摘要
常规的配电网多源信息数据辨识方法以电网运行衍生特征挖掘为主,数据辨识偏差相对较大,影响配电网的稳定运行。因此,提出基于生成对抗网络的配电网多源信息数据辨识方法。确定配电网线路阻抗参数,依靠配电网线路短时间内不变的特征,获取线路阻抗额定值,从而判断配电网的运行状态。基于生成对抗网络辨识配电网电流状态序列,根据配电网设备的电流数据变化,确定设备运行状态,实现配电网的优化运行。实验结果表明,该方法的辨识偏差较小,能够满足配电网多源信息数据辨识需求,具有较高的推广价值。
The conventional data identification method of distribution network multi-source information is mainly based on the derived feature mining of power network operation,and the data identification deviation is relatively large,which affects the stable operation of distribution network.Therefore,a multi-source information data identification method based on generative adversarial network is proposed.The impedance parameters of the distribution network line are determined,and the line impedance rating is obtained by relying on the characteristics of the distribution network line unchanged in a short time,so as to judge the operation state of the distribution network.Based on the generated adversarial network,the current state sequence of distribution network is identified,and the operating state of distribution network equipment is determined according to the current data changes of distribution network equipment,so as to realize the optimal operation of distribution network.The experimental results show that the identification deviation of this method is small,and it can meet the requirements of multisource information data identification in distribution network,and has high popularization value.
作者
安喜峰
AN Xifeng(State Grid Shaanxi Electric Power Co.,Ltd.Weinan Power Supply Company,Weinan Shaanxi 714000,China)
出处
《信息与电脑》
2023年第22期196-198,共3页
Information & Computer
关键词
生成对抗网络
配电网
多源信息数据
generate adversarial network
distribution network
multi-source information data
