径向基函数递推最小二乘算法优化学习的研究  

Research on Optimization Learning of Radial Basis Function Recursive Least Square Algorithm

作　　者：葛淑杰[1] 姜天文[1] 付岩[2] 

机构地区：[1]黑龙江科技学院计算机与信息工程系,黑龙江哈尔滨150027 [2]哈尔滨工程大学计算机科学与技术学院,黑龙江哈尔滨150001

出　　处：《计算机仿真》2007年第5期150-153,共4页Computer Simulation

摘　　要：对于广泛使用的三层感知机BP神经网络存在收敛速度慢,预测精度不高的问题,提出了基于径向基函数(RBF)递推最小二乘算法调整网络连接权值以及通过自适应学习的方法优化径向基函数形状参数的协作式自适应算法,并采用该算法进行了RBF神经网络的训练和仿真实验。结果表明:所提出的算法较BP算法以及固定α值的RBF算法有较快的收敛速度;最后,将训练后的神经网络应用于煤矿瓦斯涌出量的预测中,结果大大提高了预测的精度。因此,该算法具有很高的应用价值。In this paper, Due to slow restrain rate and lower precision accuracy in the BP neural network of three layers perception machine, A cooperative adaptive algorithm based on the Radial Basis Function （RBF） reeursive least square algorithm and adaptive learning for optimizing shape parameter of RBF is proposed, by which training of RBF neural network and simulation experiments are done. The results show that the algorithm described in this paper have more fast restrain rate of network than BP algorithm and RBF algorithm of fixed αtvalue . in the end, We apply trained neural network to estimate mine gas gushing , and precision accuracy is enhanced rapidly. So the algorithm proposed in this paper has high of application value.

关 键 词：径向基函数 递推最小二乘算法 形状参数 自适应学习 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]