Advances in Engineering Software最新文献_第2页

Deep neural network based head injury criterion estimation for more efficient pedestrian protection performance evaluation in bonnet structure design

IF 4 2区工程技术 Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Advances in Engineering Software

Pub Date : 2025-01-23 DOI: 10.1016/j.advengsoft.2025.103867

Tianci Zhang , Weiwei Wang , Bo Li , Ting Zeng , Fanliang Meng

In the early development stage of a new car model, fast evaluation of pedestrian protection performance is required to accelerate the bonnet structure design process. Current evaluation methods rely on finite element (FE) simulation to calculate the Head Injury Criterion (HIC) values at specified points on the bonnet, before proceeding to the more expensive real-world impact test using headform impactors. However, the FE based approach typically takes several days to complete the meshing, boundary condition setting and HIC calculation to evaluate a single design candidate. To further increase the evaluation efficiency, this paper presents a novel HIC estimation approach based on deep learning. An end-to-end deep neural network model is proposed which can directly generate the HIC value without resorting to FE methods. It uses seven variables pertaining to the panel height, structural difference, thickness and head type as the input based on the definition of HIC. Convolution layers are utilised to aggregate the surrounding structural information for each target point. To demonstrate the effectiveness of the proposed approach, cross validation results are presented based on a dataset of over five thousands target points collected from 28 cars. For the green target regions, the average HIC estimation accuracy is 93.1 %, which outperforms the result of 83 % reported in previous work. A comparison with the traditional support vector regression method demonstrates the advantages of the proposed approach.

{"title":"Deep neural network based head injury criterion estimation for more efficient pedestrian protection performance evaluation in bonnet structure design","authors":"Tianci Zhang , Weiwei Wang , Bo Li , Ting Zeng , Fanliang Meng","doi":"10.1016/j.advengsoft.2025.103867","DOIUrl":"10.1016/j.advengsoft.2025.103867","url":null,"abstract":"<div><div>In the early development stage of a new car model, fast evaluation of pedestrian protection performance is required to accelerate the bonnet structure design process. Current evaluation methods rely on finite element (FE) simulation to calculate the Head Injury Criterion (HIC) values at specified points on the bonnet, before proceeding to the more expensive real-world impact test using headform impactors. However, the FE based approach typically takes several days to complete the meshing, boundary condition setting and HIC calculation to evaluate a single design candidate. To further increase the evaluation efficiency, this paper presents a novel HIC estimation approach based on deep learning. An end-to-end deep neural network model is proposed which can directly generate the HIC value without resorting to FE methods. It uses seven variables pertaining to the panel height, structural difference, thickness and head type as the input based on the definition of HIC. Convolution layers are utilised to aggregate the surrounding structural information for each target point. To demonstrate the effectiveness of the proposed approach, cross validation results are presented based on a dataset of over five thousands target points collected from 28 cars. For the green target regions, the average HIC estimation accuracy is 93.1 %, which outperforms the result of 83 % reported in previous work. A comparison with the traditional support vector regression method demonstrates the advantages of the proposed approach.</div></div>","PeriodicalId":50866,"journal":{"name":"Advances in Engineering Software","volume":"202 ","pages":"Article 103867"},"PeriodicalIF":4.0,"publicationDate":"2025-01-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143179368","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 enhanced local damage model for 2D and 3D quasi-brittle fracture: ABAQUS-FEM implementation and comparative study on the effect of equivalent strains

IF 4 2区工程技术 Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Advances in Engineering Software

Pub Date : 2025-01-16 DOI: 10.1016/j.advengsoft.2025.103863

Quan Nhu Tran , Minh Ngoc Nguyen , Chanh Dinh Vuong , Nhung Nguyen , Tinh Quoc Bui

In this paper, the initiation and propagation of cracks in quasi-brittle materials are characterized by an enhanced local damage model. The model is implemented using UMAT, an ABAQUS user-subroutine platform, for fracture in two- and three-dimensional media. As usual, the material state is represented by a damage parameter

D

ranging from 0 (intact) to 1 (completed failure). Different from the conventional local damage models (Kachanov [1], 1999), (Lemaitre [2], 1985) both fracture energy and characteristic length of the element are incorporated into the calculation of the damage parameter, hence the inherent issue of mesh-dependency is mitigated. A comparative study on the effect of equivalent strain on the prediction of crack path as well as load–displacement curve under mixed-mode condition is conducted. For that purpose, three types of equivalent strain are adopted based on three models: the modified von Mises model, the Ottosen criterion model, and the new Mazars model (named after the work of Mazars et al., [3] (2015) in this paper). The accuracy and performance of the developed codes and proposed damage approach in association with the three types of equivalent strain are demonstrated by comparison of the computed results with experimental data as well as other numerical results reported in the literature. The experimental data is well-fit by the simulation results. A detailed description of implementation in UMAT subroutine is also provided. The source codes are provided for free, and users can amend them for their own purpose.

