基于深度学习的异步风力发电机组功率振荡抑制方法

Deep Learning Based Power Oscillation Suppression Method for Asynchronous Wind Turbine Generators

由于引起异步风力发电机组功率振荡的因素较多,难以保障功率振荡抑制的效果,为此,提出基于深度学习的异步风力发电机组功率振荡抑制方法的研究。从风力机动态特性、发电机组动态特性以及传动链动态特性3个角度分析了异步风力发电机组功率振荡状态,借助长短期记忆网络实现对振荡的有效抑制。在测试结果中,设计的振荡抑制方法应用效果不仅更高效,且对于基础振荡幅度的控制效果更好。

Due to the numerous factors that cause power oscillation in asynchronous wind turbines,the effectiveness of power oscillation suppression is difficult to guarantee.Therefore,a deep learming based method for suppressing power oscillation in asynchronous wind turbines is proposed.After analyzing the power oscillation state of asynchronous wind turbines from three perspectives:Dynamic characteristics of wind turbines,dynamic characteristics of generator units,and dynamic characteristics of transmission chains,effective suppression of oscillation is achieved through the use of long short-term memory networks.In the test results,the application effect of designing oscillation suppression methods is not only more efficient,but also better for controlling the basic oscillation amplitude.