Pub Date : 2024-11-04DOI: 10.1016/j.advengsoft.2024.103806
Hyeon-Gyeong Lee, Hyun-Gyu Kim
In this paper, a novel method is proposed to improve the accuracy of local solutions of the PODI-RBF method for real-time finite element (FE) simulations of indenter contact problems. In the offline stage, proper orthogonal decomposition (POD) basis vectors and coefficients are extracted from solution snapshots collected from full-order FE simulations of indenter contact problems with training contact locations and indentation depths. In the online stage, RBF interpolation is used to estimate POD basis vectors and coefficients for a new contact loading. Although the POD with interpolation (PODI) method using RBFs is very useful for obtaining FE solutions of indenter contact problems in real time, local solutions near a new contact location are less accurate when a new contact location is not close to the training contact locations. To improve the accuracy of local solutions near a new contact location, the first POD basis vector is replaced by the shifted first POD basis vector for the closest training contact location to the new contact location. Numerical results show that the modified PODI-RBF method is efficient and effective to achieve real-time FE simulations of indenter contact problems while improving the accuracy of local solutions near contact locations.
本文提出了一种新方法，以提高用于压头接触问题实时有限元（FE）模拟的 PODI-RBF 方法局部解的精度。在离线阶段，适当的正交分解（POD）基向量和系数是从训练接触位置和压痕深度的压头接触问题全阶有限元模拟的解快照中提取的。在联机阶段，使用 RBF 插值估计新接触载荷的 POD 基向量和系数。虽然使用 RBF 的插值 POD（PODI）方法对于实时获得压头接触问题的有限元求解非常有用，但当新接触位置与训练接触位置不相近时，新接触位置附近的局部求解精度较低。为了提高新接触位置附近局部解的精确度，第一 POD 基向量由与新接触位置最近的训练接触位置的移位第一 POD 基向量代替。数值结果表明，改进后的 PODI-RBF 方法能高效实现压头接触问题的实时有限元模拟，同时提高接触位置附近局部解的精度。
{"title":"A modified PODI-RBF method to improve the accuracy of local solutions for real-time finite element simulations of indenter contact problems","authors":"Hyeon-Gyeong Lee, Hyun-Gyu Kim","doi":"10.1016/j.advengsoft.2024.103806","DOIUrl":"10.1016/j.advengsoft.2024.103806","url":null,"abstract":"<div><div>In this paper, a novel method is proposed to improve the accuracy of local solutions of the PODI-RBF method for real-time finite element (FE) simulations of indenter contact problems. In the offline stage, proper orthogonal decomposition (POD) basis vectors and coefficients are extracted from solution snapshots collected from full-order FE simulations of indenter contact problems with training contact locations and indentation depths. In the online stage, RBF interpolation is used to estimate POD basis vectors and coefficients for a new contact loading. Although the POD with interpolation (PODI) method using RBFs is very useful for obtaining FE solutions of indenter contact problems in real time, local solutions near a new contact location are less accurate when a new contact location is not close to the training contact locations. To improve the accuracy of local solutions near a new contact location, the first POD basis vector is replaced by the shifted first POD basis vector for the closest training contact location to the new contact location. Numerical results show that the modified PODI-RBF method is efficient and effective to achieve real-time FE simulations of indenter contact problems while improving the accuracy of local solutions near contact locations.</div></div>","PeriodicalId":50866,"journal":{"name":"Advances in Engineering Software","volume":"199 ","pages":"Article 103806"},"PeriodicalIF":4.0,"publicationDate":"2024-11-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142578513","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}
Pub Date : 2024-10-30DOI: 10.1016/j.advengsoft.2024.103803
Haoqing Ding , Bingwen Qian , Yutao Hu , Changli Wang , Xin Zhang , Ruqi Sun , Bin Xu
The isogeometric analysis of variable-stiffness structures with curvilinear fibers has gained considerable research attention. However, dealing with structures that have complex cutouts poses challenges for isogeometric analysis. Additionally, the thermal-elastic behavior of variable-stiffness structures must be carefully considered, as they often operate in thermal environments. This study introduces a novel trimmed non-uniform rational basis spline (NURBS) method to address these challenges and investigate the thermal buckling behavior of variable-stiffness plates. The method generates trimmed NURBS elements using a level-set function on the initial NURBS mesh to describe complex geometries. Segmented density interpolation formulas are proposed to capture the contributions of different NURBS elements and to prevent localized eigenmodes. An artificial shear correction factor is introduced to mitigate shear locking. Several numerical examples with various boundary conditions and fiber configurations, are presented to demonstrate the high accuracy and low computational costs of the proposed method.
