Applied Mathematical Modelling最新文献_第10页

Multi-strategy secretary bird optimization algorithm for threshold segmentation of cardiac CT images 心脏CT图像阈值分割的多策略秘书鸟优化算法

IF 4.4 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Applied Mathematical Modelling

Pub Date : 2026-06-01 Epub Date: 2025-12-14 DOI: 10.1016/j.apm.2025.116679

Zhaoqiang Wang , Hao Liu , Liangping Tu

The Secretary Bird Optimization Algorithm is a meta-heuristic algorithm based on the mathematical modeling of the predatory behavior of secretary birds toward snakes. Owing to its outstanding performance, it has been widely applied in various fields. However, the original algorithm also suffers from issues such as an imbalance between exploration and exploitation, as well as limited capability in handling complex optimization problems. Therefore, this study proposes a Multi-strategy Adaptive Hybrid Secretary Bird Optimization Algorithm to enhance its performance. This algorithm integrates four key strategies: the Levy-differential adaptive predation strategy, which combines Levy flight with differential evolution to enhance global exploration capabilities; the golden sine perturbation attack strategy, which uses golden sine guidance to balance exploration and exploitation; the adaptive sine step-size escape strategy, which dynamically adjusts escape probability and step size to prevent premature convergence; and the adaptive gene exchange mechanism, which enhances population diversity through random gene exchange. Experiments conducted on the benchmark test suites of CEC2022 and CEC2017 demonstrated that the proposed algorithm outperforms the original algorithm and other advanced algorithms in terms of convergence speed, solution quality, and stability. When combined with a multi-threshold segmentation model based on the Otsu method and applied to cardiac CT image segmentation after noise reduction, experiments showed that the proposed algorithm outperformed various comparison algorithms in terms of objective function values and metrics, such as peak signal-to-noise ratio, feature similarity index, and structural similarity index. The advantage becomes more pronounced as the number of thresholds increases, effectively validating the performance in this task.

秘书鸟优化算法是一种基于秘书鸟对蛇捕食行为数学建模的元启发式算法。由于其优异的性能，在各个领域得到了广泛的应用。然而，原有算法也存在着探索与开发不平衡、处理复杂优化问题能力有限等问题。为此，本文提出了一种多策略自适应混合秘书鸟优化算法来提高其性能。该算法集成了四个关键策略：Levy-差分自适应捕食策略，该策略将Levy飞行与差分进化相结合，增强了全局搜索能力；金正弦摄动攻击策略，利用金正弦导引平衡勘探与开发；自适应正弦步长逃逸策略，动态调整逃逸概率和步长，防止过早收敛；适应性基因交换机制，通过随机基因交换增强种群多样性。在CEC2022和CEC2017的基准测试套件上进行的实验表明，本文算法在收敛速度、解质量和稳定性方面都优于原始算法和其他先进算法。结合基于Otsu方法的多阈值分割模型，应用于降噪后的心脏CT图像分割，实验表明，本文算法在峰值信噪比、特征相似度、结构相似度等目标函数值和指标上均优于各种比较算法。随着阈值数量的增加，优势变得更加明显，从而有效地验证该任务中的性能。

{"title":"Multi-strategy secretary bird optimization algorithm for threshold segmentation of cardiac CT images","authors":"Zhaoqiang Wang , Hao Liu , Liangping Tu","doi":"10.1016/j.apm.2025.116679","DOIUrl":"10.1016/j.apm.2025.116679","url":null,"abstract":"<div><div>The Secretary Bird Optimization Algorithm is a meta-heuristic algorithm based on the mathematical modeling of the predatory behavior of secretary birds toward snakes. Owing to its outstanding performance, it has been widely applied in various fields. However, the original algorithm also suffers from issues such as an imbalance between exploration and exploitation, as well as limited capability in handling complex optimization problems. Therefore, this study proposes a Multi-strategy Adaptive Hybrid Secretary Bird Optimization Algorithm to enhance its performance. This algorithm integrates four key strategies: the Levy-differential adaptive predation strategy, which combines Levy flight with differential evolution to enhance global exploration capabilities; the golden sine perturbation attack strategy, which uses golden sine guidance to balance exploration and exploitation; the adaptive sine step-size escape strategy, which dynamically adjusts escape probability and step size to prevent premature convergence; and the adaptive gene exchange mechanism, which enhances population diversity through random gene exchange. Experiments conducted on the benchmark test suites of CEC2022 and CEC2017 demonstrated that the proposed algorithm outperforms the original algorithm and other advanced algorithms in terms of convergence speed, solution quality, and stability. When combined with a multi-threshold segmentation model based on the Otsu method and applied to cardiac CT image segmentation after noise reduction, experiments showed that the proposed algorithm outperformed various comparison algorithms in terms of objective function values and metrics, such as peak signal-to-noise ratio, feature similarity index, and structural similarity index. The advantage becomes more pronounced as the number of thresholds increases, effectively validating the performance in this task.</div></div>","PeriodicalId":50980,"journal":{"name":"Applied Mathematical Modelling","volume":"154 ","pages":"Article 116679"},"PeriodicalIF":4.4,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145753282","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Modeling and proactive security strategies for CAV platoon lane changing with optimal energy consumption 最优能耗CAV排变道建模及主动安全策略

