基于长短期记忆网络的电网同调机群快速辨识  

A Fast Prediction Method of Coherent Generators Based on Long Short-term Memory Network

作　　者：毛煜 尚海昆[1] 于卓琦 MAO Yu;SHANG Haikun;YU Zhuoqi(School of Electrical Engineering,Northeast Electric Power University,Jilin 132012;State Grid Zhejiang Hangzhou Fuyang Power Supply Company Co.,Ltd.,Hangzhou 311400)

机构地区：[1]东北电力大学电气工程学院,吉林132012 [2]国网浙江省电力有限公司杭州市富阳区供电公司,杭州311400

出　　处：《电气工程学报》2022年第2期201-207,共7页Journal of Electrical Engineering

摘　　要：基于长短期记忆网络(Long short-term memory,LSTM)提出了一种电网同调机群的快速辨识方法。首先针对两机振荡模型,挖掘相量平面内电压相量轨迹的分类特性,为机组的同调性辨识提供了依据;其次,基于短时响应数据,利用LSTM分别对机端电压实、虚部时序轨迹进行预测,并依据复合而成的相量轨迹判断机组的分群情况;最后,利用扩展等面积法则(Extended equal area criterion,EEAC)对上述分群情况进行验证,进而给出同调机群的最终辨识结果。IEEE-39节点系统算例验证了方法的有效性,具有较好的工程应用价值。Based on the long short-term memory network(LSTM),a fast prediction method of coherent generators is proposed.Firstly,the classification characteristics of bus voltage phase trajectories are extracted to provide a new way for the identification of generator coherency.Secondly,based on the short-term response data,the real and imaginary parts of the generator terminal voltage phase are predicted respectively by using LSTM,and the coherent generators are identified based on the fitted voltage phase trajectories.Finally,the extended equal area criterion(EEAC)is used to further verify the coherency of the identified generator groups.The proposed method is validated used in the IEEE-39 bus system,and the simulation results show that the method has the advantages of higher engineering practice value.

关 键 词：同调机群辨识 电压相量轨迹 长短期记忆网络(LSTM) 扩展等面积法则(EEAC) 

分 类 号：TM561[电气工程—电器]