本文采用增强型局部损伤模型来描述准脆性材料中裂纹的产生和扩展。该模型使用 ABAQUS 用户子程序平台 UMAT 实现，适用于二维和三维介质的断裂。与往常一样，材料状态由损伤参数 D 表示，范围从 0（完好无损）到 1（完全破坏）。与传统的局部损伤模型（Kachanov [1], 1999）和（Lemaitre [2], 1985）不同的是，在计算损伤参数时将断裂能和元素的特征长度都考虑在内，因此减轻了固有的网格依赖问题。我们对混合模式条件下等效应变对裂纹路径预测以及荷载-位移曲线的影响进行了比较研究。为此，基于三种模型采用了三种等效应变：改进的 von Mises 模型、Ottosen 准则模型和新的 Mazars 模型（本文以 Mazars 等人的研究成果命名，[3] (2015)）。通过将计算结果与实验数据以及文献中报道的其他数值结果进行比较，证明了所开发的代码和所提出的损伤方法与三种等效应变相关联的准确性和性能。模拟结果很好地拟合了实验数据。此外，还提供了 UMAT 子程序实施的详细说明。源代码免费提供，用户可根据自己的需要进行修改。

{"title":"An enhanced local damage model for 2D and 3D quasi-brittle fracture: ABAQUS-FEM implementation and comparative study on the effect of equivalent strains","authors":"Quan Nhu Tran , Minh Ngoc Nguyen , Chanh Dinh Vuong , Nhung Nguyen , Tinh Quoc Bui","doi":"10.1016/j.advengsoft.2025.103863","DOIUrl":"10.1016/j.advengsoft.2025.103863","url":null,"abstract":"<div><div>In this paper, the initiation and propagation of cracks in quasi-brittle materials are characterized by an enhanced local damage model. The model is implemented using UMAT, an ABAQUS user-subroutine platform, for fracture in two- and three-dimensional media. As usual, the material state is represented by a damage parameter <span><math><mi>D</mi></math></span> ranging from 0 (intact) to 1 (completed failure). Different from the conventional local damage models (Kachanov <span><span>[1]</span></span>, 1999), (Lemaitre <span><span>[2]</span></span>, 1985) both fracture energy and characteristic length of the element are incorporated into the calculation of the damage parameter, hence the inherent issue of mesh-dependency is mitigated. A comparative study on the effect of equivalent strain on the prediction of crack path as well as load–displacement curve under mixed-mode condition is conducted. For that purpose, three types of equivalent strain are adopted based on three models: the modified von Mises model, the Ottosen criterion model, and the new Mazars model (named after the work of Mazars et al., <span><span>[3]</span></span> (2015) in this paper). The accuracy and performance of the developed codes and proposed damage approach in association with the three types of equivalent strain are demonstrated by comparison of the computed results with experimental data as well as other numerical results reported in the literature. The experimental data is well-fit by the simulation results. A detailed description of implementation in UMAT subroutine is also provided. The source codes are provided for free, and users can amend them for their own purpose.</div></div>","PeriodicalId":50866,"journal":{"name":"Advances in Engineering Software","volume":"202 ","pages":"Article 103863"},"PeriodicalIF":4.0,"publicationDate":"2025-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143180443","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

Enhancing speedup in multifidelity training process by design automation to generate online reduced-order models in Convection–Diffusion–Reaction systems

IF 4 2区工程技术 Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Advances in Engineering Software

Pub Date : 2025-01-14 DOI: 10.1016/j.advengsoft.2025.103864

Feng Bai

In this article, a new intelligent framework is proposed by utilizing design automation in multifidelity training process to generate online reduced-order models (ROMs) in Convection–Diffusion–Reaction systems (CDR-PDE), aiming to enhance computing speedup in training and extract the representation of state–space training data. In the design automation techniques, the entire training process is divided into two layers of time divisions: (1) in the first layer, there are several large sections with equal size; (2) in the second layer, each large section includes two sub-intervals with different sizes (or may be equal) to enable significant speedup by elongating ROMs and shortening full-order models (FOMs) as much as possible. Each of sub-intervals is simulated in either FOM or ROM with least-squares methods on-the-fly by the switch criteria and algorithms in which the POD modes can be automatically selected. The main goal of this research is to enhance computing speedup in training process in CDR-PDEs without loss of the model accuracy in a robust way. During the multifidelity simulation in training process, the numbers of POD modes are upgraded automatically at the end of each sub-interval in FOM, meaning that the users do not need to determine the POD numbers in a priori. Three typical numerical examples in CDR-PDE are investigated. According to the observation from numerical studies, the POD modes are upgraded in incremental SVD and the numbers of POD modes in training process increase with the numbers of update procedures; beside that the computing speedup is obviously enhanced and excellent model accuracy can be achieved except the strong boundary layer area.

