基于LSTM多类型机组风电场等值建模研究

Study on the Equivalent Modeling of Wind Power Farm Based on LSTM Multi-type Wind Turbine Generators

为了对包含低电压穿越环节的风电机组的风电场建立有效的等值模型,本文讨论了通用模型参数对于风电机组动态特性的影响,提出了基于长短期记忆网络的等值建模方法,以降低限幅环节对建模准确性的影响。以多类型的风电场为例,建立两种算例系统,将文中方法与支持向量机算法进行对比。该方法能有效地提高风电场等值建模的有效性,有一定的参考价值。

To establish an effective equivalent model for wind farms with wind turbines which had the ability of low voltage ride through, the influence of the parameters of generic models on the dynamic characteristics of wind turbines was discussed, and an equivalent modeling method based on long-short term memory neural network was proposed to improve the accuracy, especially to reduce the influence of limit steps. Taking multi-type wind farms as an example, two kinds of example systems were established, and the proposed method was compared with the support vector machine(SVM) method. The results showed that this algorithm effectively improved the efficiency of equivalent modeling in wind farms and had certain reference value.