Journal of The Franklin Institute-engineering and Applied Mathematics最新文献_第4页

Toward emotionally intelligent e-learning: Real-time engagement detection through fine-tuned deep facial emotion recognition 面向情商高的电子学习：通过微调深度面部情感识别进行实时参与检测

IF 4.2 3区计算机科学

Journal of The Franklin Institute-engineering and Applied Mathematics Pub Date : 2026-03-15 Epub Date: 2026-02-09 DOI: 10.1016/j.jfranklin.2026.108477

Hamada El Kabtane , Fatima Zohra Ennaji , Khaoula Elkabtane , Maryam Knouzi , Amal Ourdou

{"title":"Toward emotionally intelligent e-learning: Real-time engagement detection through fine-tuned deep facial emotion recognition","authors":"Hamada El Kabtane , Fatima Zohra Ennaji , Khaoula Elkabtane , Maryam Knouzi , Amal Ourdou","doi":"10.1016/j.jfranklin.2026.108477","DOIUrl":"10.1016/j.jfranklin.2026.108477","url":null,"abstract":"<div><div>Enhancing e-learning systems through engagement monitoring is a critical priority. Adapting e-learning platforms to be responsive to learners’ emotional states is viewed as an innovative approach in this domain. However, a key challenge faced by researchers is determining how to effectively measure and assess student engagement based on their emotional cues and behaviors. Thus, several methods have been developed to evaluate student engagement in online learning environments. Yet, many of these existing methods have focused primarily on detecting the engagement level without adequately considering whether the student is actively engaged in the learning process itself. This paper presents a novel system for real-time engagement detection in online learning environments using facial emotion recognition based on a fine-tuned VGG-19. The proposed system utilizes a camera to capture learners’ facial expressions and classify them into seven emotional states: Angry, Disgust, Fear, Happy, Neutral, Sad, and Surprise. Additionally, the system detects four other states: No face detected, Not looking forward, Sleepiness, and None (devoid of human presence frame). Our experimental evaluations include two tests: the first without assigned weights for the learners’ emotional and states indicators, achieving an accuracy of 87.09%, while the second, with weighted coefficients, resulting in a significant improvement to 95.12%. The weighted approach enhances the system’s ability to prioritize emotional cues, demonstrating its effectiveness in accurately identifying engagement levels. The proposed system has potential applications in enhancing online learning experiences by providing real-time feedback to educators and improving learner engagement.</div></div>","PeriodicalId":17283,"journal":{"name":"Journal of The Franklin Institute-engineering and Applied Mathematics","volume":"363 5","pages":"Article 108477"},"PeriodicalIF":4.2,"publicationDate":"2026-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146172569","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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 parameter-free filled function method for global optimization and solving nonlinear equation systems 非线性方程组全局优化与求解的无参数填充函数新方法

IF 4.2 3区计算机科学

Journal of The Franklin Institute-engineering and Applied Mathematics Pub Date : 2026-03-15 Epub Date: 2026-02-10 DOI: 10.1016/j.jfranklin.2026.108479

Jinlan Liu , Helei Kang , Renyun Liu , Dongpo Xu

{"title":"A novel parameter-free filled function method for global optimization and solving nonlinear equation systems","authors":"Jinlan Liu , Helei Kang , Renyun Liu , Dongpo Xu","doi":"10.1016/j.jfranklin.2026.108479","DOIUrl":"10.1016/j.jfranklin.2026.108479","url":null,"abstract":"<div><div>Global optimization is essential for addressing complex real-world problems characterized by non-convexity and numerous local minima. Traditional gradient-based methods often struggle in such scenarios due to poor convergence to local optima rather than the global solution. To overcome this limitation, we propose a novel parameter-free filled function, an auxiliary construct designed to enhance global optimization algorithms by escaping local optima. The proposed function introduces an inverse cosine component to create a continuously differentiable and completely parameter-free formulation that effectively overcomes the common limitations of false smoothing and numerical instability found in existing filled function methods. Theoretical analysis demonstrates the function’s continuous differentiability and its effectiveness in escaping local optima. Building on this formulation, we develop the Non-Parametric Filled Function algorithm (NPFFF), which alternately minimizes the original objective function and the filled function to traverse the search space. NPFFF eliminates the need for manual parameter tuning, simplifying implementation and enhancing robustness. Complexity analysis indicates that constructing the filled function requires only linear time. Experimental results on standard benchmark functions demonstrate that NPFFF achieves faster convergence and higher accuracy than existing methods. Furthermore, NPFFF successfully solves nonlinear systems of equations by transforming them into global optimization tasks.</div></div>","PeriodicalId":17283,"journal":{"name":"Journal of The Franklin Institute-engineering and Applied Mathematics","volume":"363 5","pages":"Article 108479"},"PeriodicalIF":4.2,"publicationDate":"2026-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146172573","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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

