| 引用本文: | 李华瑞,李文博,李 铮,等.基于生成对抗网络与度量学习的数据驱动频率安全评估[J].电力系统保护与控制,2024,52(18):101-111.[点击复制] |
| LI Huarui,LI Wenbo,LI Zheng,et al.Data-driven frequency security assessment based on generative adversarial networks and metric learning[J].Power System Protection and Control,2024,52(18):101-111[点击复制] |
|
| 摘要: |
| 随着大容量远距离高压直流输电工程的建设和大规模可再生能源的接入，电力系统的频率安全面临严峻挑战。为了对频率安全进行快速准确的在线评估，提出一种基于度量学习与生成对抗网络技术的数据驱动频率安全评估模型。首先，选取关键频率安全指标作为模型输出，并构建输入特征集。然后，使用改进的基于Wasserstein距离度量的生成对抗网络(Wasserstein generative adversarial network, WGAN)学习电力系统历史运行场景分布信息，生成覆盖系统典型运行方式的运行场景以构建训练样本集。计及电力系统复杂运行方式下单个机器学习模型对频率安全评估的不适用性，基于核回归度量学习(metric learning for kernel regression, MLKR)算法构建由多个子模型构成的频率安全组合评估模型。最后使用简化的山东电网算例，验证了所提方法的有效性。 |
| 关键词: 频率安全 机器学习 数据驱动 生成对抗网络 度量学习 |
| DOI：10.19783/j.cnki.pspc.231542 |
| 投稿时间：2023-12-27修订日期：2024-03-15 |
| 基金项目:江苏省基础研究计划自然科学基金项目资助(BK20230167)；北京市自然科学基金青年基金项目资助(3234063)；国网江苏省电力有限公司科技项目资助(J2022016) |
|
| Data-driven frequency security assessment based on generative adversarial networks and metric learning |
| LI Huarui1,LI Wenbo1,LI Zheng1,JIA Yuqiao1,LIU Quan1,MIAO Deyang1,LI Yaran1,WANG Baocai2 |
| (1. State Grid Jiangsu Electric Power Co., Ltd. Electric Power Research Institute, Nanjing 211100, China; 2. State Key Laboratory
of Power Grid Safety and Energy Conservation (China Electric Power Research Institute), Beijing 100192, China) |
| Abstract: |
| With the construction of large-capacity long-distance high-voltage direct current transmission projects and the large-scale integration of renewable energy, frequency security of the power system is facing severe challenges. For fast and accurate online assessment of frequency security, a data-driven model based on a metric learning (ML) and generative adversarial network (GAN) is proposed. First, the key frequency security indicators are selected as the outputs of the model, and an input feature set is constructed. Then, the improved Wasserstein generative adversarial network (WGAN) based on the Wasserstein distance metric is used to learn the distribution information of historical operation scenarios of power systems. This generates operational scenarios covering typical modes to build the training sample set. Considering the inapplicability of a single machine learning model to frequency security assessment with complicated operational modes of power systems, a combined assessment model for assessment composed of multiple sub-models is constructed based on metric learning for a kernel regression (MLKR) method. Finally, a simplified Shandong power system example is used to verify the effectiveness of the proposed method. |
| Key words: frequency security machine learning data-driven generative adversarial network metric learning |
