RBF神经网络在飞艇单发停车故障中的应用  

Application of RBF Neural Networks in Airship Engine Flameout Failure

作　　者：张洪涛[1] 陈澜[1] 王志峰[1] 

机构地区：[1]西北工业大学自动化学院,西安710129

出　　处：《火力与指挥控制》2012年第12期80-82,86,共4页Fire Control & Command Control

基　　金：航空科学基金资助项目(20100753007)

摘　　要：针对飞艇单发停车故障,提出了一种基于径向基神经网络的模型跟随非线性重构控制策略,并使用遗传算法对RBF网络进行了优化。该方法通过响应误差构造神经网络的学习规则,神经网络的输出驱动系统向着消除误差的方向运动。理论分析和仿真验证表明,该控制策略具有良好的重构性能和鲁棒性。A new type of non-linear reconfigurable control strategy based on model-following method using Radical Basis Function（RBF） neural networks for the airship＇s engine flameout failure is presented in this paper.Also genetic algorithm is applied to optimize the RBF networks.The learning rules of the neural networks are fixed by the response errors in this method,and the errors of the control system are decreased because of the outputs of the NN.Theoretical analysis and simulation results reveal that this method has good reconfigurable performance and robustness.

关 键 词：遗传算法 模型跟随 径向基神经网络 单发停车 

分 类 号：V249[航空宇航科学与技术—飞行器设计]