Premium
A combined fuzzy logic and artificial neural network approach for non‐linear identification of IPMC actuators with hysteresis modification
Author(s) -
Zamyad Hojat,
Naghavi Nadia,
Barmaki Hasan
Publication year - 2018
Publication title -
expert systems
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.365
H-Index - 38
eISSN - 1468-0394
pISSN - 0266-4720
DOI - 10.1111/exsy.12283
Subject(s) - nonlinear autoregressive exogenous model , actuator , adaptive neuro fuzzy inference system , computer science , artificial muscle , control theory (sociology) , hysteresis , artificial neural network , displacement (psychology) , neuro fuzzy , fuzzy logic , nonlinear system , artificial intelligence , fuzzy control system , physics , psychology , control (management) , quantum mechanics , psychotherapist
Ionic polymer–metal composite (IPMC) actuators are one of the most prominent electroactive polymers with expected widespread use in the future. IPMC actuators exhibit hysteresis, which causes non‐linearity in bending behaviour of them. In this paper, a modified adaptive neuro‐fuzzy inference system and a non‐linear autoregressive with exogenous input (ANFIS–NARX) approach is presented for non‐linear identification of IPMC actuators. The proposed method utilizes a hysteresis operator, which increases the accuracy of the IPMC identification in combination with an ANFIS–NARX structure. The proposed model has a flexible structure to estimate the output (IPMC displacement as an actuator) for different training and testing data sets. Experimental results are provided to show the effectiveness of the accurate tracking capability of the proposed method to capture the real mechanical displacement features of the IPMC actuator.