检验和Wilcoxon秩和检验，证明了它与十种最先进的算法相比的有效性和竞争力。并将该方法应用于某40台发电机组电力系统的经济负荷调度问题，取得了满意的结果。

{"title":"An adaptive differential evolution algorithm with exponential crossover based on a learning strategy within the difference vector","authors":"Junpeng Chen , Zhenyu Meng","doi":"10.1016/j.swevo.2025.102247","DOIUrl":"10.1016/j.swevo.2025.102247","url":null,"abstract":"<div><div>Most Differential Evolution (DE) researchers tend to adopt the binomial crossover operation in tackling optimization problems. However, we find that the DE variants using exponential crossover can also achieve superior performance to those using binomial crossover, as long as appropriate parameter control strategies are applied. Therefore, this paper proposes a new DE algorithm, an adaptive Differential Evolution algorithm with exponential crossover based on Learning Strategy within Difference vector (DLS-DE), to fill the gap in this field. The main contributions of this work are summarized as follows: First, a two-phase parameter control strategy is designed to regulate the scale factor <span><math><mi>F</mi></math></span> for balancing exploration and exploitation. In addition, considering the dispersion of effective parameter values, an adaptive <span><math><msub><mrow><mi>σ</mi></mrow><mrow><mi>F</mi></mrow></msub></math></span> strategy is proposed to adjust the sampling distribution and enhance parameter adaptability. Second, a differential vector learning strategy is developed to identify and incorporate promising difference vector information during an individual’s stagnation, enabling the search direction to adapt based on its past performance. Finally, the algorithm employs exponential crossover, where the crossover rate <span><math><mrow><mi>C</mi><mi>R</mi></mrow></math></span> is automatically generated, and a fitness-independent parameter weight update mechanism is adopted to mitigate premature convergence. The performance of DLS-DE is evaluated on 88 benchmark functions from the CEC2013, CEC2014, and CEC2017 test suites. Statistical analyses, including the Friedman test and the Wilcoxon rank-sum test, demonstrate its effectiveness and competitiveness compared with ten state-of-the-art algorithms. In addition, DLS-DE is applied to an Economic Load Dispatch (ELD) problem in a power system with 40 generating units, achieving satisfactory results.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"100 ","pages":"Article 102247"},"PeriodicalIF":8.5,"publicationDate":"2025-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145684934","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A target-guided adaptive path planning method for air-ground cross-domain unmanned swarm 空地跨域无人机群目标制导自适应路径规划方法

IF 8.5 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Swarm and Evolutionary Computation

Pub Date : 2025-12-03 DOI: 10.1016/j.swevo.2025.102220

Qiang Peng , Husheng Wu , Renjun Zhan , Jingyi Geng , Yuanqing Xia , Lining Xing

The air-ground cross-domain unmanned swarm is widely used in military reconnaissance, disaster rescue and other fields, but its path planning faces severe challenges such as heterogeneous swarm coordination and highly dynamic obstacle avoidance in complex dynamic environments. Aiming at these technical difficulties, this paper proposes a target-guided adaptive path planning method (TAPP) inspired by the running behaviour of wolf packs in nature. The method adopts a parallel graph search strategy, dynamically fuses target location information to guide individual path planning, and simulates the autonomous decision-making mechanism of individual wolves. A local avoidance mechanism based on dynamic prioritization is also introduced to avoid motion conflicts while ensuring the efficiency of the swarm task. In order to comprehensively evaluate the effectiveness of the method, a complex air-ground cross-domain cooperative simulation scenario is designed, and systematic comparison experiments between the proposed method and several advanced path planning algorithms are conducted. The experimental data show that the proposed method demonstrates significant advantages in computational efficiency, path optimization, and adaptability, especially when dealing with large-scale and highly dynamic unmanned swarm tasks. The proposed method provides a solution for unmanned swarm path planning, which has a good application prospect.

地空跨域无人群在军事侦察、灾害救援等领域有着广泛的应用，但其路径规划面临着复杂动态环境下异构群协调和高动态避障等严峻挑战。针对这些技术难点，本文提出了一种受自然界狼群奔跑行为启发的目标导向自适应路径规划方法（TAPP）。该方法采用并行图搜索策略，动态融合目标位置信息指导个体路径规划，模拟个体狼的自主决策机制。引入基于动态优先级的局部回避机制，在保证群任务效率的同时避免运动冲突。为了全面评价该方法的有效性，设计了复杂的空地跨域协同仿真场景，并与几种先进的路径规划算法进行了系统的对比实验。实验数据表明，该方法在计算效率、路径优化和适应性方面具有显著优势，特别是在处理大规模、高动态的无人群体任务时。该方法为无人驾驶群体路径规划提供了一种解决方案，具有良好的应用前景。

