中文摘要：

为提高电力系统暂态稳定评估单个模型的准确率,研究了基于元学习策略的暂态稳定评估问题,提出了支持向量机、决策树、朴素贝叶斯和K最近邻法作为基学习算法,线性回归为元学习算法的Stacking评估模型。该模型将上述基学习算法的概率输出作为新训练数据的输入特征,同时保留原始的类标识。线性回归算法在新训练集上学习得到最终暂态稳定评估结果。新英格兰39节点测试系统和IEEE50机测试系统上仿真实现了该模型,仿真结果证明所提模型比单个模型的评估性能更好,为电力系统暂态稳定评估提供了新的思路。

英文摘要：

In order to increase singe model＇s accuracy in transient stability assessment of power systems,the transient stability assessment based on meta-learning strategy is studied,and a Stacking assessment model is presented.The base learning includes support vector machines,decision trees,naive Bayesian and K-nearest neighbor classifier.Linear regression is adopted as Stacking assessment model of meta-learning.The model uses the probabilistic output of base learning algorithm as input attributes in a new training set,and keeps the original class labels.The final transient stability assessment result is acquired after learning in the new training set by linear regression.The simulations on New England 39-bus and IEEE 50-machcine test systems show that the assessment performance of the proposed approach is better than that of the single models and provides a new way to assess power system transient stability.This work is supported by National Natural Science Foundation of China（No.90610026）.