半月刊

ISSN 1000-1026

CN 32-1180/TP

+高级检索 English
基于两阶段迁移学习的电力系统暂态稳定评估框架
作者:
作者单位:

1.电网智能化调度与控制教育部重点实验室(山东大学),山东省济南市 250061;2.中国电力科学研究院有限公司,北京市 100192

摘要:

为提高基于数据驱动的暂态稳定评估模型对电网的自适应性,将深度迁移学习引入更新过程,提出一种基于两阶段迁移学习的暂态稳定评估框架。所提框架根据时间尺度分为2个阶段:在第1阶段,利用深度子领域自适应网络挖掘无标注数据信息,将模型的评估性能快速提升到相对可靠的水平,得到迁移模型,结合时域仿真法进行综合判稳,提高电网变化初期模型的可用性;在第2阶段,利用迁移模型筛选高价值样本集,并结合样本迁移和微调技术进行二次更新,使评估性能恢复到较高水平,降低更新时间成本。在IEEE 39节点系统和中国某省级电网模型上进行测试,结果表明所提框架具有完备性,可使模型快速响应电网的变化。

关键词:

基金项目:

国家重点研发计划资助项目(响应驱动的大电网稳定性智能增强分析与控制技术,2021YFB2400800)。

通信作者:

作者简介:

李保罗(1996—),男,博士研究生,主要研究方向:深度学习在暂态稳定评估中的应用。E-mail:378011236@qq.com
孙华东(1975—),男,通信作者,博士,教授级高级工程师,博士生导师,主要研究方向:电力系统稳定与控制。E-mail:sunhd@epri.sgcc.com.cn
张恒旭(1975—),男,博士,教授,博士生导师,主要研究方向:电力系统稳定分析与控制、电力系统检测、电力系统数值仿真。E-mail:zhanghx@sdu.edu.cn


Transient Stability Assessment Framework of Power System Based on Two-stage Transfer Learning
Author:
Affiliation:

1.Key Laboratory of Power System Intelligent Dispatch and Control of Ministry of Education (Shandong University), Jinan 250061, China;2.China Electric Power Research Institute Co., Ltd., Beijing 100192, China

Abstract:

In order to improve the adaptability of the data-driven transient stability assessment model to the power grid, deep transfer learning is introduced into the updating process, and a transient stability assessment framework based on two-stage transfer learning is proposed. The proposed framework is divided into two stages according to the time scale. In the first stage, the deep subdomain adaptive network is used to mine unlabeled data information, and the evaluation performance of the model is rapidly improved to a relatively reliable level. The obtained transfer model is combined with the time domain simulation method for comprehensive stability judgment, which improves the usability of the model in the initial stage of power grid change. In the second stage, the transfer model is used to select the high-value sample set, and the second update is carried out by combining the sample transfer and fine-tuning technology to restore the evaluation performance to a higher level and reduce the updating time cost. The test results in the models of IEEE 39-bus system and a provincial power grid in China show that the proposed framework has completeness and can make the model quickly respond to the changes in the power grid.

Keywords:

Foundation:
This work is supported by National Key R&D Program of China (No. 2021YFB2400800).
引用本文
[1]李保罗,孙华东,张恒旭,等.基于两阶段迁移学习的电力系统暂态稳定评估框架[J].电力系统自动化,2022,46(17):176-185. DOI:10.7500/AEPS20211207005.
LI Baoluo, SUN Huadong, ZHANG Hengxu, et al. Transient Stability Assessment Framework of Power System Based on Two-stage Transfer Learning[J]. Automation of Electric Power Systems, 2022, 46(17):176-185. DOI:10.7500/AEPS20211207005.
复制
支撑数据及附录
分享
历史
  • 收稿日期:2021-12-07
  • 最后修改日期:2022-03-31
  • 录用日期:2022-03-31
  • 在线发布日期: 2022-09-16
  • 出版日期: