Computational Accuracy Research Articles

Mechanism of the Non-Kasha Fluorescence in Pyrene.

The high-energy shoulder in the gas-phase fluorescence emission spectrum of pyrene is a well-known example of non-Kasha emission. We comparatively assess two approaches, vibronic perturbation theory and nonadiabatic dynamics, in their ability to predict and explain the gas-phase fluorescence spectrum of pyrene. While both methods qualitatively capture the non-Kasha emission, they differ in their computational requirements, accuracy, and physical interpretation. Vibronic perturbation theory and nonadiabatic dynamics are complementary and can be combined in a two-step approach to non-Kasha fluorescence.

How to correct Ehrenfest nonadiabatic dynamics in open quantum systems: Ehrenfest plus random force (E + σ) dynamics.

One key challenge in the study of nonadiabatic dynamics in open quantum systems is to balance computational efficiency and accuracy. Although Ehrenfest dynamics (ED) is computationally efficient and well-suited for large complex systems, ED often yields inaccurate results. To address these limitations, we improve the accuracy of the traditional ED by adding a random force (E + σ). In this work, the construction of random forces is considered in Markovian and non-Markovian scenarios, and we ensure the dynamics satisfy the detailed balance in both scenarios. By comparing our E + σ with existing methods such as the electronic friction model and surface hopping, we furthermore validate its reliability. In addition, the E + σ model still retains the high efficiency of ED and does not incur much additional computation. We believe that this method provides an alternative to accurately describe the mixed quantum-classical dynamics in open quantum systems, particularly for large complex systems.

Artificial Intelligence-Empowered Radiology—Current Status and Critical Review

Humanity stands at a pivotal moment of technological revolution, with artificial intelligence (AI) reshaping fields traditionally reliant on human cognitive abilities. This transition, driven by advancements in artificial neural networks, has transformed data processing and evaluation, creating opportunities for addressing complex and time-consuming tasks with AI solutions. Convolutional networks (CNNs) and the adoption of GPU technology have already revolutionized image recognition by enhancing computational efficiency and accuracy. In radiology, AI applications are particularly valuable for tasks involving pattern detection and classification; for example, AI tools have enhanced diagnostic accuracy and efficiency in detecting abnormalities across imaging modalities through automated feature extraction. Our analysis reveals that neuroimaging and chest imaging, as well as CT and MRI modalities, are the primary focus areas for AI products, reflecting their high clinical demand and complexity. AI tools are also used to target high-prevalence diseases, such as lung cancer, stroke, and breast cancer, underscoring AI’s alignment with impactful diagnostic needs. The regulatory landscape is a critical factor in AI product development, with the majority of products certified under the Medical Device Directive (MDD) and Medical Device Regulation (MDR) in Class IIa or Class I categories, indicating compliance with moderate-risk standards. A rapid increase in AI product development from 2017 to 2020, peaking in 2020 and followed by recent stabilization and saturation, was identified. In this work, the authors review the advancements in AI-based imaging applications, underscoring AI’s transformative potential for enhanced diagnostic support and focusing on the critical role of CNNs, regulatory challenges, and potential threats to human labor in the field of diagnostic imaging.

Tactical flight optimizer: a novel optimization technique tested on mathematical, mechanical, and structural optimization problems

Abstract The current study presents a novel gradient-free metaheuristic search algorithm named Tactical Flight Optimizer (TFO), tailored to meet the growing need for high-performance optimization techniques in solving complex engineering and mathematical problems. The main contribution of this study is the development of a method that simulates tactical air combat formations, offering a sophisticated alternative to conventional search algorithms. In the proposed method, the location of each agent is updated based on a resultant vector derived from three updating vectors. The updating vectors incorporate total information stored by the agents in each iteration. Consequently, the navigation process is guided by a more logical mechanism rather than a simple random process. The search performance of the TFO is initially benchmarked by solving a set of constrained mathematical functions. Subsequently, it is evaluated by addressing a suite of constrained mechanical and structural problems, containing both discrete and continuous decision variables. The obtained results are compared with five other well-stablished metaheuristic techniques. Acquired numerical results indicate that the TFO algorithm can provide promising results in solving both mathematical and engineering problems in terms of computational cost, accuracy, and stability.

