.png){kind=icon}
返回
Kriging-KNN Hybrid Analysis Method for Structural Reliability Analysis
.png){kind=icon}
PRE
AI
DOI：10.1061/(ASCE)BE.1943-5592.0001837.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
When the conventional response surface method is used to solve the reliability problem in complex structures, the response surface fitting accuracy is low, and the reliability accuracy does not satisfy the requirements of the design specifications owing to the complex structure and highly nonlinear implicit functional function. Therefore, the kriging proxy model is used to construct the response surface of the implicit functional function. In addition, the kriging proxy model is combined with the k-nearest neighbor (KNN) algorithm. By improving the optimization efficiency of the model parameters, the constructed implicit function can be used to simulate the structural limit state function. Therefore, kriging-KNN hybrid analysis to calculate structure failure probability will be proposed. A numerical example will be provided to demonstrate the effectiveness of the proposed method. The results show that the proposed method utilized the kriging proxy model to construct a response surface with high fitting accuracy that used a few samples. In addition, the KNN will be used to address the inadequate accuracy and efficiency of the kriging agent model for classification; therefore, effectively improving the accuracy and efficiency of the structure reliability calculation. Compared with the traditional response surface method, the kriging-KNN hybrid analysis method reduced the error rate and improved the prediction accuracy and calculation efficiency significantly. Furthermore, the model could be easily combined with the existing general finite-element analysis software to analyze the reliability of complex structures.
Keyword:
Bridge engineering
Reliability analysis
Kriging model
Nearest neighbor algorithm
Response surface
期刊
IF:
3.5
论文数:
2.7K
被引数:
8.8K
机构
引用论文
Reliability analysis of shallow tunnels using the response surface methodology.png)
UNDERGROUND SPACE
OAAI
UNDERGROUND SPACEIF8.3
OAAI分享收藏分享收藏Comparison of artificial neural network (ANN) and response surface methodology (RSM) prediction in compressive strength of recycled concrete aggregates人工神经网络 (ANN) 和响应面法 (RSM) 预测再生混凝土骨料抗压强度的比较
PREAI
分享收藏
查看更多内容