Documents published in Scipedia

• An Enhanced Pilot Aided Channel Estimation in XL-MIMO Communication Systems Using Bitterling Swallow Fish Optimization

  M. Kamal, S. Abideen, Y. Luo, A. Hussein, H. Samkari, M. Alleyani, T. Ma

  Rev. int. métodos numér. cálc. diseño ing. (2025). Vol. 41, (4), 90

  
  

  Abstract

  The continuous evolution of 6th Generation (6G) wireless networks places Extremely Large-scale Multiple-Input Multiple-Output (XL-MIMO) schemes as a crucial enabler for ultra-reliable and high data rate communication. Channel estimation in XL-MIMO is crucial here because it allows the system to precisely recognize the wireless channel conditions between the transmitter and receiver, which is vital for enhancing signal processing, and resource allocation. Traditional pilot-aided channel estimation approaches face challenges, such as high error. Hence, this work proposes an innovative model called the Deep Kronecker Network-Bitterling Swallow Fish Optimization Algorithm (DKN-BSwaFOA) for pilot-aided channel estimation in XL-MIMO systems. Initially, the system model of XL-MIMO is contemplated. The pilot insertion is done at the transmitter, and the location of the pilot symbol is optimally selected using BSwaFOA. The signal is propagated over the hybrid field channel, where both farfield and near-field components coexist. At the receiver, the channel is estimated using DKN, which is trained using the proposed BSwaFOA. The experimental outcomes demonstrated that the DKN_BSwaFOA computed the minimum Root Mean Square Error (RMSE), Bit Error Rate (BER), and Mean Square Error (MSE) of 0.030, 0.002, and 0.001.OPEN ACCESS Received: 23/09/2025 Accepted: 11/11/2025

  Abstract
  The continuous evolution of 6th Generation (6G) wireless networks places Extremely Large-scale Multiple-Input Multiple-Output (XL-MIMO) schemes as a crucial enabler for ultra-reliable [...]

  • 472
  • 5/5
  •  read

• Entropy-Guided k-Core Pruning Balancing Redundancy Reduction and Information Preservation for Efficient CNN Compression

  Y. Yang, D. Jiang, X. Deng, L. Wang

  Rev. int. métodos numér. cálc. diseño ing. (2025). Vol. 41, (4), 88

  
  

  Abstract

  Convolutional Neural Networks (CNNs) are widely used in computer vision, but their massive computational cost and parameter redundancy hinder deployment on resource-constrained devices (e.g., edge terminals). Existing filter pruning methods often struggle to balance two critical goals: aggressive redundancy reduction and effective preservation of taskcritical information—either leading to excessive accuracy loss or insufficient compression. To address this challenge, we are the first to jointly exploit k-core decomposition and information entropy in a unified pruning criterion, and we instantiate this idea in a novel graph–entropy collaborative framework that achieves Pareto-optimal compression-accuracy trade-offs. The key steps are as follows: First, we use perceptual hashing (pHash) to calculate the similarity of output feature maps between filters, then model each filter as a node in an undirected graph—edges are established only when filter similarity exceeds a predefined threshold, forming a “redundancy graph” that quantifies inter-filter redundancy. Second, kcore decomposition is applied to this graph to identify high-order redundant substructures, which helps locate redundant filters at the structural level. Finally, information entropy is introduced to evaluate the “informational value” of each node (filter) in the k-core: only filters with low redundancy and high information content are retained, ensuring minimal loss of critical features. Extensive experiments are conducted on CIFAR10 and CIFAR-100 datasets, using representative CNN architectures (VGGNet-16, ResNet-56/110, DenseNet-40). Specifically, VGGNet-16 achieves a 65.8% reduction in floating point operations (FLOPs) and an 88.8% reduction in parameters while experiencing only a 1.24% decrease in Top-1 accuracy. ResNet-56 attains a 50.1% reduction in FLOPs with a nearly imperceptible accuracy loss of 0.03%, markedly surpassing the Fire together wire together (FTWT) method which reduces FLOPs by 54% at the cost of a 1.38% accuracy decline. DenseNet-40 accomplishes a 76.5% FLOPs reduction with a 1.55% accuracy decrease, demonstrating the method’s strong applicability for high-intensity compression of densely connected networks. Furthermore, the method’s scalability is validated on the large-scale ImageNet dataset with ResNet-50, where it achieves a 73.65% FLOPs reduction with competitive accuracy, underscoring its practicality for real-world applications. These outcomes collectively affirm the effectiveness and broad applicability of the proposed graphentropy collaborative pruning framework.

