摘要
针对电力系统暂态稳定性分析中影响系统稳定性的关键因素在向量空间有一定相似性的特点 ,提出了一种基于聚类的人工神经网络方法 ,并在此基础上提出了一种对神经网络学习参数进行有效调整的经验方法 ,通过对实际电网的计算 ,证实了所述方法的快速性和有效性。
In transient stability analysis, the features vectors affecting system stability are dramatically similar to each other inthe vector space. Aiming at this characteristic, this paper presents an artificial neural network (ANN) method based onclustering analysis. With the help of clustering analysis. the structure of ANN is simplified and the learning rate is improveddramatically. Furthermore. in order to decrease time-consumption for ANN training, this paper presents an efficient strategyfor adjusting learning rate of ANN automatically. The result of a case study illustrates its validity. Compared withconventional integral method (SBS). this method is superior in calculation speed..
出处
《电力系统自动化》
EI
CSCD
北大核心
2000年第12期1-3,共3页
Automation of Electric Power Systems
基金
国家重点基础研究专项经费资助项目! ( G1 9980 2 0 31 3)