IF 4.4 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Applied Mathematical Modelling

Pub Date : 2026-06-01 Epub Date: 2025-12-13 DOI: 10.1016/j.apm.2025.116685

Jinqi Zhang , Maode Yan , Lei Zuo , Tianwen Yan

This paper proposes a cooperative framework for connected and autonomous vehicle platoon lane changing with proactive security strategies and collaborative energy optimization during trajectory planning. Specifically, the feasible lane changing trajectory is modeled using a quintic polynomial approach. A dynamic safety-critical lane changing region is established to guarantee platoon safety, which can proactively adapt to real-time traffic and velocity variations. Within this safety-guaranteed region, a distributed energy optimization model is formulated, incorporating vehicle dynamics constraints (position, velocity, comfort) and environmental factors (wind resistance), which aims to minimize platoon energy consumption. The Karush-Kuhn-Tucker conditions are employed to solve the optimization problem, explicitly quantifying the impact of initial velocity, lane changing duration, and distance on energy efficiency. Numerical simulations across diverse traffic scenarios demonstrate that the proposed approach reduces energy consumption by 10–30 % compared to baseline methods (constant velocity offset sinusoidal function and polynomial of degree seven), while maintaining proactive collision avoidance.

提出了一种基于主动安全策略和轨迹规划协同能量优化的互联自动车辆排变道协同框架。具体而言，采用五次多项式方法对可行变道轨迹进行建模。建立动态安全关键变道区域，保证车队安全，能够主动适应实时交通和速度变化。在此安全保证区域内，结合车辆动力学约束（位置、速度、舒适性）和环境因素（风阻），建立分布式能量优化模型，以最小化车队能耗。采用Karush-Kuhn-Tucker条件求解优化问题，明确量化初始速度、变道时间和距离对能效的影响。不同交通场景的数值模拟表明，与基线方法（等速偏移正弦函数和7度多项式）相比，该方法可减少10-30%的能耗，同时保持主动避免碰撞。

{"title":"Modeling and proactive security strategies for CAV platoon lane changing with optimal energy consumption","authors":"Jinqi Zhang , Maode Yan , Lei Zuo , Tianwen Yan","doi":"10.1016/j.apm.2025.116685","DOIUrl":"10.1016/j.apm.2025.116685","url":null,"abstract":"<div><div>This paper proposes a cooperative framework for connected and autonomous vehicle platoon lane changing with proactive security strategies and collaborative energy optimization during trajectory planning. Specifically, the feasible lane changing trajectory is modeled using a quintic polynomial approach. A dynamic safety-critical lane changing region is established to guarantee platoon safety, which can proactively adapt to real-time traffic and velocity variations. Within this safety-guaranteed region, a distributed energy optimization model is formulated, incorporating vehicle dynamics constraints (position, velocity, comfort) and environmental factors (wind resistance), which aims to minimize platoon energy consumption. The Karush-Kuhn-Tucker conditions are employed to solve the optimization problem, explicitly quantifying the impact of initial velocity, lane changing duration, and distance on energy efficiency. Numerical simulations across diverse traffic scenarios demonstrate that the proposed approach reduces energy consumption by 10–30 % compared to baseline methods (constant velocity offset sinusoidal function and polynomial of degree seven), while maintaining proactive collision avoidance.</div></div>","PeriodicalId":50980,"journal":{"name":"Applied Mathematical Modelling","volume":"154 ","pages":"Article 116685"},"PeriodicalIF":4.4,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145753352","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

An innovative data-driven grey Bernoulli model for the port container throughput forecasting 基于数据驱动的港口集装箱吞吐量预测灰色伯努利模型

IF 4.4 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Applied Mathematical Modelling

Pub Date : 2026-06-01 Epub Date: 2025-12-15 DOI: 10.1016/j.apm.2025.116709

Lin Yang , Pan Yang , Xuemei Li , Yuhu Cui , Guangxue Wan

With the increasing importance of container transport in global trade and marine supply chain, it is crucial to make precise predictions of its throughput. From a data-driven perspective, this study proposes a dynamic adaptive structural prediction framework designed to identify time series patterns characterized by rising, declining, fluctuating, and stationary trends. Firstly, this framework improves the conventional grey Bernoulli model by introducing adaptive multi-weight coefficients and structural segmentation approach. Specifically, adaptive multi-weight coefficients prioritize recent data through dynamic weighting, enhancing the timeliness to optimize initial values. Furthermore, the structural segmentation approach is employed to optimize the background value so as to effectively mitigating the influence of inherent data patterns. Besides, a nonlinear parameter optimization approach for the novel model based on the Particle Swarm Optimization algorithm is proposed, with robustness test through ablation experiment analysis, sensitivity analysis, comparative analysis with other algorithms and different division interval. We verify the proposed model’s superiority by applying it to the port container throughput forecasting with mean absolute percentage errors of 0.8637 %, 0.2312 %, 4.9046 %, 8.5412 %, and 1.4073 %, respectively. Results reveal that the novel model demonstrates superior accuracy, flexibility, and broad applicability. Consequently, the novel model is deployed for container throughput predictions in 2024–2030. This study delivers a valuable tool for port decision makers to rationally plan layout, optimize resource allocation, and improve operational efficiency.

