Automatic Identification of Key Transmission Sections Considering Source-load Uncertainties

作者：Jiaqi Geng, Jiacheng Liu, Xiaoming Liu, Yao Liu, Xuetao Dong, Jun Liu · 年份：2024 · DOI：10.1109/ispec59716.2024.10892480 · 被引用次数：1 · 研究领域：Power System Reliability and Maintenance、Vibration and Dynamic Analysis、Power Systems Fault Detection

With the rapid development of renewable energy sources, more and more uncertain power sources such as wind power and photovoltaic power stations are connected to the power systems, causing changes in the operating status and even the topology structure of the power systems. As a result, the sectional power flow may fluctuate in a wide range, thus increasing the risk of some lines exceeding their limits and affecting the power flow transmission between and within regions. In serious cases, the transmission channel can be interrupted, resulting in huge losses, which brings challenges to the traditional AC transmission cross-section identification method. Therefore, this paper proposes an automatic identification method of key transmission sections considering source-load uncertainties. Firstly, we construct a sample set of output and node load of renewable energy units, simulate various scenarios under the influence of source load uncertainty. Then we process the power flow calculation results as input. Finally, we use the key cross section results identified by the margin index as a label, to train the machine learning model based on the AdaBoost algorithm. Simulation results on the IEEE-39 test system demonstrate the effectiveness and accuracy of the proposed automatic identification of key cross sections.