基于多新息随机梯度算法的网侧变流器参数辨识方法研究

Parameter identification of grid-side converter based on MISG

永磁直驱风力发电机(D-PMSG)通过双PWM全功率变流器实现和电网的连接,但在其运行过程中,网侧变流器的参数变化会对其控制性能造成影响,精确辨识网侧变流器参数对于提升发电系统控制性能具有重要意义。文章基于网侧变流器在d-q轴的动态方程,构建了网侧变流器的回归模型,设计了网侧变流器电气参数的带收敛因子的多新息遗忘随机梯度辨识算法,实现网侧变流器的电气参数辨识,并通过仿真实验验证了算法的有效性。

Permanent magnet synchronous wind power system is connected the power grid by dual PWM full-power converter. The change of grid-side converter parameters will affect the control performance of the system, and accurate parameter identification is very important to converter control and operation state monitoring. Based on the dynamical equations of d-q axis currents, the system regression model is proposed, and multi-innovation stochastic gradient algorithm with convergence index for grid-side converter parameters identification is derived. Simulation experiments results shows the effectiveness of the designed algorithm.