Voltage control of a class of underactuated electromechanical systems via an energy shaping approach 一类欠驱动机电系统的能量整形电压控制

IF 4.2 3区计算机科学

Journal of The Franklin Institute-engineering and Applied Mathematics Pub Date : 2026-03-15 Epub Date: 2026-02-10 DOI: 10.1016/j.jfranklin.2026.108512

Jesús Sandoval , Luis Cervantes-Pérez , Isaac Gandarilla , Víctor Santibáñez , César Higuera , Marco Castro-Liera

{"title":"Voltage control of a class of underactuated electromechanical systems via an energy shaping approach","authors":"Jesús Sandoval , Luis Cervantes-Pérez , Isaac Gandarilla , Víctor Santibáñez , César Higuera , Marco Castro-Liera","doi":"10.1016/j.jfranklin.2026.108512","DOIUrl":"10.1016/j.jfranklin.2026.108512","url":null,"abstract":"<div><div>The control of a class of electromechanical systems electrically-driven by brushed Direct Current (DC) motors of permanent magnet is addressed in this paper. An energy shaping approach is used for the voltage controller design, which achieves position/speed regulation of underactuated mechanical systems –those systems with more actuators than degrees of freedom– actuated by DC motors. The main contribution is a control law in terms of voltage feeding each DC motor instead of a torque/force input, which does not need voltage or armature current feedback, but only position measurements and speed. The proposed voltage controller is useful for this class of electrically-driven mechanical systems used in experimental workbenches found in numerous automatic control laboratories. The effectiveness of the proposed controller is demonstrated through two underactuated mechanical systems: the disk-on-disk and the self-balancing robot. For the disk-on-disk system, numerical simulations are provided, whereas for the self-balancing robot, real-time experimental results are displayed.</div></div>","PeriodicalId":17283,"journal":{"name":"Journal of The Franklin Institute-engineering and Applied Mathematics","volume":"363 5","pages":"Article 108512"},"PeriodicalIF":4.2,"publicationDate":"2026-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146172571","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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

Event-triggered consensus control for short memory fractional-order multiagent systems with input delay 具有输入延迟的短记忆分数阶多智能体系统的事件触发一致性控制

IF 4.2 3区计算机科学

Journal of The Franklin Institute-engineering and Applied Mathematics Pub Date : 2026-03-15 Epub Date: 2026-02-09 DOI: 10.1016/j.jfranklin.2026.108507

Taotao Hu , Miaomiao Shi , Kaibo Shi , Xianya Wuliu

{"title":"Event-triggered consensus control for short memory fractional-order multiagent systems with input delay","authors":"Taotao Hu , Miaomiao Shi , Kaibo Shi , Xianya Wuliu","doi":"10.1016/j.jfranklin.2026.108507","DOIUrl":"10.1016/j.jfranklin.2026.108507","url":null,"abstract":"<div><div>This article concentrates on the event-triggered consensus control for short memory fractional-order multiagent systems with input delay. First, considering the limitation of network communication resources, an event-triggered consensus strategy with input delay is designed, which only depend on the sampled state information of neighboring agents. It is shown that the proposed event-triggered control strategy is able to effectively adjust the frequency of the information transfer between neighboring agents. Utilizing the fractional Lyapunov stability theory and some inequality techniques, the delay-dependent consensus criteria for fractional-order multiagent systems are established. Meanwhile, analyzing the existence of the lower bound of the triggering time interval for any two continuous events, Zeno behavior is excluded. Finally, some simulations results are given to confirm the correctness of the proposed consensus approach, which show that the designed event-triggered consensus strategy can effectively reduce the communication frequency between neighboring agents compared with the known works.</div></div>","PeriodicalId":17283,"journal":{"name":"Journal of The Franklin Institute-engineering and Applied Mathematics","volume":"363 5","pages":"Article 108507"},"PeriodicalIF":4.2,"publicationDate":"2026-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146172566","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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

Finite-time mixed H2/H∞ control for stochastic T-S fuzzy networked semi-Markov jump systems with event-triggering and quantization 具有事件触发和量化的随机T-S模糊网络半马尔可夫跳跃系统的有限时间混合H2/H∞控制

IF 4.2 3区计算机科学

Journal of The Franklin Institute-engineering and Applied Mathematics Pub Date : 2026-03-15 Epub Date: 2026-02-03 DOI: 10.1016/j.jfranklin.2026.108457