{"title":"Enhancing speedup in multifidelity training process by design automation to generate online reduced-order models in Convection–Diffusion–Reaction systems","authors":"Feng Bai","doi":"10.1016/j.advengsoft.2025.103864","DOIUrl":"10.1016/j.advengsoft.2025.103864","url":null,"abstract":"<div><div>In this article, a new intelligent framework is proposed by utilizing design automation in multifidelity training process to generate online reduced-order models (ROMs) in Convection–Diffusion–Reaction systems (CDR-PDE), aiming to enhance computing speedup in training and extract the representation of state–space training data. In the design automation techniques, the entire training process is divided into two layers of time divisions: (1) in the first layer, there are several large sections with equal size; (2) in the second layer, each large section includes two sub-intervals with different sizes (or may be equal) to enable significant speedup by elongating ROMs and shortening full-order models (FOMs) as much as possible. Each of sub-intervals is simulated in either FOM or ROM with least-squares methods on-the-fly by the switch criteria and algorithms in which the POD modes can be automatically selected. The main goal of this research is to enhance computing speedup in training process in CDR-PDEs without loss of the model accuracy in a robust way. During the multifidelity simulation in training process, the numbers of POD modes are upgraded automatically at the end of each sub-interval in FOM, meaning that the users do not need to determine the POD numbers in a priori. Three typical numerical examples in CDR-PDE are investigated. According to the observation from numerical studies, the POD modes are upgraded in incremental SVD and the numbers of POD modes in training process increase with the numbers of update procedures; beside that the computing speedup is obviously enhanced and excellent model accuracy can be achieved except the strong boundary layer area.</div></div>","PeriodicalId":50866,"journal":{"name":"Advances in Engineering Software","volume":"202 ","pages":"Article 103864"},"PeriodicalIF":4.0,"publicationDate":"2025-01-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143179917","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

Data integration framework for multi-level information modelling and numerical analysis of deteriorated RC bridges

IF 4 2区工程技术 Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Advances in Engineering Software

Pub Date : 2025-01-07 DOI: 10.1016/j.advengsoft.2024.103860

Mohamed Ayham Alsharif , Georgia Thermou , Jelena Ninic , Walid Tizani

Structural assessment of deteriorated bridges is a vital step in the decision-making process for asset management. This paper proposes a novel Building Information Modelling (BIM) and Finite Element (FE) integration framework to analyse and retrofit damaged Reinforced Concrete (RC) bridges. The drive behind the framework is to increase the efficiency and accuracy of structural assessment. To this end, a software tool called Bridge Damage Information Modeller (BriDIM) is developed. Two-way automated data interpretation and exchange between BIM and FE is created to allow for the evaluation of the residual structural capacity against limit states based on a comprehensive multi-level numerical model. BriDIM automatically feedback the FE analysis results into BIM environment for visualisation. The research findings suggest that utilising BIM as a database that includes structural specifications, damage information and retrofitting aspects brings benefits in terms of consistency, modelling and computational efficiency throughout the life cycle of a structure. The proposed framework transforms the way we approach assessments by consolidating all mechanical damage-related information into one 3D visualised BIM model. Ultimately BriDIM holds potential in facilitating the development of Digital Twins for existing structures. The model can be used to create scenarios of various damage combinations and provide a reliable tool to simulate the current condition of the bridge and the expected conditions of the bridge in case of damage propagation or change in environmental conditions.

{"title":"Data integration framework for multi-level information modelling and numerical analysis of deteriorated RC bridges","authors":"Mohamed Ayham Alsharif , Georgia Thermou , Jelena Ninic , Walid Tizani","doi":"10.1016/j.advengsoft.2024.103860","DOIUrl":"10.1016/j.advengsoft.2024.103860","url":null,"abstract":"<div><div>Structural assessment of deteriorated bridges is a vital step in the decision-making process for asset management. This paper proposes a novel Building Information Modelling (BIM) and Finite Element (FE) integration framework to analyse and retrofit damaged Reinforced Concrete (RC) bridges. The drive behind the framework is to increase the efficiency and accuracy of structural assessment. To this end, a software tool called Bridge Damage Information Modeller (BriDIM) is developed. Two-way automated data interpretation and exchange between BIM and FE is created to allow for the evaluation of the residual structural capacity against limit states based on a comprehensive multi-level numerical model. BriDIM automatically feedback the FE analysis results into BIM environment for visualisation. The research findings suggest that utilising BIM as a database that includes structural specifications, damage information and retrofitting aspects brings benefits in terms of consistency, modelling and computational efficiency throughout the life cycle of a structure. The proposed framework transforms the way we approach assessments by consolidating all mechanical damage-related information into one 3D visualised BIM model. Ultimately BriDIM holds potential in facilitating the development of Digital Twins for existing structures. The model can be used to create scenarios of various damage combinations and provide a reliable tool to simulate the current condition of the bridge and the expected conditions of the bridge in case of damage propagation or change in environmental conditions.</div></div>","PeriodicalId":50866,"journal":{"name":"Advances in Engineering Software","volume":"202 ","pages":"Article 103860"},"PeriodicalIF":4.0,"publicationDate":"2025-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143179918","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

