Restricted accessResearch articleFirst published online 2026
New approach for enhanced damage identification in frame structures with limited data: leveraging integrated convolutional neural network model with transfer learning
To address the challenge of high-precision damage identification in frame structures with limited data availability, this article proposes an integrated HOLMSST-DAMIIM-2DCNN model by combing higher-order local maximum synchrosqueezing transform (HOLMSST), dual attention mechanism (DAM), improved inception module (IIM), and two-dimensional convolutional neural network (2DCNN).In this method, the HOLMSST algorithm is initially proposed to transform the one-dimensional response signals into time–frequency images. Subsequently, a new 2DCNN architecture is established by incorporating an IIM with a DAM, and then pretrained on a proportionally divided dataset. Furthermore, the pretrained model is fine-tuned to the target task through a transfer learning (TL) strategy, significantly bolstering the damage identification performance of the proposed model in scenarios characterized by data scarcity. Finally, the accuracy and generalization ability of the proposed HOLMSST-DAMIIM-2DCNN model and TL strategy are validated through four datasets. The results demonstrate that the proposed HOLMSST-DAMIIM-2DCNN model exhibits high accuracy and robust noise resistance in multicategory damage identification tasks for frame structures. Notably, even in the face of a limited number of samples, the integration of the HOLMSST-DAMIIM-2DCNN model with the TL strategy can still enhance the learning capability and damage identification accuracy significantly for frame structures.
YanBDingWJinZ, et al. Explainable machine learning-based prediction for aerodynamic interference of a low-rise building on a high-rise building. J Build Eng2024; 82: 108285.
2.
TianXSunXLiuG, et al. Optimization design of the jacket support structure for offshore wind turbine using topology optimization method. Ocean Eng2022; 243: 110084.
3.
ShiGHouLZhaoH. Numerical study on the seismic behaviour of high-strength steel moment-resisting frames: multi-scale modelling and validation. Thin Wall Struct2025; 208: 112807.
4.
ChenYLinWChenH, et al. Automatic detection method for loose bolts in steel frame structures based on wearable lightweight impedance system and multi-channel information fusion deep learning model. Struct Health Monit2025.
5.
BartschHSchaffrathSDiY, et al. Damage mechanics based failure prediction for structures under seismic action. J Constr Steel Res2020; 173: 106264.
6.
ChaYJAliRLewisJ, et al. Deep learning-based structural health monitoring. Autom Constr2024; 161: 105328.
7.
ZhangBRenYHeS, et al. A review of methods and applications in structural health monitoring (SHM) for bridges. Measurement2025; 245: 116575.
8.
SayandipGKoushikR. Damage quantification using spectral response of a multi-degree-of-freedom system with spatial and temporal stiffness variations: application to shear-type frames. Measurement2025; 242: 116109.
9.
RosafalcoLTorzoniMManzoniA, et al. Online structural health monitoring by model order reduction and deep learning algorithms. Comput Struct2021; 255: 106604.
10.
JiaJLiY. Deep learning for structural health monitoring: data, algorithms, applications, challenges, and trends. Sensors2023; 23(21): 8824.
11.
ZhangLShenJZhuB. A research on an improved Unet-based concrete crack detection algorithm. Struct Health Monit2021; 20(4): 1864–1879.
12.
PantaMHoqueMTAbdelguerfiM, et al. IterLUNet: deep learning architecture for pixel-wise crack detection in levee systems. IEEE Access2023; 11: 12249–12262.
13.
HeLReyndersESimoenE, et al. Vibration-based damage assessment of the Boirs viaduct with a multistage model updating approach. Struct Health Monit Epub ahead of print September 2025. DOI: 10.1177/14759217251374973
14.
AvciOAbdeljaberOKiranyazS, et al. A review of vibration-based damage detection in civil structures: from traditional methods to machine learning and deep learning applications. Mech Syst Signal Process2021; 147: 107077.
15.
LiuJLWangSFLiYZ, et al. Time-varying damage detection in beam structures using variational mode decomposition and continuous wavelet transform. Constr Build Mater2024; 411: 134416.
16.
AhmadzadehMZahraiSMBitarafM. An integrated deep neural network model combining 1D CNN and LSTM for structural health monitoring utilizing multisensor time-series data. Struct Health Monit2024; 24(1): 447–465.
17.
ZhangJGuoXZongS, et al. A new neural-network-based method for structural damage identification in single-layer reticulated shells. Front Struct Civ Eng2024; 18(1): 104–121.
18.
ChenDCuiHLiZ, et al. Indirect identification and analysis of bridge damage using vehicle-bridge coupled vibration and deep learning. J Perform Constr Facil2024; 38(4): 17–23.
19.
ShanDSunRTanK. Damage identification for cable-stayed model bridges on shaking table based on joint CNN & LSTM with multi-channel and multi-scale. Measurement2025; 245(6): 116572.
20.
AbdeljaberOAvciOKiranyazS, et al. Real-time vibration-based structural damage detection using one-dimensional convolutional neural networks. J Sound Vib2017; 388: 154–170.
21.
ZhangYQMiyamoriYMikamiS, et al. Vibration-based structural state identification by a 1-dimensional convolutional neural network. Comput Aided Civ Infrastruct Eng2019; 34(9): 822–839.