随着集装箱运输在全球贸易和海洋供应链中的重要性日益提高，对其吞吐量进行准确预测至关重要。从数据驱动的角度出发，本研究提出了一个动态自适应结构预测框架，旨在识别以上升、下降、波动和平稳趋势为特征的时间序列模式。首先，该框架通过引入自适应多权系数和结构分割方法对传统的灰色伯努利模型进行改进；具体而言，自适应多权系数通过动态加权对近期数据进行优先排序，增强了优化初始值的时效性。此外，采用结构分割方法对背景值进行优化，有效减轻了固有数据模式的影响。此外，提出了一种基于粒子群优化算法的新模型非线性参数优化方法，并通过烧蚀实验分析、灵敏度分析、与其他算法的对比分析和不同除法区间进行鲁棒性检验。将该模型应用于港口集装箱吞吐量预测，平均绝对百分比误差分别为0.8637%、0.2312%、4.9046%、8.5412%和1.4073%，验证了该模型的优越性。结果表明，该模型具有较高的准确性、灵活性和广泛的适用性。因此，该新模型可用于2024-2030年的集装箱吞吐量预测。本研究为港口决策者合理规划布局、优化资源配置、提高运营效率提供了有价值的工具。

{"title":"An innovative data-driven grey Bernoulli model for the port container throughput forecasting","authors":"Lin Yang , Pan Yang , Xuemei Li , Yuhu Cui , Guangxue Wan","doi":"10.1016/j.apm.2025.116709","DOIUrl":"10.1016/j.apm.2025.116709","url":null,"abstract":"<div><div>With the increasing importance of container transport in global trade and marine supply chain, it is crucial to make precise predictions of its throughput. From a data-driven perspective, this study proposes a dynamic adaptive structural prediction framework designed to identify time series patterns characterized by rising, declining, fluctuating, and stationary trends. Firstly, this framework improves the conventional grey Bernoulli model by introducing adaptive multi-weight coefficients and structural segmentation approach. Specifically, adaptive multi-weight coefficients prioritize recent data through dynamic weighting, enhancing the timeliness to optimize initial values. Furthermore, the structural segmentation approach is employed to optimize the background value so as to effectively mitigating the influence of inherent data patterns. Besides, a nonlinear parameter optimization approach for the novel model based on the Particle Swarm Optimization algorithm is proposed, with robustness test through ablation experiment analysis, sensitivity analysis, comparative analysis with other algorithms and different division interval. We verify the proposed model’s superiority by applying it to the port container throughput forecasting with mean absolute percentage errors of 0.8637 %, 0.2312 %, 4.9046 %, 8.5412 %, and 1.4073 %, respectively. Results reveal that the novel model demonstrates superior accuracy, flexibility, and broad applicability. Consequently, the novel model is deployed for container throughput predictions in 2024–2030. This study delivers a valuable tool for port decision makers to rationally plan layout, optimize resource allocation, and improve operational efficiency.</div></div>","PeriodicalId":50980,"journal":{"name":"Applied Mathematical Modelling","volume":"154 ","pages":"Article 116709"},"PeriodicalIF":4.4,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145753353","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Multi-class population maximum dispersion classifier for high-dimension low-sample size data 多维低样本量数据的多类总体最大离散分类器

IF 4.4 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Applied Mathematical Modelling

Pub Date : 2026-06-01 Epub Date: 2025-12-15 DOI: 10.1016/j.apm.2025.116693

Qingbo Yin , Sheng-Che Hung , Liran Shen

Classification of multi-class high-dimension and low-sample-size data sets poses significant challenges to many machine learning methods (or mathematical modelling and computational methods). To address this practical issue, there is an urgent need to develop new classification techniques. In this paper, we propose a new classifier, named Multi-Class No-seParated Data Maximum Dispersion Classifier. To mitigate data piling — a phenomenon of over-fitting, the intra-class difference is incorporated in the objective function. Our classifier aims to finds a projection space for each class, in which the samples of the corresponding class are scattered over a large interval, while the samples of the corresponding class and the samples of other classes are linearly separable. Additionally, the margin gap is maximized between the corresponding class and other classes. The main advantages of our classifier are as follows: (1) The objective function can be decomposed into subproblems, one for each class. (2)Each sub-problem of the objective function can be solved in the entire feature spaces to prevent the data-piling. (3)The samples of the corresponding class are only required when each sub-problem is solved, thereby the imbalance problem of multi-class data is avoided. (4) The computational complexity is linear in the number of classes. (5)The low-rank characteristic of the covariance matrix of high-dimension and low-sample-size data sets is leveraged to expedite computations. Experimental results and comprehensive discussions on 1 simulated data set and 11 real-world data sets (4 face image datasets and 7 gene expression datasets) demonstrate that our classifier exhibits superior performance with high efficiency.