Xikui Liu , Yapeng Teng , Yan Li

{"title":"Finite-time mixed H2/H∞ control for stochastic T-S fuzzy networked semi-Markov jump systems with event-triggering and quantization","authors":"Xikui Liu , Yapeng Teng , Yan Li","doi":"10.1016/j.jfranklin.2026.108457","DOIUrl":"10.1016/j.jfranklin.2026.108457","url":null,"abstract":"<div><div>This paper mainly addresses the finite-time mixed <em>H</em><sub>2</sub>/<em>H</em><sub>∞</sub> control issue of stochastic T-S fuzzy networked control systems with semi-Markov switching by constructing a dynamic output feedback controller (DOFC). Firstly, a novel criterion of finite-time stability (FTS) is defined for stochastic systems, and by employing linear matrix inequality (LMI) and the free-weighting matrix methods, sufficient conditions for the system to satisfy this definition are given. Then, a new Lyapunov-Krasovskii functional (LKF) associated with the sojourn time is constructed for semi-Markov jump systems (semi-MJSs), this function encompasses the entire Markov switching process and augments the practical applicability of the results. Furthermore, considering the transition rates (TRs) matrix of semi-MJSs is time-varying, we suppose that the TRs matrix belongs to a bounded yet unfixed polytope to overcome the computational challenges induced by time-varying TRs. Next, in order to reduce communication burden of stochastic networked semi-MJSs and save network resources, the event-triggered mechanism (ETM) and the quantized output strategy are adopted simultaneously. Finally, a solution algorithm is proposed based on the obtained conditions, and the feasibility of this work is verified through the practical application of a mass-spring-damper mechanical system (MSDMS) and a comparative example.</div></div>","PeriodicalId":17283,"journal":{"name":"Journal of The Franklin Institute-engineering and Applied Mathematics","volume":"363 5","pages":"Article 108457"},"PeriodicalIF":4.2,"publicationDate":"2026-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146172574","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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

Thermal inertia analysis and its bound-dependent active control for thermoacoustic systems 热声系统的热惯性分析及其界相关主动控制

IF 4.2 3区计算机科学

Journal of The Franklin Institute-engineering and Applied Mathematics Pub Date : 2026-03-15 Epub Date: 2026-02-03 DOI: 10.1016/j.jfranklin.2026.108463

Yuzhuo Zhao , Dan Ma

{"title":"Thermal inertia analysis and its bound-dependent active control for thermoacoustic systems","authors":"Yuzhuo Zhao , Dan Ma","doi":"10.1016/j.jfranklin.2026.108463","DOIUrl":"10.1016/j.jfranklin.2026.108463","url":null,"abstract":"<div><div>The thermoacoustic system with the thermal inertia in the Rijke tube is considered, where the thermal inertia between the unstable heat release and the velocity field in the Rijke tube refers to a time-delay. To suppress the thermoacoustic instability phenomenon, a time-delay bound-dependent active control is designed for the thermoacoustic system with the known and the unknown heat release rates. Unlike previous works, a time-delay bound-dependent active controller is proposed to make sure the velocity and pressure of the thermoacoustic system with the known and the unknown heat release rates to converge to a small neighborhood about the zero, in the case of the unknown upper bound on the unknown time-delay. Moreover, this control method can calculate the maximum allowable time-delay of the active controller. In addition, considering the known heat release rate, the maximum allowable time-delay of the active controller is related to the control gain or physical parameters of the thermoacoustic system; When considering the unknown heat release rate, the maximum allowable time-delay of the active controller is only related to the control gain. Finally, the effectiveness of the active control methods are proved through the simulation experiment.</div></div>","PeriodicalId":17283,"journal":{"name":"Journal of The Franklin Institute-engineering and Applied Mathematics","volume":"363 5","pages":"Article 108463"},"PeriodicalIF":4.2,"publicationDate":"2026-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146122638","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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

Adaptive fixed-time prescribed performance control for the upper limb exoskeleton with arbitrary initial errors 具有任意初始误差的上肢外骨骼自适应定时性能控制

IF 4.2 3区计算机科学

Journal of The Franklin Institute-engineering and Applied Mathematics Pub Date : 2026-03-15 Epub Date: 2026-02-02 DOI: 10.1016/j.jfranklin.2026.108459

Xianlei Zhang , Yan Zhang , Qing Hu , Zongyu Han

引用次数: 0

Event-triggered based finite-time adaptive control for multi-agent systems with unknown input delay and input quantization 具有未知输入延迟和输入量化的多智能体系统的事件触发有限时间自适应控制