{"title":"A trimmed-NURBS-based thermal buckling isogeometric analysis framework for the variable stiffness plate with complex cutouts","authors":"Haoqing Ding , Bingwen Qian , Yutao Hu , Changli Wang , Xin Zhang , Ruqi Sun , Bin Xu","doi":"10.1016/j.advengsoft.2024.103803","DOIUrl":"10.1016/j.advengsoft.2024.103803","url":null,"abstract":"<div><div>The isogeometric analysis of variable-stiffness structures with curvilinear fibers has gained considerable research attention. However, dealing with structures that have complex cutouts poses challenges for isogeometric analysis. Additionally, the thermal-elastic behavior of variable-stiffness structures must be carefully considered, as they often operate in thermal environments. This study introduces a novel trimmed non-uniform rational basis spline (NURBS) method to address these challenges and investigate the thermal buckling behavior of variable-stiffness plates. The method generates trimmed NURBS elements using a level-set function on the initial NURBS mesh to describe complex geometries. Segmented density interpolation formulas are proposed to capture the contributions of different NURBS elements and to prevent localized eigenmodes. An artificial shear correction factor is introduced to mitigate shear locking. Several numerical examples with various boundary conditions and fiber configurations, are presented to demonstrate the high accuracy and low computational costs of the proposed method.</div></div>","PeriodicalId":50866,"journal":{"name":"Advances in Engineering Software","volume":"199 ","pages":"Article 103803"},"PeriodicalIF":4.0,"publicationDate":"2024-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142554744","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}
Concrete cracks will greatly affect the normal use function of buildings. Traditional crack detection and image processing methods have problems such as large amounts of calculation and low detection accuracy. In this paper, the DeepLabV3+ network model is improved by introducing CBAM and ECANet attention mechanisms. The backbone stem module is changed to three 3 × 3 convolutions with larger receptive fields, and three low-level feature maps are extracted as input maps for the decoder to enhance semantic information, and finally form the C-E-DeepLabV3+ model. The method proposed in this paper is validated by integrating multiple typical crack image libraries such as Crack500. The results show that the MIoU value can reach 77.84 %, which is 4 %, 5.53 %, 6.52 %, 4.49 % and 3.44 % higher than the original model DeepLabV3+, advanced segmentation model YOLOv8x, classical segmentation models UNet, MobileNet and PSPNet, respectively. And in terms of model parameter amount, it is 39 % lower than the original DeepLabV3+ model, and compared to other traditional models, it is only slightly higher than the lightweight model MobileNet. On this basis, the orthogonal skeleton line method is used to calculate the length and width of segmented cracks. Compared with the actual measured values, the accuracy of the method in this paper can reach more than 93 %, which has good engineering applicability.