{"title":"A target-guided adaptive path planning method for air-ground cross-domain unmanned swarm","authors":"Qiang Peng , Husheng Wu , Renjun Zhan , Jingyi Geng , Yuanqing Xia , Lining Xing","doi":"10.1016/j.swevo.2025.102220","DOIUrl":"10.1016/j.swevo.2025.102220","url":null,"abstract":"<div><div>The air-ground cross-domain unmanned swarm is widely used in military reconnaissance, disaster rescue and other fields, but its path planning faces severe challenges such as heterogeneous swarm coordination and highly dynamic obstacle avoidance in complex dynamic environments. Aiming at these technical difficulties, this paper proposes a target-guided adaptive path planning method (TAPP) inspired by the running behaviour of wolf packs in nature. The method adopts a parallel graph search strategy, dynamically fuses target location information to guide individual path planning, and simulates the autonomous decision-making mechanism of individual wolves. A local avoidance mechanism based on dynamic prioritization is also introduced to avoid motion conflicts while ensuring the efficiency of the swarm task. In order to comprehensively evaluate the effectiveness of the method, a complex air-ground cross-domain cooperative simulation scenario is designed, and systematic comparison experiments between the proposed method and several advanced path planning algorithms are conducted. The experimental data show that the proposed method demonstrates significant advantages in computational efficiency, path optimization, and adaptability, especially when dealing with large-scale and highly dynamic unmanned swarm tasks. The proposed method provides a solution for unmanned swarm path planning, which has a good application prospect.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"100 ","pages":"Article 102220"},"PeriodicalIF":8.5,"publicationDate":"2025-12-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145684856","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

An Improved Tuna Swarm Optimization Algorithm based XGBOOST Classification Method for Food Risk Evaluation 基于改进金枪鱼群优化算法的XGBOOST食品风险评价分类方法

IF 8.5 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Swarm and Evolutionary Computation

Pub Date : 2025-12-03 DOI: 10.1016/j.swevo.2025.102249

Yongming Han , Fan Yu , Jiaxin Liu , Zhen Zhang , Bo Ma , Ling Wang , Zhiqiang Geng

Food safety is a major issue related to people's livelihood, which constantly affects human health. Currently, the effectiveness of many promising food safety risk evaluation methods is strongly depended on the static hyperparameter configuration due to complexity and diversity of hyperparameters. Therefore, this paper proposes a novel extreme gradient boosting (XGBOOST) classification method based on the improved tuna swarm optimization (ITSO) algorithm incorporating a dynamic adaptive mechanism to evaluate the food risk. The ITSO optimizes the initial population using an elite reverse learning strategy to prevent premature convergence. Meanwhile, a two-stage Nnonlinear function is designed to adjust the probability of spiral foraging strategy and parabolic foraging strategy, which makes the ITSO quickly explore the global search spatial in the early stage, and focus on local search spatial in the later stage. Furthermore, the Lévy Flight strategy is used to improve the abilities of searching and jumping out of the local optimum. Finally, the hybrid ITSO-XGBOOST is constructed, and the ITSO is used to adaptive optimizing the parameters of the XGBOOST to improve the food risk evaluation ability. Compared with traditional optimization algorithms, the ITSO demonstrates its superiority on seven different functions. And the experiments on a real-world multi-indicator dairy dataset verifies the proposed method. Specifically, compared with baselines, all evaluation indexes show the significant advantages of the ITSO-XGBOOST with the highest accuracy of 82.61 % and the F1 of 82.09 %, which fully verifies the stably and accurately accessing ability of the proposed method in the complex scenario of multiple indexes.

食品安全是关系民生的重大问题，不断影响着人类的健康。目前，由于超参数的复杂性和多样性，许多有前景的食品安全风险评价方法的有效性在很大程度上依赖于静态超参数配置。为此，本文提出了一种基于改进的金枪鱼群优化（ITSO）算法的极限梯度提升（XGBOOST）分类方法，并结合动态自适应机制对食品风险进行评估。ITSO使用精英逆向学习策略来优化初始种群，以防止过早收敛。同时，设计了一种两阶段非线性函数来调整螺旋觅食策略和抛物线觅食策略的概率，使ITSO在早期快速探索全局搜索空间，在后期专注于局部搜索空间。在此基础上，利用lsamvy Flight策略提高了搜索和跳出局部最优的能力。最后，构建了混合ITSO-XGBOOST，并利用ITSO对XGBOOST参数进行自适应优化，提高了食品风险评价能力。与传统的优化算法相比，ITSO在7个不同的函数上显示出其优越性。在实际多指标乳制品数据集上的实验验证了该方法的有效性。与基线相比，各评价指标均显示出ITSO-XGBOOST的显著优势，最高准确率为82.61%，F1为82.09%，充分验证了所提方法在多指标复杂场景下稳定、准确的获取能力。

{"title":"An Improved Tuna Swarm Optimization Algorithm based XGBOOST Classification Method for Food Risk Evaluation","authors":"Yongming Han , Fan Yu , Jiaxin Liu , Zhen Zhang , Bo Ma , Ling Wang , Zhiqiang Geng","doi":"10.1016/j.swevo.2025.102249","DOIUrl":"10.1016/j.swevo.2025.102249","url":null,"abstract":"<div><div>Food safety is a major issue related to people's livelihood, which constantly affects human health. Currently, the effectiveness of many promising food safety risk evaluation methods is strongly depended on the static hyperparameter configuration due to complexity and diversity of hyperparameters. Therefore, this paper proposes a novel extreme gradient boosting (XGBOOST) classification method based on the improved tuna swarm optimization (ITSO) algorithm incorporating a dynamic adaptive mechanism to evaluate the food risk. The ITSO optimizes the initial population using an elite reverse learning strategy to prevent premature convergence. Meanwhile, a two-stage Nnonlinear function is designed to adjust the probability of spiral foraging strategy and parabolic foraging strategy, which makes the ITSO quickly explore the global search spatial in the early stage, and focus on local search spatial in the later stage. Furthermore, the Lévy Flight strategy is used to improve the abilities of searching and jumping out of the local optimum. Finally, the hybrid ITSO-XGBOOST is constructed, and the ITSO is used to adaptive optimizing the parameters of the XGBOOST to improve the food risk evaluation ability. Compared with traditional optimization algorithms, the ITSO demonstrates its superiority on seven different functions. And the experiments on a real-world multi-indicator dairy dataset verifies the proposed method. Specifically, compared with baselines, all evaluation indexes show the significant advantages of the ITSO-XGBOOST with the highest accuracy of 82.61 % and the F1 of 82.09 %, which fully verifies the stably and accurately accessing ability of the proposed method in the complex scenario of multiple indexes.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"100 ","pages":"Article 102249"},"PeriodicalIF":8.5,"publicationDate":"2025-12-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145736996","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A variant of united multi-operator evolutionary algorithms with application to livestock feed ration optimization 联合多算子进化算法的一种变体及其在牲畜饲料配粮优化中的应用

IF 8.5 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Swarm and Evolutionary Computation

Pub Date : 2025-12-03 DOI: 10.1016/j.swevo.2025.102238

Libin Hong , Dongxu Zhang , Tianxiang Cui

The United Multi-Operator Evolutionary Algorithms (UMOEAs) demonstrate high performance in evolutionary computation due to their combination of various operators for Differential Evolution (DE) variants and local search mechanisms, and have evolved into three versions over the last decade. In this work, a novel variant of the UMOEAs is proposed, named UMOEAs-IV. UMOEAs-IV employs a novel calculation method for the scaling factor

F

, which generates dynamic values to adaptively control the step size of mutation operators, a mutation strategy with complementary operators that better balance exploration and exploitation, an Estimation-of-Distribution Algorithm (EDA) to learn the probabilistic distribution of promising individuals, and a stagnation strategy to help individuals escape local optima. UMOEAs-IV is compared with 15 recently proposed DE-based algorithms and tested on the IEEE Congress on Evolutionary Computation 2017 (CEC2017) benchmark functions, showing superior performance over all of them. It is also applied to livestock feed ration optimization for beef and dairy cattle, where it shows top performance in the experimental results. The source code of UMOEAs-IV is provided at https://github.com/microhard1999/CODES.

联合多算子进化算法（umoea）由于结合了差分进化（DE）变体和局部搜索机制的各种算子，在进化计算中表现出高性能，并且在过去十年中已经发展成三个版本。在这项工作中，提出了一种新的umoea变体，命名为umoea - iv。UMOEAs-IV采用了一种新颖的缩放因子F的计算方法，该方法生成动态值以自适应控制突变算子的步长，采用了一种具有互补算子的突变策略，更好地平衡了探索和开发，采用了一种分布估计算法（EDA）来学习有希望个体的概率分布，采用了一种停滞策略来帮助个体逃避局部最优。umoas - iv与最近提出的15种基于de的算法进行了比较，并在IEEE进化计算大会2017 （CEC2017）基准函数上进行了测试，显示出优于所有算法的性能。并将其应用于肉牛、奶牛的饲料配给量优化，在试验结果中表现出最佳性能。umoas - iv的源代码提供于https://github.com/microhard1999/CODES。

{"title":"A variant of united multi-operator evolutionary algorithms with application to livestock feed ration optimization","authors":"Libin Hong , Dongxu Zhang , Tianxiang Cui","doi":"10.1016/j.swevo.2025.102238","DOIUrl":"10.1016/j.swevo.2025.102238","url":null,"abstract":"<div><div>The United Multi-Operator Evolutionary Algorithms (UMOEAs) demonstrate high performance in evolutionary computation due to their combination of various operators for Differential Evolution (DE) variants and local search mechanisms, and have evolved into three versions over the last decade. In this work, a novel variant of the UMOEAs is proposed, named UMOEAs-IV. UMOEAs-IV employs a novel calculation method for the scaling factor <span><math><mi>F</mi></math></span>, which generates dynamic values to adaptively control the step size of mutation operators, a mutation strategy with complementary operators that better balance exploration and exploitation, an Estimation-of-Distribution Algorithm (EDA) to learn the probabilistic distribution of promising individuals, and a stagnation strategy to help individuals escape local optima. UMOEAs-IV is compared with 15 recently proposed DE-based algorithms and tested on the IEEE Congress on Evolutionary Computation 2017 (CEC2017) benchmark functions, showing superior performance over all of them. It is also applied to livestock feed ration optimization for beef and dairy cattle, where it shows top performance in the experimental results. The source code of UMOEAs-IV is provided at <span><span>https://github.com/microhard1999/CODES</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"100 ","pages":"Article 102238"},"PeriodicalIF":8.5,"publicationDate":"2025-12-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145684857","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Generating continuous multi-objective benchmark problems by the object-oriented method 用面向对象方法生成连续多目标基准问题

IF 8.5 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Swarm and Evolutionary Computation

Pub Date : 2025-12-02 DOI: 10.1016/j.swevo.2025.102242

Qingshan Tan , Changhe Li , Miqing Li , Shengxiang Yang

Designing multi-objective benchmark test functions is an important topic in the field of evolutionary multi-objective optimization because it can help researchers identify the strengths and weaknesses of algorithms and contribute to improving their performance. However, existing methods for constructing multi-objective test problems exhibit certain specific limitations, such as the homogeneous structure of objectives, regular shapes of the Pareto optimal sets, and so on. To address these issues, this paper proposes an object-oriented construction method that abstracts components of a multi-objective optimization problem’s fitness landscape into classes. By designing attributes of these components, such as size, shape, position, and quantity, diverse test classes can be generated. Through a combination of these attributes, test cases with varying difficulty levels and distinct characteristics can be constructed. Based on this generator, several novel features are displayed. Moreover, five classic multi-objective evolutionary algorithms are tested on them. The results show different behaviors of these algorithms on the constructed test cases. At the same time, these generated features pose significant challenges to the tested algorithms.

设计多目标基准测试函数是进化多目标优化领域的一个重要课题，因为它可以帮助研究人员识别算法的优缺点，并有助于提高算法的性能。然而，现有的构造多目标测试问题的方法存在一定的局限性，如目标的齐次结构、Pareto最优集的规则形状等。为了解决这些问题，本文提出了一种面向对象的构造方法，该方法将多目标优化问题的适应度景观的组成部分抽象为类。通过设计这些组件的属性，例如大小、形状、位置和数量，可以生成不同的测试类。通过这些属性的组合，可以构建具有不同难度级别和不同特征的测试用例。在此基础上，显示了一些新颖的特征。并对五种经典的多目标进化算法进行了测试。结果显示了这些算法在构建的测试用例上的不同行为。同时，这些生成的特征对测试算法提出了重大挑战。

{"title":"Generating continuous multi-objective benchmark problems by the object-oriented method","authors":"Qingshan Tan , Changhe Li , Miqing Li , Shengxiang Yang","doi":"10.1016/j.swevo.2025.102242","DOIUrl":"10.1016/j.swevo.2025.102242","url":null,"abstract":"<div><div>Designing multi-objective benchmark test functions is an important topic in the field of evolutionary multi-objective optimization because it can help researchers identify the strengths and weaknesses of algorithms and contribute to improving their performance. However, existing methods for constructing multi-objective test problems exhibit certain specific limitations, such as the homogeneous structure of objectives, regular shapes of the Pareto optimal sets, and so on. To address these issues, this paper proposes an object-oriented construction method that abstracts components of a multi-objective optimization problem’s fitness landscape into classes. By designing attributes of these components, such as size, shape, position, and quantity, diverse test classes can be generated. Through a combination of these attributes, test cases with varying difficulty levels and distinct characteristics can be constructed. Based on this generator, several novel features are displayed. Moreover, five classic multi-objective evolutionary algorithms are tested on them. The results show different behaviors of these algorithms on the constructed test cases. At the same time, these generated features pose significant challenges to the tested algorithms.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"100 ","pages":"Article 102242"},"PeriodicalIF":8.5,"publicationDate":"2025-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145684940","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

RR-SMO: A novel optimization algorithm for enhancing route efficiency in outdoor sports equipment RR-SMO：一种提高户外运动装备路线效率的新型优化算法

IF 8.5 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Swarm and Evolutionary Computation

Pub Date : 2025-12-01 DOI: 10.1016/j.swevo.2025.102225

Xinxin Zhou , Xiangzhi Wu , Mingwei Wang , Youzhi Sun

The rise in sedentary lifestyles of people has led to a range of health problems, illustrating the need for innovative solutions to promote physical activity. Intelligent sports equipment has become important in promoting physical activity by providing real-time feedback and personalized guidance. However, the effectiveness of such equipment is often limited by suboptimal route planning and user interface design. Traditional algorithms frequently experience premature convergence and limited route diversity, leading to suboptimal navigation paths. Therefore, we proposed a novel Random Replacement based Slime Mold Optimization (RR-SMO) algorithm with a human-centered interface design for route optimization in outdoor sports equipment. The Random Replacement Strategy (RRS) is introduced into the algorithm to selectively adjust the dimensions of the best solution. This strategy helps to avoid premature convergence, boosts convergence speed, and ensures a diverse population, which contributes to more accurate and varied route planning in intricate navigation scenarios. Simultaneously, the interactive interface design for route navigation is designed with clear iconography, optimized touch targets, and adaptive visual cues to improve user experience. The comprehensive experimental assessment of the RR-SMO algorithm against existing algorithms using diverse metrics revealed that the RR-SMO algorithm achieves a navigation efficiency of 98.75 %, waypoint accuracy of 99.32 %, convergence time of 15 seconds, algorithm robustness of 97.82 %, a response time of 112 milliseconds, and an error rate of 5 %. The RR-SMO algorithm surpassed baseline methods by identifying optimal waypoints effectively and provides a scalable, intelligent solution for outdoor sports navigation to support healthy lifestyles.

