Urmia University of Technology was formed and started to work in summer 2007. Following the principled agreement of UUT establishment and with the special consideration of president Dr. Ahmadinejad and the officials attempts, an independent funding (budget) based on Iran government proposal and the Islamic parliament approval was allocated to the university. Obtaining educational and research activities permission in Information Technology (IT), Mechanical, and Industrial engineering, UUT officially started in February, 2007.
Accurate core-scale thermal–hydraulic prediction underpins safety margin evaluation, DNBR assessment, and PWR licensing, yet full rod-level CFD of an entire VVER-1000 core is computationally prohibitive, and prior porous-media full-core studies have often left closure coefficients and convergence criteria only partially specified. This study presents a steady-state CFD model of a VVER-1000 (V-320) core, coupling a fully specified anisotropic porous-media formulation for the 163 fuel assemblies with the RANS equations and the realizable k–ε model, at 15.7 MPa, 291.4 °C inlet, and 17,880 kg/s flow (3000 MWth). A four-level mesh study (4.2–24.3 million cells) and a Grid Convergence Index analysis (GCI_fine = 0.31 %) established numerical reliability. The predicted outlet temperature (320.5 °C) and pressure drop (135.9 kPa) agreed with plant design data within 2.18 %, and the sub-model was independently validated against coolant-mixing and pressure-drop benchmark data (deviation < 0.71 %), with a combined outlet-temperature uncertainty of ±1.88 %. This reproducible closure framework and quantified uncertainty budget distinguish the present work from earlier, less transparently documented porous-media CFD studies of VVER-1000 cores.
The collapse of a cavitation bubble near a rigid boundary induces a high-speed transient liquid jet that accelerates the liquid to the boundary. In this study, the dynamics of a laser-induced bubble confined between two parallel walls is studied experimentally and numerically. The effects of confinement/gap distance H as well as the off-centre distance δ on the dynamics of the bubble are examined. The numerical results are validated against the corresponding experimental data in terms of the evolution of the bubble shape during expansion, collapse, and rebound. Furthermore, a case with a more confined situation is considered. The numerical simulation elucidates intricate aspects of the bubble dynamics around the jet impact, which are not obtainable through experiments. The numerical results are scrutinised in a more comprehensive manner in order to derive the pressure and velocity fields, thereby determining their impact on the walls. It was found that the location of the bubble with respect to the walls and the degree of confinement determine the number as well as the strength of the liquid jet. Typically, when the confinement increased (H decreased), the peak pressure at the upper wall decreased significantly (17
The field of scientific machine learning, which originally utilized multilayer perceptrons (MLPs), is increasingly adopting Kolmogorov-Arnold Networks (KANs) for data encoding. This shift is driven by the limitations of MLPs, including poor interpretability, fixed activation functions, and difficulty capturing localized or high-frequency features. KANs address these issues with enhanced flexibility, enabling efficient modeling of complex nonlinear interactions and effectively overcoming the constraints associated with conventional MLP architectures. This review categorizes recent progress in KAN-based models across three distinct perspectives: (i) data-driven learning, (ii) physics-informed modeling, and (iii) deep-operator learning. Each perspective is examined through the lens of architectural design, training strategies, application efficacy, and comparative evaluation against MLP-based counterparts. By benchmarking KANs against MLPs, we highlight consistent improvements in accuracy, convergence, and spectral representation, clarifying KANs' advantages in capturing complex dynamics. In addition to reviewing recent literature, this work also presents several comparative evaluations that clarify central characteristics of KAN modeling and hint at their potential implications for real-world applications. Finally, this review identifies critical challenges and open research questions in KAN development, particularly regarding computational efficiency, theoretical guarantees, hyperparameter tuning, and algorithm complexity. We also outline future research directions aimed at improving the robustness, scalability, and physical consistency of KAN-based frameworks.
Population growth and household lifestyles make energy management and improving energy consumption patterns global concerns. Growing energy consumption within residential section increases the direct dependency on fossil fuels. To reduce reliance, smart home technology and home energy management systems have been established. These systems plan the use of household appliances from peak hours to off-peak times to minimize energy consumption costs. Moreover, combining these systems with renewable energy sources like photovoltaic addresses environmental concerns and lowers daily grid energy usage. Incorporating an energy storage system with solar panels is essential for reducing solar energy waste, guaranteeing continuous electricity supply during blackouts, and lowering energy costs at peak hours, especially when combined with effective battery charge and discharge control. This paper presents a two-stage stochastic programming model for optimizing solar panel and energy storage sizing, along with daily energy scheduling, in a green smart home equipped with photovoltaic generation, battery storage, and electric vehicle (EV) charging under carbon emission pricing. The proposed model minimizes daily grid energy costs and consumption by utilizing day-ahead electricity prices and implementing efficient battery management strategies, while simultaneously incorporating EV charging dynamics to enhance both economic and environmental performance. The proposed model generates optimal 24-hour scheduling profiles for controllable appliances, battery charging/discharging operations, and electric vehicle charging, ensuring efficient energy utilization across all manageable household assets. To validate the efficacy of the proposed approach, three real-life case studies were conducted utilizing the embedded CPLEX solver within GAMS Studio, enabling comprehensive implementation and evaluation of the models.
Mental workload (MWL) indicates cognitive effort during a task and is a key marker of mental state. Understanding MWL is vital in neuroergonomics, cognitive neuroscience, human-machine interaction, and intelligent systems. This paper introduces a new multi-stage approach for classifying mental workload using EEG signals. The method involves four key steps: preprocessing the signals to eliminate noise; applying the multivariate synchrosqueezing transform (MSST) for precise time-frequency analysis; extracting deep features with a new convolutional neural network (CNN) architecture that includes a time-frequency attention module (CNN-TFAN); and finally, reducing feature dimensionality for classification. The approach was tested on two public datasets, STEW and MAT. Results indicated that the optimized combination of semisupervised discriminant analysis (SDA) for feature reduction and support vector machine (SVM) for classification yielded the best results, with accuracy rates of 97.1% on STEW and 98.6% on MAT. Additional analysis revealed that MSST outperformed other time-frequency methods and that the deeper, attention-enhanced network architecture significantly improved classification accuracy. These findings demonstrate the effectiveness and robustness of the proposed framework for mental workload analysis.
