Towards a fleetwide data-driven lifetime assessment methodology of offshore wind support structures based on SCADA and SHM data
de Nolasco Santos, F.; Robbelein, K.; D'Antuono, P.; Noppe, N.; Weijtjens, W.; Devriendt, C. (2023). Towards a fleetwide data-driven lifetime assessment methodology of offshore wind support structures based on SCADA and SHM data, in: Rizzo, P. et al. European Workshop on Structural Health Monitoring. EWSHM 2022 - Volume 1. pp. 123-132. https://dx.doi.org/10.1007/978-3-031-07254-3_13
In: Rizzo, P.; Milazzo, A. (Ed.) (2023). European Workshop on Structural Health Monitoring. EWSHM 2022 - Volume 1. Springer: Cham. ISBN 978-3-031-07253-6; e-ISBN 978-3-031-07254-3. XXI, 930 pp. https://dx.doi.org/10.1007/978-3-031-07254-3, meer
|
| Beschikbaar in | Auteurs |
|
Documenttype: Congresbijdrage
|
| Trefwoord |
|
| Author keywords |
Offshore wind turbine support structures; Lifetime assessment; Neural networks |
| Auteurs | | Top |
- de Nolasco Santos, F., meer
- Robbelein, K., meer
- D'Antuono, P., meer
|
|
|
| Abstract |
In recent years there has been an increased interest of the offshore wind industry to use structural health monitoring (SHM) data in the assessment of consumed lifetime and lifetime extension for an entire wind farm. In order for operators, certifying bodies, insurance entities and government agencies to agree on a lifetime extension, a commonly accepted lifetime assessment strategy with proven results is required. This paper aims to provide such an answer through a data-driven lifetime assessment approach using SHM and SCADA data. The research involves training neural network (NN) models using SCADA and SHM data to estimate the fore-aft damage equivalent moment (DEM) at the tower interface level on a 10-min basis for implementation in a data-driven lifetime assessment. The NN are trained and validated based on one instrumented turbine (the fleetleader) and cross-validated based on another instrumented turbine. A DEM representative for the lifetime of the asset is calculated based on the 10-min DEM’s. An analysis of the NN models’ performance (error of 10-min DEM estimation in relation to DEM derived from SHM data) and accuracy (lifetime DEM error) is undertaken. The DEM representative for the lifetime of the assets is benchmarked with the as-designed DEM to assess the lifetime. |
|
