基于压缩感知和堆叠降噪自动编码器的配电网扰动事件智能溯源方法

Intelligent Traceability Method of Distribution Network Disturbance Events Based on CS-SDAE

鉴于配电网扰动事件的溯源分析有利于准确了解扰动原因、保障电力系统的安全稳定运行,提出了一种基于压缩感知和堆叠降噪自动编码器的配电网扰动事件智能溯源方法,首先利用压缩感知方法将添加了噪声的原始数据映射到压缩域,在保留扰动特征的同时提高了分析效率;然后将压缩采样数据作为堆叠降噪自动编码器的输入,通过堆叠降噪自动编码器的特征自学习能力,学习得到扰动数据中的鲁棒性特征,实现特征与不同配电网扰动事件的关联,构造扰动事件智能溯源模型。通过PSCAD/EMTDC中搭建的IEEE 14节点模型获得的仿真数据进行验证,表明所提方法能准确溯源配电网扰动事件。

The traceability analysis of distribution network disturbance events is conducive to accurately understanding the cause of the disturbance, ensuring the safe and stable operation of the power systems. This paper proposes an intelligent traceability method for distribution network disturbance events based on compressed sensing and stacked denoising autoencoder(SDAE). Firstly, the compressed sensing(CS) is used to map the noise-added original data to the compressed domain to improve the analysis efficiency while retaining the disturbance characteristics. Then, the compressed sampling data is used as the input of SDAE, and the robustness characteristics in the distribution network disturbance data are learned to realize the correlation between the characteristics and different power disturbance events. The simulation data obtained by the IEEE 14-node model built in PSCAD/EMTDC is used to verify the performance of the proposed method. The simulation experiment shows that the proposed method can accurately trace the source of distribution network disturbance events.