多类高维和低样本数据集的分类对许多机器学习方法（或数学建模和计算方法）提出了重大挑战。为了解决这一实际问题，迫切需要开发新的分类技术。本文提出了一种新的分类器，即多类无分离数据最大离散分类器。为了减轻数据堆积-一种过度拟合现象，在目标函数中加入了类内差异。我们的分类器的目标是为每个类找到一个投影空间，在这个投影空间中，对应类的样本分散在一个大的间隔上，而对应类的样本与其他类的样本是线性可分的。另外，相应的类和其他类之间的边际差是最大的。我们的分类器的主要优点如下：(1)目标函数可以分解成子问题，每个类一个子问题。(2)目标函数的每个子问题都可以在整个特征空间中求解，防止数据堆积。(3)求解每个子问题时只需要对应类的样本，从而避免了多类数据的不平衡问题。(4)计算复杂度在类数上呈线性关系。(5)利用高维、低样本数据集协方差矩阵的低秩特性，加快计算速度。在1个模拟数据集和11个真实数据集（4个人脸图像数据集和7个基因表达数据集）上的实验结果和综合讨论表明，我们的分类器性能优越，效率高。

{"title":"Multi-class population maximum dispersion classifier for high-dimension low-sample size data","authors":"Qingbo Yin , Sheng-Che Hung , Liran Shen","doi":"10.1016/j.apm.2025.116693","DOIUrl":"10.1016/j.apm.2025.116693","url":null,"abstract":"<div><div>Classification of multi-class high-dimension and low-sample-size data sets poses significant challenges to many machine learning methods (or mathematical modelling and computational methods). To address this practical issue, there is an urgent need to develop new classification techniques. In this paper, we propose a new classifier, named Multi-Class No-seParated Data Maximum Dispersion Classifier. To mitigate data piling — a phenomenon of over-fitting, the intra-class difference is incorporated in the objective function. Our classifier aims to finds a projection space for each class, in which the samples of the corresponding class are scattered over a large interval, while the samples of the corresponding class and the samples of other classes are linearly separable. Additionally, the margin gap is maximized between the corresponding class and other classes. The main advantages of our classifier are as follows: (1) The objective function can be decomposed into subproblems, one for each class. (2)Each sub-problem of the objective function can be solved in the entire feature spaces to prevent the data-piling. (3)The samples of the corresponding class are only required when each sub-problem is solved, thereby the imbalance problem of multi-class data is avoided. (4) The computational complexity is linear in the number of classes. (5)The low-rank characteristic of the covariance matrix of high-dimension and low-sample-size data sets is leveraged to expedite computations. Experimental results and comprehensive discussions on 1 simulated data set and 11 real-world data sets (4 face image datasets and 7 gene expression datasets) demonstrate that our classifier exhibits superior performance with high efficiency.</div></div>","PeriodicalId":50980,"journal":{"name":"Applied Mathematical Modelling","volume":"154 ","pages":"Article 116693"},"PeriodicalIF":4.4,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145784738","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Approximate solutions to the shrinking core model and their interpretation 收缩核模型的近似解及其解释

IF 4.4 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Applied Mathematical Modelling

Pub Date : 2026-06-01 Epub Date: 2025-12-20 DOI: 10.1016/j.apm.2025.116715

Cristian Moreno-Pulido , Rachael Olwande , Tim Myers , Francesc Font

The shrinking core model (SCM) describes the reaction of a solid particle with a surrounding fluid. In this work, we revisit the SCM by deriving it from the underlying physical processes and performing a careful non-dimensionalisation, which highlights the limitations of the commonly used pseudo-steady-state approximation, particularly in liquid-solid systems where fluid and solid densities are comparable. To address these limitations, we derive approximate analytical solutions using a perturbation method that improves upon the pseudo-steady-state model. We also obtain a small-time solution capturing early transient behaviour. A semi-implicit finite difference scheme is implemented to solve the full model numerically and benchmark the analytical approximations. We demonstrate that the perturbation solution provides significantly improved accuracy over the pseudo-steady-state model, especially in diffusion-limited regimes. Finally, we propose a simple fitting procedure combining the perturbation with the early-time solutions to estimate physical parameters from experimental data at minimal computational cost.

收缩核模型（SCM）描述了固体颗粒与周围流体的反应。在这项工作中，我们通过从潜在的物理过程中推导SCM并进行仔细的无量纲化来重新审视SCM，这突出了常用的伪稳态近似的局限性，特别是在流体和固体密度可比较的液-固系统中。为了解决这些限制，我们使用一种改进了伪稳态模型的微扰方法推导出近似解析解。我们还获得了捕获早期瞬态行为的小时间解。采用半隐式有限差分格式对全模型进行数值求解，并对解析近似进行基准检验。我们证明了微扰解在伪稳态模型上提供了显著提高的精度，特别是在扩散受限的情况下。最后，我们提出了一个简单的拟合程序，将扰动与早期解结合起来，以最小的计算成本从实验数据中估计物理参数。

