基于混沌线性回归与Elman神经网络的组合预测

Combined Prediction Based on Chaotic Linear Regression and Elman Neural Network

摘要: 针对单一模型对混沌时间序列预测能力不足的问题，本文将混沌线性回归模型和Elman神经网络模型相结合，设计一种组合预测模型，并将其应用于Lorenz混沌时间序列。首先建立混沌线性回归模型，然后建立Elman神经网络模型，最后建立基于加权算术平均算子的混沌线性回归模型和Elman神经网络的组合预测模型。通过Lorenz混沌时间序列检验了组合模型在实际预测中的有效性。仿真实验表明，组合预测模型在预测精度方面明显优于这两个单项预测模型，验证了本文提出的组合模型具有良好的预测效能。

Abstract: Aiming at the problem that the single model has insufficient ability to predict chaotic time series, this paper combines a chaotic linear regression model and an Elman neural network model to de-sign a combined prediction model and applies it to Lorenz chaotic time series. First, a chaotic linear regression model is established, then an Elman neural network model is established, and finally, a combined prediction model of a chaotic linear regression model based on a weighted arithmetic average operator and an Elman neural network is established. The validity of the combined model in actual prediction is tested by Lorenz chaotic time series. Simulation experiments show that the combined prediction model is significantly better than these two single-item prediction models in terms of prediction accuracy, which verifies that the proposed combination model has good predic-tion performance.