摘要
本文介绍了动态对角递归网络,并针对BP算法收敛慢的缺点,将递推预报误差学习算法应用到神经网络权值和域值的训练。同时,将动态对角递归网络引入到电力变压器的故障诊断中,通过改良三比值方法来实现故障诊断,提高了神经网络故障诊断收敛性及增量学习能力。最后,仿真诊断实例表明了应用这种方法实现的变压器故障诊断系统的有效性。
A simple recurrent neural network named as diagonal recurrent neural network is studied in this paper. To overcome the slow convergence of the BP algorithm, recursive prediction error algorithm is proposed, which can train both the weight and the bias. In addition, the recurrent neural network trained with RPE algorithm is used in fault diagnosis of power transformer based on improved three ratio method . It can improve the astringency and increment learning ability. At last, the simulation diagnosis demonstrates the effectiveness of the proposed algorithm.
出处
《微计算机信息》
北大核心
2007年第06S期214-215,共2页
Control & Automation
基金
吉林省科技厅项目(200505263)
关键词
动态对角递归网络
递推预报误差
故障诊断
气相色谱分析法
diagonal recurrent neural network
recursive prediction error algorithm
fault diagnosis
dissolved gas analysis
