Research On Doubly Fed Induction Generator Dynamic Modeling And Application

With the rapid development of wind power industry and continuous improvement of wind power technology, the form of wind farms connected to the power grid is converted from distribution network to transmission network.Owing to variable speed constant frequency operation characteristics and reactive power regulation function, doubly fed induction generator(DFIG)is drawing more and more attention.However, the dynamic mathematical models of DFIG is not unifiedso that intensive study is needed. With the increasing capacity percentage, wind farmshave significant influence to power grid. New static stability problem will occur after doubly fed wind farms access. In addition, DFIG has complex control system and higher-order mathematical model, which will cause "dimension disaster"in wind farm analysis.In consequence, researching how to reduce the order of DFIGmathematical model accurately appears very important.Focusing on the situation that mosttraditional DFIG simplified mathematical models for small signal stability analysis neglecting the grid side controllerdynamic, this paper firstly established the DFIG mathematical model include grid side controllerdynamic, and compared eigenvalue calculation resultswith traditional models. It can be found that part of the dominant modes (weak damping modes and oscillation modes) are omitted after ignoring grid side controllerdynamic, which directlyaffects small signal stability analysis calculation resultsaccuracy.Therefore, grid side controller dynamic should be take into consideration.As for weak damping oscillation modes excited by DFIG, sensitivityanalysis method is adopted in this paper to find the most relevant parameters and then optimization scheme is put forward to enhance the system small signal stability. It can greatly reduce the workload compared with traditional parameters optimization methods.Finally, in view of the established particular DFIG mathematical model, this paper brought up DFIG reduced order modelsunder different operation conditions for wind farms level analysis by dominant mode shapes method. Meanwhile, time domain simulation analysis method is adopted to verify the accuracy and applicability of the reduced order model.