Engineering Analysis with Boundary Elements最新文献_第2页

Mechanisms of non-uniform propagation of hydraulic fractures: A comprehensive numerical investigation 水力裂缝非均匀扩展机制：综合数值研究

IF 4.2 2区工程技术

Engineering Analysis with Boundary Elements Pub Date : 2025-05-26 DOI: 10.1016/j.enganabound.2025.106307

Changxin Yang , Zhaozhong Yang , Hehua Wang , Liping Jiang , Liangping Yi , Yi Cheng , Duo Yi

{"title":"Mechanisms of non-uniform propagation of hydraulic fractures: A comprehensive numerical investigation","authors":"Changxin Yang , Zhaozhong Yang , Hehua Wang , Liping Jiang , Liangping Yi , Yi Cheng , Duo Yi","doi":"10.1016/j.enganabound.2025.106307","DOIUrl":"10.1016/j.enganabound.2025.106307","url":null,"abstract":"<div><div>Based on the displacement discontinuity method (DDM) and the finite volume method (FVM), a non-planar three-dimensional integrated simulator for simulating hydraulic fracture (HF) propagation is proposed to study the competitive propagation mechanisms of HF in the presence of stress interference and natural fractures (NFs). The embedded discrete fracture model (EDFM) is introduced to describe the fluid flow between fracture and matrix, and the fully coupled equations for fracture propagation and fluid flow are derived. An optimized methodology for calculating and assembling of the influence coefficient matrix is introduced. Based on the established numerical model, the discrepancies in fracture geometries and fluid pressure during multi-cluster fracturing in the horizontal well are systematically analyzed, both with and without considering the impact of fracture deflection. The basis for selecting planar versus non-planar three-dimensional models to simulate the competitive propagation of multiple fractures is presented. Moreover, the asymmetrical distribution of NFs leads to non-uniform propagation of the wings of HF. The degree of non-uniformity in HF propagation is proportional to the differences between NFs. A reduction in the length and approach angle of NFs results in a greater propagation distance for HF.</div></div>","PeriodicalId":51039,"journal":{"name":"Engineering Analysis with Boundary Elements","volume":"178 ","pages":"Article 106307"},"PeriodicalIF":4.2,"publicationDate":"2025-05-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144138030","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Physics-informed neural network model using natural gradient descent with Dirichlet distribution 基于Dirichlet分布的自然梯度下降神经网络模型

IF 4.2 2区工程技术

Engineering Analysis with Boundary Elements Pub Date : 2025-05-24 DOI: 10.1016/j.enganabound.2025.106282

R. Abdulkadirov , P. Lyakhov , V. Baboshina

{"title":"Physics-informed neural network model using natural gradient descent with Dirichlet distribution","authors":"R. Abdulkadirov , P. Lyakhov , V. Baboshina","doi":"10.1016/j.enganabound.2025.106282","DOIUrl":"10.1016/j.enganabound.2025.106282","url":null,"abstract":"<div><div>In this article we propose the physics-informed neural network model which contains the natural gradient descent with Dirichlet distribution. Such an optimizer can more accurately converge in the global minimum of the loss function in a short number of iterations. Due to natural gradient, one considers not only the gradient directions but also convexity of the loss function. Using the Dirichlet distribution, natural gradient allows for a reduction in time consumption comparing with the second order approaches. The proposed physics-informed neural model increases the accuracy of solving initial and boundary value problems for partial differential equations, such as the heat and Burgers equation, on <span><math><mrow><mn>0</mn><mtext>%</mtext><mo>−</mo><mn>10</mn><mtext>%</mtext></mrow></math></span> Gaussian noised data. Compared with the state-of-the-art optimization methods, the proposed natural gradient descent with Dirichlet distribution achieves the more accurate solution by <span><math><mrow><mn>9</mn><mtext>%</mtext><mo>−</mo><mn>62</mn><mtext>%</mtext></mrow></math></span>, estimated by mean squared error and <span><math><msup><mrow><mi>L</mi></mrow><mrow><mn>2</mn></mrow></msup></math></span> error.</div></div>","PeriodicalId":51039,"journal":{"name":"Engineering Analysis with Boundary Elements","volume":"178 ","pages":"Article 106282"},"PeriodicalIF":4.2,"publicationDate":"2025-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144123774","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A customized CNN architecture for detecting double and multiple JPEG compression in small blocks 一个定制的CNN架构，用于检测小块中的双重和多重JPEG压缩

