A Neuro-Fuzzy Inference System Through Integration of Fuzzy Logic and Extreme Learning Machines

Abstract

This paper investigates the feasibility of applying a relatively novel neural network technique, i.e., extreme learning machine (ELM), to realize a neuro-fuzzy Takagi-Sugeno-Kang (TSK) fuzzy inference system. The proposed method is an improved version of the regular neuro-fuzzy TSK fuzzy inference system. For the proposed method, first, the data that are processed are grouped by the k-means clustering method. The membership of arbitrary input for each fuzzy rule is then derived through an ELM, followed by a normalization method. At the same time, the consequent part of the fuzzy rules is obtained by multiple ELMs. At last, the approximate prediction value is determined by a weight computation scheme. For the ELM-based TSK fuzzy inference system, two extensions are also proposed to improve its accuracy. The proposed methods can avoid the curse of dimensionality that is encountered in backpropagation and hybrid adaptive neuro-fuzzy inference system (ANFIS) methods. Moreover, the proposed methods have a competitive performance in training time and accuracy compared to three ANFIS methods.

ELM-ANFIS Based Controller for Plug-In Electric Vehicle to Grid Integration

“…The ELM [19][20][21] is a new concept learning algorithm, which works on single hidden layer feed forward neural network (SLFN) that randomly chooses hidden nodes and the output weights of SLFN is determined analytically. The outputs are learned and updated in a single step and it really makes it as a linear network model.…”

“…Based on the concept of ELM [20,21], the new e-ANFIS learning algorithm is developed. The ELM concept is used in variety of applications such as signal classifications [22][23][24], electricity forecasting [25] etc.…”

Introduction to Pattern Recognition and Bioinformatics

“…Neuro-fuzzy approach is perhaps the most visible hybrid paradigm [51, [228][229][230]254, 261], realized so far, in soft computing framework. Besides the generic advantages, the neuro-fuzzy approach provides the corresponding application specific merits [76,111,120,330,368,388,394]. Rough-fuzzy [209,250,265] and neuro-rough [68,158,264] hybridizations are also proving to be fruitful frameworks for modeling human perceptions and providing means for computing with words.…”

A Neural Fuzzy System for Vibration Control in Flexible Structures

“…In fact, fuzzy reasoning in (1) is a TS1 paradigm [14]. Test results have shown that the TS1 NF control provides more smooth control effects than a TS0 paradigm because it contains more linear consequent parameters than in the TS0 controller.…”

“…Once the NF control paradigm is established, its parameters should be trained properly so as to achieve optimal control performance. In this case, in each training epoch, the nonlinear system parameters of the MFs are trained by the gradient method in the backward pass, whereas the linear consequent parameters are fine-tuned by the LSE in the forward pass [14].…”