基于BP神经网络和EKF神经网络在曲线拟合性能上的对比分析研究

Comparison Analysis of the Performance on the Curve Fitting of BP Neural Network and EKF Neural Network

本文首先介绍了曲线拟合的相关含义,接着讨论了BP神经网络和EKF神经网络的原理,研究对比分析了BP神经网络和EKF神经网络在曲线拟合性能,通过Matlab仿真实验表明,EKF算法能更好的对系统进行辨识,具有较高的收敛性。

This paper firstly describes the underlying meaning of curve fitting, then discuss the principle of BP neural network and EKF neural network. The paper studies the curve fitting performance of BP neural network and EKF neural network, through Matlab simulation results, the EKF algorithm can better identifies the system and has a higher convergence.