{"title":"Concrete crack recognition and geometric parameter evaluation based on deep learning","authors":"Wang Shaowei, Xu Jiangbo, Wu Xiong, Zhang Jiajun, Zhang Zixuan, Chen Xinyu","doi":"10.1016/j.advengsoft.2024.103800","DOIUrl":"10.1016/j.advengsoft.2024.103800","url":null,"abstract":"<div><div>Concrete cracks will greatly affect the normal use function of buildings. Traditional crack detection and image processing methods have problems such as large amounts of calculation and low detection accuracy. In this paper, the DeepLabV3+ network model is improved by introducing CBAM and ECANet attention mechanisms. The backbone stem module is changed to three 3 × 3 convolutions with larger receptive fields, and three low-level feature maps are extracted as input maps for the decoder to enhance semantic information, and finally form the C-E-DeepLabV3+ model. The method proposed in this paper is validated by integrating multiple typical crack image libraries such as Crack500. The results show that the MIoU value can reach 77.84 %, which is 4 %, 5.53 %, 6.52 %, 4.49 % and 3.44 % higher than the original model DeepLabV3+, advanced segmentation model YOLOv8x, classical segmentation models UNet, MobileNet and PSPNet, respectively. And in terms of model parameter amount, it is 39 % lower than the original DeepLabV3+ model, and compared to other traditional models, it is only slightly higher than the lightweight model MobileNet. On this basis, the orthogonal skeleton line method is used to calculate the length and width of segmented cracks. Compared with the actual measured values, the accuracy of the method in this paper can reach more than 93 %, which has good engineering applicability.</div></div>","PeriodicalId":50866,"journal":{"name":"Advances in Engineering Software","volume":"199 ","pages":"Article 103800"},"PeriodicalIF":4.0,"publicationDate":"2024-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142554743","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}
Pub Date : 2024-10-29DOI: 10.1016/j.advengsoft.2024.103802
Lang Xu , Zhiping Wen , Huaizhi Su , Simonetta Cola , Nicola Fabbian , Yanming Feng , Shanshan Yang
Since one of the main threats to the safety of earth-rock dams is leakage, its timely and accurate identification is crucial. Distributed fiber optic sensing system (DFOS) is considered as one of the ideal methods for leakage monitoring in earth-rock dams. However, the working conditions of earth-rock dams are complex, and the identification of fiber optic temperature measurements has issues such as low efficiency and high misjudgment rate. For improving the identification efficiency and accuracy of fiber optic temperature measurements in earth-rock dams, a signal identification method integrating least squares generative adversarial network (LSGAN), one-dimensional convolutional neural network (1DCNN), and white shark optimization (WSO) algorithm is presented. Firstly, the LSGAN model is used to augment the signals of different categories to reduce the effect of data set unbalance on the identification result. According to the variation characteristics of fiber optic temperature measurement signals in earth-rock dams, a 1DCNN model is designed to extract signal features for classification. To reduce the blindness in hyperparameter setting of 1DCNN model, the WSO algorithm is introduced to optimize its key hyperparameters, which further enhances the identification accuracy of the model. The new method is applied to a data set specifically acquired with tests on a physical model of an earth-rock dam. The identification accuracy obtained with the new method reaches 99.76 %, which is better than the accuracy of other commonly used identification methods. Upon completion of the pre-training, the new method can fulfill the practical needs of fast identification and has promising applications.
{"title":"An innovative method integrating two deep learning networks and hyperparameter optimization for identifying fiber optic temperature measurements in earth-rock dams","authors":"Lang Xu , Zhiping Wen , Huaizhi Su , Simonetta Cola , Nicola Fabbian , Yanming Feng , Shanshan Yang","doi":"10.1016/j.advengsoft.2024.103802","DOIUrl":"10.1016/j.advengsoft.2024.103802","url":null,"abstract":"<div><div>Since one of the main threats to the safety of earth-rock dams is leakage, its timely and accurate identification is crucial. Distributed fiber optic sensing system (DFOS) is considered as one of the ideal methods for leakage monitoring in earth-rock dams. However, the working conditions of earth-rock dams are complex, and the identification of fiber optic temperature measurements has issues such as low efficiency and high misjudgment rate. For improving the identification efficiency and accuracy of fiber optic temperature measurements in earth-rock dams, a signal identification method integrating least squares generative adversarial network (LSGAN), one-dimensional convolutional neural network (1DCNN), and white shark optimization (WSO) algorithm is presented. Firstly, the LSGAN model is used to augment the signals of different categories to reduce the effect of data set unbalance on the identification result. According to the variation characteristics of fiber optic temperature measurement signals in earth-rock dams, a 1DCNN model is designed to extract signal features for classification. To reduce the blindness in hyperparameter setting of 1DCNN model, the WSO algorithm is introduced to optimize its key hyperparameters, which further enhances the identification accuracy of the model. The new method is applied to a data set specifically acquired with tests on a physical model of an earth-rock dam. The identification accuracy obtained with the new method reaches 99.76 %, which is better than the accuracy of other commonly used identification methods. Upon completion of the pre-training, the new method can fulfill the practical needs of fast identification and has promising applications.