[Poster Presentation]Convolutional Deep Leaning-Based Distribution System Topology Identification with Renewables - 2nd China International Youth Conference on Electrical Engineering (CIYCEE 2021)

Convolutional Deep Leaning-Based Distribution System Topology Identification with Renewables

ID：61 Submission ID：340 View Protection：ATTENDEE Updated Time：2021-12-04 17:21:24 Hits：363 Poster Presentation

Start Time：2021-12-17 15:45 （Asia/Shanghai）

Duration：5min

Session：[Z] Poster Session » [Z9] Poster Session 9: Power system and automation

Video No Permission Presentation File

Tips: The file permissions under this presentation are only for participants. You have not logged in yet and cannot view it temporarily.

Abstract

Obtaining the distribution system topology states timely is critical for system monitoring while challenged by correlations brought by high penetrated renewable energy sources (RES). To address this issue, a deep learning model is proposed for distribution system topology identification considering the underlying complex correlations of renewables. Specifically, to remove the dependence of the power system model parameters like line impedance, the input of the model only consists of the voltage magnitudes. Then, this is fed into the proposed deep learning model (DLM), which can fully capture the data features and thus classify the topology of the grid to hedge against the correlations of the RES and thus enhance the identification accuracy. The simulation results demonstrate the accuracy and efficiency of the proposed model in the IEEE 33-node distribution system.