RT Journal Article
SR Electronic
A1 Banerjee, Prabas
A1 Karmakar, Arnab
A1 Hui, Nirmal
T1 Application of Adaptive Neuro-Fuzzy Inference System for Prediction of Mean Thickness in Backward Metal Flow Forming Process
JF Acta Mechanica Slovaca
YR 2021
VO 25
IS 2
SP 46
OP 55
DO 10.21496/ams.2021.018
UL https://www.actamechanica.sk/artkey/ams-202102-0005.php
AB In the backward metal flow forming process, the mean thickness of flow formed tubes is crucial in determining the quality. Consequently, predicting the mean thickness in incremental forming and correlating these values with the forming parameters can be useful to control this vital target. Accordingly, four different techniques of adaptive neuro-fuzzy inference system (ANFIS) is used to predict the mean thickness of parts produced by the backward metal flow forming process. The objective is to determine the best membership function of the different approaches used in ANFIS. Also, the efficiency of the developed predictive models is compared statistically to determine the best technique.