</div></div>","PeriodicalId":50866,"journal":{"name":"Advances in Engineering Software","volume":"199 ","pages":"Article 103802"},"PeriodicalIF":4.0,"publicationDate":"2024-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142540037","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}
Pub Date : 2024-10-26DOI: 10.1016/j.advengsoft.2024.103798
Bin Wang , Jiantao Bai , Shanbin Lu , Wenjie Zuo
Contact phenomenon widely exists in engineering, which is a high nonlinearity problem. However, the majority of open source contact finite element codes are written in C++, which are difficult for junior researchers to adopt and use. Therefore, this paper provides an open source 528-line MATLAB code and detailed interpretation for frictional contact finite element analysis considering large deformation, which is easy to learn and use by newcomers. This paper describes the contact projection, contact nodal forces and contact tangent stiffness matrices. The nonlinear equations are solved by the Newton–Raphson method. Numerical examples demonstrate the effectiveness of the MATLAB codes. The displacement, Cauchy stress and contact traction results are compared with the open-source software FEBIO.
{"title":"An open source MATLAB solver for contact finite element analysis","authors":"Bin Wang , Jiantao Bai , Shanbin Lu , Wenjie Zuo","doi":"10.1016/j.advengsoft.2024.103798","DOIUrl":"10.1016/j.advengsoft.2024.103798","url":null,"abstract":"<div><div>Contact phenomenon widely exists in engineering, which is a high nonlinearity problem. However, the majority of open source contact finite element codes are written in C++, which are difficult for junior researchers to adopt and use. Therefore, this paper provides an open source 528-line MATLAB code and detailed interpretation for frictional contact finite element analysis considering large deformation, which is easy to learn and use by newcomers. This paper describes the contact projection, contact nodal forces and contact tangent stiffness matrices. The nonlinear equations are solved by the Newton–Raphson method. Numerical examples demonstrate the effectiveness of the MATLAB codes. The displacement, Cauchy stress and contact traction results are compared with the open-source software FEBIO.</div></div>","PeriodicalId":50866,"journal":{"name":"Advances in Engineering Software","volume":"199 ","pages":"Article 103798"},"PeriodicalIF":4.0,"publicationDate":"2024-10-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142526651","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}
Pub Date : 2024-10-24DOI: 10.1016/j.advengsoft.2024.103797
Haifeng Chen, Ping Yu, Jiangqi Long
The optimal design of automobile seats plays an important role in passenger safety in high-speed accidents. In order to enhance the accuracy of the prediction of the input variables and output response of the seat, a hybrid machine learning prediction model that combines the improved gray wolf optimizer (IGWO) and back propagation neural network (BPNN) has been proposed, and the prediction effect of the model was validated using the seat simulation data. Initially, based on the experimental data, finite element models were developed for eight typical working conditions of automobile seats and their accuracy was validated. Subsequently, the energy absorption to mass ratio method was employed to screen the design variables, resulting in the selection of 17 thickness variables and 15 material variables. Thereafter, the gray wolf optimizer (GWO) algorithm underwent enhancement through the incorporation of the dynamic leadership hierarchy (DLH) mechanism and the revision of the positional formula, yielding the IGWO algorithm. Following this, the IGWO algorithm was applied to optimize the hyperparameters of BPNN, culminating in the establishment of the IGWO-BPNN model. Ultimately, the seat multi-objective optimization design process was addressed using multi-objective gray wolf optimizer (MOGWO) to achieve the Pareto frontier, while the decision-making was conducted using the combined compromise solution (CoCoSo) method to determine the best trade-off solution. Furthermore, the effectiveness of the proposed optimal design method is evidenced by comparing the baseline design, simulation analysis, and optimal design methods. The results indicate that the optimized automotive seat frame achieves a reduction in cost by 20.7 % and mass by 22.9 %, simultaneously maintaining safety performance. Consequently, the proposed optimization design methodology is demonstrated to be highly effective for the multi-objective optimization design of automotive seat frames.