IF 4.2 3区计算机科学

Journal of The Franklin Institute-engineering and Applied Mathematics Pub Date : 2026-03-15 Epub Date: 2025-12-21 DOI: 10.1016/j.jfranklin.2025.108351

Xiaonan Xia, Zihang Tong, Tianping Zhang, Jianwen Pan

引用次数: 0

Distributed optimization algorithm design and analysis on cooperation-competition network based on PID control 基于PID控制的协同竞争网络分布式优化算法设计与分析

IF 4.2 3区计算机科学

Journal of The Franklin Institute-engineering and Applied Mathematics Pub Date : 2026-03-15 Epub Date: 2026-02-09 DOI: 10.1016/j.jfranklin.2026.108483

Jiaxu Liu , Pengkai Wang , Song Chen , Shengze Cai , Chao Xu

{"title":"Distributed optimization algorithm design and analysis on cooperation-competition network based on PID control","authors":"Jiaxu Liu , Pengkai Wang , Song Chen , Shengze Cai , Chao Xu","doi":"10.1016/j.jfranklin.2026.108483","DOIUrl":"10.1016/j.jfranklin.2026.108483","url":null,"abstract":"<div><div>This paper addresses distributed optimization problems with equality constraints within the framework of cooperation-competition network. Motivated by the exceptional performance of proportional-integral-derivative (PID) controllers, we propose an accelerated distributed optimization algorithm to decompose the distributed optimization problem on the cooperation-competition network into two cooperative subnetworks and address them. By leveraging the Lyapunov stability theorem, we establish the exponential convergence of our algorithm over undirected connected and structurally balanced cooperation-competition graphs when the parameters (e.g., <em>k<sub>p</sub>, k<sub>i</sub>, k<sub>d</sub></em>) are selected within certain ranges, assuming that the local objective functions are smooth and strongly convex. Additionally, we provide guidelines for selecting appropriate parameter values (e.g., <em>k<sub>p</sub>, k<sub>i</sub>, k<sub>d</sub></em>). Furthermore, we show that the D-PID-CCN has great potential for nonconvex distributed optimization over the cooperation-competition network. Finally, we present the effectiveness and superiority of our proposed algorithms on several numerical simulations.</div></div>","PeriodicalId":17283,"journal":{"name":"Journal of The Franklin Institute-engineering and Applied Mathematics","volume":"363 5","pages":"Article 108483"},"PeriodicalIF":4.2,"publicationDate":"2026-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146172565","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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

Less conservative stability conditions of fractional-order time-delay systems using covering sets and filters 使用覆盖集和滤波器的分数阶时滞系统的小保守稳定性条件

IF 4.2 3区计算机科学

Journal of The Franklin Institute-engineering and Applied Mathematics Pub Date : 2026-03-15 Epub Date: 2026-02-09 DOI: 10.1016/j.jfranklin.2026.108497

Qing-Hao Zhang , Xiao-Chuang Jin , Jun-Guo Lu , Zhen Zhu

{"title":"Less conservative stability conditions of fractional-order time-delay systems using covering sets and filters","authors":"Qing-Hao Zhang , Xiao-Chuang Jin , Jun-Guo Lu , Zhen Zhu","doi":"10.1016/j.jfranklin.2026.108497","DOIUrl":"10.1016/j.jfranklin.2026.108497","url":null,"abstract":"<div><div>Complex Industrial Internet systems are often challenged by communication delays, posing significant threats to the stability of the systems. This paper conducts an in-depth investigation into the stability problems of a representative model, fractional-order time-delay systems, in Industrial Internet. The paper aims to provide theoretical support for enhancing the stability in interconnected industrial systems. To conclude this goal, the stability problem of fractional-order time-delay systems is transformed to the robust stability problem of delay-free interconnected uncertain systems. By applying the small gain theorem and fractional-order bounded real lemma, a delay-dependent stability condition is derived. Then, different covering sets and well-designed filters that capture the phase and gain information of the uncertainties are utilized to further reduce the conservatism of the proposed stability condition. The results are formulated as linear/bilinear matrix inequalities (LMIs/BMIs), and the proposed LMI/BMI criteria yield enlarged delay margins compared with previously reported conditions in the literature. Finally, two examples are analyzed and the results indicate that the new proposed conditions are effective and have less conservatism than other existing criteria.</div></div>","PeriodicalId":17283,"journal":{"name":"Journal of The Franklin Institute-engineering and Applied Mathematics","volume":"363 5","pages":"Article 108497"},"PeriodicalIF":4.2,"publicationDate":"2026-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146172564","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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