Simulation Modelling Practice and Theory最新文献_第4页

Dynamic firefighting route planning for efficient evacuation in complex subway stations: A deep learning-enhanced robust optimization approach 复杂地铁站高效疏散的动态消防路线规划：一种深度学习增强的鲁棒优化方法

IF 3.5 2区计算机科学 Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Simulation Modelling Practice and Theory

Pub Date : 2025-11-07 DOI: 10.1016/j.simpat.2025.103223

Jinli Wei , Chunyue Cui , Xiaoxia Yang

The enclosed spaces and high-density population in subway stations significantly complicate evacuation during fires, thus increasing the difficulty of emergency response. To enhance fire rescue capabilities, this study conducts robust optimization modeling for firefighting routes from costs of station facility layout, passenger flow distribution, smoke propagation patterns, and human resource expenditure. Firstly, the BKA-GRU deep learning method is designed to calculate passenger passage time at critical nodes such as gates, improving the rationality of firefighting route design. Secondly, a firefighting value function based on the importance of fire nodes is constructed, making the firefighting routes more conducive to efficient and safe passenger evacuation. Thirdly, a box-based intersection polyhedron uncertainty set is employed to model the uncertainties in firefighting travel time and firefighting time, enhancing the adaptability and robustness of the routes. Fourthly, the advanced Ivy algorithm combined with Gurobi is adopted to solve the developed robust optimization model, enabling rapid identification of efficient and stable firefighting routes in complex environments. Finally, both quantitative and qualitative analyses are used to comprehensively evaluate firefighting effectiveness. The results indicate that: (i) The BKA-GRU prediction model exhibits high accuracy and reliability in predicting node passage time. (ii) The robust optimization model for firefighting routes significantly reduces fire by-products, shortens passenger evacuation time, and mitigates congestion. (iii) The firefighting route design achieves significant improvements in temperature control and visibility enhancement, effectively improving the fire environment and enhancing rescue efficiency and safety. This study provides an innovative solution for fire rescue in complex environments.

地铁车站空间封闭、人口密集，使火灾时的疏散变得更加复杂，增加了应急响应的难度。为了提高消防救援能力，本研究从车站设施布局成本、客流分布成本、烟雾传播方式成本和人力资源支出成本等方面对消防路线进行鲁棒优化建模。首先，设计了BKA-GRU深度学习方法，计算登机口等关键节点的旅客通过时间，提高消防路线设计的合理性；其次，构建基于火灾节点重要性的消防价值函数，使消防路线更有利于高效、安全的乘客疏散。第三，采用基于框的交叉口多面体不确定性集对消防行程时间和消防时间的不确定性进行建模，增强了路径的自适应性和鲁棒性；第四，采用先进的Ivy算法结合Gurobi算法对所建立的鲁棒优化模型进行求解，实现了在复杂环境下快速识别高效稳定的消防路线。最后，采用定量分析和定性分析相结合的方法对消防效能进行综合评价。结果表明：(1)BKA-GRU预测模型在预测节点通过时间方面具有较高的准确性和可靠性。（ii）稳健的消防路线优化模型显著减少了火灾副产物，缩短了乘客疏散时间，缓解了拥堵。（三）消防路线设计在温度控制和能见度增强方面有明显改善，有效改善了火灾环境，提高了救援效率和安全性。本研究为复杂环境下的火灾救援提供了一种创新的解决方案。

{"title":"Dynamic firefighting route planning for efficient evacuation in complex subway stations: A deep learning-enhanced robust optimization approach","authors":"Jinli Wei , Chunyue Cui , Xiaoxia Yang","doi":"10.1016/j.simpat.2025.103223","DOIUrl":"10.1016/j.simpat.2025.103223","url":null,"abstract":"<div><div>The enclosed spaces and high-density population in subway stations significantly complicate evacuation during fires, thus increasing the difficulty of emergency response. To enhance fire rescue capabilities, this study conducts robust optimization modeling for firefighting routes from costs of station facility layout, passenger flow distribution, smoke propagation patterns, and human resource expenditure. Firstly, the BKA-GRU deep learning method is designed to calculate passenger passage time at critical nodes such as gates, improving the rationality of firefighting route design. Secondly, a firefighting value function based on the importance of fire nodes is constructed, making the firefighting routes more conducive to efficient and safe passenger evacuation. Thirdly, a box-based intersection polyhedron uncertainty set is employed to model the uncertainties in firefighting travel time and firefighting time, enhancing the adaptability and robustness of the routes. Fourthly, the advanced Ivy algorithm combined with Gurobi is adopted to solve the developed robust optimization model, enabling rapid identification of efficient and stable firefighting routes in complex environments. Finally, both quantitative and qualitative analyses are used to comprehensively evaluate firefighting effectiveness. The results indicate that: (i) The BKA-GRU prediction model exhibits high accuracy and reliability in predicting node passage time. (ii) The robust optimization model for firefighting routes significantly reduces fire by-products, shortens passenger evacuation time, and mitigates congestion. (iii) The firefighting route design achieves significant improvements in temperature control and visibility enhancement, effectively improving the fire environment and enhancing rescue efficiency and safety. This study provides an innovative solution for fire rescue in complex environments.</div></div>","PeriodicalId":49518,"journal":{"name":"Simulation Modelling Practice and Theory","volume":"146 ","pages":"Article 103223"},"PeriodicalIF":3.5,"publicationDate":"2025-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145529299","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

