论文部分内容阅读
ＦＫＣＮ（Ｆｕｚｚｙ Ｋｏｈｏｎｅｎ ｃｌｕｓｔｅｒ ｎｅｔｗ ｏｒｋ）将模糊隶属度的概念用于Ｋｏｈｏｎｅｎ 神经网络的学习和更新策略中，改善了Ｋｏｈｏｎｅｎ 网络的性能，是一种更为快速有效的聚类网络。作者将ＦＫＣＮ用于优化ＲＢＦ（Ｒａｄｉａｌｂａｓｉｃ ｆｕｎｃｔｉｏｎ）神经网络基函数的中心，并将优化后的ＲＢＦ网络用于曲线拟合和非线性时间序列预测，同时与基于Ｃ－ＭＥＡＮＳ的ＲＢＦ网络进行比较。实验结果表明：采用ＦＫＣＮ优化的ＲＢＦ网络具有更好的拟合和预测能力，尤其在曲线拟合实验中，ＦＫＣＮ优化的ＲＢＦ网络可以达到最小学习误差，比Ｃ－ＭＥＡＮＳ的网络小一个数量级，可见用ＦＫＣＮ优化ＲＢＦ神经网络可以较好地提高ＲＢＦ神经网络的性能。
FKCN (Fuzzy Kohonen cluster netw ork) uses the concept of fuzzy membership degree for Kohonen neural network learning and updating strategy to improve the performance of Kohonen network. It is a more efficient and efficient clustering network. The author uses FKCN to optimize the center of Radial Basic function (RBF) neural network basis function, and uses the optimized RBF network for curve fitting and nonlinear time series prediction, at the same time compares with C-MEANS based RBF network. The experimental results show that the FKCN-optimized RBF network has better fitting and predictive ability. Especially in the curve fitting experiment, the FKCN-optimized RBF network can achieve the minimum learning error, which is an order of magnitude smaller than that of the C-MEANS network. Optimizing RBF neural network with FKCN can improve the performance of RBF neural network.