Quantification of the Information Loss Resulting from Temporal Aggregation of Wind Turbine Operating Data

Abstract

SCADA operating data are more and more used across the wind energy domain, both as a basis for power output prediction and turbine health status monitoring. Current industry practice to work with this data is by aggregating the signals at coarse resolution of typically 10-min averages, in order to reduce data transmission and storage costs. However, aggregation, i.e., downsampling, induces an inevitable loss of information and is one of the main causes of skepticism towards the use of SCADA operating data to model complex systems such as wind turbines. This research aims to quantify the amount of information that is lost due to this downsampling of SCADA operating data and characterize it with respect to the external factors that might influence it. The issue of information loss is framed by three key questions addressing effects on the local and global scale as well as the influence of external conditions. Moreover, recommendations both for wind farm operators and researchers are provided with the aim to improve the information content. We present a methodology to determine the ideal signal resolution that minimized storage footprint, while guaranteeing high quality of the signal. Data related to the wind, electrical signals, and temperatures of the gearbox resulted as the critical signals that are largely affected by an information loss upon aggregation and turned out to be best recorded and stored at high resolutions. All analyses were carried out using more than one year of 1 Hz SCADA data of onshore wind farm counting 12 turbines located in the UK.

Early Detection and Diagnosis of Wind Turbine Abnormal Conditions Using an Interpretable Supervised Variational Autoencoder Model

“…The cross-correlation coefficient between each pair among these three measures is one; thus, only T BAX−BOX1 was retained to describe the battery box temperature. Second, the 10-min average erases any significant information from measures varying at a very high frequency, such as the electrical frequency and voltage (both neglected in the model) [37]. Thirdly, some variables are too scarce, i.e., their time series present many missing or nonnumerical entries [38].…”

Wind speed time series synthesis using a parametrized power spectral density function

“…Many engineering tools and financial models rely on 10 minutes (IEC, 2017) or hourly data. However, such an industry-standard practice comes with significant information loss (Beretta et al, 2021). The 10-minute resource data is inadequate for sizing and dispatch of integrated storage (Poudel et al, 2021), and the power system simulation study requires data at a resolution of 1 second or better.…”

Nonstationarity in correlation matrices for wind turbine SCADA‐data

“…Second, Ulmer et al, 10 who apply convolutional neural networks for failure detection, mention that the 10-min averaging process naturally leads to a loss of information. This effect is specifically studied in Beretta et al 27 Some researchers have tried to avoid these problems by using simulated high frequency data 24,28 while the industry uses additionally installed vibration sensors to increase monitoring quality. Stetco et al 29 provide a review on approaches using such Condition Monitoring Systems (CMS).…”