| 引用本文: | 王海东,程 杉,徐其平,刘 烨,王 灿.基于深度学习融合网络的含噪电能质量扰动识别方法[J].电力系统保护与控制,2024,52(10):11-20.[点击复制] |
| WANG Haidong,CHENG Shan,XU Qiping,LIU Ye,WANG Can.Identification of power quality disturbance with noises based on an integrated deep learning network[J].Power System Protection and Control,2024,52(10):11-20[点击复制] |
|
| 摘要: |
| 针对强噪声环境下电能质量扰动识别精度不高的问题，提出一种自适应小波降噪和深度学习相结合的电能质量扰动识别方法。首先，通过改进峰和比分层自适应阈值和能量优化的阈值函数算法对含噪扰动信号进行降噪处理。然后，通过残差神经网络对降噪后的扰动信号进行深层特征提取，在此基础上融入多头注意力机制下的双向长短时记忆网络，建立时序特征依赖关系，构成适用于噪声环境下的扰动识别框架。最后，在不同强度噪声环境下对20类扰动信号进行仿真实验。由仿真结果可知，该方法具有良好的噪声鲁棒性，在不同噪声环境下均有较高的识别正确率。 |
| 关键词: 电能质量扰动 自适应小波降噪 残差神经网络 多头注意力 双向长短时记忆网络 |
| DOI：10.19783/j.cnki.pspc.231503 |
| 投稿时间：2023-11-27修订日期：2024-01-29 |
| 基金项目:国家自然科学基金项目资助(52107108) |
|
| Identification of power quality disturbance with noises based on an integrated deep learning network |
| WANG Haidong1,CHENG Shan1,XU Qiping1,LIU Ye2,WANG Can1 |
| (1. College of Electrical Engineering and New Energy, China Three Gorges University, Yichang 443002, China;
2. Jiaxing Power Supply Company, State Grid Zhejiang Electric Power Co., Ltd., Jiaxing 314000, China) |
| Abstract: |
| A novel method combined with adaptive wavelet threshold noise reduction and deep learning is proposed to improve the accuracy of identifying power quality disturbances in strong-noise environments. First, the noise-containing disturbance signals are noise-reduced by a threshold function algorithm based on an improved peak and score level adaptive thresholding and energy optimization. Then, the residual network is used to extract deep features from the noise-reduced disturbance signals, based on which the bidirectional long short term memory network under the multi-attention mechanism is incorporated to establish temporal feature dependency. This constitutes a framework applicable to the recognition of disturbance signals in a noisy environment. Finally, numerical simulations are carried out on 20 types of disturbance signals in different noise environments. It can be seen from the results that the proposed method has good noise robustness and high recognition accuracy in different noise environments |
| Key words: power quality disturbances adaptive wavelet threshold residual neural network multi-headed attention bidirectional long-short term memory network |
