基于人工神经元网络的电力系统暂稳极限功率快速确定

Fast estimation of electric power system transient stability limits based on artificial neural network

提出用神经元网络来实现暂态能量裕度、发电机输出功率及故障切除时间三者之间的非线性映射关系，从而实现暂态稳定极限功率的快速精确求解，并将其研究范围扩大到同时考虑故障切除时间变化的影响，突破了灵敏度分析法的局限性。文中还将电力系统动态等值技术与Ｌｙａｐｕｎｏｖ直接法相结合，极大地提高了学习样本的提取速度。系统算例表明，文中所提出的暂稳极限功率确定方法速度快、精度高，并具有较强的模型适应性和较好的工程实用背景。

In this paper, the artificial neural network (ANN) was used to establish the nonlinear mapping of the transient energy margin, generator power and the fault clearing time. Lyapunov’s direct method with the system dynamic equivalents was used in order to obtain the training set of ANN. The transient stability limits of the generator at different fault clearing time were estimated very fast by ANN after being trained. The proposed approach was tested on two power systems, and the results were found to be quite accurate.