Advanced doubly fed induction generator variable speed wind turbine and wind farm control method for grid connection

Research Doctorate - Doctor of Philosophy (PhD)Owing to the global concerns on climate change and world-wide carbon pricing policies, the use of conventional fossil fuels has been descending these years. Instead, renewable energy is getting increasing popularity in the power industry. Specifically, wind power occupies a large proportion in the current energy context due to the relatively mature generation and operation techniques. With the increasing penetration of wind power into the grid, new challenges are emerging in front of the grid administrators. A series of requirement, such as power quality, frequency support, and voltage criterion, are needed to be defined so that the penetration level of wind power can be improved. To achieve these requirements, one way is to raise the reserve capacity of wind farms. Under this circumstance, the instalment of battery storage systems and reactive power compensation devices are needed, therefore bringing a heavy economic burden. Another way is to improve the performance of each wind turbine to meet the requirement by control strategies. Hence, the focus of this research is to enhance wind turbines capability of meeting grid code requirements by advanced control methods. Particularly, the high controllability of variable speed wind turbines (VSWT) enables them so adaptive to the modern power system that their control methodologies for grid connection are studied in this thesis. This research focuses on the advanced control methodologies for VSWT to improve its grid connection capability. Specifically, advanced methods are developed for three essential objectives: 1) improve wind turbines performance to ride through the low voltage condition during the fault period; 2) develop active power control strategies in a VSWT under the normal operational condition; 3) find a smart control scheme for a large scale wind farm to generate active power as the system operator desired. In the area of low voltage ride through (LVRT), an advanced control method for VSWT being able to ride through the voltage dip event is proposed. Firstly, a review of conventional LVRT methods with auxiliary devices engaged is presented. Such auxiliary devices as crowbar circuits and DC chopper are not energy efficient and may exert thermal stress in the circuit, bringing the energy storage system being more promising. A new method considering the Abstract capability of variable speed WT itself to enhance LVRT capability is presented. In fact, the excessive generation during the fault condition can be stored in the rotor inertia, and released later when the fault is cleared. Therefore, a modified method, which coordinates the control of battery device and kinetic energy, is delivered to improve LVRT performance. Meanwhile, the switch strategy for battery device is also discussed through the dynamic simulations in Matlab/SimpowerSystem. In the area of active power control strategies in a VSWT under the normal operational condition, active power control for VSWT is significant for wind power being more dispatchable and generating power as required. The active power control is achieved by making VSWT operate under the deloading mode, i.e. compromising acquired wind power through tip speed ratio adjustment from original MPPT point. How much the compromised power is should be discussed regarding to different objectives. In this research, the smoothing power output during a certain time horizon is the control objective and obtained with the optimization approach. In the area of active power control in a large-scale wind farm, a cooperation-driven distributed control is proposed in this project. It is far more important to enhance the integrating performance of a wind farm compared with a single wind turbine. For a dispatchable wind farm, system operators assign an estimated generation command to wind farm central module. In this case, finding a proper method to distribute the generation task among the widely spread WTs is increasingly urgent with wind farm scale growing. In this research, a distributed control scheme is proposed to meet with system request. Compared with the conventional centralized method, the distributed one can ease the computation burden to a large extend, and is more robust and feasible to the topology change. Moreover, the kinetic energy storage potential in a wind turbine is tapped to provide a buffer for power dispatch. The proposed control schemes have been numerically verified on a number of benchmark test systems. Their effectiveness and feasibility are verified by simulation results, showing satisfactory performances