| 引用本文: | 刘钊瑞,高云鹏,郭建波,等.基于深度自编码器高斯混合模型的窃电行为检测[J].电力系统保护与控制,2022,50(18):83-91. |
| LIU Zhaorui,GAO Yunpeng,GUO Jianbo,et al.Abnormal detection of electricity theft using a deep auto-encoder Gaussian mixture model[J].Power System Protection and Control,2022,50(18):83-91 |
|
| 摘要: |
| 针对用户侧窃电检测背景下无监督方法的适用性，研究如何解决特征提取和异常检测间的解耦问题，提出基于深度自编码器高斯混合模型(Deep Auto-encoder Gaussian Mixture Model, DAGMM)的用户窃电行为检测方法。首先对数据进行增广迪基-福勒检验，获取具有平稳性的用电数据维度。然后通过压缩网络提取数据潜在特征，利用估计网络及高斯混合模型获取反映异常程度的样本能量。最后基于端对端的学习方式对网络参数联合优化以避免模型解耦，将样本能量超过异常阈值的用户识别为窃电，据此实现用户窃电行为检测。实验结果表明，基于深度自编码器高斯混合模型的窃电行为检测方法受窃电样本影响小，提取的特征可有效反映用户用电规律，具有更高的检测准确率。相比于现有方法，其检出率、误检率、F1测度及AUC等评价指标均有显著提高。 |
| 关键词: 窃电行为 无监督学习 深度自编码器高斯混合模型 增广迪基-福勒检验 解耦 |
| DOI：DOI: 10.19783/j.cnki.pspc.211659 |
| 分类号: |
| 基金项目:国家自然科学基金项目资助(51777061)；广西电网科技项目资助(GXKJXM20200020) |
|
| Abnormal detection of electricity theft using a deep auto-encoder Gaussian mixture model |
|
LIU Zhaorui1, GAO Yunpeng1, GUO Jianbo2, LI Yunfeng1, GU Dexi1, WEN Yizhang1
|
|
(1. College of Electrical and Information Engineering, Hunan University, Changsha 410082, China;
2. Hangzhou Haixing Electrical Co., Ltd, Hangzhou 310011, China)
|
| Abstract: |
| Considering the applicability of unsupervised methods for user-side electricity theft detection, this paper studies how to solve the decoupling problem between feature extraction and anomaly detection. It proposes a user-side electricity theft detection method based on the deep auto encoder Gaussian mixture model (DAGMM). First, the electricity consumption data dimension with stationarity is obtained according to the augmented Dickey Fuller test. Then, potential characteristics of data are extracted by compressing the network. An estimation network and Gaussian mixture model are used to obtain sample energy. This reflects the degree of anomaly. Finally, network parameters are optimized jointly based on end-to-end learning to avoid model decoupling, and identify users whose sample energy exceeds the abnormal threshold as electricity thief. In this way theft of electricity can be detected. The experimental results show that the detection method based on DAGMM is less affected by the sample of electricity theft, and the extracted features can effectively reflect the user's electricity consumption law with higher detection accuracy. Compared with the existing methods, the detection rate, false detection rate, F1 measurement and AUC of the proposed method are significantly improved.
This work is supported by the National Natural Science Foundation of China (No. 51777061). |
| Key words: stealing electricity unsupervised learning deep auto-encoder Gaussian mixture model augmented Dickey Fuller test decoupling |