IF 4.2 2区工程技术

Engineering Analysis with Boundary Elements Pub Date : 2025-05-20 DOI: 10.1016/j.enganabound.2025.106290

Israr Hussain , Naeem Hussain , Shunquan Tan

{"title":"A customized CNN architecture for detecting double and multiple JPEG compression in small blocks","authors":"Israr Hussain , Naeem Hussain , Shunquan Tan","doi":"10.1016/j.enganabound.2025.106290","DOIUrl":"10.1016/j.enganabound.2025.106290","url":null,"abstract":"<div><div>Detecting double JPEG compression is crucial for verifying the authenticity and integrity of JPEG images. However, this poses significant challenges, particularly when the quality factor (QF) of the first compression is substantially higher than of the second compression, and when working with small-sized blocks. To overcome these challenges, we propose a customized convolutional neural network (CNN) architecture specifically designed for detecting double and multiple JPEG compression. The main contribution of our work is the innovative use of raw DCT coefficients as input to the CNN, enabling end-to-end feature extraction and enhancing the detection of subtle compression artifacts that might be missed by traditional techniques. The framework incorporates an ABS layer after the first convolution to eliminate the symmetry information that are not useful for JPEG compression detection, applies batch normalization (BN) to stabilize training, and reduces model complexity by employing 1 × 1 convolutions in deeper layers. Experimental results show that the proposed customized architecture significantly improves detection accuracy compared to other state-of-the-art methods, particularly in challenging small-block scenarios, and demonstrates effectiveness in detecting both double and multiple JPEG compression.</div></div>","PeriodicalId":51039,"journal":{"name":"Engineering Analysis with Boundary Elements","volume":"178 ","pages":"Article 106290"},"PeriodicalIF":4.2,"publicationDate":"2025-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144089890","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Investigation of hydraulic fracture propagation patterns and hydro-mechanical coupling mechanisms through DEM analysis 基于DEM分析的水力裂缝扩展模式及水-力耦合机理研究

IF 4.2 2区工程技术

Engineering Analysis with Boundary Elements Pub Date : 2025-05-20 DOI: 10.1016/j.enganabound.2025.106298

Q. Dong , Y. Wang

{"title":"Investigation of hydraulic fracture propagation patterns and hydro-mechanical coupling mechanisms through DEM analysis","authors":"Q. Dong , Y. Wang","doi":"10.1016/j.enganabound.2025.106298","DOIUrl":"10.1016/j.enganabound.2025.106298","url":null,"abstract":"<div><div>Accurate prediction of fracture propagation morphology contributes to the success of hydraulic fracturing operations and the estimation of oil and gas production capacity. Various factors, including fluid injection characteristics, in-situ stresses, and pre-existing natural fractures, exert significant influence on the fracture morphology. The discrete element method (DEM) captures inter-particle interactions and exhibits distinct advantages in handling the propagation and interaction of multiple fractures. In this study, we employ a simulation approach that combines DEM with the pipe network flow model. Initially, a comprehensive coupling enhancement of the pressure-updating equation is implemented, ensuring the constant satisfaction of the principle of flow conservation. This leads to an accurate fluid pressure distribution during the process of fracture propagation, which serves as the driving force for fracture development. Building upon this foundation, an analysis is conducted regarding the fracture propagation patterns and underlying microscopic mechanisms under varying fluid viscosities, pre-existing natural fractures, and gas fracturing. The findings reveal that low-viscosity fluids exhibit higher penetration as fractures extend, promoting the propensity for complex branching of fractures. When interacting with pre-existing natural fractures, the model effectively simulates interactions such as cross, offset, and capture types for different interaction angles and in-situ stress ratios. During gas fracturing, the high compressibility of gas prominently leads to the occurrence of complex multiple fractures within the particle assembly, and the effects of burn rate, duration, and the in-situ stress ratio on the morphology of gas fracturing are conducted.</div></div>","PeriodicalId":51039,"journal":{"name":"Engineering Analysis with Boundary Elements","volume":"178 ","pages":"Article 106298"},"PeriodicalIF":4.2,"publicationDate":"2025-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144098643","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Cut-cell Cartesian meshes for incompressible laminar and turbulent flows based on n-sided Cell-based Smoothed FEM 不可压缩层流和湍流的n面单元光滑有限元切单元笛卡尔网格