SimEdgeAI: A deep reinforcement learning framework for simulating task offloading in resource-constrained IoT networks SimEdgeAI：用于模拟资源受限物联网网络中任务卸载的深度强化学习框架

IF 3.5 2区计算机科学 Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Simulation Modelling Practice and Theory

Pub Date : 2025-11-07 DOI: 10.1016/j.simpat.2025.103226

Waseem Abbass , Nasim Abbas , Uzma Majeed

The rapid growth of latency-sensitive Internet of Things (IoT) applications necessitates intelligent and scalable task offloading strategies in edge computing environments operating under dynamic workloads and limited energy resources. This paper introduces SimEdgeAI, a novel Deep Reinforcement Learning (DRL) framework that formulates task offloading as a stochastic decision-making problem over a multi-discrete action space, effectively capturing the trade-offs among local execution, edge offloading, and task dropping. The framework adopts an actor–critic architecture enhanced with a Gumbel–Softmax-based policy representation, enabling differentiable and stable learning over discrete actions. The actor network produces temperature-controlled stochastic policies, while the critic estimates long-term utilities based on system-wide features such as queue lengths, transmission delays, and energy states. A multi-objective reward function penalizing latency violations, excessive energy use, and fairness deviations guides the agent towards globally efficient and equitable offloading decisions. Extensive evaluations demonstrate that SimEdgeAI reduces average task latency by up to 35% and energy consumption by 25% compared to baseline methods including Deep Deterministic Policy Gradient (DDPG), Centralized DQN (C-DQN), and Greedy policies. It achieves over 91% deadline satisfaction and superior fairness measured by Jain’s index across edge clients. Ablation and sensitivity analyses confirm the contribution of each architectural component, while visualization studies underline the framework’s multi-objective consistency. These results highlight SimEdgeAI as an effective and fair solution for real-time, large-scale IoT–edge task offloading problems.

对延迟敏感的物联网（IoT）应用的快速增长需要在动态工作负载和有限能源下运行的边缘计算环境中采用智能和可扩展的任务卸载策略。本文介绍了SimEdgeAI，这是一个新颖的深度强化学习（DRL）框架，它将任务卸载作为一个多离散动作空间上的随机决策问题，有效地捕获了局部执行、边缘卸载和任务丢弃之间的权衡。该框架采用基于gumbel - softmax的策略表示增强的参与者-评论家体系结构，使离散行为的可微分和稳定学习成为可能。参与者网络产生温度控制的随机策略，而评论家则根据系统范围的特征（如队列长度、传输延迟和能量状态）估计长期效用。一个惩罚延迟违规、过度能源使用和公平性偏差的多目标奖励函数引导智能体做出全局高效和公平的卸载决策。广泛的评估表明，与包括深度确定性策略梯度（DDPG）、集中式DQN （C-DQN）和贪婪策略在内的基线方法相比，SimEdgeAI将平均任务延迟减少了35%，能耗减少了25%。它达到了超过91%的最后期限满意度和卓越的公平性，由Jain的跨边缘客户指数衡量。消融和敏感性分析证实了每个架构组件的贡献，而可视化研究强调了框架的多目标一致性。这些结果表明，SimEdgeAI是实时、大规模物联网边缘任务卸载问题的有效和公平的解决方案。

{"title":"SimEdgeAI: A deep reinforcement learning framework for simulating task offloading in resource-constrained IoT networks","authors":"Waseem Abbass , Nasim Abbas , Uzma Majeed","doi":"10.1016/j.simpat.2025.103226","DOIUrl":"10.1016/j.simpat.2025.103226","url":null,"abstract":"<div><div>The rapid growth of latency-sensitive Internet of Things (IoT) applications necessitates intelligent and scalable task offloading strategies in edge computing environments operating under dynamic workloads and limited energy resources. This paper introduces SimEdgeAI, a novel Deep Reinforcement Learning (DRL) framework that formulates task offloading as a stochastic decision-making problem over a multi-discrete action space, effectively capturing the trade-offs among local execution, edge offloading, and task dropping. The framework adopts an actor–critic architecture enhanced with a Gumbel–Softmax-based policy representation, enabling differentiable and stable learning over discrete actions. The actor network produces temperature-controlled stochastic policies, while the critic estimates long-term utilities based on system-wide features such as queue lengths, transmission delays, and energy states. A multi-objective reward function penalizing latency violations, excessive energy use, and fairness deviations guides the agent towards globally efficient and equitable offloading decisions. Extensive evaluations demonstrate that SimEdgeAI reduces average task latency by up to 35% and energy consumption by 25% compared to baseline methods including Deep Deterministic Policy Gradient (DDPG), Centralized DQN (C-DQN), and Greedy policies. It achieves over 91% deadline satisfaction and superior fairness measured by Jain’s index across edge clients. Ablation and sensitivity analyses confirm the contribution of each architectural component, while visualization studies underline the framework’s multi-objective consistency. These results highlight SimEdgeAI as an effective and fair solution for real-time, large-scale IoT–edge task offloading problems.</div></div>","PeriodicalId":49518,"journal":{"name":"Simulation Modelling Practice and Theory","volume":"146 ","pages":"Article 103226"},"PeriodicalIF":3.5,"publicationDate":"2025-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145529300","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

Ground surface settlements and deformation behavior of in-service high-speed railway tunnel induced by obliquely undercrossed TBM tunnelling 在役高速铁路隧道斜下穿隧道掘进引起的地表沉降及变形行为

