提出了一种利用人工神经网络进行电力系统暂态稳定分析的方法。该神经网络取故障后系统暂态量为特征量，采用改进ＢＰ算法进行训练。将样本空间进行模式分类，并对不同类样本作不同处理。以实际系统为例，将选用暂态特征与选稳态特征进行比较，验证了选用暂态特征的准确性和有效性 A method of transient stability analysis of power system based on artificial neural network is proposed. The neural network takes the system transient quantity as feature quantity after fault, and adopts improved BP algorithm to train. The sample space for pattern classification, and different types of samples for different treatment. Taking the actual system as an example, the selection of transient features and the steady-state characteristics are compared to verify the accuracy and validity of the selected transient features