Answers are grounded in this paper and cite their sources.
Supervised deep-learning methods are data driven and widely used for wind turbine blade icing detection (ID). Data-driven methods generally require a complete dictionary of labeled sensor data. However, labeling sensor data increases engineering costs and can introduce costly errors such as incorrect data labels. In addition, the reported data-driven approaches ignore the contribution of structural properties of multivariate sensor data to failure patterns identification. To address these shortcomings, the current study proposes a beta variational graph attention autoencoder (β-VGATAE) for blade ID. Theβ-VGATAE model employs a beta variational autoencoder (β-VAE) architecture to achieve unsupervised learning. A graph attention network is used as a spatial feature extractor within theβ-VAE architecture since it considers the spatial structure of the sensor data. Actual sensor data from supervisory control and data acquisition systems were used to validate the proposed model. Specifically, we verified the rationality of designing each component in theβ-VGATAE. Experimental results show that the highest levels of accuracy achieved were 90.9% and 93.4% for the respective scenarios involving two wind turbines; theβ-VGATAE detection model has high accuracy and excellent generalization ability.
No takes yet. Share an insight, caveat, or question.
Wang et al. (2023) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: