Current Signal-Based Wind Turbine Fault Diagnosis
Power and Energy Systems Lab
University of Nebraska–Lincoln
Abstract
Wind turbines are usually installed on high towers at remote sites, distributed over large geographic regions, and operate in harsh environments. Therefore, compared with the turbine generators used in traditional power plants, wind turbines are subject to relatively higher failure rates and higher maintenance costs. To reduce the failure rate and downtime, improve the reliability, and reduce the maintenance costs of wind turbines, condition monitoring systems that use various types of signals for wind turbine fault diagnosis have been employed. Compared with the most widely used vibration monitoring technique, the recently developed electrical signal-based condition monitoring technique is nonintrusive and has the advantages of lower cost and easier implementation.
This webinar will present some new developments of using generator current signals for wind turbine fault diagnosis. Specifically, a current signal-based wind turbine condition monitoring system consisting of signal conditioning, feature extraction, and fault diagnosis will presented. The proposed signal conditioning technique solved the spectrum smearing problem caused by the time-varying rotational speeds of wind turbines, which facilitated the extraction of the fault-related features from the current signals collected in nonstationary operating conditions. The fault diagnosis is based on the extracted fault features and a machine learning technique, which enables automated diagnosis with the minimal human effort. Moreover, the condition monitoring system itself, especially sensors, are subject to failures. Therefore, sensor fault detection and isolation, which is helpful to improve the robustness of the condition monitoring system, also will be presented in this webinar. The proposed condition monitoring system does not need installation of additional sensors because current signals are already available in wind turbine control systems. Moreover, the proposed condition monitoring system works for wind turbines in all operating conditions, particularly nonstationary operating conditions. Therefore, it provides a cost-effective, promising solution to improving the reliability and reducing the levelized cost of energy of wind turbines.
Date and Time
Location
Hosts
Registration
- Date: 15 Sep 2020
- Time: 05:00 PM to 06:30 PM
- All times are (GMT-05:00) US/Eastern
-
Add Event to Calendar
- Raleigh, North Carolina
- United States
- Contact Event Host
-
Contact organizer
John J. Shea
jjshea@ieee.org
- Starts 31 August 2020 09:00 AM
- Ends 15 September 2020 04:00 PM
- All times are (GMT-05:00) US/Eastern
- No Admission Charge
Speakers
Mr. Yayu Peng of University of Nebraska–Lincoln
Resilient Collaborative Distributed Tertiary Controls of Microgrids
Biography:
Yayu Peng received the B. Eng. degree in electrical engineering from Chongqing University, Chongqing, China, in 2013. He is currently working toward the Ph.D. degree in electrical engineering with the Department of Electrical and Computer Engineering, University of Nebraska–Lincoln, Lincoln, NE, USA.
He was a research and development intern at the Global Energy Interconnection Research Institute North America (GEIRINA) and New York Power Authority (NYPA) in 2018 and 2020, respectively. His research interests include renewable energy systems, condition-based maintenance, and intelligent fault diagnosis and prognosis.
Agenda
5:00-5:15 PM Chapter meeting and round table introductions
5:15-6:15 PM Presentation
6:15-6:30 PM Q&A