基于改进最大相关最小冗余判据的暂态稳定评价特征选择-arXiv.PDF

基于改进最大相关最小冗余判据的 暂态稳定评估特征选择 李扬，顾雪平 (华北电力大学电气与电子工程学院，河北省 保定市 071003) Feature Selection for Transient Stability Assessment Based on Improved Maximal Relevance and Minimal Redundancy Criterion LI Yang, GU Xueping (School of Electrical and Electronic Engineering, North China Electric Power University, Baoding 071003, Hebei Province, China) ABSTRACT: A new feature selection method based on an 嵌套的候选特征子集，并使用支持向量机分类器验证各候选 improved maximal relevance and minimal redundancy (mRMR) 特征子集的分类性能，选择得到具有最大分类正确率的特征 criterion was proposed for power system transient stability 子集。基于新英格兰39 节点系统和IEEE 50 机测试系统的 assessment. First, the standard mRMR was improved by 算例结果验证了所提特征选择方法的有效性。 introducing a weight coefficient in the evaluation criteria to 关键词：暂态稳定评估；特征选择；最大相关最小冗余；支 refine the measurement of the features correlation and 持向量机；相量测量单元 redundancy. Then, the possible real-time information provided by phasor measurement unit (PMU) considered, a group of 0 引言 system-level classification features were extracted from the 电力系统暂态稳定评估(transient stability power system operation parameters to build the original feature assessment，TSA)是保证电力系统安全稳定运行的 set, and the improved mRMR was employed to evaluate the [1] 基础性手段 。近年来随着计算技术的快速发展， classification capability of the original features for feature selection. A group of nested candidate feature subsets were 基于人工神经网络(artificial neural network ， obtained by using the incremental search technique, and each A