Dynamic Stability Control Strategy For Micro Grid

Nowadays,conventional energy resources come to an exhaustion.With the pressure of environmental pollution and the growing demand for power energy,new renewable energy technologies are met with challenges and opportunities.Micro grid can operate in an islaned mode and a grid-connected mode and it can integrate various forms of distributed power supply by means of intelligent control.Various forms of distributed power supply can be effectively coordinated through the unique topology of micro grid,which will make full use of the advantages of micro grid and thus will improve the power quality.Various forms of micro power and various energy storage components can be connected in series to the power network through inverters.Accordingly,the development of inverter control methods has an important influence on the safety,efficiency,economy and other characteristics of the micro grid.In this paper,the inverter in the micro power supply is taken as the research object.Stable operation and control method are researched carefully.Firstly,the islanded mode and the grid-connected mode of the micro grid has been analysed.Basic method of constant power control,constant voltage constant frequency control and droop control is analysed in detail.The control system combining fuzzy PI controller and double closed loop control scheme,greatly improves the control effect of micro grid.Secondly,the mathematical model of the three-inverters system is established.The control method of the droop characteristic of inverters in micro grid is studied and designed.Thirdly,for the topology of single micro grid connected inverter and uncertain factors and random interference in the operation of the whole system,an H?robust control method is proposed.LMI theory is used to design and implement the controller,which makes the inverter not only operated stably for a long time,but also gained an anti-interference characteristic of a certain degree.Finally,correlation models have been built in the matlab simulink and the proposed method is verified.