System identification using a coefficient learning mechanism via a hopfield neural network

Abstract

本論文,針對非線性離散時間系統利用霍普菲爾類神經網路(HNN)作為係數學習的機制,在高斯基底函數的集合中藉由此學習機制獲得其最佳的係數,並且在取樣步幅(Sampled step size)趨進近於零的情況下利用學習的模型可以完全近似離散化的霍普菲爾學習模型。本篇論文的主要貢獻在於推導出離散型霍普菲爾學習模型的收歛性條件，最後以模擬的結果來說明本論文所提出之方法的有效性。

In this paper, an identification method is proposed for discrete-time nonlinear systems using a Hopfield neural network (HNN) as a coefficient learning mechanism to obtain optimized coefficients over a set of Gaussian basis functions. The outputs of the HNN, which are coefficients over a set of Gaussian basis functions, are discretized to be a discrete Hopfield learning model and completely approximated by the learning model if the sampled step size approaches zero. The main contribution of this paper is that the convergence condition of the discrete Hopfield learning model is derived. Finally, to demonstrate the effectiveness of the proposed methods, simulation results are illustrated in this paper.