| Summary: | Small-signal stability and dynamic performance are of great concern for AC power grids with high penetration of power converters. The interactions between these converters may lead to performance degradation or even system instability and failure at certain conditions. To deal with such problems, global modelling and integrated control are proposed. However, because of the highly integration of power grids in commercial industry environment, the lack of information about parameters and methods of the embedded converters, impedes the development of global models for system analysis and control design.
To fill the gap, this research investigated the utility of system identification techniques to estimate a state space model of the unknown power grid, and proposed an approach to incorporate it into the design of local converters. Hence interactions between the grid and the to be designed converters could be taken into account and issues mitigated.
Specifically in this research, the proposed method is applied to the control design of a grid-connected active front end (AFE). In a notional system, a voltage source inverter (VSI) is included to emulate the unknown grid and supplies power to an AFE feeding a constant power load (CPL). Firstly, a state space model of the grid is identified through perturbation and response test at the point of common coupling (PCC) in a specially designed experiment. It is then combined with the open loop model of the AFE to build a global model of the grid-AFE system. The plant for the control design will then be not only represented by the AFE's dynamics, but will also include that of the identified grid at the PCC. Implementation of the identification experiment involved and mathematical manipulations used to merge the two subsystem models are presented in detail.
The global model is utilized to synthesize a state feedback controller, denoted as `optimal $H_2$ controller' in this thesis for the AFE by the use of a structured $H_2$ algorithm, which optimizes the dynamic performance of AFE while intrinsically ensuring stability of the grid-AFE system.
Effectiveness and advantages of the proposed control design method is validated by simulations and experiments. The grid-AFE system performance when the AFE adopts the optimal $H_2$ controller or best-tuned proportional-integral (PI) controllers is compared. The use of optimal $H_2$ controller outperforms with faster dynamic response and greater stability margin the PI based solution. Scalability of the proposed method in more complex power grids and its robustness against system parameters drifting are also discussed.
|
|---|