Deep learning-based synthetic CT for dosimetric monitoring of combined conventional radiotherapy and lattice boost in large lung tumors

PurposeConventional radiotherapy (CRT) has limited local control and poses a high risk of severe toxicity in large lung tumors. This study aimed to develop an integrated treatment plan that combines CRT with lattice boost radiotherapy (LRT) and monitors its dosimetric characteristics.MethodsThis study employed cone-beam computed tomography from 115 lung cancer patients to develop a U-Net + + deep learning model for generating synthetic CT (sCT). The clinical feasibility of sCT was thoroughly evaluated in terms of image clarity, Hounsfield Unit (HU) consistency, and computational accuracy. For large lung tumors, accumulated doses to the gross tumor volume (GTV) and organs at risk (OARs) during 20 fractions of CRT were precisely monitored using matrices derived from the deformable registration of sCT and planning CT (pCT). Additionally, for patients with minimal tumor shrinkage during CRT, an sCT-based adaptive LRT boost plan was introduced, with its dosimetric properties, treatment safety in high dose regions, and delivery accuracy quantitatively assessed.ResultsThe image quality and HU consistency of sCT improved significantly, with dose deviations ranging from 0.15% to 1.25%. These results indicated that sCT is feasible for inter-fraction dose monitoring and adaptive planning. After rigid and hybrid deformable registration of sCT and pCT, the mean distance-to-agreement was 0.80 ± 0.18 mm, and the mean Dice similarity coefficient was 0.97 ± 0.01. Monitoring dose accumulation over 20 CRT fractions showed an increase in high-dose regions of the GTV (P < 0.05) and a reduction in low-dose regions (P < 0.05). Dosimetric parameters of all OARs were significantly higher than those in the original treatment plan (P < 0.01). The sCT based adaptive LRT boost plan, when combined with CRT, significantly reduced the dose to OARs compared to CRT alone (P < 0.05). In LRT plan, high-dose regions for the GTV and D95% exhibited displacements greater than 5 mm from the tumor boundary in 19 randomly scanned sCT sequences under free breathing conditions. Validation of dose delivery using TLD phantom measurements showed that more than half of the dose points in the sCT based LRT plan had deviations below 2%, with a maximum deviation of 5.89%.ConclusionsThe sCT generated by the U-Net + + model enhanced the accuracy of monitoring the actual accumulated dose, thereby facilitating the evaluation of therapeutic efficacy and toxicity. Additionally, the sCT-based LRT boost plan, combined with CRT, further minimized the dose delivered to OARs while ensuring safe and precise treatment delivery.

Research and Application of ROM Based on Res-PINNs Neural Network in Fluid System

In the design of fluid systems, rapid iteration and simulation verification are essential, and reduced-order modeling techniques can significantly improve computational efficiency and accuracy. However, traditional Physics-Informed Neural Networks (PINNs) often face challenges such as vanishing or exploding gradients when learning flow field characteristics, limiting their ability to capture complex fluid dynamics. This study presents an enhanced reduced-order model (ROM): Physics-Informed Neural Networks based on Residual Networks (Res-PINNs). By integrating a Residual Network (ResNet) module into the PINN architecture, the proposed model improves training stability while preserving physical constraints. Additionally, the model’s ability to capture and learn flow field states is further enhanced by the design of a symmetric parallel neural network structure. To evaluate the effectiveness of the Res-PINNs model, two classic fluid dynamics problems—flow around a cylinder and Vortex-Induced Vibration (VIV)—were selected for comparative testing. The results demonstrate that the Res-PINNs model not only reconstructs flow field states with higher accuracy but also effectively addresses limitations of traditional PINN methods, such as vanishing gradients, exploding gradients, and insufficient learning capacity. Compared to existing approaches, the proposed Res-PINNs provide a more stable and efficient solution for deep learning-based reduced-order modeling in fluid system design.

A Novel Approach for Solving Fractional Differential Equations via a Multistage Telescoping Decomposition Method

This study presents a novel algorithm for solving nonlinear fractional initial value problems, utilizing a multistage telescoping decomposition method (FTDM). By combining the MFTDM with the Elzaki transform, this method significantly improves computational efficiency and accuracy. Through a series of numerical experiments, the proposed approach is demonstrated to be highly practical, reliable and straightforward, offering a robust framework for solving nonlinear fractional differential equations effectively.