Study of influencing factors of performance in novel vertical roller mills

IF 4 2区工程技术 Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Advances in Engineering Software

Pub Date : 2025-01-04 DOI: 10.1016/j.advengsoft.2024.103858

Hailiang Hu , Yiming Li , Yunlong Lu , Xuejun Wang , Guiqiu Song

To optimize the particle motion characteristics in a Vertical Roller Mill (VRM), this study proposes incorporating spiral blades to the outer walls of the ash bucket and the outside of the separator. This design utilizes the space between the ash bucket, separator, middle shell, and upper shell to create specific channels for discharging particles outside the mill. The study employs computational fluid dynamics and powder classification methods to perform a comprehensive numerical analysis of the new VRM. By evaluating the flow field distribution, particle motion characteristics, and utilizing the Q criterion, the research identifies three critical parameters that improve the mill's performance: the width, angle, and number of spiral blades. Numerical analysis results reveal that as the width of the spiral blades increases, both the airflow trajectory and flow field distribution improve, thereby facilitating particle transport. When the angle of the spiral blades decreases, the airflow trajectory aligns more closely with the rotation direction of the blades, which is more conducive to discharging particles from the VRM. Furthermore, as the number of spiral blades increases, the airflow velocity within the spiral channel rises, leading to enhanced particle motion characteristics affected by the fluid. When the spiral blades are fully enclosed, with two turns and eight blades, the vortex distribution becomes more regular and the flow field stabilizes, which reduces unnecessary material recirculation. This study provides valuable guidance for optimizing the structure of the VRM and offers references for improving its internal flow fields, enhancing separation performance, and reducing energy consumption.

{"title":"Study of influencing factors of performance in novel vertical roller mills","authors":"Hailiang Hu , Yiming Li , Yunlong Lu , Xuejun Wang , Guiqiu Song","doi":"10.1016/j.advengsoft.2024.103858","DOIUrl":"10.1016/j.advengsoft.2024.103858","url":null,"abstract":"<div><div>To optimize the particle motion characteristics in a Vertical Roller Mill (VRM), this study proposes incorporating spiral blades to the outer walls of the ash bucket and the outside of the separator. This design utilizes the space between the ash bucket, separator, middle shell, and upper shell to create specific channels for discharging particles outside the mill. The study employs computational fluid dynamics and powder classification methods to perform a comprehensive numerical analysis of the new VRM. By evaluating the flow field distribution, particle motion characteristics, and utilizing the <strong><em>Q</em></strong> criterion, the research identifies three critical parameters that improve the mill's performance: the width, angle, and number of spiral blades. Numerical analysis results reveal that as the width of the spiral blades increases, both the airflow trajectory and flow field distribution improve, thereby facilitating particle transport. When the angle of the spiral blades decreases, the airflow trajectory aligns more closely with the rotation direction of the blades, which is more conducive to discharging particles from the VRM. Furthermore, as the number of spiral blades increases, the airflow velocity within the spiral channel rises, leading to enhanced particle motion characteristics affected by the fluid. When the spiral blades are fully enclosed, with two turns and eight blades, the vortex distribution becomes more regular and the flow field stabilizes, which reduces unnecessary material recirculation. This study provides valuable guidance for optimizing the structure of the VRM and offers references for improving its internal flow fields, enhancing separation performance, and reducing energy consumption.</div></div>","PeriodicalId":50866,"journal":{"name":"Advances in Engineering Software","volume":"202 ","pages":"Article 103858"},"PeriodicalIF":4.0,"publicationDate":"2025-01-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143179919","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

Implementation of a three-dimensional numerical model for the filling process in Liquid Composite Molding on polyhedral meshes

IF 4 2区工程技术 Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Advances in Engineering Software

Pub Date : 2024-12-30 DOI: 10.1016/j.advengsoft.2024.103859

Junchun Ding , Hao Luo , Yibo Wu , Wuyang Yue , Xianyang Li , Helezi Zhou , Zhigao Huang , Huamin Zhou

In this paper, a three-dimensional, two-phase numerical model based on polyhedral meshes was developed to simulate the resin-filling process in Liquid Composite Molding (LCM). The Algebraic Volume of Fluid (VOF) method tracked the resin flow and the finite volume method (FVM) was applied for stable numerical discretization and solution. The numerical model was validated by unidirectional flow experiments, with the maximum error observed at the flow front being within 5.00 %. Subsequently, the performance of polyhedral meshes in LCM simulations was compared with that of hexahedral and tetrahedral meshes, following the mesh-independence analysis. The results show that the accuracy of the model using polyhedral cells is close to hexahedral cells and higher than tetrahedral cells. The cell number of polyhedral cases is about half that of hexahedral cases and a quarter to a third of tetrahedral cases, assuming similar computational accuracy and mesh size. Polyhedral meshes consume the least computational resources, slightly less than hexahedral meshes and approximately one-third of tetrahedral meshes. Furthermore, polyhedral meshes have a similar level of mesh generation to that of the tetrahedra and are better adapted to complex geometries. Numerical modeling utilizing polyhedral meshes is advantageous for large-scale and complex-shaped parts.

