针对传统卷积神经网络(convolutional neural network,CNN)方法在时空特征提取存在不足,提出了一种改进的Incep-tion 与双向长短期记忆(bi-directional long short-term memory,BiLSTM)联合模型,以全面学习振动信号中的空间和时序信息.首先,构建具有多尺度感受野的Inception模块,自适应地提取不同尺度下的空间特征;其次,BiLSTM序列化处理时间特征,以深度挖掘时间相关性;最后,通过全局平均池化和Softmax分类器来实现钢框架结构的损伤识别.为评估该模型对噪声的鲁棒性,引入高斯白噪声作为干扰.此外,采用迁移学习策略来评估模型在不同强度激励和小样本下的泛化能力,确保适用于不同的损伤识别任务.结果表明,与传统的CNN方法相比,该模型在无噪声条件下及信噪比超过25 dB时保持了 100％的识别精度.该方法解决了土木工程应用中样本量不足和不同强度激励的实际挑战.通过微调预训练模型的参数,实现了在不同强度激励和小样本情况下的知识迁移与泛化,从而增强了模型的实际适用性.
Structural Damage Detection Study Based on Inception-BiLSTM and Transfer Learning
In response to the limitations of the traditional convolutional neural network(CNN)method in spatiotemporal feature ex-traction,an improved joint model combining Inception and bi-directional long short-term memory(BiLSTM)was proposed to compre-hensively learn spatial and temporal information from vibration signals.First,an Inception module with multi-scale receptive fields was constructed to adaptively extract spatial features at different scales.Secondly,temporal features were serialized using BiLSTM to deeply mine temporal correlations.Finally,damage identification of steel frame structures was achieved through global average pooling and a softmax classifier.To evaluate the model's robustness to noise,Gaussian white noise was introduced as interference.Additionally,a transfer learning strategy was employed to assess the model's generalization ability under different intensity excitations and small sample sizes,ensuring applicability to diverse damage identification tasks.The results demonstrate that compared with the traditional CNN method,the model maintains 100％recognition accuracy under noise-free conditions and when the signal-to-noise ratio exceeds 25 dB.This approach effectively addresses the practical challenges of insufficient sample size and varying excitation intensities in civil engi-neering applications.Through fine-tuning the parameters of the pre-trained model,knowledge transfer and generalization under different intensity incentives and small sample sizes are achieved,thereby enhancing the model's practical applicability.
steel framedamage identificationInceptionbi-directional long short-term memorytransfer learning
NETL
NSTL
万方数据
