Optimization of Process Parameters for Advanced High-Strength Steel JSC980Y Automotive Part Using Finite Element Simulation and Deep Neural Network

Guardado en:
Bibliografiske detaljer
Udgivet i:Journal of Manufacturing and Materials Processing vol. 9, no. 6 (2025), p. 197-219
Hovedforfatter: Sunanta Aekkapon
Andre forfattere: Suranuntchai Surasak
Udgivet:
MDPI AG
Fag:
Online adgang:Citation/Abstract
Full Text + Graphics
Full Text - PDF
Tags: Tilføj Tag
Ingen Tags, Vær først til at tagge denne postø!

MARC

LEADER 00000nab a2200000uu 4500
001 3223913865
003 UK-CbPIL
022 |a 2504-4494 
024 7 |a 10.3390/jmmp9060197  |2 doi 
035 |a 3223913865 
045 2 |b d20250101  |b d20251231 
100 1 |a Sunanta Aekkapon 
245 1 |a Optimization of Process Parameters for Advanced High-Strength Steel JSC980Y Automotive Part Using Finite Element Simulation and Deep Neural Network 
260 |b MDPI AG  |c 2025 
513 |a Journal Article 
520 3 |a In the stamping process of automotive parts, springback is a major problem when using Advanced High-Strength Steel (AHSS). This phenomenon significantly impacts the shape accuracy of products and is difficult to control. This study aims to optimize process parameters such as blank holder force (BHF), die clearance, and blank width to minimize springback in the workpiece. Using optimal process parameters will enhance the efficiency of die compensation processes. The study uses the Finite Element Method (FEM) simulation to predict forming behavior. The case study, Reinforcement-CTR PLR, is made from AHSS grade JSC980Y with a thickness of 1 mm. Four material model combinations were evaluated against actual experiment results to select the most accurate springback prediction model. A full factorial design was used for experiments with varied process parameters. The optimization process used regression and various Artificial Neural Networks (ANNs). From the result, a Deep Neural Network (DNN) with two hidden layers performed with the highest accuracy compared to the other models. The optimal process parameters were identified as 27.62 tons BHF, 1 mm die clearance, and a 290 mm blank width. These optimal results achieved 98.05% of the part area within a displacement tolerance of −1 to 1 mm, closely matching FEM-based validation. 
653 |a Mechanical properties 
653 |a Clearances 
653 |a Finite element method 
653 |a Accuracy 
653 |a Design of experiments 
653 |a Workpieces 
653 |a Automotive parts 
653 |a Metal forming 
653 |a Optimization techniques 
653 |a Artificial neural networks 
653 |a High strength steel 
653 |a Manufacturing 
653 |a High strength steels 
653 |a Steel 
653 |a Computer simulation 
653 |a Springback 
653 |a Simulation 
653 |a Parameter identification 
653 |a Artificial intelligence 
653 |a Prediction models 
653 |a Neural networks 
653 |a Optimization 
653 |a Factorial design 
653 |a Blankholders 
653 |a Mathematical models 
653 |a Algorithms 
653 |a Popularity 
653 |a Process parameters 
653 |a Dies 
700 1 |a Suranuntchai Surasak 
773 0 |t Journal of Manufacturing and Materials Processing  |g vol. 9, no. 6 (2025), p. 197-219 
786 0 |d ProQuest  |t ABI/INFORM Global 
856 4 1 |3 Citation/Abstract  |u https://www.proquest.com/docview/3223913865/abstract/embedded/7BTGNMKEMPT1V9Z2?source=fedsrch 
856 4 0 |3 Full Text + Graphics  |u https://www.proquest.com/docview/3223913865/fulltextwithgraphics/embedded/7BTGNMKEMPT1V9Z2?source=fedsrch 
856 4 0 |3 Full Text - PDF  |u https://www.proquest.com/docview/3223913865/fulltextPDF/embedded/7BTGNMKEMPT1V9Z2?source=fedsrch