IF 3.5 2区计算机科学 Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Simulation Modelling Practice and Theory

Pub Date : 2025-11-07 DOI: 10.1016/j.simpat.2025.103224

Jiajun Feng, Panpan Guo, Penghui Xue, Siyao Liu, Gan Wang, Yixian Wang

This study investigates ground-surface settlement and tunnel deformation induced by the construction of a TBM driven tunnel that obliquely undercrosses in-service high-speed railway tunnels. An analytical solution for predicting surface settlement is proposed by introducing the undercrossing angle and high-speed train load correction coefficients into the classical Peck formula. We validate the model’s applicability to oblique undercrossing with numerical simulations and field measurements. Building on these insights, we conduct three-dimensional finite-element (FE) modelling to quantify the effects of undercrossing angle (50°, 78°, 90°), tunnel clear distance (17.3, 13.3, 9.3 m), and excavation staging (10, 50, 100 steps) on surface settlement. The influence mechanism of train load on the deformation of the railway tunnel is analyzed. The results show that the proposed analytical solution improves surface-settlement prediction, keeping the error within 15 %. Specifically, larger undercrossing angles narrow the settlement trough and reduce the maximum settlement. Decreasing the clear distance from 17.3 to 9.3 m increases surface settlement by 65.96 %. Under train loading, surface settlement increases progressively with the number of TBM excavation steps. Train loading markedly amplifies overall tunnel deformation, increasing longitudinal deformation by 150 % and intensifying non-uniformity. The integrated analytical–numerical framework provides a practical basis for safety assessment and for optimising protective measures in similar undercrossing projects.

本文研究了在役高速铁路隧道斜下穿隧道掘进机施工引起的地表沉降和隧道变形。在经典Peck公式中引入下穿角和高速列车荷载修正系数，提出了地表沉降预测的解析解。通过数值模拟和现场实测验证了该模型对斜交下穿的适用性。基于这些见解，我们进行了三维有限元（FE）建模，以量化下穿角（50°，78°，90°），隧道清理距离（17.3,13.3,9.3 m）和开挖阶段（10,50,100步）对地表沉降的影响。分析了列车荷载对铁路隧道变形的影响机理。结果表明，所提出的解析解提高了地表沉降预测精度，误差控制在15%以内。具体而言，较大的下穿角使沉降槽变窄，减小了最大沉降。将净空距离从17.3 m减少到9.3 m，地表沉降增加65.96%。列车荷载作用下，地表沉降随掘进机开挖步数的增加而逐渐增大。列车荷载显著放大隧道整体变形，纵向变形增加150%，不均匀性加剧。该综合分析-数值框架为类似地下穿道桥的安全评价和防护措施优化提供了实践依据。

{"title":"Ground surface settlements and deformation behavior of in-service high-speed railway tunnel induced by obliquely undercrossed TBM tunnelling","authors":"Jiajun Feng, Panpan Guo, Penghui Xue, Siyao Liu, Gan Wang, Yixian Wang","doi":"10.1016/j.simpat.2025.103224","DOIUrl":"10.1016/j.simpat.2025.103224","url":null,"abstract":"<div><div>This study investigates ground-surface settlement and tunnel deformation induced by the construction of a TBM driven tunnel that obliquely undercrosses in-service high-speed railway tunnels. An analytical solution for predicting surface settlement is proposed by introducing the undercrossing angle and high-speed train load correction coefficients into the classical Peck formula. We validate the model’s applicability to oblique undercrossing with numerical simulations and field measurements. Building on these insights, we conduct three-dimensional finite-element (FE) modelling to quantify the effects of undercrossing angle (50°, 78°, 90°), tunnel clear distance (17.3, 13.3, 9.3 m), and excavation staging (10, 50, 100 steps) on surface settlement. The influence mechanism of train load on the deformation of the railway tunnel is analyzed. The results show that the proposed analytical solution improves surface-settlement prediction, keeping the error within 15 %. Specifically, larger undercrossing angles narrow the settlement trough and reduce the maximum settlement. Decreasing the clear distance from 17.3 to 9.3 m increases surface settlement by 65.96 %. Under train loading, surface settlement increases progressively with the number of TBM excavation steps. Train loading markedly amplifies overall tunnel deformation, increasing longitudinal deformation by 150 % and intensifying non<strong>-</strong>uniformity. The integrated analytical–numerical framework provides a practical basis for safety assessment and for optimising protective measures in similar undercrossing projects.</div></div>","PeriodicalId":49518,"journal":{"name":"Simulation Modelling Practice and Theory","volume":"146 ","pages":"Article 103224"},"PeriodicalIF":3.5,"publicationDate":"2025-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145529384","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

Enhancing 6G wireless performance through advanced MIMO techniques 通过先进的MIMO技术增强6G无线性能

IF 3.5 2区计算机科学 Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Simulation Modelling Practice and Theory

Pub Date : 2025-11-02 DOI: 10.1016/j.simpat.2025.103222

Arun Ananthanarayanan , S. Kanithan , Sathish Kumar Hari , Naeem Ahmed , Nadeem Pasha