{"title":"Multi-objective optimization of automotive seat frames using machine learning","authors":"Haifeng Chen, Ping Yu, Jiangqi Long","doi":"10.1016/j.advengsoft.2024.103797","DOIUrl":"10.1016/j.advengsoft.2024.103797","url":null,"abstract":"<div><div>The optimal design of automobile seats plays an important role in passenger safety in high-speed accidents. In order to enhance the accuracy of the prediction of the input variables and output response of the seat, a hybrid machine learning prediction model that combines the improved gray wolf optimizer (IGWO) and back propagation neural network (BPNN) has been proposed, and the prediction effect of the model was validated using the seat simulation data. Initially, based on the experimental data, finite element models were developed for eight typical working conditions of automobile seats and their accuracy was validated. Subsequently, the energy absorption to mass ratio method was employed to screen the design variables, resulting in the selection of 17 thickness variables and 15 material variables. Thereafter, the gray wolf optimizer (GWO) algorithm underwent enhancement through the incorporation of the dynamic leadership hierarchy (DLH) mechanism and the revision of the positional formula, yielding the IGWO algorithm. Following this, the IGWO algorithm was applied to optimize the hyperparameters of BPNN, culminating in the establishment of the IGWO-BPNN model. Ultimately, the seat multi-objective optimization design process was addressed using multi-objective gray wolf optimizer (MOGWO) to achieve the Pareto frontier, while the decision-making was conducted using the combined compromise solution (CoCoSo) method to determine the best trade-off solution. Furthermore, the effectiveness of the proposed optimal design method is evidenced by comparing the baseline design, simulation analysis, and optimal design methods. The results indicate that the optimized automotive seat frame achieves a reduction in cost by 20.7 % and mass by 22.9 %, simultaneously maintaining safety performance. Consequently, the proposed optimization design methodology is demonstrated to be highly effective for the multi-objective optimization design of automotive seat frames.</div></div>","PeriodicalId":50866,"journal":{"name":"Advances in Engineering Software","volume":"199 ","pages":"Article 103797"},"PeriodicalIF":4.0,"publicationDate":"2024-10-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142526650","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}
Pub Date : 2024-10-22DOI: 10.1016/j.advengsoft.2024.103793
Jialing Yan , Gang Hu , Bin Shu
The Human Memory Optimization (HMO) algorithm is a newly released metaheuristic algorithm based on humans in 2023, which can effectively solve most optimization problems. However, when dealing with complex optimization problems, HMO has limitations such as insufficient convergence accuracy and susceptibility to local optimal solutions. To this end, we integrated chaotic mapping, Cauchy mutation, Gaussian mutation, differential mutation, and parameter dynamic adjustment strategies into the original algorithm and developed an enhanced MGCHMO algorithm. Firstly, in the initialization phase of the MGCHMO, the Tent mapping chaotic mapping mechanism is introduced to enhance the diversity and search ability of the initial population through the traversal and randomness characteristics of chaos. Secondly, in the memory generation phase, we added the Cauchy mutation strategy, which effectively expanded the search range of the algorithm, helped the algorithm escape from local optima, and explored a broader solution space. Then, during the recall phase, Gaussian mutation and differential mutation are added. Among them, Gaussian mutation enables the algorithm to perform more refined searches within a local range. Differential mutation, on the other hand, guides the algorithm to explore towards a more optimal solution through the information of individual differences. Finally, the parameters of the algorithm are dynamically adjusted to enhance its optimization performance, ensuring that the algorithm maintains optimal search performance at different phases, thereby accelerating the convergence process and improving the quality of the solution.