本文开发了一种基于多面体网格的三维两相数值模型，用于模拟液体复合材料成型（LCM）中的树脂填充过程。流体代数体积法（VOF）跟踪树脂流动，有限体积法（FVM）用于稳定的数值离散和求解。数值模型通过单向流动实验进行了验证，在流动前沿观察到的最大误差在 5.00% 以内。随后，根据网格独立性分析，比较了多面体网格与六面体和四面体网格在 LCM 模拟中的性能。结果表明，使用多面体单元的模型精度接近六面体单元，高于四面体单元。在计算精度和网格大小相似的情况下，多面体案例的单元数约为六面体案例的一半，四面体案例的四分之一到三分之一。多面体网格消耗的计算资源最少，略低于六面体网格，约为四面体网格的三分之一。此外，多面体网格的网格生成水平与四面体网格相似，更适合复杂的几何形状。利用多面体网格进行数值建模，对于大型和复杂形状的零件非常有利。

{"title":"Implementation of a three-dimensional numerical model for the filling process in Liquid Composite Molding on polyhedral meshes","authors":"Junchun Ding , Hao Luo , Yibo Wu , Wuyang Yue , Xianyang Li , Helezi Zhou , Zhigao Huang , Huamin Zhou","doi":"10.1016/j.advengsoft.2024.103859","DOIUrl":"10.1016/j.advengsoft.2024.103859","url":null,"abstract":"<div><div>In this paper, a three-dimensional, two-phase numerical model based on polyhedral meshes was developed to simulate the resin-filling process in Liquid Composite Molding (LCM). The Algebraic Volume of Fluid (VOF) method tracked the resin flow and the finite volume method (FVM) was applied for stable numerical discretization and solution. The numerical model was validated by unidirectional flow experiments, with the maximum error observed at the flow front being within 5.00 %. Subsequently, the performance of polyhedral meshes in LCM simulations was compared with that of hexahedral and tetrahedral meshes, following the mesh-independence analysis. The results show that the accuracy of the model using polyhedral cells is close to hexahedral cells and higher than tetrahedral cells. The cell number of polyhedral cases is about half that of hexahedral cases and a quarter to a third of tetrahedral cases, assuming similar computational accuracy and mesh size. Polyhedral meshes consume the least computational resources, slightly less than hexahedral meshes and approximately one-third of tetrahedral meshes. Furthermore, polyhedral meshes have a similar level of mesh generation to that of the tetrahedra and are better adapted to complex geometries. Numerical modeling utilizing polyhedral meshes is advantageous for large-scale and complex-shaped parts.</div></div>","PeriodicalId":50866,"journal":{"name":"Advances in Engineering Software","volume":"201 ","pages":"Article 103859"},"PeriodicalIF":4.0,"publicationDate":"2024-12-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143183453","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

Stability of rectangular tunnels in cohesive-frictional soil under surcharge loading using isogeometric analysis and Bayesian neural networks

IF 4 2区工程技术 Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Advances in Engineering Software

Pub Date : 2024-12-30 DOI: 10.1016/j.advengsoft.2024.103861

Minh-Toan Nguyen , Tram-Ngoc Bui , Jim Shiau , Tan Nguyen , Thoi-Trung Nguyen

This study evaluates the stability of rectangular tunnels in cohesive-frictional soils under surcharge loading using a combination of IsoGeometric Analysis and artificial neural networks. A dataset of 12,946 samples was generated automatically to analyze a wide range of soil profiles and tunnel geometries. Stability solutions were derived using IsoGeometric Analysis coupled with second-order cone programming, enabling precise and efficient assessments of ultimate surcharge loading. A key contribution of this study is the development of a closed-form solution through a Bayesian regularized neural network, which significantly improves accuracy compared to existing methods. Advanced data visualization techniques, including two- and three-dimensional partial dependency plots, were used to reveal complex relationships among design parameters. Sensitivity analyses provided valuable insights for optimizing tunnel designs, enhancing decision-making processes in geotechnical engineering. This study aims to equip engineers with practical tools for designing rectangular tunnels in real-world applications.