IF 4.2 2区工程技术

Engineering Analysis with Boundary Elements Pub Date : 2025-05-19 DOI: 10.1016/j.enganabound.2025.106302

Chen Jiang , Chen Hong , Guo Zhou , Zhi-yang Song , Ming-yang Liu

{"title":"Cut-cell Cartesian meshes for incompressible laminar and turbulent flows based on n-sided Cell-based Smoothed FEM","authors":"Chen Jiang , Chen Hong , Guo Zhou , Zhi-yang Song , Ming-yang Liu","doi":"10.1016/j.enganabound.2025.106302","DOIUrl":"10.1016/j.enganabound.2025.106302","url":null,"abstract":"<div><div>The cut-cell Cartesian mesh is widely favored in Computational Fluid Dynamics (CFD) due to its ease of generation, body fitting and high quality. However, the presence of hanging nodes and irregular body-fitted polygonal elements makes it challenging to construct shape functions in traditional Finite Element Methods (FEM). In this work, the n-sided Cell-based Smoothed Finite Element Method (nCS-FEM) is developed, introducing detailed construction techniques of smoothed domains in various cut-cell Cartesian meshes and solving incompressible laminar and turbulent flows. The Streamline-Upwind/Petrov–Galerkin stabilization, combined with the Stabilized Pressure Gradient Projection (SUPG/SPGP) method, is employed to mitigate convection and pressure oscillations in nCS-FEM when solving the Unsteady Reynolds–Averaged Navier–Stokes (URANS) equations. Several numerical examples—including analytical solutions for laminar flows and benchmark results from aerospace applications—are presented to validate the applicability and accuracy of the nCS-FEM in solving problems with cut-cell Cartesian meshes (nCS-FEM-Cart). The results demonstrate excellent performance in predicting flow characteristics. Compared to Q4 elements, cut-cell elements facilitate the generation of variable-density meshes, thereby enabling faster computations. Furthermore, a complex structure, i.e. multistage Tesla valves, is employed to evaluate the capability of the realizable <em>k</em>-<em>ε</em> model in simulating the flow separation and forced convection, highlighting its significant potential in complex engineering problems.</div></div>","PeriodicalId":51039,"journal":{"name":"Engineering Analysis with Boundary Elements","volume":"178 ","pages":"Article 106302"},"PeriodicalIF":4.2,"publicationDate":"2025-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144083759","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Numerical investigation on shear mechanical characteristics of rock joints filled with clay-rich fillings 富粘土充填岩体节理剪切力学特性数值研究

IF 4.2 2区工程技术

Engineering Analysis with Boundary Elements Pub Date : 2025-05-17 DOI: 10.1016/j.enganabound.2025.106303

