| 引用本文: | 吴思婕,王怀远.基于集成学习的时间自适应电力系统暂态稳定评估方法[J].电力系统保护与控制,2022,50(24):112-119.[点击复制] |
| WU Sijie,WANG Huaiyuan.Transient stability assessment of power system with time-adaptive method based on ensemble learning[J].Power System Protection and Control,2022,50(24):112-119[点击复制] |
|
| 摘要: |
| 对于电力系统暂态稳定评估而言，在故障清除后的早期阶段，临界样本间的特征差异不明显，预测准确率低。随着时间推移，准确率提高，但难以保证评估的及时性。针对暂态稳定评估的评估准确性与及时性之间的矛盾，提出了基于集成学习的时间自适应电力系统暂态稳定评估方法。首先，通过EasyEnsemble算法对不平衡数据进行采样，训练出多个不同评估时刻的集成长短期记忆网络分类器，输出样本在不同评估时刻的稳定性预测结果。其次，将评估时刻进行划分，提出了多阶段阈值分类规则，自适应调整阈值，对样本预测结果进行可信度评估。最后，预测结果评估为不可信的样本交由下一评估时刻的模型继续判断，直到可信度达到阈值后输出。在IEEE 39节点系统的仿真结果表明，所提方法相较于其他时间自适应方法具有更优的评估性能，在样本不平衡的情况下该方法实现了更好的修正效果。 |
| 关键词: 深度学习 暂态稳定评估 集成学习 时间自适应评估 样本不平衡 |
| DOI：DOI: 10.19783/j.cnki.pspc.220282 |
| 投稿时间：2022-03-06修订日期：2022-05-31 |
| 基金项目:国家自然科学基金项目资助(51707040) |
|
| Transient stability assessment of power system with time-adaptive method based on ensemble learning |
| WU Sijie,WANG Huaiyuan |
| (Key Laboratory of New Energy Generation and Power Conversion Fuzhou University), Fuzhou 350116, China)) |
| Abstract: |
| For power system transient stability assessment, the characteristic differences of critical samples are not obvious in the early stage after fault clearance, and the prediction accuracy is low. Over time, the evaluation accuracy improves, while the timeliness of the evaluation is difficult to ensure. Aiming at the contradiction between the accuracy and timeliness of transient stability assessment, a time-adaptive transient stability assessment method of power system based on ensemble learning is proposed. First, the unbalanced data are sampled by the EasyEnsemble algorithm. Ensemble LSTM classifiers with different evaluation cycles are trained. Thus, the stability prediction results of samples in different evaluation cycles are output. Second, the evaluation moments are divided and multi-stage threshold classification rules are proposed. The threshold is adjusted adaptively to evaluate the reliability of the prediction results. Finally, the samples whose prediction results are evaluated as unreliable are handed over to the model of the next evaluation cycle for judgment until the reliability reaches the threshold. The simulation results in the IEEE39 bus system show that the proposed method has better performance than other time-adaptive methods; in the case of unbalanced samples, the method achieves a better correction effect.
This work is supported by the National Natural Science Foundation of China (No. 51707040). |
| Key words: deep learning transient stability assessment ensemble learning time-adaptive assessment imbalanced samples |
