Dr. Benyamin Khoshnevisan, currently affiliated with University of Southern Denmark, Department of Green Technology, SDU Life Cycle, is a seasoned researcher with over 10 years of experience in the field of sustainability assessment. Dr. Khoshnevisan has established himself as an expert in various domains, including renewable energy systems, biofuel production, waste management, wastewater treatment, and agricultural systems. His interdisciplinary approach involves the application of sustainability assessment, data mining, modeling, optimization, and multi-criteria decision analysis. This comprehensive methodology enables him to offer well-informed solutions to decision-makers. He is offering his expeties to several Horizon Europe projects including “TAKE-OFF”, “AgriLoop”, “WALNUT”, and “Agro4Agri”. Dr. Khoshnevisan’s commitment to advancing sustainable practices is evident in his contributions to research and academia. For more detailed insights into his research profile, you can visit his Google Scholar.
Pr. Benyamin KHOSHNEVISAN, University of Southern Denmark
Sustainability Opportunities and Challenges of Smart Systems and Green Processes
Abdeljalil Gattal is a professor in Computer Science at the Department of Mathematics and Computer Science in University of Tebessa, Algeria. He received his PhD in 2016 from Ecole Nationale Supérieure d’Informatique (ESI-Algeria) and his research focuses on segmentation-verification for handwritten digit recognition. He is currently an associate professor at the Department of Mathematics and Computer Science in Larbi Tebessi University, Tébessa, Algeria. He has published several papers and has supervised many Master and License students. His research interests include image analysis, pattern recognition, and recognition of handwriting.
Pr Abdeljalil GATTAL, University of Tebessa, ALGERIA
Handwriting Analysis and Recognition: Techniques and Applications »
Handwriting analysis and recognition have become crucial areas in biometrics, forensics, and document processing, with a wide range of applications including personality assessment, script identification, Handwriting recognition, keyword spotting, gender classification, writer identification, and signature verification. This presentation delves into the methodologies, challenges, and real-world implementations of these handwriting-based technologies. Traditional approaches rely on handcrafted features such as Gabor filters and Local Binary Patterns (LBP), while modern deep learning techniques, such as Convolutional Neural Networks (CNNs) and Siamese Networks, have significantly improved accuracy in tasks such as writer identification and offline signature verification. In addition, graphology-based personality assessment and gender classification utilize stylistic features such as stroke pressure, slant, and word spacing to infer behavioral traits. Despite these advancements, challenges still exist, including variability in handwriting styles, limited datasets (especially for historical scripts), and adversarial attacks in forgery detection. Future research directions focus on few-shot learning, explainable AI for graphology, and cross-script generalization. This comprehensive overview highlights the interdisciplinary nature of handwriting analysis, showcasing its potential in security, psychology, and digital archiving, while addressing open challenges and future trends in the field.
Mejdi Azaïez is a prominent researcher and professor in numerical fluid mechanics and applied mathematics. Born in Paris, France, he earned his Ph.D. from the University of Paris and began his career at the University of Toulouse III – Paul Sabatier. Since 2010, he has been a full professor at the Bordeaux Institute of Technology, heading the TREFLE department. His research focuses on the numerical solution of the Navier-Stokes equations, Proper Orthogonal Decomposition (POD) methods, and thermal and phase change phenomena. Notable publications include « A finite element model for the data completion problem » (2011), « An intrinsic Proper Generalized Decomposition for parametric symmetric elliptic problems » (2018), and « Two Phases Stefan Problem with Smoothed Enthalpy » (2016). He is the author of « Finite Element Methods for Incompressible Fluids ». In 2025, he introduced PODNO, an innovative method for solving PDEs. His work continues to advance the field and inspire future researchers.
Pr. Mejdi AZAIEZ, University of Bordeaux, France
A monolithic numerical model of solid-liquid phase change problem
In latent heat storage, certain non-phase change materials (non-PCMs) with high thermal conductivity are incorporated into the phase change materials (PCMs) with the aim of enhancing the efficiency of heat/cold storage. We term this type of non-PCMs as “enhancer”, which includes materials like graphite and copper foam, usually with a complex skeleton structure. In this talk, we propose a phase field model to describe the solidification and melting phenomena of PCMs with enhancer from a microscopic point of view. Our model is governed by the energy equation coupled with the Allen-Cahn equation. A penalty technique is applied in the Allen-Cahn equation to describe the complex structure of the non-PCMs. We use the concept of thermal resistance to define the boundary condition on the contact interface of two materials to ensure the temperature jump. Thanks to the hybrid dual formulation, the temperature can be solved as a monolithic function while satisfying the temperature jump on the material interface. In temporal discretization, a numerical scheme is developed to decouple the phase field from the temperature. In the spatial discretization, the hybrid finite element method, the Raviart-Thomas elements are used to solve the temperature and to satisfy the temperature jump on the interface. 2D and 3D simulations are carried out for both melting and solidification processes of a fossil based organic PCM, RODATHERM60 in the graphite skeleton on different porous structures to validate our model.
Pr. Abdesselem Boulkroune is a prominent researcher and professor at the University of Jijel, Algeria. He earned his Ph.D. in Applied Mathematics and has been a key figure in the field of control theory and nonlinear systems. Since joining the University of Jijel, he has led the Laboratory of Analysis, Modeling, and Simulation (LAMS). His research focuses on adaptive control, fuzzy control, and synchronization of chaotic systems. Notable publications include « Adaptive fuzzy control for a class of nonlinear systems with unknown control directions » (2015) and « Synchronization of chaotic systems with unknown parameters using adaptive control » (2017). He has also authored the book « Adaptive Control and Synchronization of Nonlinear Systems ». His contributions have significantly advanced the field of control theory and nonlinear dynamics, and he continues to inspire and mentor students and researchers.
Pr Abdesselem BOULKROUNE, University of Jijel, ALGERIA<
A Fixed-Time Adaptive Control Framework for Nonlinear Systems
with Actuator Faults, Delays, and Bounded Disturbances
This presentation focuses on the design of a fixed-time control scheme for uncertain nonlinear systems with non-strict feedback, addressing several critical challenges commonly encountered in practical applications. The proposed approach specifically considers actuator faults, input delays, model uncertainties, actuator saturation, and bounded unmatched disturbances—all of which can significantly degrade system performance and stability in the absence of appropriate compensation mechanisms.
The control strategy integrates adaptive control using a fuzzy approximator with the Dynamic Surface Control (DSC) technique. The fuzzy approximator facilitates the modeling and compensation of unknown system nonlinearities, while the DSC framework reduces the computational burden associated with conventional feedback methods, thereby enabling practical implementation.
To enhance robustness against actuator faults and disturbances, adaptive observers are developed to estimate these effects in real-time, allowing for effective compensation and improved resilience of the closed-loop system. A rigorous Lyapunov stability analysis confirms that all signals in the closed-loop system remain bounded. Moreover, the proposed control scheme guarantees that the output tracking error converges to a neighborhood of the origin within a fixed-time, independent of the initial conditions.
This fixed-time convergence property is especially valuable in critical applications that demand fast and predictable transient performance. Detailed simulations validate the effectiveness and superiority of the proposed method, demonstrating notable improvements in tracking accuracy, robustness, and fault tolerance under various adverse conditions.
