Abstract
Abstract. A reliable load history is crucial for a fatigue assessment of wind turbines. However, installing strain sensors on every wind turbine is not economically feasible. In this paper, a technique is proposed to reconstruct the thrust load history of a wind turbine based on high-frequency Supervisory Control and Data Acquisition (SCADA) data. Strain measurements recorded during a short period of time are used to train a neural network. The selection of appropriate input parameters is performed based on Pearson correlation and mutual information. Once the training is done, the model can be used to predict the thrust load based on SCADA data only. The technique is validated on two different datasets, one consisting of simulation data (using the software FAST v8, created by Jonkman and Jonkman, 2016) obtained in a controllable environment and one consisting of measurements taken at an offshore wind turbine. In general, the relative error between simulated or measured and predicted thrust load barely exceeds 15 % during normal operation.
Highlights
As the older wind farms slowly reach their designed lifetime, topics concerning fatigue, remaining useful lifetime and a possible lifetime extension gain importance
The first one is obtained by simulation in FAST, while the second one is obtained thanks to a measurement campaign performed at an offshore wind turbine
The simulated data are obtained by using the software FAST v8 (Jonkman and Jonkman, 2016), offered by the National Renewable Energy Laboratory (NREL)
Summary
As the older wind farms slowly reach their designed lifetime, topics concerning fatigue, remaining useful lifetime and a possible lifetime extension gain importance. Fatigue assessments of support structures are often based on measurements of the load history (Loraux and Brühwiler, 2016; Iliopoulos et al, 2017; Schedat et al, 2016; Ziegler et al, 2017). For several reasons accelerometers are preferred over strain gauges, they are not suited to measuring quasi-static loads. In the research presented by Iliopoulos et al (2017), the strain gauges are crucial to capture the quasi-static part of the loading. The research presented in this paper aims to replace the use of strain gauges for the estimation of quasistatic loads. Existing approaches to estimate thrust loads are based on simulations and additional design information (e.g., thrust coefficient) or acceleration measurements (Baudisch, 2012; Cosack, 2010)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