人们久坐不动的生活方式的增加导致了一系列健康问题，表明需要创新的解决方案来促进身体活动。智能运动设备通过提供实时反馈和个性化指导，在促进身体活动方面变得非常重要。然而，这种设备的有效性往往受到次优路线规划和用户界面设计的限制。传统算法往往存在过早收敛和路径多样性有限的问题，导致导航路径次优。为此，我们提出了一种新的基于随机替换的黏菌优化算法（RR-SMO），该算法具有以人为中心的界面设计，用于户外运动器材的路线优化。该算法引入随机替换策略（RRS），选择性地调整最优解的维数。该策略避免了过早收敛，提高了收敛速度，保证了种群的多样性，有助于在复杂的导航场景中更准确、更多样化地规划路线。同时，对路线导航的交互界面设计进行了清晰的图标设计、优化的触摸目标和自适应的视觉提示，以提高用户体验。采用多种指标对RR-SMO算法与现有算法进行综合实验评估，结果表明，RR-SMO算法的导航效率为98.75%，航点精度为99.32%，收敛时间为15秒，鲁棒性为97.82%，响应时间为112毫秒，错误率为5%。RR-SMO算法通过有效地识别最佳路径点，超越了基线方法，为户外运动导航提供了可扩展的智能解决方案，以支持健康的生活方式。

{"title":"RR-SMO: A novel optimization algorithm for enhancing route efficiency in outdoor sports equipment","authors":"Xinxin Zhou , Xiangzhi Wu , Mingwei Wang , Youzhi Sun","doi":"10.1016/j.swevo.2025.102225","DOIUrl":"10.1016/j.swevo.2025.102225","url":null,"abstract":"<div><div>The rise in sedentary lifestyles of people has led to a range of health problems, illustrating the need for innovative solutions to promote physical activity. Intelligent sports equipment has become important in promoting physical activity by providing real-time feedback and personalized guidance. However, the effectiveness of such equipment is often limited by suboptimal route planning and user interface design. Traditional algorithms frequently experience premature convergence and limited route diversity, leading to suboptimal navigation paths. Therefore, we proposed a novel Random Replacement based Slime Mold Optimization (RR-SMO) algorithm with a human-centered interface design for route optimization in outdoor sports equipment. The Random Replacement Strategy (RRS) is introduced into the algorithm to selectively adjust the dimensions of the best solution. This strategy helps to avoid premature convergence, boosts convergence speed, and ensures a diverse population, which contributes to more accurate and varied route planning in intricate navigation scenarios. Simultaneously, the interactive interface design for route navigation is designed with clear iconography, optimized touch targets, and adaptive visual cues to improve user experience. The comprehensive experimental assessment of the RR-SMO algorithm against existing algorithms using diverse metrics revealed that the RR-SMO algorithm achieves a navigation efficiency of 98.75 %, waypoint accuracy of 99.32 %, convergence time of 15 seconds, algorithm robustness of 97.82 %, a response time of 112 milliseconds, and an error rate of 5 %. The RR-SMO algorithm surpassed baseline methods by identifying optimal waypoints effectively and provides a scalable, intelligent solution for outdoor sports navigation to support healthy lifestyles.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"100 ","pages":"Article 102225"},"PeriodicalIF":8.5,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145684938","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Evolutionary reinforcement learning with density-based behavioral diversity enhancement for berth allocation and crane assignment 基于密度行为多样性增强的进化强化学习在泊位分配和起重机分配中的应用

IF 8.5 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Swarm and Evolutionary Computation

Pub Date : 2025-12-01 DOI: 10.1016/j.swevo.2025.102241

Shengchang Li , Peilan Xu , Zhenglong Ding , Ziqian Kong , Wenjian Luo

The berth allocation and crane assignment problems (BACAPs) present significant challenges in marine transportation. While various mathematical solvers and heuristic algorithms exist for addressing these issues, they often suffer from time-consuming planning processes in complex and uncertain environments. Despite the introduction of reinforcement learning (RL), these studies typically focus solely on discrete berth allocation problems, disregarding crane assignment considerations. To facilitate an RL-based solution, this paper formulates a mathematical model for the continuous BACAP to define the decision environment. The model incorporates consecutive vessel berth positions, time-variant quay crane (QC) assignment, the maximum number of QCs per vessel, the cross-movement of QCs, and the safe distance between vessels. Subsequently, we propose an evolutionary reinforcement learning (ERL) framework as a novel solution to complex BACAPs. We establish an event-triggered Markov decision process (ET-MDP) and devise a feasibility-aware policy network using gated recurrent units (GRUs) and attention mechanisms. Additionally, we propose an evolution strategy based on density-based behavioral diversity enhancement (ES-DDE) as an optimizer for the policy network. This mechanism employs a niching strategy to construct multiple sub-populations and establish density landscapes. It drives each sub-population to explore low-density areas to increase the agent’s behavioral diversity and mitigate bias induced by deceptive rewards. Finally, we evaluate the performance of the proposed ERL-DDE on a set of 15 BACAP cases. Comparative analysis against several state-of-the-art heuristics and RL algorithms demonstrates the superior performance of ERL-DDE.

