Premium
Nonlinear identification of a spark ignition engine torque based on ANFIS with NARX method
Author(s) -
Togun Necla,
Baysec Sedat
Publication year - 2016
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.12172
Subject(s) - nonlinear system , nonlinear autoregressive exogenous model , computer science , spark (programming language) , control theory (sociology) , adaptive neuro fuzzy inference system , ignition system , identification (biology) , system identification , autoregressive model , nonlinear system identification , torque , spark ignition engine , inference system , fuzzy logic , control engineering , artificial intelligence , fuzzy control system , artificial neural network , data modeling , engineering , mathematics , physics , botany , database , econometrics , aerospace engineering , biology , control (management) , quantum mechanics , thermodynamics , programming language
Spark ignition (SI) engines have a nonlinear dynamic system with inherent uncertainties and unpredictable disturbances. The identification of a nonlinear system is vital in many fields of engineering. In this study, SI engine torque is identified from an input–output measurement. This study aims to propose a dynamic nonlinear model that uses an adaptive neuro‐fuzzy inference system and a nonlinear auto‐regressive with exogenous input structure to identify the dynamic nonlinear behavior of an SI engine. Considerable good performance is achieved using the adaptive neuro‐fuzzy inference system nonlinear auto‐regressive with exogenous input method. For model validation, the proposed method is compared with the more conventional identification approach called the Hammerstein method. The results show that the two methods are in excellent agreement. The Hammerstein model was chosen because its identification result of the SI system was studied previously by the author. Validation results prove that the ability of the proposed model can capture the highly nonlinear behavior of the SI system.