To apply efficient beamforming, we need to be able to estimate channel state information (CSI) accurately. It is an essential factor that determines the success of high-data-rate, reliable communication in modern wireless networks. However, classic approaches tend to be inefficient in complex and fast-changing environments. This paper proposes a Deep Single-Carrier Orthogonal Frequency Division Multiplexing (DS-OFDM) to solve these difficulties. Division Multiplexing (Deep SCOFDM) framework, which incorporates Convolutional Neural End-to-End Long Short Term Memory (LSTM) & CNN networks for adaptive networks. Signal processing for 6 G systems. The proposed model simultaneously performs modulation and equalization, overcoming the drawbacks of standard OFDM systems — such as high PAPR and poor interference tolerance — by leveraging CNNs' spatial feature extraction and LSTMs' temporal feature extraction. The identifier can minimize signal degradation and increase symbol detection accuracy, as demonstrated by simulation results. In addition, it shows that the Deep SCOFDM framework exhibits lower PAPR with improved BER performance. Thus, our proposed approach outperforms other deep learning based MIMO and beamforming methods in terms of performance, faster convergence, and higher spectral efficiency. These findings suggest that the proposed approach is highly suitable for selecting intelligent and energy-efficient transceiver architectures in future 6 G networks.

为了实现高效的波束形成，我们需要能够准确地估计信道状态信息（CSI）。它是决定现代无线网络能否实现高数据速率、可靠通信的关键因素。然而，经典方法在复杂和快速变化的环境中往往效率低下。本文提出了一种深度单载波正交频分复用（DS-OFDM）技术来解决这些问题。分复用（Deep SCOFDM）框架，该框架结合了卷积神经端到端长短期记忆（LSTM）和CNN网络，用于自适应网络。6g系统的信号处理。该模型利用cnn的空间特征提取和LSTMs的时间特征提取，同时实现调制和均衡，克服了标准OFDM系统PAPR高、干扰容错性差的缺点。仿真结果表明，该标识符能最大限度地减少信号退化，提高符号检测精度。此外，深度SCOFDM框架具有较低的PAPR和较好的误码率性能。因此，我们提出的方法在性能、更快的收敛速度和更高的频谱效率方面优于其他基于深度学习的MIMO和波束形成方法。这些研究结果表明，该方法非常适合在未来的6g网络中选择智能和节能的收发器架构。

{"title":"Enhancing 6G wireless performance through advanced MIMO techniques","authors":"Arun Ananthanarayanan , S. Kanithan , Sathish Kumar Hari , Naeem Ahmed , Nadeem Pasha","doi":"10.1016/j.simpat.2025.103222","DOIUrl":"10.1016/j.simpat.2025.103222","url":null,"abstract":"<div><div>To apply efficient beamforming, we need to be able to estimate channel state information (CSI) accurately. It is an essential factor that determines the success of high-data-rate, reliable communication in modern wireless networks. However, classic approaches tend to be inefficient in complex and fast-changing environments. This paper proposes a Deep Single-Carrier Orthogonal Frequency Division Multiplexing (DS-OFDM) to solve these difficulties. Division Multiplexing (Deep SC<img>OFDM) framework, which incorporates Convolutional Neural End-to-End Long Short Term Memory (LSTM) & CNN networks for adaptive networks. Signal processing for 6 G systems. The proposed model simultaneously performs modulation and equalization, overcoming the drawbacks of standard OFDM systems — such as high PAPR and poor interference tolerance — by leveraging CNNs' spatial feature extraction and LSTMs' temporal feature extraction. The identifier can minimize signal degradation and increase symbol detection accuracy, as demonstrated by simulation results. In addition, it shows that the Deep SC<img>OFDM framework exhibits lower PAPR with improved BER performance. Thus, our proposed approach outperforms other deep learning based MIMO and beamforming methods in terms of performance, faster convergence, and higher spectral efficiency. These findings suggest that the proposed approach is highly suitable for selecting intelligent and energy-efficient transceiver architectures in future 6 G networks.</div></div>","PeriodicalId":49518,"journal":{"name":"Simulation Modelling Practice and Theory","volume":"146 ","pages":"Article 103222"},"PeriodicalIF":3.5,"publicationDate":"2025-11-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145623772","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

Experimental investigation and 3D finite element simulation of the turning process for AISI304 stainless steel AISI304不锈钢车削过程的实验研究及三维有限元模拟

IF 3.5 2区计算机科学 Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Simulation Modelling Practice and Theory

Pub Date : 2025-11-01 DOI: 10.1016/j.simpat.2025.103221

Hang Thi Pham , Tien-Thinh Le , Panagiotis G. Asteris

The aim of this study is to investigate the influence of main machining parameters during the turning process of AISI304 stainless steel through experimental work and finite element simulation employing a realistic 3D cylindrical workpiece model. The application of a realistic 3D model facilitates a strong correlation between simulation and experimental findings concerning chip morphology and temperature distribution. Both experimental and simulation results reveal the formation of elongated helical-shaped chips. An increase in cutting depth induces higher stress and equivalent plastic deformation. Meanwhile, a higher cutting speed leads to lower stress distributed on the chip and the machined workpiece. The temperature trends near the cutting tool nose and along the main cutting edge differ considerably. The highest temperature is concentrated on the main cutting edge of the cutting tool during the machining process, reaching up to 1000 °K in the case of high cutting speed and large cutting depth. In contrast, the temperature on the chip and machined surface are about 330 °K and 300 °K, respectively.

