Wind turbine fault detection using quantum long-short term memory network

Zhang, Zhefeng and Ma, Xiandong (2025) Wind turbine fault detection using quantum long-short term memory network. In: 2025 30th International Conference on Automation and Computing (ICAC) :. IEEE. ISBN 9798331525453

Text (ICAC2025_fault_detection_using_qlstm-final)
ICAC2025_fault_detection_using_qlstm-final.pdf - Accepted Version
Available under License Creative Commons Attribution.
Download (2MB)

Abstract

Fault detection plays a critical role in ensuring the reliability and safety of wind turbine operation. With the growing availability of operational data, data-driven approaches have become increasingly prevalent. This paper proposes a fault detection method based on the Quantum Long Short-Term Memory network (QLSTM). The model is trained by supervisory control and data acquisition (SCADA) data to capture temporal dependencies among multiple sensor signals under healthy conditions, forming a Normal Behavior Model (NBM). Residuals between predicted and practical measurement values are computed and evaluated using the T-distribution method to establish a threshold for anomaly identification. Experiments conducted on SCADA data show that the proposed method outperforms the conventional LSTM in terms of modeling accuracy, detection sensitivity, and early fault warning capability, achieving a 7.67 hours earlier fault detection and demonstrating the potential of quantum machine learning (QML) in wind turbine condition monitoring.