本研究采用等几何分析和人工神经网络相结合的方法，评估了粘性摩擦土中矩形隧道在附加荷载作用下的稳定性。研究自动生成了一个包含 12,946 个样本的数据集，用于分析各种土壤剖面和隧道几何形状。利用等几何分析法和二阶圆锥编程法得出了稳定性解决方案，从而能够对极限附加荷载进行精确有效的评估。本研究的一个主要贡献是通过贝叶斯正则化神经网络开发了闭式解决方案，与现有方法相比，该方法显著提高了准确性。先进的数据可视化技术，包括二维和三维部分依赖图，用于揭示设计参数之间的复杂关系。敏感性分析为优化隧道设计提供了有价值的见解，改进了岩土工程的决策过程。这项研究旨在为工程师在实际应用中设计矩形隧道提供实用工具。

{"title":"Stability of rectangular tunnels in cohesive-frictional soil under surcharge loading using isogeometric analysis and Bayesian neural networks","authors":"Minh-Toan Nguyen , Tram-Ngoc Bui , Jim Shiau , Tan Nguyen , Thoi-Trung Nguyen","doi":"10.1016/j.advengsoft.2024.103861","DOIUrl":"10.1016/j.advengsoft.2024.103861","url":null,"abstract":"<div><div>This study evaluates the stability of rectangular tunnels in cohesive-frictional soils under surcharge loading using a combination of IsoGeometric Analysis and artificial neural networks. A dataset of 12,946 samples was generated automatically to analyze a wide range of soil profiles and tunnel geometries. Stability solutions were derived using IsoGeometric Analysis coupled with second-order cone programming, enabling precise and efficient assessments of ultimate surcharge loading. A key contribution of this study is the development of a closed-form solution through a Bayesian regularized neural network, which significantly improves accuracy compared to existing methods. Advanced data visualization techniques, including two- and three-dimensional partial dependency plots, were used to reveal complex relationships among design parameters. Sensitivity analyses provided valuable insights for optimizing tunnel designs, enhancing decision-making processes in geotechnical engineering. This study aims to equip engineers with practical tools for designing rectangular tunnels in real-world applications.</div></div>","PeriodicalId":50866,"journal":{"name":"Advances in Engineering Software","volume":"201 ","pages":"Article 103861"},"PeriodicalIF":4.0,"publicationDate":"2024-12-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143182901","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 hunter–prey optimizer with its applications

IF 4 2区工程技术 Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Advances in Engineering Software

Pub Date : 2024-12-28 DOI: 10.1016/j.advengsoft.2024.103857

Qiuyu Yuan , Zunfeng Du , Haiming Zhu , Muxuan Han , Haitao Zhu , Yancang Li

In recent years, metaheuristic algorithms have shown great potential in solving complex optimization problems. However, when applied to multimodal optimization problems and scenarios involving large-scale, high-dimensional, and dynamic environments, existing algorithms still have limitations. Addressing the limitations of the Hunter–prey Optimizer (HPO), characterized by low optimization accuracy and susceptibility to local optima, this paper introduces an Improved Hunter–prey Optimizer (IHPO). The main improvements of IHPO include: (1) refining the adaptive parameter C to enhance the balance between global exploration and local exploitation throughout the iteration process; (2) introducing predator search behavior early in the process to boost global search capabilities; (3) adopting a dual-population interaction strategy, effectively regulating global and local search abilities through sequential initialization, and maintaining continuous information exchange between the two evolutionary populations. To extend its utility to multi-objective optimization, this paper introduces the Multi-Objective Hunter–prey Optimizer (MOHPO) and the Multi-Objective Improved Hunter–prey Optimizer (MOIHPO). To validate the efficacy of these enhancements, simulation experiments are conducted on 23 test functions, CEC-2022, CEC-2017, and CEC-2019 test suites. The optimization performance of MOIHPO is further assessed through six multi-objective test functions, demonstrating notable advantages in terms of convergence speed, accuracy, and stability. To validate IHPO's practical application in engineering optimization, truss optimization design, and an Extreme Learning Machine (ELM) regression prediction problem are considered. The results underscore IHPO's enhanced applicability in engineering optimization scenarios. The source code of IHPO is publicly availabe at https://ww2.mathworks.cn/matlabcentral/fileexchange/177049-improved-hunter-prey-optimizer-ihpo.

