基于AQPSO算法优化的RBF网络模型及应用研究

提出了自适应量子粒子群优化(adaptive quantum-behaved particle swarm optimization,AQPSO)算法,用于训练RBF(radial basis function)网络的基函数中心和宽度,并结合最小二乘法计算网络权值,改进了RBF网络的泛化能力.利用上证指数数据进行预测,实验结果表明,采用AQPSO算法获得的RBF网络模型不但具有很强的泛化能力,而且具有良好的稳定性,在股票数据预测中具有一定的实用价值.

A STUDY ON RBF NEURAL NETWORK MODEL WITH APPLICATION BASED ON AQPSO OPTIMIZATION ALGORITHM

Adaptive quantum-behaved particle swarm optimization(AQPSO) algorithm is proposed in order to improve the performance of RBF(radial basis function) network.By applying AQPSO algorithm to train the central position and width of the basis function adopted in the RBF network,and computing the weights of the network with least-square method,the generalization ability of the RBF neural network is improved.Experimental results with Shanghai stock index data sets show that obtained network model not only has good generalization properties,but also has better stability.It illustrates that RBF net with AQPSO optimization algorithm has the promising application in stock data forecasting.