  Abstract
  Convolutional Neural Networks (CNNs) are widely used in computer vision, but their massive computational cost and parameter redundancy hinder deployment on resource-constrained [...]

  • 44
  •  read

• Computational Study of Nozzle Configuration Effects on Heat Transfer and Flow Characteristics in Aero-Engine Swirling Anti-Icing Systems

  Y. Tu, Y. Wu, B. Zheng, S. Luo, W. Liu, Y. Zeng

  Rev. int. métodos numér. cálc. diseño ing. (2025). Vol. 41, (4), 86

  
  

  Abstract

  Engine inlet icing persists as a critical hazard to aviation operational safety, compromising aerodynamic performance and potentially inducing catastrophic engine failure. Aero-engine swirling anti-icing systems inject high-temperature bleed air into an annular chamber at the engine’s leading edge through tangentially positioned nozzles. This high-velocity jet entrains low-temperature air within the chamber, establishing a circulatory flow that effectively heats the lip surface to prevent ice formation. This study employs computational fluid dynamics (CFD) method to systematically evaluate the flow and heat transfer characteristics of four distinct nozzle configurations within an aero-engine anti-icing chamber: conical single-orifice, diffuser-equipped, elliptical dual-orifice, and elliptical quad-orifice nozzles. Results indicate that the conical single-orifice nozzle exhibits the highest entrainment efficiency due to its concentrated jet structure, whereas the diffuser-equipped nozzle demonstrates 16.4%– 18.1% lower efficiency, attributable to premature kinetic energy dissipation. At identical bleed air flow rates, the diffuser-equipped nozzle yields the lowest circulation velocity and pressure loss, necessitating minimal bleed air pressure. The elliptical quad-orifice nozzle optimally mitigates hot and cold spots via multi-jet energy dispersion, achieving a maximum 34.7% reduction in lip surface temperature differentials compared to the conical single-orifice design within the analyzed bleed air mass flow rate range. Nozzle configurations exert limited influence on the average Nusselt number, with a maximum relative deviation of 5.48% observed across all nozzle configurations when compared to established empirical correlations.OPEN ACCESS Received: 11/07/2025 Accepted: 21/08/2025 Published: 15/12/2025

  Abstract
  Engine inlet icing persists as a critical hazard to aviation operational safety, compromising aerodynamic performance and potentially inducing catastrophic engine failure. [...]

  • 121
  •  read

• A Picture Fuzzy Decision-Making Framework for COBOT Selection in Digital Supply Chains

  B. Kesici

  Rev. int. métodos numér. cálc. diseño ing. (2025). Vol. 41, (4), 85

  
  