To verify the optimization performance of MGCHMO, we conducted a series of detailed performance experiments on three different test sets: CEC2017, CEC2020, and CEC2022. The results showed that MGCHMO has higher convergence and stability. In addition, we tested the applicability of MGCHMO on 30 engineering examples, topology optimization design, aerospace orbit optimization, and curve shape optimization, and the results further demonstrated the significant application capability and feasibility of MGCHMO.
{"title":"MGCHMO: A dynamic differential human memory optimization with Cauchy and Gauss mutation for solving engineering problems","authors":"Jialing Yan , Gang Hu , Bin Shu","doi":"10.1016/j.advengsoft.2024.103793","DOIUrl":"10.1016/j.advengsoft.2024.103793","url":null,"abstract":"<div><div>The Human Memory Optimization (HMO) algorithm is a newly released metaheuristic algorithm based on humans in 2023, which can effectively solve most optimization problems. However, when dealing with complex optimization problems, HMO has limitations such as insufficient convergence accuracy and susceptibility to local optimal solutions. To this end, we integrated chaotic mapping, Cauchy mutation, Gaussian mutation, differential mutation, and parameter dynamic adjustment strategies into the original algorithm and developed an enhanced MGCHMO algorithm. Firstly, in the initialization phase of the MGCHMO, the Tent mapping chaotic mapping mechanism is introduced to enhance the diversity and search ability of the initial population through the traversal and randomness characteristics of chaos. Secondly, in the memory generation phase, we added the Cauchy mutation strategy, which effectively expanded the search range of the algorithm, helped the algorithm escape from local optima, and explored a broader solution space. Then, during the recall phase, Gaussian mutation and differential mutation are added. Among them, Gaussian mutation enables the algorithm to perform more refined searches within a local range. Differential mutation, on the other hand, guides the algorithm to explore towards a more optimal solution through the information of individual differences. Finally, the parameters of the algorithm are dynamically adjusted to enhance its optimization performance, ensuring that the algorithm maintains optimal search performance at different phases, thereby accelerating the convergence process and improving the quality of the solution.</div><div>To verify the optimization performance of MGCHMO, we conducted a series of detailed performance experiments on three different test sets: CEC2017, CEC2020, and CEC2022. The results showed that MGCHMO has higher convergence and stability. In addition, we tested the applicability of MGCHMO on 30 engineering examples, topology optimization design, aerospace orbit optimization, and curve shape optimization, and the results further demonstrated the significant application capability and feasibility of MGCHMO.</div></div>","PeriodicalId":50866,"journal":{"name":"Advances in Engineering Software","volume":"198 ","pages":"Article 103793"},"PeriodicalIF":4.0,"publicationDate":"2024-10-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142527792","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}
Pub Date : 2024-10-19DOI: 10.1016/j.advengsoft.2024.103795
Lan Nguyen-Ngoc , Hoa Tran-Ngoc , Thang Le-Xuan , Chi-Thanh Nguyen , Guido De Roeck , Thanh Bui-Tien , Magd Abdel Wahab
This paper presents a novel two-step approach to identifying structural damages in bridge structure through the integration of 1D Convolutional Neural Network (1DCNN) and Long Short-Term Memory (LSTM) networks, enhanced by the augmentation and transformation techniques using Symbolic Aggregate approXimation (SAX) for time-series data analysis. In the first step, the time-series data of the bridge is diversified and quantified by augmentation techniques to make the model more robust and increase its generalization capabilities. After that, SAX is implemented to reduce the volume and categorize time series data through the transformation of continuous time series into discrete symbols, thereby decreasing the size of the data for more efficient training performance. In the second step, an advanced DL model combining 1DCNN and LSTM is proposed to tackle the damage identification problems of the processed data. By leveraging the strengths of CNNs in feature extraction and LSTMs in sequence learning, combined with advanced techniques for data augmentation, our methodology offers a robust solution not only for improving the model's training process but also for enabling it to learn from a more diverse and comprehensive dataset that mimics different damage scenarios, allowing more accurate detection of damages within bridge structures. Validation of the proposed method is conducted using time-series data collected from Chuong Duong Bridge structure. The effectiveness of the proposed method is compared with other models, such as 1DCNN, LSTM, and the combined 1DCNN-LSTM. The results show that the proposed 1DCNN-LSTM-SAX outperforms the other methods in terms of accuracy and, thus, can be used extensively to deal with the damage identification problems of bridges using time-series data.
