Research Of New Type Static Synchronous Compensator-STATCOM

The balance of reactive power is the preconcerted condition of maintaining the grid stability,but the stable level of the grid is the key problem to restrict the development of power system.As one kind of advanced reactive power compensator in flexibility alternating current transmission system (FACTS) family,STATCOM has slightly volume,high efficiency,fast dynamic response,good control stability,small harmonics component and so on,and it has gradually become a main direction of development for static var compensation technology.Firstly,aiming at harmonic problem exist in high power STATCOM,this thesis raises two main circuit topology—Unit cascaded with zigzag transformer output structure and two-layer three-level with phase-shift transformer output structure.Assisted by Matlab,the two topological structure are validated then,low harmonic component,small THD and prefect symmetric waveforms reflect the validity and rationality of the two topology.Secondly,according to multi-objective control request of power system,this thesis has designed single-objective controller for maintaining node voltage,damping system oscillation and improving transient stability separately.And then the multi-objective fuzzy controller of STATCOM combined with fuzzy control theory and multi-state linear feedback is designed.The multi-objective fuzzy controller can adapt the change of variables and identify the main objective from different operating states of power system.Thus the correct control measure can be taken and the optimum control result can be obtained.At last,according to the dynamic mathematic model of STATCOM,this thesis founds its simulative model.A particular simulative analysis is carried out by putting it into a two-machine power system with two control strategy—multi-objective fuzzy control and reversible system PI control.The simulative result shows that,either reversible system PI control or multi-objective fuzzy control can all improve the stability of power system But compared with the former,the dynamic performance of latter is obviously better.