  Abstract

  Driven by digitalization in supply chains, the use of Collaborative Robots (COBOTs) has become increasingly widespread in recent years. They significantly contribute to process efficiency by working in place of, or in collaboration with, humans in a variety of operations, including welding, painting, assembly and disassembly, transportation, packaging, and palletizing. However, when uncertainty and different criteria are taken into account, decision support systems that compare practical robots based on their suitability for specific needs are inadequate. This study presents a comprehensive multi-criteria decision-making (MCDM) framework for prioritizing COBOTs with different features used in digital supply chain processes. Based on in-depth research in the literature and the opinions of experts working in companies that use relevant robots in the industry, the criteria to be evaluated when selecting COBOT types are identified. The importance of these criteria was determined using the Picture Fuzzy Step-wise Weight Assessment Ratio Analysis (PiF-SWARA) method, which effectively captures the uncertainty in experts’ decisionmaking processes. Subsequently, alternative COBOT types were ranked using the Picture Fuzzy Combinative Distance-Based Assessment (PiFCODAS) approach. This case study, which evaluates the PiF-SWARACODAS concept, reveals that according to expert assessments, cost is the most important criterion in COBOT selection, followed by process quality and space utilization. The findings about the selection of types emphasize that high-efficiency articulated robots operating at high speeds under mass production conditions are the primary priority. These robots are followed by humanoid robots. The third most important are power and force-limiting robots. The fourth and fifth types of COBOTs are hand-guided and safety-monitored stop robots. Validation and sensitivity analyses confirmed the robustness of the results. Overall, the proposed framework not only clarifies the key priorities for manufacturing facilities but also provides a validated decision support tool to align digitalization strategies with the most appropriate COBOT investments.OPEN ACCESS Received: 25/08/2025 Accepted: 15/10/2025 Published: 15/12/2025 An extensive literature review was conducted to examine previous research on COBOT applications in the supply chain, to identify evaluation criteria and alternatives for this study, and to highlight the originality of this study. The concept of COBOT was included in books, journals and conference proceedings accessible through the SCOPUS database. However, the literature search with the related keyword found too many studies. Therefore, the scope was further customized to include different types of COBOT models that can be used throughout the supply chain. To make the literature review process more systematic, the PRISMA methodology, which stands for “Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA)”, was used. The PRISMA approach, designed by Moher et al. [10], has become widely used among academics in recent years. Sources are analyzed and evaluated according to the determined eligibility criteria. Then, appropriate studies are selected. Table 1lists the keywords used to perform a PRISMA method literature review, along with the number of studies found using these Table 1: Literature search with A total of 1179 studies were found in SCOPUS after searching with the were found. The results were duplicated because some

  Abstract
  Driven by digitalization in supply chains, the use of Collaborative Robots (COBOTs) has become increasingly widespread in recent years. They significantly contribute to process [...]

  • 418
  •  read

• Q-Learning-Based Lightweight Task Orchestrator: A Lightweight Q-Learning-Based Scheduler for Service Level Agreement-Aware Container Placement in Heterogeneous Clusters

  V. Veeramani, M. Balamurugan, M. Oberoi

  Rev. int. métodos numér. cálc. diseño ing. (2025). Vol. 41, (4), 83

  
  

  Abstract

  Efficient management of workloads in diverse container clusters requires maintaining a balance between Service Level Agreement (SLA) compliance, Quality of Service (QoS), energy efficiency, and security, despite differences in resources and architectures. This research introduces Federated Q-Learning–based Lightweight Task Orchestrator (F-Q-LiTO), a compact and intelligent orchestration framework that combines predictive modeling, approximate data structures, and security filtering to enable adaptive task placement across distributed environments. Unlike complex deep reinforcement learning models such as Deep Reinforcement Model (DeepRM) or traditional heuristic schedulers like Kubernetes BinPacking, F-Q-LiTO uses tabular Q-learning enhanced with federated aggregation, which significantly reduces computational and communication overhead, making it ideal for edge computing environments with limited resources. The framework incorporates a Long Short-Term Memory (LSTM)–based predictor for proactive resource forecasting, a Count-Min Sketch for scalable resource utilization estimation, and an XOR filter for efficient and lightweight security enforcement. Experimental results demonstrate that F-Q-LiTO achieves 98.6% task completion, 96.8% SLA satisfaction, and reduces energy consumption to 180.5 kilowatt-hours (kWh). It outperforms DeepRM and Kubernetes by achieving 34% fewer missed deadlines and up to 30% lower energy imbalance. The system converges quickly—by Episode 6—and maintains cluster fairness (Jain’s Fairness Index= 0.98) along with priority-aware placement accuracy of 93.2%. Security analysis shows that F-Q-LiTO successfully blocks 98.5% of unauthorized task placements while using only 0.3 megabytes (MB) of memory. Overall, F-Q-LiTO demonstrates that a federated and lightweight reinforcement learning approach can deliver scalable, secure, and QoS-aware orchestration for modern edge and multi-cloud computing environments without compromising performance or efficiency.OPEN ACCESS Received: 24/07/2025 Accepted: 14/10/2025 Published: 15/12/2025

  Abstract
  Efficient management of workloads in diverse container clusters requires maintaining a balance between Service Level Agreement (SLA) compliance, Quality of Service (QoS), [...]