{"title":"Approximate solutions to the shrinking core model and their interpretation","authors":"Cristian Moreno-Pulido , Rachael Olwande , Tim Myers , Francesc Font","doi":"10.1016/j.apm.2025.116715","DOIUrl":"10.1016/j.apm.2025.116715","url":null,"abstract":"<div><div>The shrinking core model (SCM) describes the reaction of a solid particle with a surrounding fluid. In this work, we revisit the SCM by deriving it from the underlying physical processes and performing a careful non-dimensionalisation, which highlights the limitations of the commonly used pseudo-steady-state approximation, particularly in liquid-solid systems where fluid and solid densities are comparable. To address these limitations, we derive approximate analytical solutions using a perturbation method that improves upon the pseudo-steady-state model. We also obtain a small-time solution capturing early transient behaviour. A semi-implicit finite difference scheme is implemented to solve the full model numerically and benchmark the analytical approximations. We demonstrate that the perturbation solution provides significantly improved accuracy over the pseudo-steady-state model, especially in diffusion-limited regimes. Finally, we propose a simple fitting procedure combining the perturbation with the early-time solutions to estimate physical parameters from experimental data at minimal computational cost.</div></div>","PeriodicalId":50980,"journal":{"name":"Applied Mathematical Modelling","volume":"154 ","pages":"Article 116715"},"PeriodicalIF":4.4,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145796141","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Receptor-driven cycloidal optimization: Integrating biological control models with geometric search trajectories for engineering design optimization 受体驱动摆线优化：集成生物控制模型与几何搜索轨迹的工程设计优化

IF 4.4 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Applied Mathematical Modelling

Pub Date : 2026-06-01 Epub Date: 2026-02-06 DOI: 10.1016/j.apm.2026.116827

Lin Yang, Hongwen Xu

The optimization of complex engineering systems presents persistent computational challenges involving high-dimensional spaces, nonlinear objectives, and multiple local optima. Metaheuristic algorithms have become essential tools for such problems, with their effectiveness fundamentally dependent on the balance between exploration and exploitation. However, many existing approaches employ control mechanisms with limited responsiveness to search state changes and operators lacking functional specialization. This paper introduces the Receptor-Driven Cycloidal Optimization algorithm to address these limitations. The algorithm integrates a Hill-equation-based control mechanism that enables threshold-sensitive behavioral switching, with cycloidal operators where bounded hypocycloids serve exploitation and expansive epicycloids serve exploration. Experimental evaluation on CEC2017 and CEC2022 benchmarks demonstrates competitive performance, with advantages most evident on multimodal and composition functions. Engineering design applications validate practical utility. The source code is available at: https://github.com/yanggit0624/RDCO.

复杂工程系统的优化提出了持续的计算挑战，涉及高维空间、非线性目标和多个局部最优。元启发式算法已成为解决此类问题的重要工具，其有效性从根本上取决于探索与开发之间的平衡。然而，许多现有方法使用的控制机制对搜索状态变化的响应有限，操作符缺乏功能专门化。本文介绍了受体驱动摆线优化算法来解决这些限制。该算法集成了基于hill方程的控制机制，实现阈值敏感的行为切换，以及摆线算子，其中有界的次摆线用于开发，膨胀的外摆线用于勘探。在CEC2017和CEC2022基准上的实验评估显示出具有竞争力的性能，在多模态和组合功能上优势最为明显。工程设计应用验证了实际效用。源代码可从https://github.com/yanggit0624/RDCO获得。

{"title":"Receptor-driven cycloidal optimization: Integrating biological control models with geometric search trajectories for engineering design optimization","authors":"Lin Yang, Hongwen Xu","doi":"10.1016/j.apm.2026.116827","DOIUrl":"10.1016/j.apm.2026.116827","url":null,"abstract":"<div><div>The optimization of complex engineering systems presents persistent computational challenges involving high-dimensional spaces, nonlinear objectives, and multiple local optima. Metaheuristic algorithms have become essential tools for such problems, with their effectiveness fundamentally dependent on the balance between exploration and exploitation. However, many existing approaches employ control mechanisms with limited responsiveness to search state changes and operators lacking functional specialization. This paper introduces the Receptor-Driven Cycloidal Optimization algorithm to address these limitations. The algorithm integrates a Hill-equation-based control mechanism that enables threshold-sensitive behavioral switching, with cycloidal operators where bounded hypocycloids serve exploitation and expansive epicycloids serve exploration. Experimental evaluation on CEC2017 and CEC2022 benchmarks demonstrates competitive performance, with advantages most evident on multimodal and composition functions. Engineering design applications validate practical utility. The source code is available at: <span><span>https://github.com/yanggit0624/RDCO</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50980,"journal":{"name":"Applied Mathematical Modelling","volume":"154 ","pages":"Article 116827"},"PeriodicalIF":4.4,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146134725","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Unified strain-gradient frameworks for micro/nano-beams: Bridging the gap between strain-gradient and stress-driven nonlocal theories 微/纳米梁的统一应变梯度框架：弥合应变梯度和应力驱动非局部理论之间的差距

IF 4.4 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Applied Mathematical Modelling

Pub Date : 2026-06-01 Epub Date: 2026-02-04 DOI: 10.1016/j.apm.2026.116810

S.K. Lai , Z.W. Song

Although both stress-driven nonlocal theory (SDNT) and strain-gradient theory (SGT) can capture the stiffness hardening effect of micro/nano-scale beams and share identical governing equations, it remains uncertain whether these two theories can be unified within a single theoretical framework. In this work, we introduce virtual higher-order springs into the conventional SGTs, enabling unified strain-gradient frameworks that represent the differential forms of SDNT and SGT for Euler and Timoshenko micro/nano-scale beams. Furthermore, the constitutive boundary conditions in SDNT and the higher-order boundary conditions in SGT can be unified as higher-order elastic boundary conditions through virtual higher-order springs. Corresponding integral models of these unified strain-gradient frameworks are also developed to provide integral formulations for SDNT and SGT. The similarities and differences between them are systematically analyzed. More importantly, this study is the first to propose unified strain-gradient Euler and Timoshenko beam frameworks for both SDNT and SGT, which can elucidate their relationship and fill the knowledge gap in nonlocal modeling.

