基于WAMS受扰电压轨迹簇特征的电力系统暂态稳定性预测

Transient Stability Prediction of Power System Based on WAMS Characteristic of Perturbed Voltage Trajectory Clusters

为了能够实现快速实时的暂态稳定性预测,提出了一种基于故障后发电机端电压受扰轨迹簇特征的电力系统暂态稳定性预测方法。首先通过WAMS系统获取故障后发电机端电压受扰轨迹簇信息,利用电压轨迹簇的几何属性定义了29个特征指标,然后利用Relief算法从中筛选出与系统暂态稳定性密切相关的广域故障特征指标子集,并利用支持向量机(SVM)构造电力系统暂态稳定性预测器。以新英格兰10机39节点系统为例,对所提出的暂稳预测器的性能进行了大量测试。结果表明,相较于已有研究应用故障后发电机电压数据进行的暂态稳定性预测,所提出的基于广域故障特征的暂稳预测器能有效提高电力系统暂态稳定性预测的准确性(预测准确率由94%提高到96%以上)。另外,特别是对于不完全WAMS信息的暂稳预测、对未知运行方式和未知拓扑结构的电力系统暂稳预测具有独特的优势,显著提高了暂稳预测系统的鲁棒性(预测准确率均能达到94%以上)。

In order to achieve rapid prediction of real-time transient stability, a power system transient stability prediction method based on the post-fault voltage trajectory cluster features of the generators is proposed. First, trajectory cluster information of the generator voltages after disturbance is obtained by the WAMS system, then, 29 feature indexes are defined according to the geometric property of voltage trajectory clusters. The Relief algorithm is used to sort out index subset of wide-area fault features which are closely related to the transient stability status. The support vector machines（SVM） is utilized to establish the transient stability predictor. Moreover, the performance of the presented predictor is demonstrated by conducting extensively test carried out on the New England 39-bus system. Investigations show that, compared to the existing transient stability prediction studies using the post-fault generator voltage data, the proposed wide-area fault feature extraction method and the transient stability predictor can effectively improve the prediction accuracy（The prediction accuracy can be increased from 94% to more than 96%）. Especially, the proposed method has distinct advantages for transient stability prediction based on incomplete WAMS information, unknown operating conditions, and unknown topologies, which significantly improves the robustness of the transient stability prediction system（prediction accuracy can reach more than 94%）.