Skip to Main Content
 

Global Search Box

 
 
 
 

Files

File List


9059.pdf (13.4 MB)

ETD Abstract Container

Abstract Header

An Integrated Framework of Performance Assessment and Drivetrain Prognostics for Wind Turbines

Zhao, Wenyu
http://rave.ohiolink.edu/etdc/view?acc_num=ucin1393237878

Abstract Details

2014, MS, University of Cincinnati, Engineering and Applied Science: Mechanical Engineering.
The growing demand for wind energy and the advancement of turbine technologies have proliferated global adoption and expansion of wind farms over the past years. Due to various causes, including logistics difficulties and lack of predictive analytics, failures and downtime occur and lead to reduced asset availability and revenue. Prognostics and Health Management (PHM) methodologies and techniques are considered as critical technologies, where the capability of diagnosis and prognosis for turbine degradation and failure can be considerably beneficial to prevent unexpected failures, optimize maintenance decision-making and enhance overall system performance. However, there are impeding challenges for the application of PHM techniques in wind turbine area: what is the foundation to incorporate with commonly used vibration data and activate component-level maintenance; how to apply state-of-the-art signal processing, diagnosis and prognosis techniques while wind turbine components are known to be working under dynamic operating regimes constantly; how to design a systematic approach and implement suitable algorithms on a reconfigurable platform. This thesis conducts a comprehensive review of existing data systems and analytical methods to monitor wind turbine health condition. The thesis presents a framework that integrates two most commonly used data systems in wind power area: a Supervisory Control And Data Acquisition (SCADA) system and a Condition Monitoring System (CMS). A system-level, overall turbine performance is assessed with purely SCADA data, whereas degradation of drivetrain components is assessed and fault detection & localization is achieved with SCADA and CMS data. The assessment of turbine performance generates a confidence value (CV) for the turbine unit, which interprets the capability to convert wind power to electrical power under varying conditions. A systematic approach is designed to pre-process data, cluster data based on a multi-regime method, Gaussian Mixture Models (GMM), and evaluate the cluster deviation over time. The drivetrain prognostics process combines SCADA variables with features extracted from CMS data, to evaluate the overall degradation of the drivetrain, which consists of rotor, main shaft, gearbox and generator, with a Self-organizing Maps (SOM) method. The minimum quantization error (MQE) metric is used for detecting drivetrain fault and locating the fault at the component-level The proposed framework is validated with two case studies, a 2MW onshore turbine for performance assessment and a 3MW offshore turbine for drivetrain prognosis.
Jay Lee, Ph.D. (Committee Chair)
Raj Bhatnagar, Ph.D. (Committee Member)
Mark Schulz, Ph.D. (Committee Member)
David Thompson, Ph.D. (Committee Member)
100 p.
Mechanics
Wind Turbine; Prognostics; Performance Assessment; Drivetrain Prognostics; SCADA; CMS

Recommended Citations

Citations

  • Zhao, W. (2014). An Integrated Framework of Performance Assessment and Drivetrain Prognostics for Wind Turbines [Master's thesis, University of Cincinnati]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1393237878

    APA Style (7th edition)

  • Zhao, Wenyu. An Integrated Framework of Performance Assessment and Drivetrain Prognostics for Wind Turbines. 2014. University of Cincinnati, Master's thesis. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=ucin1393237878.

    MLA Style (8th edition)

  • Zhao, Wenyu. "An Integrated Framework of Performance Assessment and Drivetrain Prognostics for Wind Turbines." Master's thesis, University of Cincinnati, 2014. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1393237878

    Chicago Manual of Style (17th edition)

ucin1393237878
951
© 2014, all rights reserved.
This open access ETD is published by University of Cincinnati and OhioLINK.