{"title":"A two-step approach for damage identification in bridge structure using convolutional Long Short-Term Memory with augmented time-series data","authors":"Lan Nguyen-Ngoc , Hoa Tran-Ngoc , Thang Le-Xuan , Chi-Thanh Nguyen , Guido De Roeck , Thanh Bui-Tien , Magd Abdel Wahab","doi":"10.1016/j.advengsoft.2024.103795","DOIUrl":"10.1016/j.advengsoft.2024.103795","url":null,"abstract":"<div><div>This paper presents a novel two-step approach to identifying structural damages in bridge structure through the integration of 1D Convolutional Neural Network (1DCNN) and Long Short-Term Memory (LSTM) networks, enhanced by the augmentation and transformation techniques using Symbolic Aggregate approXimation (SAX) for time-series data analysis. In the first step, the time-series data of the bridge is diversified and quantified by augmentation techniques to make the model more robust and increase its generalization capabilities. After that, SAX is implemented to reduce the volume and categorize time series data through the transformation of continuous time series into discrete symbols, thereby decreasing the size of the data for more efficient training performance. In the second step, an advanced DL model combining 1DCNN and LSTM is proposed to tackle the damage identification problems of the processed data. By leveraging the strengths of CNNs in feature extraction and LSTMs in sequence learning, combined with advanced techniques for data augmentation, our methodology offers a robust solution not only for improving the model's training process but also for enabling it to learn from a more diverse and comprehensive dataset that mimics different damage scenarios, allowing more accurate detection of damages within bridge structures. Validation of the proposed method is conducted using time-series data collected from Chuong Duong Bridge structure. The effectiveness of the proposed method is compared with other models, such as 1DCNN, LSTM, and the combined 1DCNN-LSTM. The results show that the proposed 1DCNN-LSTM-SAX outperforms the other methods in terms of accuracy and, thus, can be used extensively to deal with the damage identification problems of bridges using time-series data.</div></div>","PeriodicalId":50866,"journal":{"name":"Advances in Engineering Software","volume":"198 ","pages":"Article 103795"},"PeriodicalIF":4.0,"publicationDate":"2024-10-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142527791","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}
Pub Date : 2024-10-19DOI: 10.1016/j.advengsoft.2024.103799
Prashant Kumar , Izaz Raouf , Jinwoo Song , Prince , Heung Soo Kim
The bearing is an indispensable part of mechanical systems. Fault diagnosis of bearing faults is vital for uninterrupted operations of the system, and to prevent catastrophic failure. Artificial intelligence implementation has revolutionized the bearing fault diagnosis method. Application of deep learning has eliminated manual feature extraction and selection requirements. While conventional convolutional neural networks have demonstrated potential in diagnosing faults, considering a more extensive variety of spatial variables can further optimize their performance. This paper proposes a multi-wide-kernel convolutional neural network-based model for bearing fault diagnosis. We propose wide kernels in the neural network's convolutional layers, which enable the model to learn broader patterns from the input for bearing fault diagnosis. The wide-kernel design enables the network to obtain local and global features more effectively, improving the network's capacity to distinguish between healthy and faulty bearings. We train and validate the proposed multi-wide-kernel convolutional neural networks using an extensive dataset of vibration signals collected from bearings under diverse scenarios. Because of its increased sensitivity to subtle fault patterns, the proposed model offers better accuracy. The model's efficacy is further confirmed by comparing it with existing cutting-edge techniques for diagnosing bearing faults.