{"title":"An improved hunter–prey optimizer with its applications","authors":"Qiuyu Yuan , Zunfeng Du , Haiming Zhu , Muxuan Han , Haitao Zhu , Yancang Li","doi":"10.1016/j.advengsoft.2024.103857","DOIUrl":"10.1016/j.advengsoft.2024.103857","url":null,"abstract":"<div><div>In recent years, metaheuristic algorithms have shown great potential in solving complex optimization problems. However, when applied to multimodal optimization problems and scenarios involving large-scale, high-dimensional, and dynamic environments, existing algorithms still have limitations. Addressing the limitations of the Hunter–prey Optimizer (HPO), characterized by low optimization accuracy and susceptibility to local optima, this paper introduces an Improved Hunter–prey Optimizer (IHPO). The main improvements of IHPO include: (1) refining the adaptive parameter C to enhance the balance between global exploration and local exploitation throughout the iteration process; (2) introducing predator search behavior early in the process to boost global search capabilities; (3) adopting a dual-population interaction strategy, effectively regulating global and local search abilities through sequential initialization, and maintaining continuous information exchange between the two evolutionary populations. To extend its utility to multi-objective optimization, this paper introduces the Multi-Objective Hunter–prey Optimizer (MOHPO) and the Multi-Objective Improved Hunter–prey Optimizer (MOIHPO). To validate the efficacy of these enhancements, simulation experiments are conducted on 23 test functions, CEC-2022, CEC-2017, and CEC-2019 test suites. The optimization performance of MOIHPO is further assessed through six multi-objective test functions, demonstrating notable advantages in terms of convergence speed, accuracy, and stability. To validate IHPO's practical application in engineering optimization, truss optimization design, and an Extreme Learning Machine (ELM) regression prediction problem are considered. The results underscore IHPO's enhanced applicability in engineering optimization scenarios. The source code of IHPO is publicly availabe at <span><span>https://ww2.mathworks.cn/matlabcentral/fileexchange/177049-improved-hunter-prey-optimizer-ihpo</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50866,"journal":{"name":"Advances in Engineering Software","volume":"201 ","pages":"Article 103857"},"PeriodicalIF":4.0,"publicationDate":"2024-12-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143182900","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 dual-channel deep neural network for tunnel boring machine slurry circulation system data prediction 用于隧道掘进机泥浆循环系统数据预测的新型双通道深度神经网络

IF 4 2区工程技术 Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Advances in Engineering Software

Pub Date : 2024-12-27 DOI: 10.1016/j.advengsoft.2024.103853

Rui Zhu , Qingchao Sun , Xuezhi Han , Huqiang Wang , Maolin Shi

The slurry circulation system is a crucial component of the Slurry Pressure Balance Tunnel Boring Machine (SPB TBM),with the pressure and flow at the inlet and outlet sections pipelines significant parameters for SPB TBMs.Accurate prediction of these parameters is essential for maintaining face pressure and preventing surface settlement or heave, providing a reference for TBM control adjustments.This research proposes a novel Dual-channel Hybrid Model based on Variational Mode Decomposition and Self-attention Temporal Convolutional Networks (DHM-VSATCN) to address this issue.This multi-input multi-output model is designed to forecast pressure and flow in slurry pipelines accurately.This method encompasses several key components, including data preprocessing,signal decomposition, an enhanced dual-channel deep learning model,a loss function, and evaluation metrics to ensure prediction accuracy. Validation of the model using a real SPB TBM operation dataset demonstrates that the model achieves excellent performance for five pressure and flow rate parameters, with low Mean Absolute Errors (MAE) ranging from 0.0032 to 4.01,R² values above 0.95, and Mean Absolute Percentage Errors (MAPE) consistently below 0.23 %. The comparative analysis highlights the superior performance of the proposed DHM-VSATCN method over models such as SVR, XGB, FTS, ARIMA, RNN, LSTM and iTransformer. Furthermore,in the context of multi-output prediction problems,the proposed dual-channel modeling strategy not only ensures prediction accuracy but also reduces training time compared to existing modeling strategies. The proposed DHM-VSATCN achieves an all-MAPE of only 0.7253 % across five parameters,with a model training time of just 1212.8 s.Therefore, this method is an effective solution for predicting TBM performance and offers valuable insights for other engineering scenarios requiring the prediction of multiple related outputs using the same input.

{"title":"A novel dual-channel deep neural network for tunnel boring machine slurry circulation system data prediction","authors":"Rui Zhu , Qingchao Sun , Xuezhi Han , Huqiang Wang , Maolin Shi","doi":"10.1016/j.advengsoft.2024.103853","DOIUrl":"10.1016/j.advengsoft.2024.103853","url":null,"abstract":"<div><div>The slurry circulation system is a crucial component of the Slurry Pressure Balance Tunnel Boring Machine (SPB TBM),with the pressure and flow at the inlet and outlet sections pipelines significant parameters for SPB TBMs.Accurate prediction of these parameters is essential for maintaining face pressure and preventing surface settlement or heave, providing a reference for TBM control adjustments.This research proposes a novel Dual-channel Hybrid Model based on Variational Mode Decomposition and Self-attention Temporal Convolutional Networks (DHM-VSATCN) to address this issue.This multi-input multi-output model is designed to forecast pressure and flow in slurry pipelines accurately.This method encompasses several key components, including data preprocessing,signal decomposition, an enhanced dual-channel deep learning model,a loss function, and evaluation metrics to ensure prediction accuracy. Validation of the model using a real SPB TBM operation dataset demonstrates that the model achieves excellent performance for five pressure and flow rate parameters, with low Mean Absolute Errors (MAE) ranging from 0.0032 to 4.01,<em>R</em><sup>2</sup> values above 0.95, and Mean Absolute Percentage Errors (MAPE) consistently below 0.23 %. The comparative analysis highlights the superior performance of the proposed DHM-VSATCN method over models such as SVR, XGB, FTS, ARIMA, RNN, LSTM and iTransformer. Furthermore,in the context of multi-output prediction problems,the proposed dual-channel modeling strategy not only ensures prediction accuracy but also reduces training time compared to existing modeling strategies. The proposed DHM-VSATCN achieves an all-MAPE of only 0.7253 % across five parameters,with a model training time of just 1212.8 s.Therefore, this method is an effective solution for predicting TBM performance and offers valuable insights for other engineering scenarios requiring the prediction of multiple related outputs using the same input.</div></div>","PeriodicalId":50866,"journal":{"name":"Advances in Engineering Software","volume":"201 ","pages":"Article 103853"},"PeriodicalIF":4.0,"publicationDate":"2024-12-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143182912","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

