引用本文: | 刘恒勇,史帅彬,徐旭辉,等.一种关联RNN模型的非侵入式负荷辨识方法[J].电力系统保护与控制,2019,47(13):162-170.[点击复制] |
LIU Hengyong,SHI Shuaibin,XU Xuhui,et al.A non-intrusive load identification method based on RNN model[J].Power System Protection and Control,2019,47(13):162-170[点击复制] |
|
摘要: |
为了进一步拓展监督学习方法在非侵入式负荷辨识中的应用,提出了一种关联循环神经网络(Recurrent Neural Networks, RNN)模型的负荷辨识方法。在该方法中,首先引入了时间窗负荷事件检测方法,提取谐波分量作为负荷特征,并将负荷特征作为RNN模型的输入。然后根据其对历史输入特征量的记忆建立由输入映射到输出的内在关联,从而建立面向时间序列输入的RNN负荷辨识方法。进一步地,为了避免“梯度消失”问题,选择了最佳的激活函数和损失函数。最后,通过单负荷辨识、多负荷辨识的实测实验,证实了所提关联RNN模型的负荷辨识方法能够有效地实现用户内部负荷设备状态的辨识要求。 |
关键词: 非侵入式 负荷辨识 深度学习 事件检测 RNN |
DOI:10.7667/PSPC20191322 |
投稿时间:2018-06-29修订日期:2018-08-31 |
基金项目:中国南方电网有限责任公司科技项目资助(090000KK52160049) |
|
A non-intrusive load identification method based on RNN model |
LIU Hengyong,SHI Shuaibin,XU Xuhui,ZHOU Dongguo,MIN Ruolin,HU Wenshan |
(Shenzhen Power Supply Bureau Co., Ltd., Shenzhen 518048, China;School of Power and Mechanical Engineering, Wuhan University, Wuhan 430072, China) |
Abstract: |
In order to extend the application by using supervised learning methods in non-intrusive load identification, a load identification method based on Recurrent Neural Networks (RNN) model is proposed. Firstly, the time window for detecting load event is introduced, and then the harmonic components are taken as the load characteristics to be used as the inputs of the RNN model. According to the memory of its memory history input feature quantity, the internal mapping of the input to the output as well as the RNN load identification method for time series inputs are established. Furthermore, in this model, the suitable activation function and loss function are selected in order to avoid the "gradient disappearance" problem. Finally, the experiment on the single and multi-load identification demonstrates the model can effectively realize the identification of the load status. This work is supported by Science and Technology Project of China Southern Power Grid Co., Ltd. (No. 090000KK52160049). |
Key words: non-intrusive load identification deep learning event detection RNN |