摘要
分析了传统的利用神经网络进行非线性辨识的方法在系统有噪声数据时鲁棒性差且系统辨识结构不准确的问题 ,提出了一种基于模糊关系的系统辨识与建模方法 .该方法将模糊控制与神经网络有机结合 ,在优化系统辨识结构的同时有效降低了噪声数据对系统的干扰 .
The problem of learning algorithm of the neural networks for system identification is discussed. Given the problem that robust of the algorithm is bad and the construction of system identification is inaccurate, a new modeling method based on fuzzy relation is presented. The new method combines fuzzy control with neural networks, which not only diminishes noise perturbation but also improves the construction of system identification.
出处
《北方工业大学学报》
2003年第3期54-58,共5页
Journal of North China University of Technology
