Calculation Module Research Articles

Thermodynamic model of coal direct chemical looping combustion

ABSTRACT Chemical looping combustion (CLC) technology offers cost-effective CO2 emission reduction. Coal direct chemical looping (CDCL) emerges as a promising solid fuel CLC technology owing to its ability to directly utilize coal without prior gasification, enhancing system integration and efficiency. Despite significant progress in CDCL, challenges persist, including carbon deposition and insufficient oxidation and reduction of oxygen carriers (OCs). This study conducted a thermodynamic equilibrium analysis of a CDCL combustion reactor using Cantera 3.0, a chemical calculation module developed at the California Institute of Technology. The analysis explored the impact of different temperatures, OC-to-coal molar ratios (θ), and OCs types on combustion products. By employing the minimum Gibbs function method and analyzing intrinsic reaction mechanisms, the study provides insights for predicting and optimizing system performance. Thermodynamic equilibrium analysis revealed that with CuO as the OC, insufficient OC (θ < 0.6) led to Cu as the reduction product, while sufficient OC (0.6 < θ < 1.3) yielded Cu and CuO, and excess OC (θ > 1.3) resulted in Cu2O. To optimize CO2 capture, a molar ratio of θ of at least 1.4 and a temperature of at least 800°C are recommended. When Fe2O3 served as the OC, insufficient OC (θ < 0.45) produced Fe and FeO, while excess OC generated FeO and Fe3O4. Optimal operational performance and CO2 capture efficiency required an OC-to-coal molar ratio exceeding 1.4 and a temperature higher than 700°C. This study demonstrates that in CDCL combustion systems, selecting appropriate OCs, temperatures, and OC-to-coal molar ratios ensures stable and efficient combustion and CO2 capture. These findings offer crucial guidance for optimizing system operation.

Analysis and selection of the structure of a multiprocessor computing system according to the performance criterion

Objectives. Analysis of the various architectures of computing systems (CSs) used in recent decades has allowed us to identify the most common structures. One of the key features is the use of mass-produced equipment to create data processing subsystems (for example, multicore processors and high-capacity semiconductor memory), as well as network equipment to build communication subsystems. This reduces hardware costs and allows typical or cluster configurations to be created, which is especially important for expensive CSs. The desire to achieve high computational speed and performance in such CSs requires minimizing the time to complete the task and balancing time delays both in data processing subsystems and in the communication subsystem which provides data transmission inside the CS. The aim of this work is to analyze computing modules (CMs) and structures on the basis of which the construction of cluster CSs is carried out.Methods. The main results of the work were obtained using methods of mathematical analysis and modeling.Results. The study considers the structure of modern multicore microprocessors as the basis for building CMs of cluster CSs. As the number of cores in the microprocessor structure increases, the communication network which unites them into a single structure becomes more complicated. It has been shown that in new developments of microprocessors, communication between cores is performed in the form of a network. The microprocessors themselves are MIMD structures in accordance with the well-known Flynn classification.Conclusions. The proposed method of selecting an effective structure of a CS allows us to obtain the optimal structure of a CS according to the criterion of performance.

Network traffic anomaly detection model based on feature grouping and multi‐autoencoders integration

AbstractThis paper presents a network traffic anomaly detection model based on feature grouping and multiple autoencoders (multi‐AEs) integration. This model comprises four modules: feature grouping module, feature learning module, AUC and optimal threshold calculation module, and anomaly detection application module. In the feature grouping module, multiple group features are constructed by selecting the different features according to their attributes and variances. In the feature learning module, the group features of normal traffic are learned based on multi‐AEs. In the AUC and optimal threshold calculation module, the AUC of each AE is calculated according to the ROC curve of the verification data, and the optimal thresholds for each AE are determined using the Youden index. In the anomaly detection application module, the AEs that participated in fusion are selected and their weights are calculated by analysing AUC value, and the scores of unknown traffic in each AE are evaluated considering both the reconstruction error distribution and the optimal threshold. Finally, the anomaly detection result can be obtained by the fusion of these multiple scores. Through validation on the UNSW‐NB15 and CICIDS2017 datasets, the accuracy of the proposed model is improved by 12.04% and 10.52%, respectively, compared to the baseline model.