Review and novel formulae for transmittance and reflectance of wedged thin films on absorbing substrates

Abstract Historically, spectroscopic techniques have been essential for studying the optical properties of thin solid films. However, existing formulae for both normal transmission and reflection spectroscopy often rely on simplified theoretical assumptions, which may not accurately align with real-world conditions. For instance, it is common to assume (1) that the thin solid layers are deposited on completely transparent thick substrates and (2) that the film surface forms a specular plane with a relatively small wedge angle. While recent studies have addressed these assumptions separately, this work presents an integrated framework that eliminates both assumptions simultaneously. In addition, the current work presents a deep review of various formulae from the literature, each with their corresponding levels of complexity. Our review analysis highlights a critical trade-off between computational complexity and expression accuracy, where the newly developed formulae offer enhanced accuracy at the expense of increased computational time. Our user-friendly code, which includes several classical transmittance and reflectance formulae from the literature and our newly proposed expressions, is publicly available in both Python and Matlab at this link.

Dynamical structure factors of warm dense matter from time-dependent orbital-free and mixed-stochastic-deterministic density functional theory

Abstract We present the first calculations of the inelastic part of the dynamical structure factor (DSF) for warm dense matter (WDM) using Time-Dependent Orbital-Free Density Functional Theory (TD-OF-DFT) and Mixed-Stochastic-Deterministic (mixed) Kohn Sham TD-DFT (KS TD-DFT). WDM is an intermediate phase of matter found in planetary cores and laser-driven experiments, where the accurate calculation of the DSF is critical for interpreting X-ray Thomson scattering (XRTS) measurements. Traditional TD-DFT methods, while highly accurate, are computationally expensive, motivating the exploration of TD-OF-DFT and mixed TD-KS-DFT as more efficient alternatives. We applied these methods to experimentally measured WDM systems, including solid-density aluminum and beryllium, compressed beryllium, and carbon-hydrogen mixtures. Our results show that TD-OF-DFT requires a dynamical kinetic energy potential in order to qualitatively capture the plasmon response. Additionally, it struggles with capturing bound electron contributions and accurately modeling plasmon dynamics without the inclusion of a dynamic kinetic energy potential. In contrast, mixed TD-KS-DFT offers greater accuracy in distinguishing bound and free electron effects, aligning well with experimental data, though at a higher computational cost. This study highlights the trade-offs between computational efficiency and accuracy, demonstrating that TD-OF-DFT remains a valuable tool for rapid scans of parameter space, while mixed TD-KS-DFT should be preferred for high-fidelity simulations. Our findings provide insight into the future development of DFT methods for WDM and suggest potential improvements for TD-OF-DFT.&amp;#xD;

Gaussian Process Regression with Soft Equality Constraints

This study introduces a novel Gaussian process (GP) regression framework that probabilistically enforces physical constraints, with a particular focus on equality conditions. The GP model is trained using the quantum-inspired Hamiltonian Monte Carlo (QHMC) algorithm, which efficiently samples from a wide range of distributions by allowing a particle’s mass matrix to vary according to a probability distribution. By integrating QHMC into the GP regression with probabilistic handling of the constraints, this approach balances the computational cost and accuracy in the resulting GP model, as the probabilistic nature of the method contributes to shorter execution times compared with existing GP-based approaches. Additionally, we introduce an adaptive learning algorithm to optimize the selection of constraint locations to further enhance the flexibility of the method. We demonstrate the effectiveness and robustness of our algorithm on synthetic examples, including 2-dimensional and 10-dimensional GP models under noisy conditions, as well as a practical application involving the reconstruction of a sparsely observed steady-state heat transport problem. The proposed approach reduces the posterior variance in the resulting model, achieving stable and accurate sampling results across all test cases while maintaining computational efficiency.

A Fully Explicit–Explicit Staggered Algorithm for the Dynamic Response of Fluid-Saturated Porous Media

The analysis of the dynamic response of fluid-saturated porous media under stress waves is of great practical value in the fields of geotechnical engineering, geophysics, earthquake engineering and marine engineering. In this research, a new fully explicit finite element algorithm for solving the coupled soil-pore fluid dynamic problem is developed based on the u–p formulation. The proposed method is a staggered algorithm that is realized with the central difference method used to solve the displacement u and the improved Euler method used to solve the fluid pressure p, it has a second-order theoretical accuracy since both the central difference method and the improved Euler method are second-order methods. The correctness and efficiency of the proposed method are investigated through two numerical examples. The computational results of the one-dimensional soil column model shows that the numerical solution of the proposed algorithm agree well with the theoretical analytical solution. The differences of the computational accuracy and efficiency between the proposed method and the other existing methods are also discussed in the numerical examples, and the results show that the proposed explicit method is higher in calculation efficiency than the implicit algorithm, and its accuracy is higher than that of the existing first-order explicit method.

Accelerating Quantum Anharmonic Vibrational Calculations by Atom-Specific Hybrid Basis Set-Based Potential Energy Surface Approach.

The development of accurate yet fast quantum mechanical methods to calculate the anharmonic vibrational spectra of large molecules is one of the major goals of ongoing developments in this field. This study extensively explores and validates a hybrid electronic basis set approach for anharmonic vibrational calculations, where the molecule is segregated into different computational layers, and such layers are then treated with different levels of electronic basis sets. Following the system-bath model, the atoms corresponding to the active sites are treated in more accurate but computationally slower, large basis set and the rest of the atoms in less accurate but computationally faster, small basis set to construct the anharmonic hybrid potential energy surface (PES). Such a hybrid protocol for constructing an anharmonic PES is named as the atom-specific hybrid basis set (ASHBS) approach. The accuracy of the ASHBS approach is tested and established by evaluating the harmonic and anharmonic frequencies of a set of four prototype molecules. Following the ASHBS approach, for a chosen active site, the transitions of the corresponding modes are found to be closer to that of a high basis set, with a majority of target modes displaying a mean absolute error of around 3.3 cm-1, while achieving the computational acceleration of 2-3 times. This study also provides insights into determining the optimal layer size to balance computational efficiency and accuracy, offering a suitable alternative, particularly for large molecules.

COAST-PROSIM: A Model for Predicting Shoreline Evolution and Assessing the Impacts of Coastal Defence Structures

Coastal zones, at the interface between land and sea, face increasing challenges from erosion, sea-level rise, and anthropogenic interventions, necessitating innovative tools for effective management and protection. This study introduces COAST-PROSIM, a novel numerical model specifically designed to predict shoreline evolution and assess the impacts of coastal defence structures on coastal morphology. Unlike existing models that often face a trade-off between computational efficiency and physical accuracy, COAST-PROSIM balances these demands by integrating two-dimensional wave propagation routines with advanced shoreline evolution equations. The model evaluates the effects of interventions such as breakwaters and groynes, enabling simulations of shoreline dynamics with reduced computational effort. By using high-resolution input data, COAST-PROSIM captures the interplay between hydrodynamics, sediment transport, and structural impacts. Tested on real-world case studies along the coasts of San Leone, Porto Empedocle, and Villafranca Tirrena, the model demonstrates its adaptability to diverse coastal environments. The results highlight its potential as a reliable tool for sustainable coastal management, allowing stakeholders to anticipate long-term changes in coastal morphology and design targeted mitigation strategies.

Development of a Modular Virtual Calibration Platform for Power Machinery

The hardware-in-the-loop (HIL)-based virtual calibration technology for power machinery significantly enhances the efficiency and flexibility of electronic controller calibration, reducing dependency on physical experiments and lowering development costs. This paper presents an independently developed modular virtual calibration platform, designed based on the ISO/IEC 42010 architecture approach and the V-Model development process. The platform adopts a modular layered architecture comprising the physical layer, communication layer, model layer, software layer, and application layer. Each layer is independently developed and seamlessly integrated, ensuring excellent scalability and maintainability. Through performance validation conducted under steady-state and transient operating conditions of the diesel/natural gas dual-fuel engine electronic control unit (ECU), experimental results demonstrated that the platform possesses high computational accuracy, rapid response capability, and stable operation across different scenarios. The experiments validated that the platform can effectively replicate the real behavior of the controlled units in a virtual environment, confirming its adaptability and reliability. The platform offers an efficient and reliable solution for power machinery controller calibration.

