Use this URL to cite or link to this record in EThOS:
Title: Methods and tools for optimisation of operational reliability and condition monitoring of large-scale industrial wind and tidal turbines
Author: Roshanmanesh, Sanaz
ISNI:       0000 0004 9346 8306
Awarding Body: University of Birmingham
Current Institution: University of Birmingham
Date of Award: 2019
Availability of Full Text:
Access from EThOS:
Access from Institution:
Industrial wind and tidal turbines are complex systems consisting of several different components and subsystems. Many of wind and tidal turbines use gearbox in their systems that is one of the most important components in the system. Gearboxes are designed to operate for the entire lifetime of a wind or tidal turbine or the equivalent of 20 years. However, very few gearboxes do achieve their intended lifetime without significant refurbishment or even replacement at least once or twice within this 20-year period. Offshore wind turbine in particular, suffer from relatively high gearbox failure rate due to the harsh sea environment. Any unpredicted gearbox failure in the wind or tidal turbines can results in considerable downtime as a result of limited access caused by weather condition. This results in loss of energy production and excessive maintenance costs that increases the total energy production costs. A reliable and effective condition monitoring system can play a significant role in reduction of operation and maintenance costs by providing early fault diagnostics and consequently longer time frame for planning required maintenance and repairs. Thus, preventing catastrophic failures due to unpredicted gearbox damages. This study presents an experimental investigation assessing the effectiveness of acoustic emission and vibration analysis in identifying different types of defects in wind and tidal turbine gearbox with means of advanced signal processing techniques, such as Spectral Kurtosis, Kurtogram analysis, Empirical Mode Decomposition and cyclo-stationary analysis to assess their effectiveness. Advanced machine learning techniques such as support vector machine (SVM) and artificial neural network (ANN) are also explored to evaluate the possibility of the fault detection using existing Supervisory Control and Data Acquisition (SCADA) condition monitoring data. Finally, combination of advanced signal processing techniques are employed to enhance the fault detection, by removing signal contamination and extraction of hidden features in the vibration and acoustic signal for better fault detection.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: Q Science (General) ; TA Engineering (General). Civil engineering (General) ; TJ Mechanical engineering and machinery ; TK Electrical engineering. Electronics Nuclear engineering