Recursive prototypical network with coordinate attention: A model for few-shot cross-condition bearing fault diagnosis

In a practical industrial scenario, the variability in bearing operating conditions complicates the collection of a sufficient number of labeled samples, thereby limiting the effectiveness of traditional deep learning-based fault diagnosis methods. In addition, the influence of abnormal samples on the prototype features also severely limits the performance of prototypical network in few-shot fault diagnosis. To address the above issues, a recursive prototypical network based on meta-learning is proposed for few-shot cross-condition bearing fault diagnosis. Firstly, a feature extractor with coordinate attention mechanism is developed, which is able to deeply extract effective features in complex vibration signals. Furthermore, a recursive prototype computation module is introduced to alleviate prototype bias arising from abnormal samples, thereby achieving a more precise representation of prototypes. Finally, a metric module is utilized to obtain the similarity between the prototypes and the query set samples to achieve an accurate classification of faults. To verify the efficacy and superiority of the proposed method, its performance was evaluated on two bearing vibration datasets. The experimental results demonstrated that the method is significantly better than other deep learning methods with high accuracy and generalization, and greater suitability for few-shot cross-condition bearing fault diagnosis tasks.

The Study of Radioactive Fallout Source of Low-Equivalent Nuclear Bursts Based on Nuclear Cloud Simulation Using the CFD-DPM

The activity-height distribution of radioactive particles in the stabilization cloud of a nuclear burst plays a crucial role in the radioactive fallout prediction model, serving as the source for transport, diffusion, and dose rate calculation modules. A gas-particle multiphase flow solver was developed using the OpenFOAM Computational Fluid Dynamics (CFD) library and discrete phase method (DPM) library under a two-way coupling regime to simulate the U.S. standard atmosphere of 1976 with good stability. The accuracy of the numerical model was verified through low-equivalent nuclear weapons tests, including RANGER-Able and BUSTER-JANGLE-Sugar, depicting reasonable spatio-temporal changes in cloud profiles. The initialization module of the Defense Land Fallout Interpretative Code (DELFIC) and activity-size distribution, which considered fractionation, were employed for nuclear fireball and radioactive particle initialization. Simulations indicated that the activity-height distribution of the stabilization cloud mainly concentrated on the lower third of air burst cloud caps, while settling near the burst center for surface or near-surface bursts. This study has confirmed the effectiveness of the gas-particle flow solver based on the CFD-DPM method in simulating low-equivalent nuclear clouds and enriching research on radioactive fallout prediction models.

A Comprehensive Framework for Transportation Infrastructure Digitalization: TJYRoad-Net for Enhanced Point Cloud Segmentation.

This research introduces a cutting-edge approach to traffic infrastructure digitization, integrating UAV oblique photography with LiDAR point clouds for high-precision, lightweight 3D road modeling. The proposed method addresses the challenge of accurately capturing the current state of infrastructure while minimizing redundancy and optimizing computational efficiency. A key innovation is the development of the TJYRoad-Net model, which achieves over 85% mIoU segmentation accuracy by including a traffic feature computing (TFC) module composed of three critical components: the Regional Coordinate Encoder (RCE), the Context-Aware Aggregation Unit (CAU), and the Hierarchical Expansion Block. Comparative analysis segments the point clouds into road and non-road categories, achieving centimeter-level registration accuracy with RANSAC and ICP. Two lightweight surface reconstruction techniques are implemented: (1) algorithmic reconstruction, which delivers a 6.3 mm elevation error at 95% confidence in complex intersections, and (2) template matching, which replaces road markings, poles, and vegetation using bounding boxes. These methods ensure accurate results with minimal memory overhead. The optimized 3D models have been successfully applied in driving simulation and traffic flow analysis, providing a practical and scalable solution for real-world infrastructure modeling and analysis. These applications demonstrate the versatility and efficiency of the proposed methods in modern traffic system simulations.