  • 88
  •  read

• Study on the Mechanical Characteristics of Concrete Pavement Corner Detachment and Grouting Rein­ forcement Using the FEM­DEM Coupling Approach

  X. ZHANG, B. Mei, D. Chen, J. Feng, T. Wang

  Rev. int. métodos numér. cálc. diseño ing. (2025). Vol. 41, (4), 81

  
  

  Abstract

  Cement concrete pavements have significant advantages in the construction of transportation infrastructure. However, the disease of slab bottom voids affects their performance and service life. In this study, the Finite Element Method-Discrete Element Method (FEM-DEM) coupling method was used to deeply explore the mechanical characteristics of concrete pavement corner voids and grouting reinforcement. First, a discrete element model for corner voids was constructed, and uniaxial compression simulations were carried out to calibrate the mesoscopic parameters of the concrete surface layer and the base course. Subsequently, an FEM-DEM coupling model was established to simulate the mechanical responses of the pavement slab under different working conditions. The research found that when there is a corner void, the cracking load is 75 kN, and the peak load is 96.8 kN. After exceeding the peak load, the cracks expand rapidly. When the strain reaches 0.2, the crack growth slows down. The displacement expands in a triangular shape, and the failure mode is shear failure. After grouting reinforcement, the peak strength increases by 53% to 150 kN, the cracking pressure is 75 kN, the cracks expand rapidly first and then stabilize, and the failure form is rectangular. In addition, the load transfer, crack distribution, force chain distribution, etc., show different laws in the void and reinforced states. The FEMDEM model overcomes traditional numerical limits, precisely simulating structure-void interactions and reinforcing mechanics. It fills a mesomacro research gap, offering new insights for pavement engineering and supporting corner void grouting treatments.OPEN ACCESS Received: 20/05/2025 Accepted: 18/07/2025 Published: 15/12/2025

  Abstract
  Cement concrete pavements have significant advantages in the construction of transportation infrastructure. However, the disease of slab bottom voids affects their performance [...]

  • 12
  •  read

• Study on the Annular Pressure Prediction Method for Ultra-Deep Well Hydraulic Lift Dual-Gradient Drilling

  Y. Chen, J. Yuan, K. Liu, J. Zhu, M. Li, H. Wang, Y. Li

  Rev. int. métodos numér. cálc. diseño ing. (2025). Vol. 41, (4), 87

  
  

  Abstract

  With the ongoing advancement of oil and gas exploration into deep and ultra-deep formations in China, precise wellbore pressure control under complex geological conditions has become a critical technical challenge for safe drilling operations. To overcome the limitations of existing dual-gradient drilling (DGD) technologies—particularly their poor applicability and limited pressure regulation capability in land-based drilling—this study introduces an innovative hydraulic-lift dual-gradient drilling annular flow model, tailored for ultra-deep vertical wells. The model accounts for solid–liquid phase separation flow characteristics and the hydraulic-lift effect of downhole dual-gradient pumps. The Stability Enhancing Two-Step (SETS) method is employed to solve the strongly nonlinear, coupled governing equations, significantly improving computational stability and efficiency. Experimental validation reveals that the model’s predicted pressure distribution closely matches measured data, with a maximum average error of only 16.4%, confirming the model’s accuracy and applicability. Additionally, this study systematically analyzes the impact of key parameters—such as drilling fluid flow rate, viscosity, lift pump speed, and the number of pump sections—on bottomhole pressure regulation, providing valuable insights into their influence on annular pressure behavior. The findings offer a solid theoretical foundation for optimizing drilling parameters and ensuring safe, efficient drilling in ultra-deep wells under challenging geological conditions.OPEN ACCESS Received: 16/07/2025 Accepted: 19/08/2025

  Abstract
  With the ongoing advancement of oil and gas exploration into deep and ultra-deep formations in China, precise wellbore pressure control under complex geological conditions [...]

  • 40
  •  read

• Freshwater Snail Optimizer: A Bio-Inspired Optimizer for Engineering Design Problems

  G. Liu, J. Pan, S. Chu, R. Wang, B. Yang, Y. Yu, S. Mirjalili

  Rev. int. métodos numér. cálc. diseño ing. (2025). Vol. 41, (4), 84

  
  