泊位分配和起重机分配问题（BACAPs）是海洋运输中的一个重大挑战。虽然存在各种数学求解器和启发式算法来解决这些问题，但它们通常在复杂和不确定的环境中受到耗时的规划过程的影响。尽管引入了强化学习（RL），但这些研究通常只关注离散泊位分配问题，而忽略了起重机分配的考虑。为了便于基于强化学习的解决方案，本文为连续BACAP制定了一个数学模型来定义决策环境。该模型考虑了船舶连续泊位、时变码头起重机分配、每艘船舶最大码头起重机数量、码头起重机的交叉移动以及船舶之间的安全距离。随后，我们提出了一种进化强化学习（ERL）框架作为复杂bacap的新解决方案。我们建立了一个事件触发的马尔可夫决策过程（ET-MDP），并使用门控循环单元（gru）和注意机制设计了一个可行性感知策略网络。此外，我们提出了一种基于基于密度的行为多样性增强（ES-DDE）的进化策略作为策略网络的优化器。该机制采用生态位策略来构建多个亚种群并建立密度景观。它驱动每个子种群探索低密度区域，以增加代理的行为多样性，并减轻由欺骗性奖励引起的偏见。最后，我们在一组15个BACAP案例中评估了所提出的ERL-DDE的性能。通过与几种最先进的启发式算法和强化学习算法的比较分析，证明了ERL-DDE的优越性能。

{"title":"Evolutionary reinforcement learning with density-based behavioral diversity enhancement for berth allocation and crane assignment","authors":"Shengchang Li , Peilan Xu , Zhenglong Ding , Ziqian Kong , Wenjian Luo","doi":"10.1016/j.swevo.2025.102241","DOIUrl":"10.1016/j.swevo.2025.102241","url":null,"abstract":"<div><div>The berth allocation and crane assignment problems (BACAPs) present significant challenges in marine transportation. While various mathematical solvers and heuristic algorithms exist for addressing these issues, they often suffer from time-consuming planning processes in complex and uncertain environments. Despite the introduction of reinforcement learning (RL), these studies typically focus solely on discrete berth allocation problems, disregarding crane assignment considerations. To facilitate an RL-based solution, this paper formulates a mathematical model for the continuous BACAP to define the decision environment. The model incorporates consecutive vessel berth positions, time-variant quay crane (QC) assignment, the maximum number of QCs per vessel, the cross-movement of QCs, and the safe distance between vessels. Subsequently, we propose an evolutionary reinforcement learning (ERL) framework as a novel solution to complex BACAPs. We establish an event-triggered Markov decision process (ET-MDP) and devise a feasibility-aware policy network using gated recurrent units (GRUs) and attention mechanisms. Additionally, we propose an evolution strategy based on density-based behavioral diversity enhancement (ES-DDE) as an optimizer for the policy network. This mechanism employs a niching strategy to construct multiple sub-populations and establish density landscapes. It drives each sub-population to explore low-density areas to increase the agent’s behavioral diversity and mitigate bias induced by deceptive rewards. Finally, we evaluate the performance of the proposed ERL-DDE on a set of 15 BACAP cases. Comparative analysis against several state-of-the-art heuristics and RL algorithms demonstrates the superior performance of ERL-DDE.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"100 ","pages":"Article 102241"},"PeriodicalIF":8.5,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145684939","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A hybrid genetic and variable neighborhood search algorithm for the hybrid flow-shop scheduling problem in pyrotechnic manufacturing industry with batch processing machines and lot-streaming 针对具有批量加工机器和批量流的烟火制造业混合流水车间调度问题，提出了一种混合遗传和可变邻域搜索算法

IF 8.5 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Swarm and Evolutionary Computation

Pub Date : 2025-11-29 DOI: 10.1016/j.swevo.2025.102234

Xiuli Wu, Ziqi Chen

Solid propellants play a critical role in national defense industries. The manufacturing process of solid rocket propellant is a complex process, in which materials of different components are mixed, stirred, casted, consolidated, demolded and shaped to form a solid propellant. Given the characteristics of lot-streaming and batch processing existed during manufacturing, the solid propellant production scheduling problem (SPPSP) can be defined as a hybrid flow-shop scheduling problem incorporating batch processing machines and lot-streaming. A mixed-integer linear programming (MILP) model with the optimization objective of minimizing makespan is established, and a hybrid genetic algorithm-variable neighborhood search (GA-VNS) algorithm is proposed to address this problem. To generate a high-quality initial population, an NEH-based initialization method is developed. Furthermore, considering the distinct characteristics of different production stages in SPPSP, a process-based neighborhood search strategy is introduced to enhance the efficiency of the local search process. Experimental results demonstrate that both the NEH-based initialization strategy and the process-based neighborhood search strategy significantly improve the algorithm's performance, confirming that the proposed algorithm can solve the SPPSP effectively.

