摘要
针对目前Chebyshev神经网络所存在的不足 ,提出一种改进的Chebyshev神经网络 ,它使用多输入多输出神经网络结构与使用改进的Chebyshev正交多项式。因此改进的神经网络不仅扩大了网络辨识模型的能力与学习适应性 ,而且算法简单 ,学习收敛速度快 ,有线性、非线性逼近精度高等优异特性。文中给出两个应用实例 。
A kind of improved Chebyshev neural network was presented according to existing insufficiency of Chebyshev neural network. The neural network presented in this paper has used the multi-input and multi-output network structure, and the improved Chebyshev orthogonal polynomial. Therefore, it not only expand the identification ability and learning adaptation of the neural network, but also has a simple algorithm, a high speed convergence of learning process, and excellent characteristics in the linear and nonlinear accurate approximation. Two application examples were given in this paper. The simulation results showed the efficiency of the improved neural network. The high speed learning algorithm and convergent speed for the neural network gained an advantage over BP network.
出处
《机床与液压》
北大核心
2003年第3期63-64,共2页
Machine Tool & Hydraulics
基金
浙江省自然科学基金资助项目 ( 5 0 0 0 3 0 )