摘要
针对传统BP神经网络初始权值和阈值随机产生、易陷于局部最优化、收敛速度慢以及隐含层的神经元数量不易确定等问题,采用遗传算法对BP神经网络的初始权值和阈值空间进行遗传优化,获取最优权值矩阵和阈值矩阵,并由此进行误差反向前馈神经网络的训练学习,同时采用试错法,结合相关公式,缩小隐含层神经元数量范围,寻找最优神经元数量,建立GA-BP神经网络模型,对输电线路舞动的发生进行预警。通过对相关地区输电线路舞动历史数据进行了算例分析,对比其他机器学习算法的预测结果准确性,结果表明:改进的GA-BP神经网络能更准确有效地预测输电线路舞动的发生情况;为防止大规模舞动灾害提供有力了保障,进一步提高了电网抵御自然灾害的能力。
The traditional back-propagation neural network(BP)is prone to random generation of initial weights and thresholds,easy to fall into the local optimization,with slow convergence rate and hard to confirm the number of neurons in the hidden layer.In this paper,the Genetic Algorithm(GA)is utilized to optimize the initial weights and thresholds space of the BP neural network.To obtain the optimal weight matrix and threshold matrix,the error-forward-feedback neural network training is carried out by using the data of transmission line galloping.An optimized GA-BP neural network model is established.The historical data of the transmission lines galloping in the related areas is analyzed by the optimized GABP neural network model.The validity and practicability of the proposed GA-BP neural network model is tested and verified.The simulation results show that the GA-BP neural network module could predict the galloping situation of transmission lines more accurately and effectively.As a result,it provides a strong guarantee for preventing large-scale grid fault disasters,and further improves the power grid’s ability to withstand natural disasters.
作者
汉京善
吕海平
李丹煜
李征
李蛟
邓元靖
HAN Jingshan;Lü Haiping;LI Danyu;LI Zheng;LI Jiao;DENG Yuanjing(China Electric Power Research Institute,Beijing 100192,China;State Grid Corporation of China,Beijing 100031,China;Northwest Branch of the SGCC,Xi’an 710048,Shaanxi,China)
出处
《电网与清洁能源》
北大核心
2021年第4期1-7,14,共8页
Power System and Clean Energy
基金
国家重点研发计划资助项目(2018YFC0809400)
国家电网有限公司科技项目(GCB11202002264)。
关键词
遗传算法
BP神经网络
机器学习
隐含层
导线舞动
genetic algorithm
BP neural network
machine learning method
hidden layer
transmission line galloping