{"title":"Multi-size wide kernel convolutional neural network for bearing fault diagnosis","authors":"Prashant Kumar , Izaz Raouf , Jinwoo Song , Prince , Heung Soo Kim","doi":"10.1016/j.advengsoft.2024.103799","DOIUrl":"10.1016/j.advengsoft.2024.103799","url":null,"abstract":"<div><div>The bearing is an indispensable part of mechanical systems. Fault diagnosis of bearing faults is vital for uninterrupted operations of the system, and to prevent catastrophic failure. Artificial intelligence implementation has revolutionized the bearing fault diagnosis method. Application of deep learning has eliminated manual feature extraction and selection requirements. While conventional convolutional neural networks have demonstrated potential in diagnosing faults, considering a more extensive variety of spatial variables can further optimize their performance. This paper proposes a multi-wide-kernel convolutional neural network-based model for bearing fault diagnosis. We propose wide kernels in the neural network's convolutional layers, which enable the model to learn broader patterns from the input for bearing fault diagnosis. The wide-kernel design enables the network to obtain local and global features more effectively, improving the network's capacity to distinguish between healthy and faulty bearings. We train and validate the proposed multi-wide-kernel convolutional neural networks using an extensive dataset of vibration signals collected from bearings under diverse scenarios. Because of its increased sensitivity to subtle fault patterns, the proposed model offers better accuracy. The model's efficacy is further confirmed by comparing it with existing cutting-edge techniques for diagnosing bearing faults.</div></div>","PeriodicalId":50866,"journal":{"name":"Advances in Engineering Software","volume":"198 ","pages":"Article 103799"},"PeriodicalIF":4.0,"publicationDate":"2024-10-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142527790","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}
Pub Date : 2024-10-18DOI: 10.1016/j.advengsoft.2024.103792
Junxiang Li , Xiwei Guo , Longchao Cao , Xinxin Zhang
The performance of engineering structures often varies over time due to the randomness and time variability of material properties, environmental conditions and load effects. This paper proposes phase-type (PH) distribution-based methods for efficient time-variant reliability analysis. The core of the proposed methods is to approximate the extreme value of a stochastic process as a PH distributed random variable, and treat the time parameter as a uniformly distributed variable. Consequently, the time-variant reliability problem is transformed into a time-invariant one. Three representative time-invariant reliability methods, first-order reliability method (FORM), importance sampling (IS) and adaptive Kriging (AK) surrogate model-based IS method (AK-IS), are integrated with the PH distribution-based approximation strategy to form the proposed methods, namely PH-FORM, PH-IS and PH-AKIS. The efficiency and accuracy of these methods are demonstrated through three examples. All codes in the study are implemented in MATLAB and provided as supplementary materials.
{"title":"Time-variant reliability analysis using phase-type distribution-based methods","authors":"Junxiang Li , Xiwei Guo , Longchao Cao , Xinxin Zhang","doi":"10.1016/j.advengsoft.2024.103792","DOIUrl":"10.1016/j.advengsoft.2024.103792","url":null,"abstract":"<div><div>The performance of engineering structures often varies over time due to the randomness and time variability of material properties, environmental conditions and load effects. This paper proposes phase-type (PH) distribution-based methods for efficient time-variant reliability analysis. The core of the proposed methods is to approximate the extreme value of a stochastic process as a PH distributed random variable, and treat the time parameter as a uniformly distributed variable. Consequently, the time-variant reliability problem is transformed into a time-invariant one. Three representative time-invariant reliability methods, first-order reliability method (FORM), importance sampling (IS) and adaptive Kriging (AK) surrogate model-based IS method (AK-IS), are integrated with the PH distribution-based approximation strategy to form the proposed methods, namely PH-FORM, PH-IS and PH-AKIS. The efficiency and accuracy of these methods are demonstrated through three examples. All codes in the study are implemented in MATLAB and provided as supplementary materials.</div></div>","PeriodicalId":50866,"journal":{"name":"Advances in Engineering Software","volume":"198 ","pages":"Article 103792"},"PeriodicalIF":4.0,"publicationDate":"2024-10-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142444664","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}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号