| A microgrid is a small power system consisting of distributed generation units,energy storage units,and loads,and has the characteristics including low inertia and large uncertainties.The tra-ditional microgrid control strategies mainly consider steady-state performance,whereas transient performance is not quantitatively analyzed.With the load changing in microgrids,the system out-puts,such as voltages and frequencies,may be greatly deviated from their set points,resulting in the abnormal operation of the system.Therefore,how to enhance the transient control performanceof microgrids has always been the focus of researches.By introducing the transient performance enhancement method,the key indices' constraints of the generation and energy storage units can be guaranteed during the transient process,and thus the microgrids can be operated stably and efficiently in the presence of load changes.However,there exist some problems when applying the transient performance control methods to microgrids.Firstly,in the presence of the uncertainties of generation and energy storage units,the steady-state errors of the systems can only be guaranteed to be bounded with the existing transient performance enhancement methods,i.e.,the accurate tracking performance cannot be achieved.Secondly,the transient performance enhancement methods can only ensure the satisfaction of microgrids' con-straints.The control optimality is not considered.Thirdly,for a multi-agent system consisting of different kinds of generation and energy storage units,although the existing control algorithms can achieve the consensus of agent states in steady-state,how to improve the transient performance of the agent outputs needs to be investigated.Fourthly,for the generation and energy storage systems in microgrids,most of the existing control methods can only assure the steady-state performance,the transient performance is not quantitatively analyzed.In this thesis,to solve the aforementioned theoretical and practical problems,advanced control methods for transient performance enhance-ment of microgrids are designed.The main contributions are as follows:1.Considering the transient performance requirement of microgrids,a neural network and ro-bust integral of the sign of the error based continuous control strategy is presented for gen-eration and energy storage units described by uncertain nonlinear systems to achieve the asymptotic tracking performance.The system outputs are ensured to be restricted in user-defined time-varying constraints,which improves the systems' transient performance.The control signal of the proposed controller is continuous,and thus the chattering phenomenon can be avoided.The prior knowledge of system uncertainties is not required.In theory,the asymptotic tracking performance of the system is guaranteed via rigorous stability analysis.2.Taking into account the control optimality requirement of microgrids,an adaptive dynamic programming method for continuous-time nonlinear systems is proposed for generation and energy storage units.To cope with the unknown internal dynamics of microgrids,a ro-bust adaptive dynamic programming algorithm based optimal control method is introduced,which avoids solving the Hamilton-Jacobi-Bellman partial differential equations directly.With the help of the small-gain theory,the system is proved to be asymptotically stable.3.Considering that microgrids can be described as multi-agent systems consisting of generation and energy storage units with different dynamics,a distributed consensus control algorithm is designed to achieve the consensus control of non-affine multi-agent systems with unknown control directions.The system transient performance is improved by placing time-varying constraints on the agents' outputs.The designed algorithm does not require information about system dynamics and modeling errors.Through theoretical analysis,all the agent states will converge to consensus asymptotically.4.A class of excitation control methods is developed for synchronous generators such as the diesel ones in microgrids to ensure the satisfaction of the time-varying constraints on system filtered errors,and to enhance the transient performance.The system transient performance can be shaped quantitatively by selecting the time-varying constraints of the system filtered errors.To deal with the uncertain damping and inertia constants,an online adaptive mecha-nism is designed to compensate for their negative impact.Through the Lyapunov synthesis,the system errors can be guaranteed to be asymptotically stable.5.To deal with the deployment problem of pulsed power loads in shipboard microgrids,a strat-egy that the pulsed power loads are supplied by flywheel energy storage systems is proposed.The impact on the system during the usage of pulsed power loads is avoided.Besides,an adaptive control algorithm is designed to improve the transient performance of the shipboard microgrid during the charging process of the flywheel energy storage system.By construct-ing the whole problem as a control problem of output-constrained systems,the system tran-sient performance can be regulated quantitatively.Via Lyapunov analysis,the closed-loop system errors are proved to be asymptotically stable theoretically.At the end of this thesis,concluding remarks are provided and future works are discussed. |
PDF Full Text Request