固体推进剂在国防工业中起着至关重要的作用。固体火箭推进剂的制造过程是一个复杂的过程，在这个过程中，不同组分的材料经过混合、搅拌、铸造、固结、脱模和成型，形成固体推进剂。考虑到固体推进剂生产过程中存在批量加工和批量加工的特点，将固体推进剂生产调度问题定义为包含批量加工机器和批量加工机器的混合流水车间调度问题。建立了以最大完成时间最小化为优化目标的混合整数线性规划（MILP）模型，并提出了遗传算法-变量邻域搜索（GA-VNS）混合算法来解决该问题。为了生成高质量的初始种群，提出了一种基于neh的初始化方法。此外，考虑到SPPSP不同生产阶段的特点，引入了基于过程的邻域搜索策略，提高了局部搜索过程的效率。实验结果表明，基于neh的初始化策略和基于过程的邻域搜索策略均显著提高了算法的性能，验证了所提算法能够有效解决SPPSP问题。

{"title":"A hybrid genetic and variable neighborhood search algorithm for the hybrid flow-shop scheduling problem in pyrotechnic manufacturing industry with batch processing machines and lot-streaming","authors":"Xiuli Wu, Ziqi Chen","doi":"10.1016/j.swevo.2025.102234","DOIUrl":"10.1016/j.swevo.2025.102234","url":null,"abstract":"<div><div>Solid propellants play a critical role in national defense industries. The manufacturing process of solid rocket propellant is a complex process, in which materials of different components are mixed, stirred, casted, consolidated, demolded and shaped to form a solid propellant. Given the characteristics of lot-streaming and batch processing existed during manufacturing, the solid propellant production scheduling problem (SPPSP) can be defined as a hybrid flow-shop scheduling problem incorporating batch processing machines and lot-streaming. A mixed-integer linear programming (MILP) model with the optimization objective of minimizing makespan is established, and a hybrid genetic algorithm-variable neighborhood search (GA-VNS) algorithm is proposed to address this problem. To generate a high-quality initial population, an NEH-based initialization method is developed. Furthermore, considering the distinct characteristics of different production stages in SPPSP, a process-based neighborhood search strategy is introduced to enhance the efficiency of the local search process. Experimental results demonstrate that both the NEH-based initialization strategy and the process-based neighborhood search strategy significantly improve the algorithm's performance, confirming that the proposed algorithm can solve the SPPSP effectively.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"100 ","pages":"Article 102234"},"PeriodicalIF":8.5,"publicationDate":"2025-11-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145616107","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Evolutionary Neural Architecture Search for Super-Resolution: Benchmarking SynFlow and model-based predictors 超分辨率的进化神经架构搜索：基准SynFlow和基于模型的预测器

IF 8.5 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Swarm and Evolutionary Computation

Pub Date : 2025-11-29 DOI: 10.1016/j.swevo.2025.102236

Sergio Sarmiento-Rosales, Víctor Adrían Sosa Hernández, Raúl Monroy

Neural Architecture Search (NAS) has emerged as a powerful tool for designing deep learning networks, particularly in image classification tasks. However, its application to Super-Resolution (SR) remains computationally demanding due to the high cost of architecture evaluations. To address this challenge, this study proposed a multi-objective NAS approach for SR, comparing performance predictors. Specifically, SynFlow serves as a zero-shot approach, providing a baseline for rapid, training-free performance estimation. At the same time, model-based predictors such as Support Vector Regression, Extra Trees, and XGBoost provide learned estimations from a subset of partially trained architectures. We employ the Non-Dominated Sorting Genetic Algorithm-III (NSGA-III) to search for architectures that maximize Peak Signal-to-Noise Ratio (PSNR) while minimizing computational complexity, measured in terms of parameter number and FLOPs. Our findings reveal that model-based methods yield more accurate performance estimations, resulting in better-balanced and more evenly distributed solutions across the Pareto front. This study contributes to the advancement of efficient NAS methodologies for SR applications by highlighting the trade-offs between computational efficiency and predictive similarity in evaluation methods.

神经结构搜索（NAS）已经成为设计深度学习网络的强大工具，特别是在图像分类任务中。然而，由于架构评估的高成本，它在超分辨率（SR）中的应用仍然需要大量的计算。为了应对这一挑战，本研究提出了一种多目标NAS方法来比较SR的性能预测指标。具体来说，SynFlow可以作为一种零射击方法，为快速、无需训练的性能评估提供基线。与此同时，基于模型的预测器，如支持向量回归、额外树和XGBoost，从部分训练的架构子集中提供学习估计。我们采用非支配排序遗传算法- iii （NSGA-III）来搜索最大化峰值信噪比（PSNR）同时最小化计算复杂性的架构，以参数数量和FLOPs来衡量。我们的研究结果表明，基于模型的方法产生更准确的性能估计，从而在整个帕累托前沿产生更好的平衡和更均匀分布的解决方案。本研究通过强调评估方法中计算效率和预测相似性之间的权衡，有助于提高高效的NAS方法在SR应用中的应用。

{"title":"Evolutionary Neural Architecture Search for Super-Resolution: Benchmarking SynFlow and model-based predictors","authors":"Sergio Sarmiento-Rosales, Víctor Adrían Sosa Hernández, Raúl Monroy","doi":"10.1016/j.swevo.2025.102236","DOIUrl":"10.1016/j.swevo.2025.102236","url":null,"abstract":"<div><div>Neural Architecture Search (NAS) has emerged as a powerful tool for designing deep learning networks, particularly in image classification tasks. However, its application to Super-Resolution (SR) remains computationally demanding due to the high cost of architecture evaluations. To address this challenge, this study proposed a multi-objective NAS approach for SR, comparing performance predictors. Specifically, SynFlow serves as a zero-shot approach, providing a baseline for rapid, training-free performance estimation. At the same time, model-based predictors such as Support Vector Regression, Extra Trees, and XGBoost provide learned estimations from a subset of partially trained architectures. We employ the Non-Dominated Sorting Genetic Algorithm-III (NSGA-III) to search for architectures that maximize Peak Signal-to-Noise Ratio (PSNR) while minimizing computational complexity, measured in terms of parameter number and FLOPs. Our findings reveal that model-based methods yield more accurate performance estimations, resulting in better-balanced and more evenly distributed solutions across the Pareto front. This study contributes to the advancement of efficient NAS methodologies for SR applications by highlighting the trade-offs between computational efficiency and predictive similarity in evaluation methods.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"100 ","pages":"Article 102236"},"PeriodicalIF":8.5,"publicationDate":"2025-11-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145616108","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A novel encoding/decoding algorithm for constraint handling in a horizontal-axis wind turbine blade optimization process 一种新的水平轴风力机叶片优化过程约束处理编解码算法

