Article Per Year (5 Year)
p-Index From 2019 - 2024
0.23
P-Index
This Author published in this journals
All Journal Rekayasa Mesin
Ryandhi Rofifu Chazim
Universitas Gadjah Mada
Mechanical Engineering

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
PERMODELAN INVERSI PEREDAM MAGNET-REOLOGI BERBASIS JARINGAN SARAF TIRUAN UNTUK SISTEM KENDALI Rafly Asprilla Alwi; Irfan Bahiuddin; Ryandhi Rofifu Chazim; Agustinus Winarno; Fitrian Imaduddin
Jurnal Rekayasa Mesin Vol. 13 No. 2 (2022)
Publisher : Jurusan Teknik Mesin, Fakultas Teknik, Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21776/jrm.v13i2.962

Abstract

The application of artificial neural network (ANN) models in magnet-rheological damper modeling is of great interest in recently challenges. Therefore, this study aims to propose a solution to overcome this problem by conducting inverse modeling using an artificial neural network. This inverse model is applied to a meandering magnet-rheological valve damper to predict the current to produce the appropriate damping force. The simulation scheme is selected with current as output and damping force, velocity, and displacement as input. The best model is formulated by varying the architecture of the artificial neural network. The best artificial neural network architecture is obtained after doing these variations. The data is divided into 80% training data, 10% validation data, and 10% test data. The activation function used is a logsig function using three hidden layers with the number of neurons in each layer [30-20-30]. The algorithm used in the chosen architecture is Levenberg-Marquardt. The regression value of 0.991 and the MSE value of 0.001 were obtained from the modeling results. The required damping force is ensured that it can be predicted well using the selected artificial neural network. The test proves that the results of the regression constant are 0.999 and the MSE value is 0.0005 when the current output value is inverted to the damper artificial neural network.
Co-Authors Agustinus Winarno Fitrian Imaduddin Irfan Bahiuddin Rafly Asprilla Alwi