基于贝叶斯正则化BP神经网络的电力系统暂态稳定定量评估研究  

Quantitative Evaluation of Transient Stability in Power System Based on Bayesian Regularized BP Neural Network

作　　者：秦国鑫 于锡平 张师[2] QIN Guo-xin;YU Xi-ping;ZHANG Shi(State Grid Jilin Electric Power Co.Ltd.,Changchun 130000,China;School of Electrical Engineering,Northeast Electric Power University,Jilin 132012,China)

机构地区：[1]国网吉林省电力有限公司超高压公司,吉林长春130000 [2]东北电力大学电气工程学院,吉林吉林132012

出　　处：《电气开关》2025年第1期55-58,62,共5页Electric Switchgear

摘　　要：电力系统的暂态稳定性与扰动大小、扰动位置、扰动时间、机组抗扰能力、运行工况等因素息息相关,是一项复杂且值得深入研究的工作。研究了BP神经网络的原理,并基于贝叶斯正则化BP神经网络建立了电力系统暂态稳定评估的模型,最后进行算例分析。通过分析可知,采用该方法可以有效评估电力系统暂态稳定性,极限切除时间计算结果略大于实际结果。The transient stability of power system is closely related to the disturbance size,disturbance position,disturbance time,disturbance resistance of unit and operating conditions,which is a complex work worthy of further study.In this paper,we study the principle of BP neural network,and establish a model for transient stability evaluation of power system based on Bayesian regularized BP neural network,and finally make example analysis.The analysis shows that the transient stability of the power system can be effectively evaluated,and the limit resection time calculation result is slightly larger than the actual result.

关 键 词：BP神经网络 暂态稳定 极限切除时间 输入量 

分 类 号：TM71[电气工程—电力系统及自动化]