Dictionary learning approach to monitoring of wind turbine drivetrain bearings

Abstract

Condition monitoring is central to the efficient operation of wind farms due to the challenging operating conditions, rapid technology development and large number of aging wind turbines. In particular, predictive maintenance planning requires the early detection of faults with few false positives. Achieving this type of detection is a challenging problem due to the complex and weak signatures of some faults, particularly the faults that occur in some of the drivetrain bearings. Here, we investigate recently proposed condition monitoring methods based on unsupervised dictionary learning using vibration data recorded over 46 months under typical industrial operations. Thus, we contribute novel test results and real-world data that are made publicly available. The results of former studies addressing condition-monitoring tasks using dictionary learning indicate that unsupervised feature learning is useful for diagnosis and anomaly detection purposes. However, these studies are based on small sets of labeled data from test rigs operating under controlled conditions that focus on classification tasks, which are useful for quantitative method comparisons but gives little insight into how useful these approaches are in practice. In this study, dictionaries are learned from gearbox vibrations in six different turbines, and the dictionaries are subsequently propagated over a few years of monitoring data when faults are known to occur. We perform the experiment using two different sparse coding algorithms to investigate if the algorithm selected affects the features of abnormal conditions. We calculate the dictionary distance between the initial and propagated dictionaries and find the time periods of abnormal dictionary adaptation starting six months before a drivetrain bearing replacement and one year before the resulting gearbox replacement. In addition, we investigate the distance between dictionaries learned from geographically close turbines of the same type under healthy conditions. We find that the features learned are similar and that a dictionary learned from one turbine can be useful for monitoring a similar turbine.

LSTM-Autoencoder Based Anomaly Detection Using Vibration Data of Wind Turbines

“…The sampling rate was 12.8 kHz, and the length of each signal segment was 1.28 s (16,384 samples). Signal segments were approximately 12 h apart for 46 consecutive months over the past 10 years [51].…”

Unsupervised Ranking of Outliers in Wind Turbines via Isolation Forest with Dictionary Learning

“…The atoms are normalized after each learning iteration. Details of the wind turbine dataset and data collection conditions, as well as method evaluation setup is available in (Martin-del-Campo et al, 2019).…”

“…Our interest here focuses on the fidelity of the sparse representation, which can be used to reconstruct the original vibration signal, and in the dictionary distance measure of learned dictionary as it propagates over time. Sparse coding with dictionary learning has proven useful for online monitoring (Martin-del-Campo et al, 2013) and identification of bearing characteristic frequencies Martin-del-Campo et al (2019). The work presented here is novel because it presents a score that can be used to rank wind turbines on how much the operation is diverging from the population without incorporating prior information on the state of the machine.…”