Development of Software for Hydrometallurgical Calculation of Metal Extraction

Hydrometallurgy plays a critical role in the metallurgical industry by providing an efficient method for extracting metals from ores and secondary materials using aqueous solutions. This approach is particularly advantageous for processing low-grade and complex ores, as well as secondary resources that cannot be effectively processed by traditional pyrometallurgical methods. The objective of this study is to develop specialized software to automate and optimize the calculations necessary for metal extraction in hydrometallurgical processes. The software integrates a complete computational framework for automating the various stages of the hydrometallurgical process, including ore composition initialization, total element mass calculations, and the determination of metal concentrations in both metal products (matte) and by-products (slag). The methodology involves the design and implementation of a web-based application using Django for the user interface and MySQL for robust data storage. The computational module, written in Python, automates the mathematical operations required to simulate complex chemical reactions and metal extraction processes. This module supports real-time processing and ensures accurate calculations for each stage of metal extraction, including leaching, extraction, re-extraction, sorption, and electrolysis. The software also features detailed tables and dynamic graphs that allow users to analyze the distribution of valuable metals and assess the influence of key operational parameters on extraction efficiency. The results show that the developed software successfully manages large datasets, enhances the precision of hydrometallurgical calculations, and minimizes the risk of human error. The implementation of the software leads to significant improvements in the economic efficiency of production by optimizing metal recovery rates and reducing operational costs. Additionally, it supports comprehensive process control by providing actionable insights into each stage of extraction, thus improving the quality of the final metal product. Overall, the software represents a significant technological advancement in the field, offering a scalable solution for industries aiming to streamline hydrometallurgical processes.

Research of Large Inflow Angles BEMT-Based Analytical–Numerical Performance Evaluation Model

This paper presents a comprehensive analytical–numerical algorithm constructed for proprotor performance evaluation, focusing on accommodating large inflow angles. The algorithm’s design, range, and analytical features are clarified, indicating its potential to improve performance analysis, particularly for blades with substantial pitch variations. The Stahlhut model has not been validated against the conventional BEMT small-inflow angle methodology. This paper implements a modified Stahlhut model, coupled with the conventional BEMT. Preliminary validations of the model demonstrate promising results, with deviations reduced to −3% to 4% compared to conventional BEMT methods exhibiting deviations as high as 20% to 88% against experimental data for a highly twisted proprotor. The reconsideration of the computational module carries considerable implications for the design and refinement of proprotors, providing alternative analysis methods that could improve operational effectiveness across a range of flight scenarios. Drawing upon the theoretical framework presented by Stahlhut, the algorithm enables a more complex understanding of proprotor dynamics, facilitating accurate predictions of the loads at each blade section. The introduced algorithm emerges as a valuable asset for evaluating proprotor performance during the early stages of design and certification, offering both low computational cost and medium to high reliability.

Mechanistic Simulation of Salt‐Affected Soil‐Plant‐Atmosphere Continuum Dynamics in Seasonally Frozen Regions

AbstractThe salt‐affected Soil‐Plant‐Atmosphere Continuum (SPAC) is a dynamic, interactive system that is particularly complex in seasonally frozen regions where salt transport, precipitation‐dissolution, and soil freeze‐thaw processes play crucial, interrelated roles. Understanding these coupled processes and representing them with mathematical models is critical for effective management of SPAC systems. This study presents a new mechanistic approach and an improved model that integrates a chemical equilibrium module within a mechanistic‐based transport computational module (modified SHAW model). The chemical equilibrium module determines salt precipitation‐dissolution using thermodynamic theory and explains the effects of efflorescence and subflorescence on system dynamics. The model enables simultaneous solutions for heat, water, and salt transport with chemical equilibrium throughout non‐freezing and freezing seasons, as well as plant growth dynamics. Assessment of the model using laboratory experiments and field studies showed good performance, with coefficient of determination values exceeding 0.65 for simulated and measured evaporation rate, leaf area index, soil water content, salt content, and temperature. Furthermore, a comparison between simulation results considering and neglecting the impact of salt precipitation‐dissolution highlights potential inaccuracies in soil heat‐water‐salt dynamics and plant water use resulting from the omission of this process in mechanistic models.