本研究通过实验工作和采用真实三维圆柱形工件模型的有限元模拟，探讨了主要加工参数对AISI304不锈钢车削加工过程的影响。逼真的三维模型的应用促进了芯片形态和温度分布的模拟和实验结果之间的强相关性。实验和仿真结果均显示了细长螺旋形切屑的形成。随着切削深度的增加，应力增大，等效塑性变形增大。同时，切削速度越快，切屑和被加工工件上的应力分布越小。靠近刀头和沿主切削刃的温度变化趋势差别很大。在加工过程中，最高温度集中在刀具的主切削刃处，在切削速度快、切削深度大的情况下，最高温度可达1000°K。相比之下，芯片和加工表面的温度分别约为330°K和300°K。

{"title":"Experimental investigation and 3D finite element simulation of the turning process for AISI304 stainless steel","authors":"Hang Thi Pham , Tien-Thinh Le , Panagiotis G. Asteris","doi":"10.1016/j.simpat.2025.103221","DOIUrl":"10.1016/j.simpat.2025.103221","url":null,"abstract":"<div><div>The aim of this study is to investigate the influence of main machining parameters during the turning process of AISI304 stainless steel through experimental work and finite element simulation employing a realistic 3D cylindrical workpiece model. The application of a realistic 3D model facilitates a strong correlation between simulation and experimental findings concerning chip morphology and temperature distribution. Both experimental and simulation results reveal the formation of elongated helical-shaped chips. An increase in cutting depth induces higher stress and equivalent plastic deformation. Meanwhile, a higher cutting speed leads to lower stress distributed on the chip and the machined workpiece. The temperature trends near the cutting tool nose and along the main cutting edge differ considerably. The highest temperature is concentrated on the main cutting edge of the cutting tool during the machining process, reaching up to 1000 °K in the case of high cutting speed and large cutting depth. In contrast, the temperature on the chip and machined surface are about 330 °K and 300 °K, respectively.</div></div>","PeriodicalId":49518,"journal":{"name":"Simulation Modelling Practice and Theory","volume":"146 ","pages":"Article 103221"},"PeriodicalIF":3.5,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145475187","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

Evolving realistic topologies for wireless mesh network simulation with EvoTopo 演化现实拓扑的无线网状网络仿真与EvoTopo

IF 3.5 2区计算机科学 Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Simulation Modelling Practice and Theory

Pub Date : 2025-10-30 DOI: 10.1016/j.simpat.2025.103217

Martina Umlauft, Wilfried Elmenreich, Udo Schilcher

The topology model is an important aspect of wireless mesh network simulation. To test a new protocol typically several different topologies are necessary to perform a statistically significant number of simulation runs. Simply using topologies of real-world networks directly is not sufficient because the number of topology data available is less than the number of topologies required for simulation. Therefore, artificially generated topologies are used. Unfortunately, many simulators use either a uniform node distribution or even just a simple grid topology which both differ significantly from real-world topologies.

We first revisit the differences between uniform and grid topologies vs. real-world topologies using four different real-world networks to motivate the need for such a tool and then present EvoTopo, a new approach to generate more realistic topologies. EvoTopo uses a genetic algorithm to create the desired number of simulation topologies from the node positions of one real-world network. The generated topologies are evolved to be “similar” w.r.t. homogeneity of the node distribution, nearest neighbor distance, and node density. We evaluate our algorithm analyzing average overall node distance, the degree distribution of nodes, and the performance of a simple flooding algorithm and compare our algorithm to other approaches. The EvoTopo tool and the four sample topologies can be downloaded from our homepage; generated topologies are written to a simple text file which can be imported into a simulator of choice.

拓扑模型是无线网状网络仿真的一个重要方面。为了测试一个新协议，通常需要几个不同的拓扑来执行统计上显著数量的模拟运行。简单地直接使用实际网络的拓扑是不够的，因为可用的拓扑数据数量少于仿真所需的拓扑数量。因此，使用人工生成的拓扑。不幸的是，许多模拟器要么使用统一的节点分布，要么甚至只是一个简单的网格拓扑，这两者都与现实世界的拓扑结构有很大的不同。我们首先使用四种不同的现实网络来回顾统一和网格拓扑与现实拓扑之间的差异，以激发对这种工具的需求，然后介绍EvoTopo，一种生成更现实拓扑的新方法。EvoTopo使用遗传算法从一个现实世界网络的节点位置创建所需数量的模拟拓扑。生成的拓扑被演化为“相似”的拓扑，即节点分布、最近邻距离和节点密度的均匀性。我们通过分析平均总体节点距离、节点度分布和简单泛洪算法的性能来评估我们的算法，并将我们的算法与其他方法进行比较。EvoTopo工具和四个拓扑样例可从我们的主页下载；生成的拓扑被写入一个简单的文本文件，该文件可以导入到所选择的模拟器中。

{"title":"Evolving realistic topologies for wireless mesh network simulation with EvoTopo","authors":"Martina Umlauft, Wilfried Elmenreich, Udo Schilcher","doi":"10.1016/j.simpat.2025.103217","DOIUrl":"10.1016/j.simpat.2025.103217","url":null,"abstract":"<div><div>The topology model is an important aspect of wireless mesh network simulation. To test a new protocol typically several different topologies are necessary to perform a statistically significant number of simulation runs. Simply using topologies of real-world networks directly is not sufficient because the number of topology data available is less than the number of topologies required for simulation. Therefore, artificially generated topologies are used. Unfortunately, many simulators use either a uniform node distribution or even just a simple grid topology which both differ significantly from real-world topologies.</div><div>We first revisit the differences between uniform and grid topologies vs. real-world topologies using four different real-world networks to motivate the need for such a tool and then present <span>EvoTopo</span>, a new approach to generate more realistic topologies. <span>EvoTopo</span> uses a genetic algorithm to create the desired number of simulation topologies from the node positions of one real-world network. The generated topologies are evolved to be “similar” w.r.t. homogeneity of the node distribution, nearest neighbor distance, and node density. We evaluate our algorithm analyzing average overall node distance, the degree distribution of nodes, and the performance of a simple flooding algorithm and compare our algorithm to other approaches. The <span>EvoTopo</span> tool and the four sample topologies can be downloaded from our homepage; generated topologies are written to a simple text file which can be imported into a simulator of choice.</div></div>","PeriodicalId":49518,"journal":{"name":"Simulation Modelling Practice and Theory","volume":"146 ","pages":"Article 103217"},"PeriodicalIF":3.5,"publicationDate":"2025-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145475186","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