虽然应力驱动非局部理论（SDNT）和应变梯度理论（SGT）都可以捕捉微/纳米尺度梁的刚度硬化效应，并具有相同的控制方程，但这两种理论能否统一在一个理论框架内尚不确定。在这项工作中，我们将虚拟高阶弹簧引入传统的SGT，实现统一的应变梯度框架，代表欧拉和Timoshenko微/纳米尺度梁的SDNT和SGT的不同形式。此外，SDNT的本构边界条件和SGT的高阶边界条件可以通过虚拟高阶弹簧统一为高阶弹性边界条件。建立了统一应变梯度框架的相应积分模型，给出了统一应变梯度框架与统一应变梯度框架的积分公式，并系统分析了两者的异同。更重要的是，本研究首次提出了SDNT和SGT的统一应变梯度Euler和Timoshenko梁框架，可以阐明它们之间的关系，填补了非局部建模的知识空白。

{"title":"Unified strain-gradient frameworks for micro/nano-beams: Bridging the gap between strain-gradient and stress-driven nonlocal theories","authors":"S.K. Lai , Z.W. Song","doi":"10.1016/j.apm.2026.116810","DOIUrl":"10.1016/j.apm.2026.116810","url":null,"abstract":"<div><div>Although both stress-driven nonlocal theory (SDNT) and strain-gradient theory (SGT) can capture the stiffness hardening effect of micro/nano-scale beams and share identical governing equations, it remains uncertain whether these two theories can be unified within a single theoretical framework. In this work, we introduce virtual higher-order springs into the conventional SGTs, enabling unified strain-gradient frameworks that represent the differential forms of SDNT and SGT for Euler and Timoshenko micro/nano-scale beams. Furthermore, the constitutive boundary conditions in SDNT and the higher-order boundary conditions in SGT can be unified as higher-order elastic boundary conditions through virtual higher-order springs. Corresponding integral models of these unified strain-gradient frameworks are also developed to provide integral formulations for SDNT and SGT. The similarities and differences between them are systematically analyzed. More importantly, this study is the first to propose unified strain-gradient Euler and Timoshenko beam frameworks for both SDNT and SGT, which can elucidate their relationship and fill the knowledge gap in nonlocal modeling.</div></div>","PeriodicalId":50980,"journal":{"name":"Applied Mathematical Modelling","volume":"154 ","pages":"Article 116810"},"PeriodicalIF":4.4,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146134743","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Debye screening in surface waves in microstructured semiconductors with Drude-modelled carrier transport: A Robin boundary generalization 微结构半导体表面波的德拜筛选与德鲁德模型载流子输运：罗宾边界推广

IF 4.4 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Applied Mathematical Modelling

Pub Date : 2026-06-01 Epub Date: 2025-12-17 DOI: 10.1016/j.apm.2025.116702

Mahargha Biswas

This work develops a comprehensive theoretical model for Rayleigh-type surface waves in a semi-infinite n-type piezoelectric-flexoelectric semiconductor subjected to a static magnetic field. Two previously underexplored mechanisms are emphasized: Debye-Hückel electrostatic screening and Hall current-induced magnetoelectric anisotropy. The intrinsic inhomogeneity introduced by spatially non-uniform carrier distributions near the surface arising from doping effects is captured through a nontrivial Debye screening layer, which needs a shift from classical electrostatic boundary assumptions. Accordingly, generalized Robin-type boundary conditions are derived, synthesizing the limiting cases of the Dirichlet and Neumann formulations (corresponding to electrically short and open cases). The Hall effect is incorporated through a modified current continuity relation, based on the Drude model, wherein Lorentz-force-mediated charge carrier deflection introduces off-diagonal modifications drift-diffusion tensors. The governing system of Rayleigh wave is constructed from the fully coupled magneto-flexoelectro-elastic field equations, wherein flexoelectricity and strain-gradient elasticity introduce size-dependent electromechanical coupling. Dispersion relations are derived analytically in the long-wavelength limit and numerically solved for general cases, revealing a profound sensitivity of the Rayleigh wave phase velocity and attenuation to the interplay between the Debye screening scale, Hall mobility, and flexoelectric coefficients. The findings highlight the limitations of simplified boundary treatments in semiconductor acoustics and demonstrate multiple tunable wave effects resulting from multiscale interaction mechanisms. This framework provides a physically coherent and mathematically sound basis for designing and optimizing next-generation surface acoustic wave (SAW) devices.

