Redirigiendo al acceso original de articulo en 20 segundos...
Inicio  /  Applied Sciences  /  Vol: 13 Par: 22 (2023)  /  Artículo
ARTÍCULO
TITULO

Prediction of Failure Modes of Steel Tube-Reinforced Concrete Shear Walls Using Blending Fusion Model Based on Generative Adversarial Networks Data Augmentation

Guangchao Yang    
Jigang Zhang    
Zhehao Ma and Weixiao Xu    

Resumen

The steel tube-reinforced concrete (STRC) shear wall plays an important role in the seismic design of high-rise building structures. Due to the synergistic collaboration between steel tubes and concrete, they effectively enhance the ductility and energy dissipation capacity of conventional shear walls. To identify vulnerable areas prone to brittle failure and optimize the design, it is essential to develop a rapid method for identifying the failure mode of STRC shear walls. In this study, a fast identification method of STCR shear wall failure modes based on a Blending fusion model with Generative Adversarial Network (GAN) augmented data is proposed. The GAN is employed to address the issue of inadequate experimental data by generating new samples. This method combines classification boosting (Catboost), Random Forest (RF), K-Nearest Neighbors (KNN), and Least Absolute Shrinkage and Selection Operator (LASSO) to establish the Blending-CRKL fusion model to improve the prediction accuracy of the failure mode of STRC shear walls. The results reveal a significant improvement in the prediction performance of KNN, Backpropagation Neural Network (BPNN), RF, Light Gradient Boosting Machine (LightGBM), Catboost, and Blending-CRKL models after augmenting the training set with GAN. On average, the accuracy increased by 13%, precision increased by 81%, recall increased by 48%, and F1 score increased by 67%. The proposed Blending-CRKL fusion model outperforms the tested KNN, BPNN, RF, LightGBM, and Catboost models, achieving an accuracy rate of 97% in predicting the failure mode of STRC shear walls. Additionally, the stability and robustness of the Blending-CRKL model were validated, while the important features and value ranges of different failure modes were analyzed. This study provides a reference for the rapid identification of the failure mode of STRC shear walls.

 Artículos similares

       
 
Haobin Wen, Long Zhang and Jyoti K. Sinha    
On top of the condition-based maintenance (CBM) practice for rotating machinery, the robust estimation of remaining useful life (RUL) for rolling-element bearings (REB) is of particular interest. The failure of a single bearing often results in secondary... ver más
Revista: Applied Sciences

 
Evangelos Filippou, Spyridon Kilimtzidis, Athanasios Kotzakolios and Vassilis Kostopoulos    
The pursuit of more efficient transport has led engineers to develop a wide variety of aircraft configurations with the aim of reducing fuel consumption and emissions. However, these innovative designs introduce significant aeroelastic couplings that can... ver más
Revista: Aerospace

 
Naseer Muhammad Khan, Liqiang Ma, Muhammad Zaka Emad, Tariq Feroze, Qiangqiang Gao, Saad S. Alarifi, Li Sun, Sajjad Hussain and Hui Wang    
The brittleness index is one of the most integral parameters used in assessing rock bursts and catastrophic rock failures resulting from deep underground mining activities. Accurately predicting this parameter is crucial for effectively monitoring rock b... ver más
Revista: Water

 
Xiaokai Li, Xiaolong Zhang, Faming Zhang, Jian Huang, Shixiong Tang and Zhiqing Liu    
The mountainous areas of Southwest China have the characteristics of valley deep-cutting, a large topographic gradient, complex geological structures, etc. With the development of infrastructure construction in the area, the construction of bridges acros... ver más
Revista: Water

 
Evan Anderson, Budi Gunawan, James Nicholas, Mathew Ingraham and Bernadette A. Hernandez-Sanchez    
Marine energy generation technologies such as wave and tidal power have great potential in meeting the need for renewable energy in the years ahead. Yet, many challenges remain associated with marine-based systems because of the corrosive environment. Co... ver más