FIGURE 6 - available via license: Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International
Content may be subject to copyright.
Source publication
This systematic literature review explores the potential of machine learning-based approaches to detect and prevent bird collisions with wind turbines. It provides a comprehensive review of the current approaches and identifies critical gaps in the literature, which may serve as the groundwork for future research and development in this area. As a...
Context in source publication
Context 1
... 23 papers rather consider visual elements for bird identification which in fact means that VOLUME 4, 2016 only eight of 22 papers considered a further audio processing of already collected audio data (or combined data), which is mostly done with a further processing of the mel spectrum to corresponding Mel-Frequency Cepstral Coeffcients (MFCC) (Fig. 6). Both pre-emphases enable the researchers for the amplitude calibration of audio signals according to different frequency bands and thus to standardize and prepare the data for further processing. A common issue of audio recordings in the field is the differentiation of requested bird sounds with environmental signals or overlapping ...
Similar publications
Low-level jets (LLJs) are examples of non-logarithmic wind speed profiles affecting wind turbine power production, wake recovery, and structural/aerodynamic loading. However, there is no consensus regarding which definition should be applied for jet identification. In this study we argue that a shear definition is more relevant to wind energy than...
Citations
... Sound: The analysis of frequency, duration, and volume of bird vocalizations collected by microphones through passive acoustic monitoring [15,16]. ...
... Morphology: The analysis of the size of a bird, its shape (beak, wings, etc.), or plumage colour based on the video or image recordings of optical sensors [16]. ...
... The comprehensive review of machine learning methods by Principato et al. [16] includes updated information on bird-related classification approaches: preprocessing techniques (image augmentation and audio signal transforms), classification algorithms (convolutional neural network, CNN; K-nearest neighbours, kNN; RF), or classifier evaluation metrics (accuracy and precision). We supplemented the overview with a few scientific papers that were not mentioned in the survey. ...
With the expansion of green energy, more and more data show that wind turbines can pose a significant threat to some endangered bird species. The birds of prey are more frequently exposed to collision risk with the wind turbine blades due to their unique flight path patterns. This paper shows how data from a stereovision system can be used for an efficient classification of detected objects. A method for distinguishing endangered birds from common birds and other flying objects has been developed and tested. The research focused on the selection of a suitable feature extraction methodology. Both motion and visual features are extracted from the Bioseco BPS system and retested using a correlation-based and a wrapper-type approach with genetic algorithms (GAs). With optimal features and fine-tuned classifiers, birds can be distinguished from aeroplanes with a 98.6% recall and 97% accuracy, whereas endangered birds are delimited from common ones with 93.5% recall and 77.2% accuracy.