本文建立了一个在静态磁场作用下半无限n型压电-柔性半导体中瑞利型表面波的综合理论模型。强调了两个以前未充分探索的机制：debye - h ckel静电筛选和霍尔电流诱导的磁电各向异性。由掺杂效应引起的表面附近空间非均匀载流子分布引入的本征不均匀性通过非平凡德拜屏蔽层捕获，这需要从经典静电边界假设中转变。因此，综合Dirichlet和Neumann公式的极限情况（对应于电短和开情况），导出了广义robin型边界条件。霍尔效应是通过基于德鲁德模型的修正电流连续性关系来实现的，其中洛伦兹力介导的载流子偏转引入了非对角线修正漂移扩散张量。瑞利波的控制系统是由完全耦合的磁-柔-电-弹场方程构成的，其中柔电和应变梯度弹性引入了尺寸相关的机电耦合。在长波长极限下解析推导了色散关系，并对一般情况进行了数值求解，揭示了瑞利波相速度和衰减对德拜筛分尺度、霍尔迁移率和挠曲电系数之间的相互作用具有深刻的敏感性。这些发现突出了半导体声学中简化边界处理的局限性，并展示了由多尺度相互作用机制产生的多重可调谐波效应。该框架为设计和优化下一代表面声波（SAW）器件提供了物理上连贯和数学上合理的基础。

{"title":"Debye screening in surface waves in microstructured semiconductors with Drude-modelled carrier transport: A Robin boundary generalization","authors":"Mahargha Biswas","doi":"10.1016/j.apm.2025.116702","DOIUrl":"10.1016/j.apm.2025.116702","url":null,"abstract":"<div><div>This work develops a comprehensive theoretical model for Rayleigh-type surface waves in a semi-infinite n-type piezoelectric-flexoelectric semiconductor subjected to a static magnetic field. Two previously underexplored mechanisms are emphasized: Debye-Hückel electrostatic screening and Hall current-induced magnetoelectric anisotropy. The intrinsic inhomogeneity introduced by spatially non-uniform carrier distributions near the surface arising from doping effects is captured through a nontrivial Debye screening layer, which needs a shift from classical electrostatic boundary assumptions. Accordingly, generalized Robin-type boundary conditions are derived, synthesizing the limiting cases of the Dirichlet and Neumann formulations (corresponding to electrically short and open cases). The Hall effect is incorporated through a modified current continuity relation, based on the Drude model, wherein Lorentz-force-mediated charge carrier deflection introduces off-diagonal modifications drift-diffusion tensors. The governing system of Rayleigh wave is constructed from the fully coupled magneto-flexoelectro-elastic field equations, wherein flexoelectricity and strain-gradient elasticity introduce size-dependent electromechanical coupling. Dispersion relations are derived analytically in the long-wavelength limit and numerically solved for general cases, revealing a profound sensitivity of the Rayleigh wave phase velocity and attenuation to the interplay between the Debye screening scale, Hall mobility, and flexoelectric coefficients. The findings highlight the limitations of simplified boundary treatments in semiconductor acoustics and demonstrate multiple tunable wave effects resulting from multiscale interaction mechanisms. This framework provides a physically coherent and mathematically sound basis for designing and optimizing next-generation surface acoustic wave (SAW) devices.</div></div>","PeriodicalId":50980,"journal":{"name":"Applied Mathematical Modelling","volume":"154 ","pages":"Article 116702"},"PeriodicalIF":4.4,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145784733","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Warehouse storage and retrieval optimization via clustering, dynamical systems modeling, and GPU-accelerated routing 基于聚类、动态系统建模和gpu加速路由的仓库存储和检索优化

IF 4.4 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Applied Mathematical Modelling

Pub Date : 2026-06-01 Epub Date: 2025-12-18 DOI: 10.1016/j.apm.2025.116700

Magnus Bengtsson, Jens Wittsten, Jonas Waidringer

This paper introduces a warehouse optimization procedure aimed at enhancing the efficiency of product storage and retrieval. By representing product locations and order flows within a time-evolving graph structure, we employ unsupervised clustering to define and refine compact order regions, effectively reducing picking distances. The procedure is formulated as a dynamic mathematical model using tools from random dynamical systems theory, enabling a principled analysis of the system’s behavior over time even under random operational variations. For routing within this framework, we implement a parallelized Bellman-Ford algorithm, utilizing GPU acceleration to evaluate path segments efficiently. To address scalability challenges inherent in large routing graphs, we introduce a segmentation strategy that preserves performance while maintaining tractable memory requirements. Our results demonstrate significant improvements in both operational efficiency and computational feasibility in large-scale warehouse environments.

本文介绍了一种仓库优化程序，旨在提高产品的存储和检索效率。通过在时间演化的图结构中表示产品位置和订单流，我们采用无监督聚类来定义和细化紧凑的订单区域，有效地减少了拣选距离。该程序使用随机动力系统理论的工具制定为动态数学模型，即使在随机操作变化下，也可以对系统随时间的行为进行原则性分析。对于该框架内的路由，我们实现了并行Bellman-Ford算法，利用GPU加速有效地评估路径段。为了解决大型路由图中固有的可伸缩性挑战，我们引入了一种分段策略，在保持性能的同时保持可处理的内存需求。我们的研究结果表明，在大规模仓库环境中，操作效率和计算可行性都有显著提高。

引用次数: 0

Incorporating energy expenditure and rest allowance in a human-robot collaborative learning-forgetting process 人-机器人协同学习遗忘过程中能量消耗和休息补偿的研究

IF 4.4 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Applied Mathematical Modelling