Multi-agent reinforcement learning and variational inequality models for international trade networks under crisis 危机下国际贸易网络的多智能体强化学习和变分不等式模型

IF 3.5 2区计算机科学 Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Simulation Modelling Practice and Theory

Pub Date : 2025-10-29 DOI: 10.1016/j.simpat.2025.103219

Laura De Natale, Georgia Fargetta, Laura R.M. Scrimali, Sebastiano Battiato

Global disasters increasingly disrupt agricultural commodity flows, with food insecurity as a major consequence. Quantitative tools to assess such impacts are essential for resilience. We propose a hybrid Multi-Agent Reinforcement Learning (MARL) architecture to solve Variational Inequality (VI) problems in multi-commodity trade equilibria. While variational inequalities offer a rigorous method, their resolution via MARL faces stability and convergence challenges. Our actor–critic approach integrates a Gradient-based Learning Rate (GLR) scheduler, adaptive epsilon decay, prioritized replay, and a dual reward combining individual and centralized feedback. Agents representing supply and demand learn optimal strategies to reach equilibrium in simulated markets. Experiments, spanning stable conditions, dynamic price shifts, and route blockages, show faster convergence and stronger robustness than Multi-Agent Proximal Policy Optimization (MAPPO) and Multi-Agent Deep Deterministic Policy Gradient (MADDPG). Results highlight the promise of MARL for simulating economic behavior and optimizing decentralized decision-making in complex systems.

全球灾害日益扰乱农产品流通，粮食不安全是一个主要后果。评估这种影响的定量工具对复原力至关重要。我们提出了一种混合多智能体强化学习（MARL）架构来解决多商品贸易均衡中的变分不等式（VI）问题。虽然变分不等式提供了一种严格的方法，但通过MARL解决它们面临稳定性和收敛性的挑战。我们的参与者-评论家方法集成了基于梯度的学习率（GLR）调度程序、自适应epsilon衰减、优先级重播以及结合个人和集中反馈的双重奖励。代表供给和需求的代理在模拟市场中学习最优策略以达到均衡。实验表明，在稳定条件下，动态价格变化和路由阻塞的情况下，比多智能体近端策略优化（MAPPO）和多智能体深度确定性策略梯度（MADDPG）的收敛速度更快，鲁棒性更强。结果突出了MARL在模拟复杂系统中的经济行为和优化分散决策方面的前景。

{"title":"Multi-agent reinforcement learning and variational inequality models for international trade networks under crisis","authors":"Laura De Natale, Georgia Fargetta, Laura R.M. Scrimali, Sebastiano Battiato","doi":"10.1016/j.simpat.2025.103219","DOIUrl":"10.1016/j.simpat.2025.103219","url":null,"abstract":"<div><div>Global disasters increasingly disrupt agricultural commodity flows, with food insecurity as a major consequence. Quantitative tools to assess such impacts are essential for resilience. We propose a hybrid Multi-Agent Reinforcement Learning (MARL) architecture to solve Variational Inequality (VI) problems in multi-commodity trade equilibria. While variational inequalities offer a rigorous method, their resolution via MARL faces stability and convergence challenges. Our actor–critic approach integrates a Gradient-based Learning Rate (GLR) scheduler, adaptive epsilon decay, prioritized replay, and a dual reward combining individual and centralized feedback. Agents representing supply and demand learn optimal strategies to reach equilibrium in simulated markets. Experiments, spanning stable conditions, dynamic price shifts, and route blockages, show faster convergence and stronger robustness than Multi-Agent Proximal Policy Optimization (MAPPO) and Multi-Agent Deep Deterministic Policy Gradient (MADDPG). Results highlight the promise of MARL for simulating economic behavior and optimizing decentralized decision-making in complex systems.</div></div>","PeriodicalId":49518,"journal":{"name":"Simulation Modelling Practice and Theory","volume":"146 ","pages":"Article 103219"},"PeriodicalIF":3.5,"publicationDate":"2025-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145398229","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

DPNVC: A novel density-based probability VANET caching framework built upon the NDN DPNVC：一种基于NDN的基于密度的概率VANET缓存框架

IF 3.5 2区计算机科学 Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Simulation Modelling Practice and Theory

Pub Date : 2025-10-28 DOI: 10.1016/j.simpat.2025.103218

Yuanchen Li , Lin Guan , Ziyang Zhang , George Vogiatzis

Vehicular Ad Hoc Networks (VANETs) are an important component of modern network systems, supporting applications such as real-time entertainment, traffic notifications, and emergency services. However, the highly dynamic and rapidly changing topology of VANETs presents serious challenges for conventional data retrieval mechanisms designed for Mobile Ad Hoc Networks (MANETs), resulting in degraded performance. To address this issue, a novel Density-Based Probability VANET Caching Framework Built Upon the Named Data Networking (NDN) was proposed, namely DPNVC. This original framework dynamically calculates caching probabilities based on local traffic density, enabling to adapt to frequent topology changes. Additionally, the NDN communication model is applied to effectively suppress redundant packet forwarding in VANET environments. Empirical simulation results show that DPNVC significantly enhances Quality of Service (QoS) in various scenarios, including urban, highway, and city settings. Compared to baseline methods, it reduces link load by up to 25 %, decreases data retrieval time by up to 30 %, and improves the local satisfaction ratio by up to 66 %. It also maintains a competitive one-hop hit ratio performance.