Congcong Hou , Yongshui Kang , Bin Liu , Jiancheng Xiao

{"title":"Numerical investigation on shear mechanical characteristics of rock joints filled with clay-rich fillings","authors":"Congcong Hou , Yongshui Kang , Bin Liu , Jiancheng Xiao","doi":"10.1016/j.enganabound.2025.106303","DOIUrl":"10.1016/j.enganabound.2025.106303","url":null,"abstract":"<div><div>The clay filling layer can significantly influences the shear behavior of rock joints. In this study, a numerical model for the direct shear test on clay-rich filling rock masses under constant normal stress was developed using the Continuum Discontinuum Element Method (CDEM), and validated through compression-shear experiments. Subsequently, the influence mechanism of water content and various normal stresses on the shear behavior of rocks were systematically examined. From a microscopic perspective, the damage evolution of rough joint surfaces under different water contents and normal stresses was quantitatively assessed. Finally, numerical results were compared against theoretical models. The research results indicate that the CDEM can effectively simulate the nonlinear mechanical behaviors of rock joints containing clay-rich filling layers during shearing process, and it fits well with the experimental results. The peak shear strength decreases as water content increases, while it increases with the increase of normal stress. The normal stress would change the maximum static friction force when the filling layer slides, while water content would change the friction angle between the filling material particles, both of which affect the peak shear strength of the rock joint. Additionally, it was compared with the strength criteria obtained from the experiments, and the comparison results are consistent.</div></div>","PeriodicalId":51039,"journal":{"name":"Engineering Analysis with Boundary Elements","volume":"178 ","pages":"Article 106303"},"PeriodicalIF":4.2,"publicationDate":"2025-05-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144067557","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A novel hybrid machine learning and optimization approach for stochastic free vibration analysis of graphene platelets reinforced functionally graded triply periodic minimal surface microplates 一种新的混合机器学习和优化方法用于石墨烯片增强功能梯度三周期最小表面微板的随机自由振动分析

IF 4.2 2区工程技术

Engineering Analysis with Boundary Elements Pub Date : 2025-05-17 DOI: 10.1016/j.enganabound.2025.106304

Van-Thien Tran , Trung-Kien Nguyen , Thuc P. Vo

{"title":"A novel hybrid machine learning and optimization approach for stochastic free vibration analysis of graphene platelets reinforced functionally graded triply periodic minimal surface microplates","authors":"Van-Thien Tran , Trung-Kien Nguyen , Thuc P. Vo","doi":"10.1016/j.enganabound.2025.106304","DOIUrl":"10.1016/j.enganabound.2025.106304","url":null,"abstract":"<div><div>This paper proposes a new hybrid deep neural network and optimization approach for stochastic vibration analysis of graphene platelets reinforced functionally graded triply periodic minimal surface (GPLR-FG-TPMS) microplates with material properties uncertainties. A combination of the bidirectional long short-term memory model (BiLSTM), the Shrimp and Goby Association Search Algorithm (SGA) and Chebyshev polynomials of the second kind utilized in the Ritz method to improve the accuracy of the numerical solutions is developed. The deterministic fundamental frequencies of GPLR-FG-TPMS microplates are first analyzed using a combination of third-order shear deformation theory, modified couple stress theory and Ritz-type series solutions. Subsequently, their stochastic responses under material properties uncertainties are obtained using the SGA-BiLSTM model. Numerical examples are obtained to investigate the effects of the porosity coefficients, graphene platelets weight fractions, thickness-to-length ratios, length-to-material ratios, and different boundary conditions on the natural frequencies of GPLR-FG-TPMS microplates. The novel findings of this paper provide valuable insights and serve as a reference for future research.</div></div>","PeriodicalId":51039,"journal":{"name":"Engineering Analysis with Boundary Elements","volume":"178 ","pages":"Article 106304"},"PeriodicalIF":4.2,"publicationDate":"2025-05-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144067548","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

An improved MLS-based numerical manifold method for saturated-unsaturated seepage in porous media 基于改进mls的多孔介质饱和-非饱和渗流数值流形方法

IF 4.2 2区工程技术

Engineering Analysis with Boundary Elements Pub Date : 2025-05-14 DOI: 10.1016/j.enganabound.2025.106299

Yuanqiang Chen , Cheng Liu , Hong Zheng , Xiaocheng Huang , Shunkai Liu , Jian Peng

引用次数: 0

A novel hybrid contact detection algorithm for 2D FDEM: Balancing efficiency and memory consumption

IF 4.2 2区工程技术

Engineering Analysis with Boundary Elements Pub Date : 2025-05-14 DOI: 10.1016/j.enganabound.2025.106291

