Off-campus UNL users: To download campus access dissertations, please use the following link to log into our proxy server with your NU ID and password. When you are done browsing please remember to return to this page and log out.
Non-UNL users: Please talk to your librarian about requesting this dissertation through interlibrary loan.
Wind Turbine Drivetrain Fault Diagnosis Based on Compressive Sensing and Information Fusion
Abstract
The reliability and accuracy of condition monitoring and fault diagnosis of wind turbine drivetrains relies on the availability, reliability, and accuracy of sensor data. The problems of insufficient, corrupted, or missing sensor data usually cause unreliable, inaccurate, or false fault diagnosis results for wind turbines. In this dissertation, various methods were proposed to address the problems of insufficient or faulty sensor data to improve the reliability, robustness, and accuracy of the existing wind turbine drivetrain condition monitoring systems. Firstly, a vibration and current information fusion technique was proposed to overcome the limitation of relying on vibration signals only for fault diagnosis. Since the current signals are already collected for the control of the electric machine, the proposed method does not need any additional hardware cost while improving the reliability and robustness of the condition monitoring system. Secondly, sensors are also subject to failure, which may cause failure of the whole condition monitoring system. To address this problem, this dissertation proposed methods for sensor fault detection and isolation to make sure that the signals collected for condition monitoring are not corrupted. Thirdly, this work solved the inevitable data loss problem in remote condition monitoring system by a compressive sensing-based missing-data-tolerant fault detection method. The method transmits compressive measurements of condition monitoring signals for fault detection and is robust to data loss. Lastly, to reduce the computational cost and data storage burden, this work proposed a method for fault diagnosis using compressed data. The method does not need to conduct the time-consuming data decompression.
Subject Area
Electrical engineering|Alternative Energy|Information science
Recommended Citation
Peng, Yayu, "Wind Turbine Drivetrain Fault Diagnosis Based on Compressive Sensing and Information Fusion" (2021). ETD collection for University of Nebraska-Lincoln. AAI28547372.
https://digitalcommons.unl.edu/dissertations/AAI28547372