The dataset beam-signal contains the spectrum vibration signals in the frequency domain measured from a beam reinforced with masses under healthy and faulty conditions. This data is for a commonly used system in various industrial applications. The data can be used for online condition process monitoring to detect and diagnose any anomaly or faulty condition in the system. Hence, the datasets provide the geometric and experimental measurements performed on the beam reinforced with masses for various mass losses considered structural damage. The collected data included the following datasets:
- Dataset Mass-position contains 70 sampling positions for the six masses attached to the beam.
- Dataset DI contains 280 damage indexes calculated using the FRAC method.
- Dataset beam-signal includes 280 inertances responses magnitudes and respective phases considering 70 samples of healthy and 210 samples of damaged conditions.
The dataset beam-signal can be used to develop structural health monitoring techniques for detecting damage and anomalies in the structure. The dataset's Mass-position and DIs can impose parametric uncertainty in the experiment. Stochastic and damage identification metrics can be used for further insights on new monitoring and control techniques. Since the tests include paramedic uncertainty, they can also be employed in uncertainty quantification, stochastic modelling and supervised and unsupervised machine learning techniques.
Therefore, the datasets are intended to benefit the scientific community investigating the dynamics of structures and readers interested in experimental practices applied to systems and modelling. These datasets can be used for numerical model validation, identification techniques, uncertainty quantification, machine learning, and structural integrity monitoring algorithms based on experimental measurement samples on the beam reinforced with mass.
The datasets can be accessed and downloaded in \url{https://zenodo.org/records/8081690?token=eyJhbGciOiJIUzUxMiJ9.eyJpZCI6IjlmNTJiMWQ1LTRiNjUtNDU5NS05OWJmLTQ3MWViNzU3M2NiYSIsImRhdGEiOnt9LCJyYW5kb20iOiIwYTkwNmZhZDRhMDY4NWVjYzRmMzY4ZDgyNjE5OTQ5OCJ9.ks6tuHRdVm0snw8gs3Fp9atKX4njoPjtnqisDxBKHxQ6ct2hhEdLp7doZ43QTABk28PK0FHZpxsJfYO2TwqRFA}
A detailed description of the experiment can be found in
[1] Sousa, A.A.S.R., da Silva Coelho, J., Machado, M.R. et al. Multiclass Supervised Machine Learning Algorithms Applied to Damage and Assessment Using Beam Dynamic Response. J. Vib. Eng. Technol. (2023). https://doi.org/10.1007/s42417-023-01072-7
[2] Monitoramento da Integridade Estrutural de Vigas utilizando Técnicas de Aprendizado de Máquina, 2023. Mestrado em Integridade de Materiais da Engenharia - Universidade de Brasília (In portuguese)
[3]Amanda Aryda Silva Rodrigues de Sousa and Marcela Rodrigues Machado (2023) “Damage assessment of a physical beam reinforced with masses - dataset”, Multiclass Supervised Machine Learning Algorithms Applied to Damage and Assessment Using Beam Dynamic Response. Zenodo. doi: 10.5281/zenodo.8081690
Citation
Amanda Aryda Silva Rodrigues de Sousa and Marcela Rodrigues Machado (2023) “Damage assessment of a physical beam reinforced with masses - dataset”, Multiclass Supervised Machine Learning Algorithms Applied to Damage and Assessment Using Beam Dynamic Response. Zenodo. doi: 10.5281/zenodo.8081690.