ESR 6

Various attempts have been made to develop holistic wind turbine condition monitoring (CM) systems that make use of vibration measurement and analysis, in line oil particle counting systems, measured blade root bending moments, and also SCADA data. Commercial systems will soon enter the market but little is known about the effectiveness of such systems; in particular what is the successful detection rate for different times ahead of complete component failure, and also the false alarm rate.

This PhD research proposal aims to remedy this situation through laboratory testing using hardware in the loop approach, and with emphasis on gearbox condition monitoring. This requires that a suitable wind turbine simulation model is developed that can replicate the data that condition monitoring systems measure, i.e. SCADA data like component temperatures, vibration levels at sensor locations, oil particle concentrations in the gearbox oil. The inputs to the model will be wind speed incident on the turbine, and ambient temperature. The turbine model would include the modelling of key faults like a cracked gear tooth, gear tooth scuffing or pitting, excessive bearing clearance, and shaft misalignment. The turbine model will make use of GH Bladed to represent the rotor aerodynamics and overall structural response, and Romax Wind to model the drive train and the details of the gearbox (teeth, bearings, etc). The final aim is to assess commercial CM systems by feeding them with model output for these various faults and assessing whether and when the CM systems detect these and with what reliability.

The first phase of research will develop an initial model of an undamaged wind turbine based on a 2.3 MW commercial wind turbine will be validated against data collected from a large wind farm of identical turbines that is available to the research group at the University of Strathclyde. The second phase will dedicate to the modelling of the various gearbox faults listed above. Validation will be undertaken using data from the same wind farm.

Industrial

SGURR
Tentative schedule: March – June 2017

Academic

UCLM
Tentative schedule: March – July 2017