Title

DETERMINATION OF THE PROCESS WINDOW OF AUSTEMPER TREATMENT TO OBTAIN ADI THROUGH NEURAL NETWORK SIMULATION

DETERMINATION OF THE PROCESS WINDOW OF AUSTEMPER TREATMENT TO OBTAIN ADI THROUGH NEURAL NETWORK SIMULATION

Authorship

Pereira, Leonardo; Belle, Matheus Roberto; Pasini, Willian Martins; Barcellos, Vinicius Karlinski de

Pereira, Leonardo

5 Articles, Downloaded 2023 Times

Belle, Matheus Roberto

I am this author

Pasini, Willian Martins

2 Articles, Downloaded 605 Times

Barcellos, Vinicius Karlinski de

4 Articles, Downloaded 1536 Times

DOI

10.5151/1516-392X-31701

Downloads

Abstract

The objective of this work is to determine the process window of austemper treatment to obtain an Austempered Ductile Iron (ADI). A ductile iron alloy with additions of manganese, copper, nickel and molybdenum was used. The austemper cycle was consisted in austenitizing at 900°C for 90 minutes and austemper at 380°C with different times. To determine the process window, it was used a group of programs which allows, through neural network method, estimate the amount of retained austenite in function of the alloy composition and the austemper treatment cycle parameters. The results of the simulation were compared with experimental results of hardness and toughness tests. According to the simulations, the highest retained austenite content is reached through austemper performed for 35 to 40 minutes. The experimental results show hardness stabilization on austemper times of 60, 90 and 120 minutes. The neural network simulation was considered a good tool, permitting the cost reduction upon the necessary tests to determine the process window on an ADI development.

The objective of this work is to determine the process window of austemper treatment to obtain an Austempered Ductile Iron (ADI). A ductile iron alloy with additions of manganese, copper, nickel and molybdenum was used. The austemper cycle was consisted in austenitizing at 900°C for 90 minutes and austemper at 380°C with different times. To determine the process window, it was used a group of programs which allows, through neural network method, estimate the amount of retained austenite in function of the alloy composition and the austemper treatment cycle parameters. The results of the simulation were compared with experimental results of hardness and toughness tests. According to the simulations, the highest retained austenite content is reached through austemper performed for 35 to 40 minutes. The experimental results show hardness stabilization on austemper times of 60, 90 and 120 minutes. The neural network simulation was considered a good tool, permitting the cost reduction upon the necessary tests to determine the process window on an ADI development.

Keywords

Austempering Ductile Iron; Process Windows; Neural Network; Retained Austenite

Austempering Ductile Iron; Process Windows; Neural Network; Retained Austenite