Research on the Application of Echo State Network in Data Prediction

Abstract

Echo state network is often used in the analysis of industrial data. The main drawback of this approach is that the prediction results are often locally optimal. In order to further improve the prediction accuracy of the echo state network, this paper designs a global optimization strategy based on the three level network model. In the optimization process, the core parameters and optimization methods of the echo state network are improved to achieve the goal of the global optimum. The improvement measures proposed in this paper can reduce the prediction error of echo state network in the practical application of data prediction.