摘要
利用B样条小波神经网络对离散非线性系统进行辨识研究和对PWM系统进行动态建模，比用BP算法有更好的收敛性和稳定性。PWM系统用神经网络建模，弥补了原线性理论分析的不足。
This paper is concerned with the study of the identification for discrete time non-linear System by means of feedforward B-spline wavelet neural networks. The dynamic modelling method of PWM System in this paper is more effective than BP algorithm. This PWM modelling is a compensation for the preceding linear analysis theory.
出处
《上海工程技术大学学报》
CAS
1997年第1期18-23,共6页
Journal of Shanghai University of Engineering Science
关键词
调速系统
非线性系统
神经网络
PWM系统
建模
Speed control systems Non-linear systems Spline functions Neural networks