车载自组织网络（vanet）是现代网络系统的重要组成部分，支持实时娱乐、交通通知和紧急服务等应用。然而，vanet的高度动态和快速变化的拓扑结构对传统的移动自组织网络（manet）数据检索机制提出了严峻的挑战，导致性能下降。为了解决这一问题，提出了一种基于命名数据网络（NDN）的基于密度的概率VANET缓存框架，即DPNVC。这个原始框架基于本地流量密度动态计算缓存概率，从而能够适应频繁的拓扑变化。此外，应用NDN通信模型有效地抑制了VANET环境下的冗余报文转发。实证仿真结果表明，DPNVC在城市、高速公路和城市等场景下显著提高了服务质量（QoS）。与基线方法相比，该方法最多可减少25%的链路负载，最多可减少30%的数据检索时间，并将本地满意度提高66%。它还保持了具有竞争力的单跳命中率性能。

{"title":"DPNVC: A novel density-based probability VANET caching framework built upon the NDN","authors":"Yuanchen Li , Lin Guan , Ziyang Zhang , George Vogiatzis","doi":"10.1016/j.simpat.2025.103218","DOIUrl":"10.1016/j.simpat.2025.103218","url":null,"abstract":"<div><div>Vehicular Ad Hoc Networks (VANETs) are an important component of modern network systems, supporting applications such as real-time entertainment, traffic notifications, and emergency services. However, the highly dynamic and rapidly changing topology of VANETs presents serious challenges for conventional data retrieval mechanisms designed for Mobile Ad Hoc Networks (MANETs), resulting in degraded performance. To address this issue, a novel Density-Based Probability VANET Caching Framework Built Upon the Named Data Networking (NDN) was proposed, namely DPNVC. This original framework dynamically calculates caching probabilities based on local traffic density, enabling to adapt to frequent topology changes. Additionally, the NDN communication model is applied to effectively suppress redundant packet forwarding in VANET environments. Empirical simulation results show that DPNVC significantly enhances Quality of Service (QoS) in various scenarios, including urban, highway, and city settings. Compared to baseline methods, it reduces link load by up to 25 %, decreases data retrieval time by up to 30 %, and improves the local satisfaction ratio by up to 66 %. It also maintains a competitive one-hop hit ratio performance.</div></div>","PeriodicalId":49518,"journal":{"name":"Simulation Modelling Practice and Theory","volume":"146 ","pages":"Article 103218"},"PeriodicalIF":3.5,"publicationDate":"2025-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145398230","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

Prediction of load time histories on building facades under urban explosion environment 城市爆炸环境下建筑立面荷载时程预测

IF 3.5 2区计算机科学 Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Simulation Modelling Practice and Theory

Pub Date : 2025-10-28 DOI: 10.1016/j.simpat.2025.103220

Doudou Si, Zuanfeng Pan, Wendi Li

The quantification of blast load on a building constitutes a pivotal determinant in blast-resistant structural engineering. When the explosion occurs in a complex urban environment, the blast load on the building surface is conventionally obtained through scaled explosion tests or computational fluid dynamics simulations. Given the significant expenses of explosion tests and numerical simulations, this study innovatively applies the long short-term memory (LSTM) network to predict the blast load time history in urban explosion scenarios. The main idea is to establish an LSTM network to learn the temporal and spatial variations of blast loads from limited training data and predict the time history of blast loads at unmonitored locations or unknown explosion scenarios. In particular, a unidirectional multi-layer stacked LSTM architecture was used, and recursive computation was performed using a sliding window. The optimal hyperparameters for the model were determined through Bayesian optimization. The performance of the LSTM network was validated through numerical simulations of an explosion in a straight urban street. The results demonstrate that the LSTM network can accurately predict the multi-peak characteristics of blast loads in the street and the arrival times of each peak, demonstrating significant potential for blast load time histories prediction in complex environments.

建筑爆炸荷载的量化是结构抗震工程中一个关键的决定因素。当爆炸发生在复杂的城市环境中时，建筑物表面的爆炸载荷通常是通过比例爆炸试验或计算流体力学模拟来获得的。鉴于爆炸试验和数值模拟的巨大开销，本研究创新性地将长短期记忆（LSTM）网络应用于城市爆炸场景的爆炸载荷时程预测。其主要思想是建立LSTM网络，从有限的训练数据中学习爆炸载荷的时空变化，预测未监测地点或未知爆炸场景下爆炸载荷的时程。特别采用了单向多层堆叠LSTM架构，并采用滑动窗口进行递归计算。通过贝叶斯优化确定了模型的最优超参数。通过对城市直街爆炸的数值模拟，验证了LSTM网络的性能。结果表明，LSTM网络能够准确预测街道爆炸荷载的多峰特征及各峰到达时间，在复杂环境下的爆炸荷载时程预测中具有重要的应用潜力。

