Grid current control methods for MV wind energy conversion systems

Wind energy has been one of the most important renewable energies as an alternative to fossil energy. The power rating of a single wind turbine has been increasing during the past forty years. A megawatt level wind turbine has been a tendency, especially for offshore wind energy application. The current common solution is to make the two-level low voltage converter in parallel in order to achieve sufficient power rating. However, research has found that by employing medium voltage multilevel converters will reduce the cost of energy. Additionally, the grid converter is the only way to connect the renewable power generation to the power grid. Grid converter control strategies considering normal and abnormal operation have been one of most concerned points for renewable energy integration. Based on these issues, the Ph.D. project will focus on the grid converter control algorithms investigation. In order to design the control and evaluate the performance of the proposed control algorithms, system modelling is necessary and an essential task at first. Including the DC-link capacitor, grid converter, LCL filter and PCC voltage, the modelling of these parts is built in chapter 3. Besides, one scaled experiment low voltage system setup is designed and introduced as well, which will be used for experimental verification. For the control of the grid converter with LCL filter, the current control strategy becomes more complex if compared to a grid converter with simple inductive filters. Although an LCL filter can greatly increase the attenuation for the high frequency range, it also brings new problems, such as resonance issues and which current should be adopted as the feedback current, grid current or converter side current. This is discussed in chapter 4 in detail. For overcoming the resonance problem, passive and active damping are discussed in this chapter as well. For evaluating the control performance or requiring the wind turbine behavior on different scenarios, the latest grid standards are also reviewed. Some suggestions on the grid code modifications are given as well for better regulating the next generation wind turbine to behave as a traditional synchronous generator. Thermal performance is an important aspect related to the system reliability, efficiency and lifetime. The loss calculation and temperature estimation method are introduced in the third section of chapter 4. Based on the previous study in part II, system design and evaluation of wind energy conversion system, grid converter control strategies are investigated in part III, which includes three chapters, the state of the art grid converter control, model predictive control and backstepping control. Voltage orientated control as the industrial solution for grid converter control has been developed and adopted for the past forty years, which is difficult to be improved without limitation. It is necessary to find another control to meet the technical challenges, e.g., switching frequency limitation. One requirement of the MV power control is to reduce the switching frequency to decrease the switching losses. Then the semiconductor reliability might be increased and the system efficiency is improved. Through the comparison, it has been proved that the proposed multiple steps model predictive control can achieve the lowest switching frequency without compromising the current quality. In order to reduce the sensors, one Luenberger observer is designed for estimating the grid current and capacitor voltage, which is needed for the active damping. It should be pointed out that for reducing the processor calculation burden, the observer is implemented with Xilinx System Generator in FPGA for experiments. In order to duel with the parameter mismatch, distorted grid voltage, and unbalanced grid faults, a non-linear control is proposed in chapter 7, which is a recursive Lyapunov-based method with high robustness. Finally, a conclusion is given at end of this Ph.D. thesis, and some interesting further research points are listed as well.

Author dc:creator

• Yin, Hang

Advisor dc:contributor.advisor

• Dieckerhoff, Sibylle

Licence dc:rights.uri

• https://creativecommons.org/licenses/by/4.0/

Language dc:language.iso

en

Identifier URI

http://dx.doi.org/10.14279/depositonce-8790

OAI identifier oai:identifier

oai:depositonce.tu-berlin.de:11303/9757