基于RBF神经网络的无通道保护  

Non-communication protection based on RBF neural network

作　　者：陈少华[1] 余耀权[1] 郑帅[1] 叶杰宏[1] 

机构地区：[1]广东工业大学自动化学院,广东广州510090

出　　处：《福建工程学院学报》2004年第2期134-137,148,共5页Journal of Fujian University of Technology

基　　金：广东省教育厅自然科学基金(Z02033)

摘　　要：在故障分析的基础上,阐述了应用径向基函数神经网络(RBF神经网络)实现输电线全线路无通道快速保护的原理,并对各种故障类型的电磁暂态分析程序(EMTP)和matlab进行仿真测试,以证明该保护的可行性。On the basis of fault analysis, this paper describes the principle of non-communication fast protection of a whole transmission line with artificial neural network, the protection consisting of two parts: instantaneous protection within the protective range of 85% and successive tripping protection beyond the protective range of 85%. The network, a three-layered RBF model, consists of four subsidiary networks: ANN1 to determine fault directions, ANN2 to identify fault types and fault phase, ANN3 to trip breaker instantaneously and ANN4 to detect tripping situation of an adjacent breaker. Simulation results of all types of faults demonstrate that the model is feasible.

关 键 词：电流保护 无通道保护 RBF神经网络 相继动作 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]