基于扩展卡尔曼滤波的DFIG变流器控制系统参数辨识方法

A Parameters Identification Method of DFIG Converter Control System Based on Extended Kalman Filter

变流器为风电机组的重要组成部分,其控制系统对风电机组的动态输出特性具有显著影响。针对双馈风电机组(doubly fed induction generator,DFIG)变流器,提出其控制参数辨识方法。根据DFIG变流器控制的详细动态模型分析其控制参数辨识的可行性,并结合参数的轨迹灵敏度来定量分析其辨识的难易程度;基于风电机组实际运行工况下可量测获取变量及其时间序列,结合扩展卡尔曼滤波算法推导相应的状态转移矩阵及观测矩阵,并构建风电机组变流器控制的离散化迭代方程,进而建立变流器控制参数辨识模型。以并网1台DFIG的4机2区系统作为测试系统,通过量测风电机组定转子侧、电网侧电气量来实现变流器控制参数的辨识,仿真结果证明了该方法的准确性和鲁棒性。

The converter is the key part of the wind turbine model,and its control system has a significant impact on the dynamic characteristics of the wind turbine.Therefore,this paper presents a parameter identification method for doubly-fed induction generator(DFIG)convertor controllers.Firstly,based on the mathematical model of the DFIG converter control system,the feasibility of control parameter identification is analyzed,and the difficulty of identification is quantitatively analyzed combined with the trajectory sensitivity of the parameters.Secondly,the state transition matrix and observation matrix of the extended Kalman filter are derived through applying measurable electrical quantities time series in the actual operation process,and the discrete extended Kalman filter equation of the DFIG converter control system is obtained,and the control parameter identification model is established.Finally,the modified 4-machine 2-area system integrated with DFIG is used as the test system to identify the DFIG controller parameters with the voltages and currents at rotor side,stator side and grid side respectively.The simulation results have verified the robustness and accuracy of the proposed method.