Fraud App Detection using Machine Learning Algorithms

Abstract - The increasing number of mobile applications has resulted in a rise in malicious software, including fraud apps that take advantage of user permissions and system vulnerabilities to compromise sensitive personal and financial data. This paper introduces a comprehensive fraud detection system that leverages machine learning techniques to classify mobile applications as either benign or malicious. The system is implemented as a web-based platform using Flask, enabling users to upload, analyse, and classify APK (Android Package) files efficiently. Important features, such as application permissions and metadata (including app name, target SDK version, and file size), are extracted from APK files and transformed into feature vectors for classification by two machine learning models: an Artificial Neural Network (ANN) and a Support Vector Classifier (SVC). The ANN model achieves a classification accuracy of 92.26%, while the SVC model reaches an accuracy of 89%. To improve model performance further, a Genetic Algorithm (GA) is used for feature selection, which reduces the number of features and enhances both the computational efficiency and predictive accuracy of the models. The system offers an intuitive user interface that allows users to interact with the detection models, preview datasets, select classification algorithms, and view in-depth results, including safety recommendations for uploaded APK files. Additionally, the system provides visualizations of performance metrics and highlights the importance of specific features, improving the interpretability and transparency of the decision-making process.

Ply Optimization of Composite Laminates for Processing-Induced Deformation and Buckling Eigenvalues Based on Improved Genetic Algorithm.

The structure of thermoset composite laminated plates is made by stacking layers of plies with different fiber orientations. Similarly, the stiffened panel structure is assembled from components with varying ply configurations, resulting in thermal residual stresses and processing-induced deformations (PIDs) during manufacturing. To mitigate the residual stresses caused by the geometric features of corner structures and the mismatch between the stiffener-skin ply orientations, which lead to PIDs in composite-stiffened panels, this study proposes a multi-objective stacking optimization strategy based on an improved adaptive genetic algorithm (IAGA). The viscoelastic constitutive model was employed to describe the modulus variation during the curing process to ensure computational accuracy. In this study, the IAGA was proposed to optimize the ply-stacking sequence of L-shaped stiffeners in composite laminated structures. The results demonstrate a reduction in the spring-in angle to 0.12°, a 50% improvement compared to symmetric balanced stacking designs, while the buckling eigenvalues were improved by 20%. Additionally, the IAGA outperformed the traditional non-dominated sorting genetic algorithm (NSGA), achieving a threefold increase in the Pareto solution diversity under identical constraints and reducing the convergence time by 70%. These findings validate the effectiveness of asymmetric ply design and provide a robust framework for enhancing the structural performance and manufacturability of composite laminates.

Comparative Aerodynamic Analysis and Parallel Performance of 2D CFD Simulations of a VAWT Using Sliding Mesh Interface Method

ABSTRACTWith rapid advancements in computer hardware and numerical modeling methods, Computational Fluid Dynamics (CFD) has gained prominence in simulating complex flows. As parallel computation becomes an industry standard, the computational efficiency of simulations has become critical. The flow around a Vertical Axis Wind Turbine (VAWT), characterized by complex dynamics and challenging rotating geometry, serves as an intriguing case for CFD studies. This study employs the open‐source CFD solvers SU2 and OpenFOAM to simulate the incompressible, unsteady, and turbulent flow around an H‐type Darrieus VAWT in two dimensions. Spatial and temporal discretization parameters are examined to balance computational cost and accuracy, revealing notable effects on power predictions. Simulations conducted under identical conditions allow for a comparison of the predictions and parallel performances of SU2 and OpenFOAM across three distinct tip speed ratios (TSRs). The findings show that discretization parameters behave differently at various TSRs. While power predictions from SU2 and OpenFOAM generally align with experimental data and with each other, discrepancies arise at lower TSRs, with thrust predictions showing better consistency. Although OpenFOAM provides a faster solution across all parallel configurations, SU2 demonstrates superior parallel scalability, achieving higher speedup and efficiency.

