论文部分内容阅读
在相空间重构理论的基础上,将改进的遗传算法和神经网络结合起来,提出了一种混合遗传神经网络预测混沌时间序列的方法。通过复相关法和Cao方法重构混沌时间序列,利用改进的遗传算法优化神经网络的结构、初始权值和阈值,然后训练神经网络求得最优解。该算法应用到混沌时间序列的预测中,验证了该算法的有效性,并与BP和RBF算法的预测精度进行了比较,仿真结果表明该算法对混沌时间序列具有更好的非线性拟合能力和更高的预测精度。
Based on the phase space reconstruction theory, an improved genetic algorithm and neural network are combined to propose a hybrid genetic algorithm to predict the chaotic time series. The complex correlation method and the Cao method are used to reconstruct the chaotic time series. The improved genetic algorithm is used to optimize the structure, initial weight and threshold of the neural network, and then the neural network is trained to obtain the optimal solution. The algorithm is applied to the prediction of chaotic time series to verify the effectiveness of the proposed algorithm and to compare with the prediction accuracy of BP and RBF algorithms. The simulation results show that the proposed algorithm has better nonlinear fitting ability for chaotic time series And higher prediction accuracy.