论文部分内容阅读
探讨用神经网络的学习算法及模糊推理方法为非线性系统建模的问题．给出了学习模糊规则的新算法．这个算法首先用竞争学习为训练样本的输入空间进行聚类，然后为其确定区域划分边界，并按样本输入区域学习模糊规则．文中对于模糊规则提出了相应的模糊推理算法．并用算例验证了本文算法的有效性．
The problem of modeling nonlinear system with neural network learning algorithm and fuzzy inference method is discussed. A new algorithm for learning fuzzy rules is given. The algorithm first uses competitive learning to cluster the input space of training samples, and then determines the boundary for the region division and learns the fuzzy rules according to the sample input region. In this paper, the corresponding fuzzy reasoning algorithm is proposed for fuzzy rules. An example is given to verify the effectiveness of the proposed algorithm.