22.
DuranBEmoryDAzamYE, et al. A novel CNN architecture for robust structural damage identification via strain measurements and its validation via full-scale experiments. Measurement2025; 239: 115393.
23.
ChenZGrylliasKLiW. Mechanical fault diagnosis using convolutional neural networks and extreme learning machine. Mech Syst Signal Process2019; 133: 106272.
24.
NieGZhangZJiaoZ, et al. A novel intelligent bearing fault diagnosis method based on image enhancement and improved convolutional neural network. Measurement2025; 242: 116148.
25.
LiuJLChenRQiuFL, et al. Structural instantaneous frequency identification of non-stationary signals using GDAVMD and MSST. Structures2025; 72: 108234.
26.
LiCLiHGaoG, et al. An accelerating convolutional neural networks via a 2D entropy based-adaptive filter search method for image recognition. Appl Soft Comput2023; 142: 110326.
27.
ZhangXZhuXYuY, et al. Transfer learning-based structural damage identification for building structures with limited measurement data. Int J Struct Stab Dyn2025; 25(9): 2550093.
28.
LiuCXuXWuJ, et al. Deep transfer learning-based damage detection of composite structures by fusing monitoring data with physical mechanism. Eng Appl Artif Intell2023; 123: 106245.
29.
YosinskiJCluneJBengioY, et al. How transferable are features in deep neural networks. In: Proceedings of the international conference on neural information processing systems, Montréal, Canada, 2014.
30.
TengSChenXChenG, et al. Structural damage detection based on transfer learning strategy using digital twins of bridges. Mech Syst Signal Process2023; 191(1): 110160.
31.
TengSChenGYanZ, et al. Vibration-based structural damage detection using 1-D convolutional neural network and transfer learning. Struct Health Monit2023; 22(4): 2888–2909.
32.
ZhangXZhuXYuY, et al. Transfer learning-based structural damage identification for building structures with limited measurement data. Int J Struct Stab Dyn2024; 25(9): 2550093.
33.
YuGWangZZhaoP, et al. Local maximum synchrosqueezing transform: an energy-concentrated time-frequency analysis tool. Mech Syst Signal Process2019; 117: 537–552.
34.
SiYXuHZhuX, et al. SCSA: exploring the synergistic effects between spatial and channel attention. Neurocomputing2025; 634: 129866.
35.
XuYZhangLChenL, et al. Attention-guided residual spatiotemporal network with label regularization for fault diagnosis with small samples. Sensors2025; 25: 4772.
36.
TolieFHRenJElyanE. DICAM: deep inception and channel-wise attention modules for underwater image enhancement. Neurocomputing2024; 584: 127585.
37.
WangXWangSLiJ, et al. Omnidirectional image super-resolution via position attention network. Neural Netw2024; 178: 106464.
38.
GaoYXuXYiuTW, et al. Transfer learning for smart construction: advances and future directions. Autom Constr2025; 175: 106238.
39.
ZhangJPeiGZhuX, et al. Diesel engine fault diagnosis for multiple industrial scenarios based on transfer learning. Measurement2024; 228: 114338.
40.
LiuYFangSE. Cross-domain structural damage identification using transfer learning strategy. Eng Struct2024; 311: 118171.
41.
BaoYSunHXuY, et al. Recent advances in structural health diagnosis: a machine learning perspective. Adv Bridge Eng2025; 6: 7.
42.
XueYYangRChenX, et al. A review on transferability estimation in deep transfer learning. IEEE Trans Artif Intell2024; 5: 5894–5914.
43.
LiuJLChenLHWeiXJ. Damage identification for steel frame structures based on improved one-dimensional depthwise separable convolutional neural network. J Civil Struct Health Monit2025; 15: 2903–2929.
44.
FigueiredoEParkGFarrarCR, et al. Machine learning algorithms for damage detection under operational and environmental variability. Struct Health Monit2011; 10(6): 559–572.
45.
JiangKHanQDuX, et al. A decentralized unsupervised structural condition diagnosis approach using deep auto-encoders. Comput Aided Civ Infrastruct Eng2021; 36(6): 711–732.
46.
CiveraMCalamaiGFragonaraLZ. Experimental modal analysis of structural systems by using the fast relaxed vector fitting method. Struct Control Health Monit2021; 28: e2695.
47.
DessenaGCiveraMFragonaraLZ, et al. A Loewner-based system identification and structural health monitoring approach for mechanical systems. Struct Control Health Monit2023; 2023: 1891062.
48.
JohnsonEALamHFKatafygiotisLS, et al. Phase I IASC-ASCE structural health monitoring benchmark problem using simulated data. J Eng Mech2004; 130(1): 3–15.
49.
AbdeljaberOAvciOKiranyazMS, et al. 1-D CNNs for structural damage detection: Verification on a structural health monitoring benchmark data. Neurocomputing2018; 275: 1308–1317.
50.
YuGWangZHZhaoP. Multi-synchrosqueezing Transform. IEEE Trans Ind Electron2019; 66(7): 5441–5455.
51.
HochreiterS. The vanishing gradient problem during learning recurrent neural nets and problem solutions. Int J Uncertain Fuzziness Knowl Based Syst1998; 6(2): 107–116.
