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.
Current-Based Fault Diagnosis and Prognosis for Wind Turbine Drivetrain Gearboxes
Fault diagnosis and prognosis of drivetrain gearboxes is a prominent challenge in wind turbine condition monitoring. Current-based techniques have been successfully used for condition monitoring and fault diagnosis of wind turbine drivetrain gearboxes. They have shown advantages over the traditional vibration-based techniques in terms of cost, implementation, and reliability. However, there are still many challenges in using generator current signals for fault diagnosis and prognosis of wind turbine drivetrain gearboxes. First, it is hard to extract fault features from nonstationary current signals due to varying shaft rotation speeds. Moreover, the signal-to-noise ratio of fault signature in current signals is usually low due to high amplitude of fundamental frequency component, which makes the fault diagnosis difficult. Meanwhile, an accurate prediction of drivetrain gearbox remaining useful life (RUL) is important to achieve condition-based maintenance to ensure secure and reliable operations of wind turbines and reduce the cost of wind power. This, however, is challenging due to the lack of accurate physical degradation models and limited monitoring data.^ In this dissertation, current-based fault diagnosis methods are proposed for wind turbine drivetrain gearboxes. New signal conditioning methods are developed to extract fault features from current signals of doubly-fed induction generator (DFIG)-based wind turbines under nonstationary conditions. A deep classifier consists of stacked autoencoder and support vector machine (SVM) is proposed to improve the accuracy of fault classification results of the traditional SVM classifier. Moreover, a new fault prognosis and RUL prediction method for drivetrain gearboxes based on adaptive neuro fuzzy inference system (ANFIS) and particle filtering (PF) technique is proposed. Finally, an enhanced PF algorithm is proposed to improve the fault prognosis and RUL prediction accuracy of the ANFIS-PF method. The proposed methods have been validated by small-scale permanent magnet synchronous generator (PMSG)- and DFIG-based wind turbine drivetrain simulators in the lab and MW-scale DFIG-based wind turbines in the field.^
Cheng, Fangzhou, "Current-Based Fault Diagnosis and Prognosis for Wind Turbine Drivetrain Gearboxes" (2017). ETD collection for University of Nebraska - Lincoln. AAI10682503.