Expanded Neo-Fuzzy Adaptive Decayed Brain Emotional Learning Network for Online Time Series Predication

Houssen Salh Ali Milad,Jason Gu

doi:10.1109/access.2021.3076668

Abstract

The Neo-Fuzzy integrated Adaptive Decayed Brain Emotional Learning (NF-ADBEL) network has recently been proposed for online time series predicting problems. The NF-ADBEL network is suitable for online time series prediction with shorter update intervals and offers features such as fast learning, accuracy, simplicity, and lower computational complexity. However, the neo-fuzzy neuron network in NF-ADBEL was integrated only in the orbitofrontal cortex (OFC) part of the ADBEL network. This paper aims to further improve the performance of the NF-ADBEL network by integrating the neo-fuzzy neuron network into the amygdala (AMY) section as well, inspired by a fully integrated version of a neo-fuzzy-based pattern recognizer. As is known, the AMY has two outputs: one response is based on imprecise information received from the thalamus, and the second response is based on information received from the sensory cortex. In this study, the imprecise response generation is operated as previously, while the other AMY process is treated by neo-fuzzy neurons. The resultant network is called Expanded Neo-Fuzzy integrated Adaptive Decayed Brain Emotional Learning (ENF-ADBEL). The modified network is still simple and meets the requirement for online prediction problems. A few chaotic and stochastic nonlinear systems, namely the Mackey-Glass, Lorenz, Rossler, disturbance storm time index, Narendra dynamic plant identification, wind speed and wind power series, are used to evaluate the performance of the proposed network in terms of the root mean squared error (RMSE) and correlation coefficient (COR) criteria in a MATLAB programming environment.

Highlights

A number of different techniques have been applied to the chaotic times series prediction problem, with varying degrees of success
The proposed ENF-ADBEL network is tested in a MATLAB (R2014a) programming environment for online forecasting of chaotic time-series, including Mackey-Glass, Lorenz, Rossler, Narendra plant, disturbance storm time index, wind speed, and wind power
A comparison is made with NF-ADBEL and F-ADBEL and multilayer perceptron (MLP) networks driven by the near-optimal set of alpha, beta and gamma parameters

Summary

Introduction

A number of different techniques have been applied to the chaotic times series prediction problem, with varying degrees of success. The Artificial Neural Network (ANN) is probably the technique most frequently used. There is an increase in time computational complexity with this approach because there is no optimal structure for the number of neurons and number of layers, or for the activation function suitable for the. These drawbacks affect the reliability and accuracy of the prediction. The Brain Emotional Learning Neural Network (BELNN) has recently emerged as an alternative to classical artificial neural networks for approximating nonlinear functions. BELNN is inspired by both feed-forward neural networks and fast learning and has been applied to time series prediction techniques [1]–[3]

Objectives

Results

Conclusion