基于自适应小波正交基神经网络的参数辨识  

Application of Adaptive Wavelet Orthogonal Basis Neural Network for parameter Identification

作　　者：李臣明[1] 韩子鹏[1] 孔建国[2] 

机构地区：[1]南京理工大学,江苏南京210094 [2]炮兵学院南京分院,江苏南京211132

出　　处：《火力与指挥控制》2006年第6期56-59,共4页Fire Control & Command Control

摘　　要：提出了一种基于小波正交基神经网络的非线性在线辨识方法。小波变换具有良好的收敛速度和逼近精度,神经网络具有强大的非线性映射能力、自学习、自适应等优势,采用正交基小波函数作为神经网络的基函数构成小波神经网络,该网络兼有小波函数和神经网络的优点;制导航弹数字仿真结果表明,该方法对含噪声数据进行处理效果好,能很好地满足非线性在线辨识的需求。This paper presents an online identification method for guided bomb based adaptive wavelet orthogonal basis neural network. Wavelet transformation has high convergence speed and high approximation accuracy. Also, the neural network has good nonlinear mapping, self-study, adaptive ability. The wavelet neural network was constructed by using the orthogonal basis wavelet function as the basis function of the neural network, it consists of the excellence of wavelet and neural network. The simulation result of the homing bomb shows that the method possesses the advantages of wiping off the yawp which can satisfy the demand of nonlinear online identification.

关 键 词：正交基小波 小波神经网络 参数辨识 

分 类 号：V249.122.2[航空宇航科学与技术—飞行器设计]