Vibration Data for Laminated Composite Structures: Healthy and Delamination States
Description
This dataset contains vibration data collected from laminated composite structures in three health states: healthy and two delamination cases. The composites were fabricated using carbon fiber prepreg (T700SC-12k-60E) with a [0/90/0/90]s ply configuration via hot press compression molding. Delaminations were introduced using PTFE Teflon film near the fixed and free ends. Vibration data were acquired using an accelerometer (Type 4517-C) under cantilever conditions, excited by random signals. Each health state includes data from five samples, with ten random responses per sample, captured over 15 seconds at a 2.5 kHz sampling rate. This dataset supports prognostics and health monitoring (PHM), and structural health monitoring (SHM) research. Dataset Structure: The dataset is organized into three folders, each representing a different health state: healthy, delamination-1, and delamination-2. Within each folder, there are five MAT files, corresponding to data collected from five samples of the respective health state. Each MAT file contains 10 data structures (referred to as variables P01, P02,...,P10), representing 10 random vibration responses for each sample. Each response consists of two columns: the first column represents the input excitation signal sent to the shaker, and the second column captures the output vibration signal measured by the accelerometer. More details about the dataset can be found in the below references (if you use this dataset please cite): 1. Azad, Muhammad Muzammil, and Heung Soo Kim. "Hybrid deep convolutional networks for the autonomous damage diagnosis of laminated composite structures." Composite Structures 329 (2024): 117792. https://doi.org/10.1016/j.compstruct.2023.117792 2. Azad, Muhammad Muzammil, Prashant Kumar, and Heung Soo Kim. "Delamination detection in CFRP laminates using deep transfer learning with limited experimental data." Journal of Materials Research and Technology 29 (2024): 3024-3035. https://doi.org/10.1016/j.jmrt.2024.02.067 3. Azad, Muhammad Muzammil, Sungjun Kim, and Heung Soo Kim. "Autonomous data-driven delamination detection in laminated composites with limited and imbalanced data." Alexandria Engineering Journal 107 (2024): 770-785. https://doi.org/10.1016/j.aej.2024.09.004 4. Azad, Muhammad Muzammil, and Heung Soo Kim. "An explainable artificial intelligence‐based approach for reliable damage detection in polymer composite structures using deep learning." Polymer Composites (2024). https://doi.org/10.1002/pc.29055 5. Azad, Muhammad Muzammil, and Heung Soo Kim. "Noise robust damage detection of laminated composites using multichannel wavelet-enhanced deep learning model." Engineering Structures 322 (2025): 119192. https://doi.org/10.1016/j.engstruct.2024.119192