Adaptive Threshold Wave Front Sensing for Portable Adaptive Optics System Driven by Hybrid Parallel Technique

Our Portable Adaptive Optics (PAO) system designed for high-contrast imaging of exoplanets with current 2–4 m class telescopes achieves a correction speed of nearly 1000 Hz, utilizing a Shack–Hartmann Wave Front Sensor (WFS) in a 9 × 9 sub-aperture configuration. As we look towards adapting the PAO system for larger telescopes, an increase in the number of sub-apertures in the WFS and enhanced precision in wave front detection are imperative. Originally programmed in LabVIEW, our initial PAO software is based on a traditional centroid calculation module for nighttime wave front sensing and lacks adaptive processing of background noise. To address these limitations and to boost the PAO system's performance and accuracy in wave front detection, we propose a compressive neural network (Th-Net) combined with a specialized hybrid parallel programming approach for wave front detection. Our experimental results indicate that this hybrid parallel technique and Th-Net significantly enhance the PAO system's operational speed and wave front detection precision under uneven background noise. This work paves the way so that a duplicable and low-cost PAO system can be used for direct imaging of exoplanets with large telescopes.

CME Velocity Field Calculation Model Based on an Unsupervised Transformer Optical Flow Network

The optical flow algorithm (OF) is one the main methods for calculating image velocity field and has many applications in space weather. Most OF calculations are applied to the motion of labeled rigid objects and are not suitable for velocity detection of high-energy particles, such as in a coronal mass ejection (CME). Fluctuations in exposure time and the influence of space weather will lead to inconsistent brightness of the same feature point at different times. To address this problem, we propose an unsupervised multiscale optical flow network based on Vision Transformer, named UTFlowNet. The network comprises a multiscale feature extraction module and a coarse-to-fine global optical flow calculation module. The movement of high-energy particles emitted during a CME eruption follows certain physical rules. Therefore, we apply fluid motion–based loss functions to analyze the motion of high-energy particles more effectively, addressing the problem of CME motion field extraction. Our method can be applied to the real-time automatic extraction of a CME’s velocity field and performs well with inconsistent brightness, large-scale motion, and strong CME noise. Additionally, we can estimate subpixel level fine-grained velocity. Our model may be affected by overfitting during cross–data set inference, so we encourage performing a small amount of transfer learning on new data sets to mitigate this issue. In order to verify the accuracy of our method, we conducted experiments and verification on the Solar and Heliospheric Observatory LASCO C2 data and the High Altitude Observatory MLSO data. We constructed a large-scale displacement simulation data set based on LASCO C2 data and tested on it, achieving the best results.

КОМП'ЮТЕРНИЙ МОДУЛЬ ДЛЯ АНІМАЦІЇ РУХУ ШАТУННО-ПОВЗУНКОВОЇ ГРУПИ ПРОГРАМИ KDAM ПРИ ВИЗНАЧЕННЯ КІНЕМАТИЧНИХ ТА ДИНАМІЧНИХ ПАРАМЕТРІВ

