基于广义S变换和并联神经网络的结构损伤识别研究
Study on structural damage identification based on generalized S-transform and parallel neural network
李行健 1吕建达 1赵凌云 1刁延松1
作者信息
- 1. 青岛理工大学 土木工程学院,青岛 266525
- 折叠
摘要
目前在利用CNN网络提取特征的结构损伤识别研究中,仅仅利用1D-CNN和2D-CNN提取的特征进行损伤识别存在准确率低、识别效率不高等问题.提出了一种基于广义S变换和并联神经网络的结构损伤识别方法.为了丰富输入信号的特征维度,利用广义S变换将滤波后的信号转化成时频图,并同时将一维加速度响应信号和二维时频图分别输入1D-CNN和2D-CNN中进行时域和时频域特征提取,并在汇聚层进行特征拼接,然后通过FC层和Softmax层对损伤识别结果进行分类.利用IASC-ASCE SHM Benchmark结构第二阶段试验数据对所提出的并联网络模型进行验证,结果表明,所提出的网络模型与其他同类方法相比具有更高的识别精度和识别效率.
Abstract
Among the existing researches on structural damage identification that use CNN network to extract features,problems such as low accuracy and low recognition efficiency can be found when only 1D-CNN and 2D-CNN are used to extract features for damage identi-fication.Therefore,this paper proposes a structural damage identification method based on generalized S-transform and parallel neural network.In order to enrich the feature dimen-sions of the input signal,the filtered signal is converted into a time-frequency diagram by u-sing the generalized S-transform.At the same time,the one-dimensional acceleration re-sponse signal and the two-dimensional time-frequency diagram are input into 1D-CNN and 2D-CNN respectively for time-frequency and time-frequency feature extraction,and the char-acteristics are spliced in the convergence layer.Then,the damage identification results are classified through FC layer and Softmax layer.The proposed parallel network model is veri-fied by the second-stage test data of IASC-ASCE SHM Benchmark structure.The results show that the proposed network model has higher identification accuracy and efficiency than other similar methods.
关键词
损伤识别/广义S变换/卷积神经网络/时频分析/特征融合Key words
damage identification/generalized S-transform/convolution neural network/time-frequency analysis/feature fusion引用本文复制引用
基金项目
山东省自然科学基金(ZR2021ME239)
出版年
2024
NETL
NSTL
维普
万方数据