{"title":"Prediction of load time histories on building facades under urban explosion environment","authors":"Doudou Si, Zuanfeng Pan, Wendi Li","doi":"10.1016/j.simpat.2025.103220","DOIUrl":"10.1016/j.simpat.2025.103220","url":null,"abstract":"<div><div>The quantification of blast load on a building constitutes a pivotal determinant in blast-resistant structural engineering. When the explosion occurs in a complex urban environment, the blast load on the building surface is conventionally obtained through scaled explosion tests or computational fluid dynamics simulations. Given the significant expenses of explosion tests and numerical simulations, this study innovatively applies the long short-term memory (LSTM) network to predict the blast load time history in urban explosion scenarios. The main idea is to establish an LSTM network to learn the temporal and spatial variations of blast loads from limited training data and predict the time history of blast loads at unmonitored locations or unknown explosion scenarios. In particular, a unidirectional multi-layer stacked LSTM architecture was used, and recursive computation was performed using a sliding window. The optimal hyperparameters for the model were determined through Bayesian optimization. The performance of the LSTM network was validated through numerical simulations of an explosion in a straight urban street. The results demonstrate that the LSTM network can accurately predict the multi-peak characteristics of blast loads in the street and the arrival times of each peak, demonstrating significant potential for blast load time histories prediction in complex environments.</div></div>","PeriodicalId":49518,"journal":{"name":"Simulation Modelling Practice and Theory","volume":"146 ","pages":"Article 103220"},"PeriodicalIF":3.5,"publicationDate":"2025-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145475185","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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 simulation-based approach for reconstructing a diverse set of supply chain models with sparse data using a quality diversity algorithm 一种基于仿真的方法，利用质量多样性算法重构具有稀疏数据的多样化供应链模型集

IF 3.5 2区计算机科学 Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Simulation Modelling Practice and Theory

Pub Date : 2025-10-28 DOI: 10.1016/j.simpat.2025.103216

Isabelle M. van Schilt , Jan H. Kwakkel , Jelte P. Mense , Alexander Verbraeck

Data on supply chains is often sparse due to reluctance among actors to share their data, making supply chain simulation modeling difficult. As a result, supply chain simulation models suffer from parametric and structural uncertainties, and there is a large variety of plausible simulation models that would align with the sparse observations about the real-world supply chain. Constructing a diverse set of models that fit sparse data is not an easy task. A relatively unknown approach to generating this diverse set of plausible models is the Quality Diversity (QD) algorithm. This study evaluates the feasibility of using QD to generate a diverse ensemble of supply chain simulation models for a varying degree of data sparseness. The results show that QD is able to generate a diverse ensemble of supply chain models, including the ground truth. As expected, QD successfully identifies the structure of the ground truth most frequently for a low level of data sparseness. When the sparseness of the data increases, QD is prone to overfitting, identifying supply chain structures that are more complex than the ground truth. Further research should focus on reviewing the calibration metric for sparse data, to reduce the overfitting of complex network structures.

由于参与者之间不愿意共享他们的数据，供应链上的数据通常是稀疏的，这使得供应链仿真建模变得困难。因此，供应链模拟模型受到参数和结构不确定性的影响，并且有大量的合理的模拟模型可以与关于真实世界供应链的稀疏观察相一致。构建一组适合稀疏数据的不同模型并不是一件容易的事。质量多样性（QD）算法是一种相对未知的方法来生成这种多样的可信模型集。本研究评估了使用QD为不同程度的数据稀疏性生成多样化的供应链仿真模型集合的可行性。结果表明，QD能够生成供应链模型的多样化集合，包括基础事实。正如预期的那样，对于低水平的数据稀疏性，QD最频繁地成功地识别了基础真值的结构。当数据的稀疏性增加时，QD容易过度拟合，从而识别出比基本事实更复杂的供应链结构。进一步的研究应集中在审查稀疏数据的校准度量，以减少复杂网络结构的过拟合。

{"title":"A simulation-based approach for reconstructing a diverse set of supply chain models with sparse data using a quality diversity algorithm","authors":"Isabelle M. van Schilt , Jan H. Kwakkel , Jelte P. Mense , Alexander Verbraeck","doi":"10.1016/j.simpat.2025.103216","DOIUrl":"10.1016/j.simpat.2025.103216","url":null,"abstract":"<div><div>Data on supply chains is often sparse due to reluctance among actors to share their data, making supply chain simulation modeling difficult. As a result, supply chain simulation models suffer from parametric and structural uncertainties, and there is a large variety of plausible simulation models that would align with the sparse observations about the real-world supply chain. Constructing a diverse set of models that fit sparse data is not an easy task. A relatively unknown approach to generating this diverse set of plausible models is the Quality Diversity (QD) algorithm. This study evaluates the feasibility of using QD to generate a diverse ensemble of supply chain simulation models for a varying degree of data sparseness. The results show that QD is able to generate a diverse ensemble of supply chain models, including the ground truth. As expected, QD successfully identifies the structure of the ground truth most frequently for a low level of data sparseness. When the sparseness of the data increases, QD is prone to overfitting, identifying supply chain structures that are more complex than the ground truth. Further research should focus on reviewing the calibration metric for sparse data, to reduce the overfitting of complex network structures.</div></div>","PeriodicalId":49518,"journal":{"name":"Simulation Modelling Practice and Theory","volume":"146 ","pages":"Article 103216"},"PeriodicalIF":3.5,"publicationDate":"2025-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145529383","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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