The development of new schemes of the thread feeding system requires an operational assessment of the value of the tension in front of the working zone. The objective function in the tasks of optimizing technological processes is the minimum necessary tension. The development of special computer programs for determining the tension in the working area makes it possible to determine the necessary technological parameters. The variable parameter in the objective function is the sum of the angles of coverage of the working bodies by the thread. The use of a computer program allows you to determine the tension and change in relative tension in the filling zones of technological machines, which allows you to optimize the shape of the thread supply line even at the stage of designing the technological process. A computer module was developed for the animation of the movement of the connecting rod-slider group, which is a component of the KDAM program for determining the kinematic and dynamic parameters of the connecting rod-slider group of the mechanisms of textile and light industry machines, which allows at the preliminary design stage to visually evaluate the behavior of individual links of the connecting rod mechanism of the slider group and determine the coordinates, projections of velocities and accelerations of the center of mass of the connecting rod, the slider and the points of attachment of the links of the Asura groups on the connecting rod. The KDAM program is a structural element of a computer software complex for assessing the intensity of technological processes in the light and textile industry when determining the change in the relative tension of the thread in the filling zones on the technological equipment. Kinematic and dynamic analysis of flat mechanisms plays a significant role in the design of new mechanisms and modernization of existing ones. The results obtained during these studies can be used to calculate the strength of individual links, their inertial characteristics, optimization of structural parameters of mechanisms, and minimization of consumed energy. The computer module of the KDAM program for determining the kinematic and dynamic parameters of the connecting rod-slider group of mechanisms of light industry machines allows you to determine the coordinates, projections of velocities and accelerations of the center of mass of the connecting rod, slider and the point of attachment of the links of the Asura groups on the connecting rod. During the dynamic analysis of the operation of the connecting rod-slider group, the projections of the full reaction in the joints between the crank and the connecting rod, between the connecting rod and the slider, and between the slider and the fixed guide are determined.

UAV Geo-Localization Dataset and Method Based on Cross-View Matching.

The stable flight of drones relies on Global Navigation Satellite Systems (GNSS). However, in complex environments, GNSS signals are prone to interference, leading to flight instability. Inspired by cross-view machine learning, this paper introduces the VDUAV dataset and designs the VRLM network architecture, opening new avenues for cross-view geolocation. First, to address the limitations of traditional datasets with limited scenarios, we propose the VDUAV dataset. By leveraging the virtual-real mapping of latitude and longitude coordinates, we establish a digital twin platform that incorporates 3D models of real-world environments. This platform facilitates the creation of the VDUAV dataset for cross-view drone localization, significantly reducing the cost of dataset production. Second, we introduce a new baseline model for cross-view matching, the Virtual Reality Localization Method (VRLM). The model uses FocalNet as its backbone and extracts multi-scale features from both drone and satellite images through two separate branches. These features are then fused using a Similarity Computation and Feature Fusion (SCFF) module. By applying a weighted fusion of multi-scale features, the model preserves critical distinguishing features in the images, leading to substantial improvements in both processing speed and localization accuracy. Experimental results demonstrate that the VRLM model outperforms FPI on the VDUAV dataset, achieving an accuracy increase to 83.35% on the MA@20 metric and a precision of 74.13% on the RDS metric.

Large language model as parking planning agent in the context of mixed period of autonomous vehicles and Human-Driven vehicles

Autonomous vehicles (AVs) are anticipated to revolutionize future transportation, necessitating updates to traffic infrastructure, particularly parking facilities, due to the unique characteristics of AVs compared to Human-Driven Vehicles (HDVs). During the transition period in which AVs and HDVs coexist, adaptable infrastructure is essential to accommodate both vehicle types. Traditional research, typically reliant on complex mathematical models and simulations, faces challenges in adapting to diverse urban settings, requiring substantial time and resources. To address these challenges, a government-level framework was developed, enabling urban planners to quickly and accurately evaluate and optimize existing parking facilities for future AV and HDV coexistence scenarios. The framework integrates a Large Language Model (LLM) to enhance flexibility and efficiency in parking planning throughout the transitional period. Structured guidance is incorporated to enhance decision-making precision and reduce LLM hallucination risks. The flexibility, robustness, and accuracy of the framework were validated through step-by-step and end-to-end testing using real-world datasets. Specifically, the framework achieved 91.1 % comprehensiveness and 70.2 % consistency in Indicator Selection Module testing, a 68.9 % success rate in the Single Indicator Calculation Module, and a 66.7 % success rate in end-to-end testing, demonstrating its practical value in supporting cities during AV integration. Finally, the success rates of different LLM agent modules were further explored, along with a comparison of multiple LLMs and an analysis of key issues related to LLM trustworthiness in urban planning applications. The research highlights the potential of LLMs in advancing urban planning processes and optimizing existing infrastructure, contributing to smarter and more adaptable urban environments.