Pub Date : 2026-06-01 Epub Date: 2025-12-15 DOI: 10.1016/j.apm.2025.116711

Ali Keshvarparast , Mohamad Y. Jaber , Saeed Zolfaghari , Hamid Afshari

Collaborative robots (cobots) are being increasingly deployed in modern manufacturing to enhance flexibility, productivity, and safety. However, most existing production models overlook key human-centric factors, such as learning, forgetting, and fatigue, which significantly impact performance in human–robot collaboration (HRC) systems. Unlike manual environments, HRC requires precise synchronization between human adaptability and robotic consistency. Humans improve through learning, but forget over time. Meanwhile, fatigue accumulates, demanding rest that affects pacing and throughput. In parallel, physical fatigue accumulates, requiring rest periods that influence pacing and system throughput. This study, therefore, presents a Learning–Forgetting Fatigue–Recovery Model for Human–Robot Collaboration (LFFRM-HRC) that integrates human learning and forgetting dynamics with energy-based fatigue–recovery behavior within an optimization framework. The model evaluates how learning rate, number of batches, and forgetting duration affect total production time and average rest time across sixty simulation scenarios. Three system configurations are compared: manual assembly, HRC without task assignment, and a proposed HRC approach with dynamic task allocation. The results demonstrate that the proposed LFFRM-HRC with task assignment outperforms both manual and unstructured cobot setups, achieving substantial reductions in production time and rest needs. Moreover, the study shows that the optimal number of batches and shorter rest times mitigate the adverse effects of forgetting and fatigue. These findings necessitate including physical fatigue in system design, noting that cognitive fatigue, though important, is beyond the current scope. The LFFRM-HRC provides a practical, ergonomically sensitive tool for optimizing cobot-enabled production systems, supporting improved productivity while safeguarding worker well-being.

协作机器人（cobots）越来越多地应用于现代制造业，以提高灵活性、生产力和安全性。然而，大多数现有的生产模型忽略了以人为中心的关键因素，如学习、遗忘和疲劳，这些因素会显著影响人机协作（HRC）系统的性能。与人工环境不同，HRC需要在人的适应性和机器人的一致性之间进行精确的同步。人类通过学习而进步，但随着时间的推移却会遗忘。同时，疲劳积累，需要休息，影响步伐和吞吐量。同时，身体疲劳的积累，需要休息的时间，影响节奏和系统吞吐量。因此，本研究提出了一个人机协作的学习-遗忘疲劳-恢复模型（LFFRM-HRC），该模型将人类的学习和遗忘动力学与基于能量的疲劳-恢复行为整合在一个优化框架内。该模型评估学习率、批次数量和遗忘持续时间如何影响60个模拟场景中的总生产时间和平均休息时间。比较了手工装配、不分配任务的HRC和动态分配任务的HRC三种系统配置。结果表明，具有任务分配的LFFRM-HRC优于手动和非结构化协作机器人设置，实现了生产时间和休息需求的大幅减少。此外，研究表明，最佳批次数量和较短的休息时间可以减轻健忘和疲劳的不利影响。这些发现需要在系统设计中包括身体疲劳，注意到认知疲劳虽然重要，但超出了目前的范围。LFFRM-HRC提供了一种实用的、符合人体工程学的工具，用于优化协作机器人驱动的生产系统，在保障工人福祉的同时支持提高生产率。

{"title":"Incorporating energy expenditure and rest allowance in a human-robot collaborative learning-forgetting process","authors":"Ali Keshvarparast , Mohamad Y. Jaber , Saeed Zolfaghari , Hamid Afshari","doi":"10.1016/j.apm.2025.116711","DOIUrl":"10.1016/j.apm.2025.116711","url":null,"abstract":"<div><div>Collaborative robots (cobots) are being increasingly deployed in modern manufacturing to enhance flexibility, productivity, and safety. However, most existing production models overlook key human-centric factors, such as learning, forgetting, and fatigue, which significantly impact performance in human–robot collaboration (HRC) systems. Unlike manual environments, HRC requires precise synchronization between human adaptability and robotic consistency. Humans improve through learning, but forget over time. Meanwhile, fatigue accumulates, demanding rest that affects pacing and throughput. In parallel, physical fatigue accumulates, requiring rest periods that influence pacing and system throughput. This study, therefore, presents a Learning–Forgetting Fatigue–Recovery Model for Human–Robot Collaboration (LFFRM-HRC) that integrates human learning and forgetting dynamics with energy-based fatigue–recovery behavior within an optimization framework. The model evaluates how learning rate, number of batches, and forgetting duration affect total production time and average rest time across sixty simulation scenarios. Three system configurations are compared: manual assembly, HRC without task assignment, and a proposed HRC approach with dynamic task allocation. The results demonstrate that the proposed LFFRM-HRC with task assignment outperforms both manual and unstructured cobot setups, achieving substantial reductions in production time and rest needs. Moreover, the study shows that the optimal number of batches and shorter rest times mitigate the adverse effects of forgetting and fatigue. These findings necessitate including physical fatigue in system design, noting that cognitive fatigue, though important, is beyond the current scope. The LFFRM-HRC provides a practical, ergonomically sensitive tool for optimizing cobot-enabled production systems, supporting improved productivity while safeguarding worker well-being.</div></div>","PeriodicalId":50980,"journal":{"name":"Applied Mathematical Modelling","volume":"154 ","pages":"Article 116711"},"PeriodicalIF":4.4,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145784739","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0