IF 8.5 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Swarm and Evolutionary Computation

Pub Date : 2025-11-29 DOI: 10.1016/j.swevo.2025.102235

Krittatee Sangounsak, Sujin Bureerat, Akraphon Janon

The aim of this research is to improve the speed of the wind turbine blade optimization process while consuming less computation power. The traditional process using the constraint optimization technique and metaheuristics (MHs) cannot simultaneously handle many design variables, which is usually addressed by simplifying the problem. In this study, a novel encoding/decoding algorithm for constraint handling (ENDEACH) was constructed to handle all design variables prior to the application of MHs. The encoding process introduces the decision variable vector containing random numbers and boundaries for all design variables. Then the decoding process generates design variables from the vector. The technique produces design variables that are random while being within technical boundaries. Therefore, unfeasible search areas were eliminated. The proposed hybrid optimization process was applied to a 5-MW horizontal-axis wind turbine blade design. 53 design variables, including chord length, twist angle, chord distribution slope, twist distribution slope, and airfoil shape, were treated by ENDEACH. Then, several up-to-date MHs were used to optimize for the maximum power coefficient of wind turbines. The performance gains of the MHs were indicated by the highest convergence rate and minimum fitness value. Moreover, the optimal rotor had a power coefficient of up to 0.4987, 7.28 % higher than that of the National Renewable Energy Laboratory’s 5-MW reference wind turbine. The encoding/decoding algorithm together with MHs significantly outperformed the traditional method with respect to the optimization process and wind turbine blade performance, thereby enabling us to complete the optimization more quickly and obtain better results.

本研究的目的是提高风力机叶片优化过程的速度，同时减少计算量的消耗。传统的约束优化和元启发式方法不能同时处理多个设计变量，通常通过简化问题来解决这一问题。在本研究中，构建了一种新的约束处理编码/解码算法（ENDEACH）来处理mh应用前的所有设计变量。编码过程引入了包含随机数和所有设计变量边界的决策变量向量。然后解码过程从向量中生成设计变量。该技术在技术范围内产生随机的设计变量。因此，排除了不可行的搜索区域。将所提出的混合优化方法应用于5mw水平轴风力机叶片设计。采用ENDEACH对53个设计变量进行了处理，包括弦长、扭角、弦分布斜率、扭分布斜率和翼型形状。然后，利用几个最新的模型对风力发电机的最大功率系数进行了优化。最高的收敛速度和最小的适应度值表明了mhh的性能提升。最优转子功率系数高达0.4987，比国家可再生能源实验室5mw参考风力机功率系数提高7.28%。编/解码算法结合MHs在优化过程和风力机叶片性能上都明显优于传统方法，使我们能够更快地完成优化并获得更好的结果。

{"title":"A novel encoding/decoding algorithm for constraint handling in a horizontal-axis wind turbine blade optimization process","authors":"Krittatee Sangounsak, Sujin Bureerat, Akraphon Janon","doi":"10.1016/j.swevo.2025.102235","DOIUrl":"10.1016/j.swevo.2025.102235","url":null,"abstract":"<div><div>The aim of this research is to improve the speed of the wind turbine blade optimization process while consuming less computation power. The traditional process using the constraint optimization technique and metaheuristics (MHs) cannot simultaneously handle many design variables, which is usually addressed by simplifying the problem. In this study, a novel encoding/decoding algorithm for constraint handling (ENDEACH) was constructed to handle all design variables prior to the application of MHs. The encoding process introduces the decision variable vector containing random numbers and boundaries for all design variables. Then the decoding process generates design variables from the vector. The technique produces design variables that are random while being within technical boundaries. Therefore, unfeasible search areas were eliminated. The proposed hybrid optimization process was applied to a 5-MW horizontal-axis wind turbine blade design. 53 design variables, including chord length, twist angle, chord distribution slope, twist distribution slope, and airfoil shape, were treated by ENDEACH. Then, several up-to-date MHs were used to optimize for the maximum power coefficient of wind turbines. The performance gains of the MHs were indicated by the highest convergence rate and minimum fitness value. Moreover, the optimal rotor had a power coefficient of up to 0.4987, 7.28 % higher than that of the National Renewable Energy Laboratory’s 5-MW reference wind turbine. The encoding/decoding algorithm together with MHs significantly outperformed the traditional method with respect to the optimization process and wind turbine blade performance, thereby enabling us to complete the optimization more quickly and obtain better results.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"100 ","pages":"Article 102235"},"PeriodicalIF":8.5,"publicationDate":"2025-11-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145684937","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0