  Abstract

  As the scale and nonlinearity of optimization problems continue to increase, traditional deterministic solution strategies are becoming increasingly flawed in the face of the exponential growth of search space dimensions and multimodal objective functions. Metaheuristic algorithms, with their probability-driven global search capabilities and local development capabilities, have gradually become an essential tool for solving complex optimization tasks. We propose a Freshwater Snail Optimizer (FSO), inspired by the social behavior of water snails in terms of movement and collision, as a metaheuristic algorithm. FSO combines the floating of water snails’ air chambers, movement in the water, and population collisions and divides them into groups during initialization to balance exploration and development, achieving gratifying optimization results, especially in high-dimensional problems. We utilized CEC 2017 and CEC 2022 to qualitatively analyze FSO in various problems, and employed the Friedman test and Wilcoxon rank sum test for statistical testing. Experimental results show that our proposed FSO achieved 32 first-place results on 41 problems compared with 9 classic algorithms, and 27 first-place results when compared with 9 emerging algorithms that appeared in the past two years. FSO has also achieved the first comparison results in six engineering optimizations on multiple occasions, proving that FSO possesses well optimization capabilities and practicality for real-world problems. The source code accompanying this article has been released at:https://github.com/leogalaxy0603/Freshwater-SnailOptimizer(accessed on 12 October 2025).OPEN ACCESS Received: 19/08/2025 Accepted: 13/10/2025

  Abstract
  As the scale and nonlinearity of optimization problems continue to increase, traditional deterministic solution strategies are becoming increasingly flawed in the face of [...]

  • 238
  •  read

• Construction and Orthogonality of Fractional Laguerre Functions via the Caputo Derivative

  K. Abuasbeh, M. Awadalla, A. Sharif, M. Balti

  Rev. int. métodos numér. cálc. diseño ing. (2025). Vol. 41, (4), 79

  
  

  Abstract

  This paper presents a rigorous framework for generalizing Laguerre polynomials to the fractional domain using the Caputo derivative. We solve the resulting fractional Laguerre differential equation via the power series method, deriving an explicit formfor the fractional Laguerre functions. A key contribution is the identification of a novel weight function, wα(x) = x−(2α−1)e−x, which is essential to prove the orthogonality of these functions over the interval [0,∞). Comprehensive numerical validation is provided, confirming the theoretical orthogonality across a wide range of fractional orders α and demonstrating a clean reduction to the classical polynomials when α = 2. An analysis of computational feasibility confirms the practical applicability of these functions for solving fractional differential equations and other applied problems.

  Abstract
  This paper presents a rigorous framework for generalizing Laguerre polynomials to the fractional domain using the Caputo derivative. We solve the resulting fractional Laguerre [...]

  • 55
  •  read

• A Review of Earthquake Landslide Hazard Assessment Methods

  F. Xiong, J. Sun, S. Han

  Rev. int. métodos numér. cálc. diseño ing. (2025). Vol. 41, (4), 75

  
  

  Abstract

  Landslides triggered by earthquakes are large in scale and wide in scope, making them one of the most serious geological disasters. Earthquake landslide hazard assessment has become an important part of disaster reduction and prevention work. Based on existing research and assessment practices, such an assessment is divided into two levels: individual landslide assessment and regional landslide assessment. The individual assessment, mainly required by specific engineering seismic issues, serves as the foundation of earthquake landslide hazard assessment. It includes two analysis methods: qualitative analysis based on causal relationships (e.g., comprehensive indicator modeling, logistic regression, neural network modeling, information quantity evaluation) and mechanical analysis based on physical-mechanical mechanisms (e.g., quasi-static method, Newmark method, dynamic time-history method). This paper summarizes the characteristics and problems of these two methods. Regional assessment caters to regional strong earthquake geological disaster rescue deployment, future earthquake defense planning, and engineering construction strategic layout. It has two strategies—“from region to individual” (earthquake-focused, coarse-to-fine) and “from individual to region” (landslide-focused, point-to-area)—which differ in observation angles and technical routes. Currently, the individual assessment can estimate landslide hazard probability by considering potential seismic source ori-entations, but the regional assessment lags, e.g., ignoring such orientations and lacking the application of the dynamic time-history method. Thus, this paper proposes establishing slope seismic resistance fields and multi-azimuth seismic impact fields, then overlaying them to determine regional earth-quake landslide distribution probability, and points out future research directions.

  Abstract
  Landslides triggered by earthquakes are large in scale and wide in scope, making them one of the most serious geological disasters. Earthquake landslide hazard assessment [...]

  • 59
  •  read