Design and Implementation of a Geographic Scenario Computation System Based on Knowledge Base

Abstract. The promotion and expansion of Geographic Information System (GIS) applications have increased the complexity of geocomputation. The concept of a geographic scene, which encompasses both geographic entities and complex spatial interaction relationships, serves as a means of organizing and expressing geographic information. Existing methods of organizing and computing geographic scenes often focus on single geometric elements or single-category objects, neglecting the complexity of the geocomputation process and leading to biased results. To address this issue, this study designs and implements a geographic scenario computation system based on WebGIS to overcome the lack of integration of knowledge elements in traditional geographic scenario computation methods. The system consists of a data management module, a geographic scenario visualization module, and a geographic scenario computation module. It not only realizes the visualization of geographic scenarios but also introduces a knowledge base into the geographic scenario computation process, enabling accurate management of geographic scenarios. This paper uses a farmland scenario in Shandong Province, China, as a case study to verify the effectiveness and feasibility of the system. The application of this system provides new technical support for the computation process of geographic scenarios, promoting the refinement and intelligent development of geographic scenario management. This method can also be applied to similar applications of intelligent sensors and the Internet of Things in the organization and management of sensing information in the future.

Edge Caching Placement Strategy based on Evolutionary Game for Conversational Information Seeking in Edge Cloud Computing

In Internet applications, network conversation is the primary communication between the user and server. The server needs to efficiently and quickly return the corresponding service according to the conversation sent by the user to improve the users’ Quality of Service. Thus, Conversation Information Seeking (CIS) research has become a hot topic today. In Cloud Computing (CC), a central service mode, the conversation is transmitted between the user and the remote cloud over a long distance. With the explosive growth of Internet applications, network congestion, long-distance communication, and single point of failure have brought new challenges to the centralized service mode. People put forward Edge Cloud Computing (ECC) to meet the new challenges of the centralized service mode of CC. As a distributed service mode, ECC is an extension of CC. By migrating services from the remote cloud to the network edge closer to users, ECC can solve the above challenges in CC well. In ECC, people solve the problem of CIS through edge caching. The current research focuses on designing the edge cache strategy to achieve more predictable caching. In this article, we propose an edge cache placement method Evolutionary Game based Caching Placement Strategy (EG-CPS). This method consists of three modules: the user preference prediction module, the content popularity calculation module, and the cache placement decision module. To maximize the predictability of the cache strategy, we are committed to optimizing the cache hit rate and service latency. The simulation experiment compares the proposed strategy with several other cache strategies. The experimental results illustrate that EG-CPS can reduce up to 2.4% of the original average content request latency, increase the average direct cache hit rate by 1.7%, and increase the average edge cache hit rate by 3.3%.

Cross-resolution land cover classification using outdated products and transformers

ABSTRACT Large-scale, high-resolution land cover classification is a prerequisite for constructing Earth system models and addressing ecological and resource issues. Advancements in satellite sensor technology have led to improvements in spatial resolution and wider coverage areas. Nevertheless, the lack of high-resolution labelled data is still a challenge, hindering the large-scale application of land cover classification methods. In this study, a Transformer-based weakly supervised method for cross-resolution land cover classification using outdated data is proposed. First, to capture long-range dependencies without overlooking the fine-grained details of objects, a U-Net-like Transformer based on a reverse difference mechanism (RDM) using dynamic sparse attention is designed. Second, an anti-noise loss calculation module based on optimal transport (OT) is proposed. The anti-noise loss calculation identifies confident areas and vague areas based on the OT matrix, which relieves the effect of noises on outdated land cover products. By introducing a weakly supervised loss with weights and using an unsupervised loss, the RDM-based U-Net-like Transformer was trained. Remote sensing images with 1 m resolutions and the corresponding ground truths of six states in the United States were used to validate the performance of the proposed method. The experiments used outdated land cover products with 30 m resolutions from 2013 as training labels and produced land cover maps with 1 m resolutions from 2017. The results showed the superiority of the proposed method over state-of-the-art methods. The code is available at https://github.com/yu-ni1989/ANLC-Former.