Filling gaps in MODIS NDVI data using hybrid multiple imputation–Machine learning and DINCAE techniques: Case study of the State of Hawaii

IF 4 2区工程技术 Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Advances in Engineering Software

Pub Date : 2024-12-25 DOI: 10.1016/j.advengsoft.2024.103856

Trang Thi Kieu Tran , Sayed M. Bateni , Hamid Mohebzadeh , Changhyun Jun , Manish Pandey , Dongkyn Kim

Normalized difference vegetation index (NDVI) data are vital for monitoring vegetation dynamics and health. However, NDVI time-series data obtained via remote sensing often contain missing values due to factors such as cloud cover, snow, and hardware failures. To address this problem and fill gaps in NDVI data from the Moderate Resolution Imaging Spectroradiometer (MODIS), this study combines the multiple imputations by chained equations (MICE) model with three machine learning techniques: Knearest neighbor, multilayer perceptron (MLP), and boosted regression tree. Additionally, the data interpolating convolutional auto-encoder (DINCAE), a recently proposed imputation method, is employed for imputation and comparison. The performance of all these models is evaluated using MODIS NDVI data from Oahu, Hawaii for training and validation. Synthetic scenarios with gap sizes of 20 %, 40 %, 60 %, and 80 % are created to assess the models’ feasibility for each gap size. Furthermore, all models are tested using data from Hawaii Island and Maui. Results indicate that the MICE-MLP model achieves the highest accuracy in imputing missing NDVI values on Oahu, with root mean square error (RMSE) values of 0.1028, 0.1112, and 0.1224 for missing ratios of 20 %, 40 %, and 60 %, respectively. Similarly, MICE-MLP outperforms other models using Hawaii Island and Maui data at gap sizes below 80 %. While the DINCAE model demonstrates superior accuracy at an 80 % gap size, its computational speed is slower than MICE-MLP. Overall, the findings underscore the robustness and accuracy of the MICE-MLP model in imputing missing NDVI data, making it a reliable alternative to existing methods.

{"title":"Filling gaps in MODIS NDVI data using hybrid multiple imputation–Machine learning and DINCAE techniques: Case study of the State of Hawaii","authors":"Trang Thi Kieu Tran , Sayed M. Bateni , Hamid Mohebzadeh , Changhyun Jun , Manish Pandey , Dongkyn Kim","doi":"10.1016/j.advengsoft.2024.103856","DOIUrl":"10.1016/j.advengsoft.2024.103856","url":null,"abstract":"<div><div>Normalized difference vegetation index (NDVI) data are vital for monitoring vegetation dynamics and health. However, NDVI time-series data obtained via remote sensing often contain missing values due to factors such as cloud cover, snow, and hardware failures. To address this problem and fill gaps in NDVI data from the Moderate Resolution Imaging Spectroradiometer (MODIS), this study combines the multiple imputations by chained equations (MICE) model with three machine learning techniques: Knearest neighbor, multilayer perceptron (MLP), and boosted regression tree. Additionally, the data interpolating convolutional auto-encoder (DINCAE), a recently proposed imputation method, is employed for imputation and comparison. The performance of all these models is evaluated using MODIS NDVI data from Oahu, Hawaii for training and validation. Synthetic scenarios with gap sizes of 20 %, 40 %, 60 %, and 80 % are created to assess the models’ feasibility for each gap size. Furthermore, all models are tested using data from Hawaii Island and Maui. Results indicate that the MICE-MLP model achieves the highest accuracy in imputing missing NDVI values on Oahu, with root mean square error (RMSE) values of 0.1028, 0.1112, and 0.1224 for missing ratios of 20 %, 40 %, and 60 %, respectively. Similarly, MICE-MLP outperforms other models using Hawaii Island and Maui data at gap sizes below 80 %. While the DINCAE model demonstrates superior accuracy at an 80 % gap size, its computational speed is slower than MICE-MLP. Overall, the findings underscore the robustness and accuracy of the MICE-MLP model in imputing missing NDVI data, making it a reliable alternative to existing methods.</div></div>","PeriodicalId":50866,"journal":{"name":"Advances in Engineering Software","volume":"201 ","pages":"Article 103856"},"PeriodicalIF":4.0,"publicationDate":"2024-12-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143182911","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