He Liu , Panpan Zhu , Quansheng Liu , Yongchao Tian , Yiming Lei , Xin Yin , Zuliang Shao , Guicheng He

{"title":"A novel hybrid contact detection algorithm for 2D FDEM: Balancing efficiency and memory consumption","authors":"He Liu , Panpan Zhu , Quansheng Liu , Yongchao Tian , Yiming Lei , Xin Yin , Zuliang Shao , Guicheng He","doi":"10.1016/j.enganabound.2025.106291","DOIUrl":"10.1016/j.enganabound.2025.106291","url":null,"abstract":"<div><div>In this paper, a novel hybrid contact detection algorithm, AGS (Adaptive grid-based search) & GJK (Gilbert-Johnson-Keerthi), is proposed to accelerate 2D FDEM (Combined finite-discrete element method) simulations. AGS algorithm maintains computational efficiency comparable to traditional broad search methods based on uniform grid decomposition, while significantly reducing memory consumption by utilizing only effective grid cells in identifying potential contact pairs. Additionally, a modified GJK algorithm with specific initial search direction is employed for contact resolution, which is not only easy to implement but also more efficient than SAT (Separation Axis Theorem). Compared·to existing FDEM algorithms like NBS (Non-binary Search) and GGS (Global Grid-based Search) & SAT (Separation Axis Theorem), the proposed algorithm effectively overcomes their limitations, offering high computational efficiency, low memory consumption, and reduced sensitivity to mesh size. The effectiveness of the proposed algorithm is validated through multiple numerical cases, demonstrating its suitability for both quasi-static and dynamic simulations. Notably, in the numerical cases examined, the algorithm achieves a speed-up ratio of 2.76 at case scale compared to NBS, and saves 37.8 % in memory consumption compared to GGS & SAT. Furthermore, the algorithm is highly versatile and can potentially be extended to DEM (Discrete Element Method) applications involving other convex-shaped particles.</div></div>","PeriodicalId":51039,"journal":{"name":"Engineering Analysis with Boundary Elements","volume":"178 ","pages":"Article 106291"},"PeriodicalIF":4.2,"publicationDate":"2025-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143943381","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A Burton & Miller type FM/BEM for the solution of 3D stokes flow problems in scaffolds in perfusion bioreactor settings

IF 4.2 2区工程技术

Engineering Analysis with Boundary Elements Pub Date : 2025-05-13 DOI: 10.1016/j.enganabound.2025.106297

Euripides J. Sellountos , Demosthenes Polyzos , Theodore V. Gortsas

{"title":"A Burton & Miller type FM/BEM for the solution of 3D stokes flow problems in scaffolds in perfusion bioreactor settings","authors":"Euripides J. Sellountos , Demosthenes Polyzos , Theodore V. Gortsas","doi":"10.1016/j.enganabound.2025.106297","DOIUrl":"10.1016/j.enganabound.2025.106297","url":null,"abstract":"<div><div>Perfusion bioreactors involve the continuous flow of culture medium through the scaffold. While sophisticated experimental setups exist for measuring flow parameters, a deep understanding of the three-dimensional (3D) flow throughout a scaffold mounted in a bioreactor can be accomplished via computational methods. The flow in a bioreactor is characterized by low Reynold numbers and can be modeled as a Stokes flow problem. The Boundary Element Method (BEM) is a well-known tool for solving Stokes flow problems. In this work we examine the fluid flow around two tissue engineering scaffolds with designs which are typically utilized in this field, using a Fast Multipole Boundary Element Method (FM/BEM) based on the Burton & Miller type formulation. The accuracy of the method is assessed via a representative benchmark problem. The results for the pressure and stress distribution on the surface of the examined scaffolds indicate the efficacy of the FM/BEM approach and facilitate meaningful conclusions on the subject.</div></div>","PeriodicalId":51039,"journal":{"name":"Engineering Analysis with Boundary Elements","volume":"178 ","pages":"Article 106297"},"PeriodicalIF":4.2,"publicationDate":"2025-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143943258","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0