Advanced Wind Turbine Operational Modelling for Evaluating and Improving Power Electronics Reliability

  • Arsim Ahmedi

Student thesis: Phd

Abstract

There is a reliability concern for wind turbines (WTs) installed in offshore wind farms, where the harsh environment not only has a potential to lead to more failures, but could also increase the time and cost of intervention. Such unreliability problems can cause downtime and losses, which would be reflected at the cost of energy and affect all involved parties, including the customers. The work undertaken in this thesis focuses on the reliability of power electronic systems which represent a crucial element in the operation of the majority of modern WTs. One particular reliability concern in WT power converters is the early aging and degradation. To analyse the degradation of power electronic converters, there is a need for advanced simulation models that combine wind turbine operation and mission profile with the specific failure mechanisms. The focus of the thesis is the modelling and evaluation of the thermo-mechanical stress of the power electronic unit, which is believed to be the source of one of the most important failure and aging mechanisms in converter modules. A methodology for calculating the converter thermal stress and consumed lifetime for a specific design, failure mechanism, and operational profile of a WT has been developed. This methodology allows for evaluating WT operational transients that have been overlooked by similar models in the literature. The developed approach and model can be used to evaluate different design and operational aspects of WTs in the design phase while also having potential to be used to predict the lifetime and failures during operation. Such evaluation has been undertaken in this thesis using several case studies and operational scenarios: comparing machine side and grid side converters, different generator control strategies, as well as the reliability impact of reactive power transfer to the grid. The most thermally stressed part of converter is identified to be the diode on the machine side converter, and it is concluded that the dynamic transients and the response associated with the torque control of the generator significantly impact the converter lifetime. From the generator control, the strategies that deliver closer maximum power point tracking bring higher stress on the converter. After undertaking model order reduction and optimisation of the simulation, the converter aging is analysed for longer WT operational time and two days with different wind speed profile are compared using measured wind speed data.
Date of Award1 Aug 2023
Original languageEnglish
Awarding Institution
  • The University of Manchester
SupervisorMike Barnes (Supervisor) & Victor Levi (Supervisor)

Keywords

  • Power Electronics
  • Thermal Modelling
  • Thermal Cycling
  • Converter Reliability
  • Wind Turbine
  • IGBT
  • Lifetime Prediction

Cite this

'