Study of the C-γ-Reθ Transition Model Based on the NNW-HyFLOW Software

Abstract Since the release of hypersonic computational fluid software NNW-HyFLOW in 2023, its computational modules have been continuously expanded for the development needs of hypersonic vehicles. The C-γ-Reθ transition model has been developed based in the framework of the NNW-HyFLOW to expand the prediction of transition in the design of high-speed vehicles. Firstly, the calibration of two basic relations of Re θc and Flength which is originally designed for the low-speed flat plate cases have been realized and expanded to high-speed flows. Secondly, the the empirical formula of Mach number correction has been adjusted by several typical high-speed transition cases. The same correction also has been made to the turbulence Prandtl number correction following the sae procedure. A crossflow transition module has been added with the calibration of the crossflow Reynolds number and validation of the experiment data of the HyTRV lifting body. The numerical tests show that the C-γ-Reθ transition model based on NNW-HyFLOW software has been basically realized with the ability of steamwise transition and crossflow transition prediction.

Simulation of the Full‐Process Dynamics of Floating Vehicles Driven by Flash Floods

AbstractFlash flooding has become more prominent under climate change, threatening people's life and property. Post‐event investigations of recent events emphasize the role of floating debris, such as vehicles, in exacerbating damage. Few modeling methods and tools have been developed to simulate the full‐process dynamics of floating debris driven by large‐scale flood waves in real world. In this work, a fully coupled model is developed for simulating the full‐process interactive movements of vehicles driven by flash flood hydrodynamics, from entrainment, transport to deposition. The proposed coupled modeling system consists of a finite volume shock‐capturing hydrodynamic model solving the 2D shallow water equations and a 3D discrete element method (DEM) model. The proposed two‐way coupling approach estimates the hydrostatic and hydrodynamic forces acting on solid objects using the water depth and velocity predicted by the hydrodynamic model; the resulting counter forces on the fluid flow are then considered by adding extra source terms in the hydrodynamic model. A multi‐sphere method is further embedded in the DEM model to better represent vehicle shapes. New calculation modules are further implemented to represent the vehicle entrainment, contact and stopping motions. The coupled model is applied to reproduce a flash flood event hit Boscastle in the UK in 2004. Over 100 vehicles were moved and carried downstream by the highly transient flood flow. The model well predicts the hydrodynamics, interactive transport process and the final locations of vehicles. The proposed coupled model provides a new tool for simulating large‐scale flash flooding processes, including debris dynamics.

A comprehensive review of advances in physics-informed neural networks and their applications in complex fluid dynamics

Physics-informed neural networks (PINNs) represent an emerging computational paradigm that incorporates observed data patterns and the fundamental physical laws of a given problem domain. This approach provides significant advantages in addressing diverse difficulties in the field of complex fluid dynamics. We thoroughly investigated the design of the model architecture, the optimization of the convergence rate, and the development of computational modules for PINNs. However, efficiently and accurately utilizing PINNs to resolve complex fluid dynamics problems remain an enormous barrier. For instance, rapidly deriving surrogate models for turbulence from known data and accurately characterizing flow details in multiphase flow fields present substantial difficulties. Additionally, the prediction of parameters in multi-physics coupled models, achieving balance across all scales in multiscale modeling, and developing standardized test sets encompassing complex fluid dynamic problems are urgent technical breakthroughs needed. This paper discusses the latest advancements in PINNs and their potential applications in complex fluid dynamics, including turbulence, multiphase flows, multi-field coupled flows, and multiscale flows. Furthermore, we analyze the challenges that PINNs face in addressing these fluid dynamics problems and outline future trends in their growth. Our objective is to enhance the integration of deep learning and complex fluid dynamics, facilitating the resolution of more realistic and complex flow problems.