Deep-TEMNet: A Hybrid U-Net–2D LSTM Network for Efficient and Accurate 2.5D Transient Electromagnetic Forward Modeling

The transient electromagnetic (TEM) method is a crucial tool for subsurface exploration, providing essential insights into the electrical resistivity structures beneath the Earth’s surface. Traditional forward modeling approaches, such as the finite-difference time-domain (FDTD) method and the finite-element method (FEM), are computationally intensive, limiting their practicality for real-time, high-resolution, or large-scale investigations. To address these challenges, we present Deep-TEMNet, an advanced deep learning framework specifically designed for two-dimensional TEM forward modeling. Deep-TEMNet integrates the U-Net architecture with a tailored two-dimensional long short-term memory (2D LSTM) module, allowing it to effectively capture complex spatial-temporal relationships in TEM data. The U-Net component enables high-resolution spatial feature extraction, while the 2D LSTM module enhances temporal modeling by processing spatial sequences in two dimensions, thereby optimizing the representation of electromagnetic field dynamics over time. Trained on high-fidelity FEM-generated datasets, Deep-TEMNet achieves exceptional accuracy in reproducing electromagnetic field distributions across diverse geological scenarios, with a mean squared error of 0.00000134 and a root mean square percentage error of 0.002373019. The framework offers over 150 times the computational speed of traditional FEMs, with an average inference time of just 3.26 s. Extensive validation across varied geological conditions highlights Deep-TEMNet’s robustness and adaptability, establishing its potential for efficient, large-scale subsurface mapping and real-time data processing. By combining U-Net’s spatial resolution capabilities with the sequential processing strength of the 2D LSTM module, Deep-TEMNet significantly advances computational efficiency and accuracy, positioning it as a valuable tool for geophysical exploration, environmental monitoring, and other applications requiring scalable, real-time TEM analyses that are easily integrated into remote sensing workflows.

Enhanced accuracy in solar irradiance forecasting through machine learning stack-based ensemble approach

ABSTRACT Accurate solar irradiance (SI) prediction is vital for optimizing solar photovoltaic systems. This study addresses shortcomings in existing forecasting methods by exploring advanced machine-learning techniques using meteorological satellite data. We develop three novel models for SI forecasting: Stack-based Ensemble Fusion with Meta-Neural Network (SEFMNN), Extreme Gradient Boosting-Squared Error (XGB-SE), and Extreme Learning Machine (ELM). These models predict All-sky and Clear-sky shortwave solar irradiance across three Chinese provinces (Guangdong, Shandong, and Zhejiang) and one Saudi Arabian province (Najran). The SEFMNN model combines Artificial Neural Network (ANN), Random Forest (RF), and Support Vector Machine (SVM) to improve prediction accuracy. The XGB-SE model employs a specialized loss function to manage extreme values in historical data. These models are designed to mitigate overfitting and data inconsistency while balancing computational efficiency and predictive accuracy. Comparative analysis reveals that SEFMNN and XGB-SE outperform the ELM model, with SEFMNN achieving an R2 of 0.9979, MAE of 0.0231, and MSE of 0.0020 in Najran. This demonstrates that SEFMNN significantly enhances solar irradiance forecasting, aiding efficient solar photovoltaic system planning and operation.

A hybrid harmonic polynomial cell and STF strip theory method for marine hydrodynamics and ship motion analysis

ABSTRACT This paper presents a hybrid method that combines the Harmonic Polynomial Cell (HPC) method and STF strip theory to solve marine hydrodynamics and ship motion problems. The efficient and precise HPC method is extended to three-dimensional ship hydrodynamics using STF strip theory. A new single-node hybridisation condition is introduced to enhance the stability of hydrodynamic predictions. To improve computational accuracy for complex hull shapes, automatic structured grid generation and mesh refinement techniques are applied. The hybrid HPC-STF method is then used to solve 3D ship radiation and diffraction problems in regular waves with forward speed. The method’s performance is validated through comparisons of hydrodynamic coefficients, wave forces, and motion response amplitude operators (RAOs) against experimental data and results from other numerical methods. The findings demonstrate that this hybrid approach provides accurate and effective solutions for ship hydrodynamics, making it a valuable tool for analysing marine structures and motions.