Long-term fatigue estimation on offshore wind turbines interface loads through loss function physics-guided learning of neural networks

Santos, Francisco de Nolasco; D’Antuono, Pietro; Robbelein, Koen; Noppe, Nymfa; Weijtjens, Wout; Devriendt, Christof

doi:10.1016/j.renene.2023.01.093

Cited by 36 publications

(23 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A recommendation which has been confirmed in various other research on this topic, e.g. [10][13] [15].…”

Section: Available Datasupporting

confidence: 53%

“…The present analysis has been performed on a WTG support structure which is instrumented with a load monitoring setup. Future work will be to assess the influence of curtailment on loads and lifetime farm wide by implementation of the research towards farm wide DEL predictions [10].…”

Section: Discussionmentioning

confidence: 99%

“…For purpose of this work DEL is the most suitable parameter to assess the impact on lifetime. The DEL is a commonly known fatigue load parameter used in design of foundations, other structural components and subject of ongoing research for farm wide predictions as disclosed in [5] [10]. The present work particularly investigates the role of curtailment regimes with different frequency of stop and start cycles on the consumed lifetime.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Effect of curtailment scenarios on the loads and lifetime of offshore wind turbine generator support structures

Robbelein

Daems

Verstraeten

et al. 2023

J. Phys.: Conf. Ser.

Self Cite

View full text Add to dashboard Cite

Curtailment is a known phenomenon for wind turbine operators of both onshore and offshore wind turbine generators (WTG). Curtailment refers to the situation in which the power output of all WTG’s within a windfarm is forced below the expected power output at the occurring environmental conditions. A direct consequence of curtailment is the loss of power production. In the present contribution further consequences of curtailment of an offshore wind farm (OWF) are studied from the perspective of the support structure, in specific the foundation. In relation to curtailment a couple of potentially critical operational conditions impacting the fatigue consumption of the support structure can be identified. Besides the standstill during operational windspeed conditions, in specific damaging for the +7MW generation WTG’s, curtailment introduces repeated transitions between operational conditions. Since transitions between operational conditions of a WTG are known to be a cause of high fatigue loads in the structural components of the WTG, their increased occurrence due to curtailment might also have an impact on the fatigue consumption of the support structure. With the growing interest of the industry to quantify and potentially optimize the structural lifetime consumption in view of potential lifetime extension of OWF assets, any potential fatigue damaging operational condition is to be investigated. The present work focusses on the investigation of the impact these transitional load cycles may have on the structural lifetime of the WTG foundation. To assess the impact on lifetime, the assessment of the damage equivalent loads (DEL) derived from structural health monitoring (SHM) data are used as a data-driven alternative for model-based load simulations. In the present work such data-driven lifetime assessment studies the impact of curtailment regimes with different frequency of stop and start cycles on the structural lifetime. The study is performed based on 1 year of SHM data collected from two OWF’s. The assessment demonstrates that the impact of additional transitional load cycles on the structural fatigue life consumption is to be considered when defining a long-term curtailment strategy for an OWF.

show abstract

“…A recommendation which has been confirmed in various other research on this topic, e.g. [10][13] [15].…”

Section: Available Datasupporting

confidence: 53%

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Effect of curtailment scenarios on the loads and lifetime of offshore wind turbine generator support structures

Robbelein

Daems

Verstraeten

et al. 2023

J. Phys.: Conf. Ser.

Self Cite

View full text Add to dashboard Cite

show abstract

“…The validity of the PIMLs were verified through numerical arithmetic and experimental studies. A PIML for RUL prediction of offshore wind turbines is proposed by using the Minkowski logarithmic error as a loss function and introducing damages accumulation in the ANN [69]. The soft constraint methods used may not guarantee that the predictions fully satisfy the prior physical knowledge.…”

Section: Applications Of Piml In Structural Integritymentioning

confidence: 99%

“…When the input data contain a lot of noise, the physical constraints in PIML should be strengthened to guarantee the reliability of the prediction results. Moreover, for complex failure processes, due to insufficient comprehensive knowledge of the mechanism, the different physically based models as physical loss terms significantly affect the predictive performance of PIML for predictions at different time scales [69]. Therefore, an adaptive PIML incorporating multiple physical constraints needs to be developed to cope with structural integrity issues under different use conditions by strengthening or relaxing physical constraints.…”

Section: Challenges and Outlookmentioning

confidence: 99%

Physics-informed machine learning and its structural integrity applications: state of the art

Zhu,

Wang,

Luo

et al. 2023

Phil. Trans. R. Soc. A.

View full text Add to dashboard Cite

The development of machine learning (ML) provides a promising solution to guarantee the structural integrity of critical components during service period. However, considering the lack of respect for the underlying physical laws, the data hungry nature and poor extrapolation performance, the further application of pure data-driven methods in structural integrity is challenged. An emerging ML paradigm, physics-informed machine learning (PIML), attempts to overcome these limitations by embedding physical information into ML models. This paper discusses different ways of embedding physical information into ML and reviews the developments of PIML in structural integrity including failure mechanism modelling and prognostic and health management (PHM). The exploration of the application of PIML to structural integrity demonstrates the potential of PIML for improving consistency with prior knowledge, extrapolation performance, prediction accuracy, interpretability and computational efficiency and reducing dependence on training data. The analysis and findings of this work outline the limitations at this stage and provide some potential research direction of PIML to develop advanced PIML for ensuring structural integrity of engineering systems/facilities. This article is part of the theme issue 'Physics-informed machine learning and its structural integrity applications (Part 1)'.

show abstract

Farm‐wide interface fatigue loads estimation: A data‐driven approach based on accelerometers

de N Santos,

Noppe,

Weijtjens

et al. 2024

Wind Energy

Self Cite

View full text Add to dashboard Cite

Fatigue has become a major consideration factor in modern offshore wind farms as optimized design codes, and a lack of lifetime reserve has made continuous fatigue life monitoring become an operational concern. In this contribution, we discuss a data‐driven methodology for farm‐wide tower‐transition piece fatigue load estimation. We specifically debate the employment of this methodology in a real‐world farm‐wide setting and the implications of continuous monitoring. With reliable nacelle‐installed accelerometer data at all locations, along with the customary 10‐min supervisory control and data acquisition (SCADA) statistics and three strain gauge‐instrumented 'fleet‐leaders', we discuss the value of two distinct approaches: use of either fleet‐leader or population‐based data for training a physics‐guided neural network model with a built‐in conservative bias, with the latter taking precedence. In the context of continuous monitoring, we touch on the importance of data imputation, working under the assumption that if data are missing, then its fatigue loads should be modeled as under idling. With this knowledge at hand, we analyzed the errors of the trained model over a period of 9 months, with monthly accumulated errors always kept below . A particular focus was given to performance under high loads, where higher errors were found. The cause for this error was identified as being inherent to the use of 10‐min statistics, but mitigation strategies have been identified. Finally, the farm‐wide results are presented on fatigue load estimation, which allowed to identify outliers, whose behavior we correlated with the operational conditions. Finally, the continuous data‐driven, population‐based approach here presented can serve as a springboard for further lifetime‐based decision‐making.

show abstract

Long-term fatigue estimation on offshore wind turbines interface loads through loss function physics-guided learning of neural networks

Cited by 36 publications

References 28 publications

Effect of curtailment scenarios on the loads and lifetime of offshore wind turbine generator support structures

Effect of curtailment scenarios on the loads and lifetime of offshore wind turbine generator support structures

Physics-informed machine learning and its structural integrity applications: state of the art

Farm‐wide interface fatigue loads estimation: A data‐driven approach based on accelerometers

Contact Info

Product

Resources

About