Multi-parameter optimization of cylinder upsetting using FE and ANN
Source abstract
Abstract
Abstract Cold cylinder upsetting of 16-MnCr5 low-alloy steel is a critical bulk metal forming operation in which energy efficiency and workpiece mechanical integrity are strongly governed by process parameter selection. This study presents a systematic finite element (FE) simulation-based investigation of the combined effects of friction coefficient ( μ = 0.05–0.20), upper die velocity ( v = 15–30 mm s −1 ), and workpiece initial temperature ( T 0 = 25 °C–100°C) _ all within the cold forging regime ( T / T m = 0.172–0.215) _ on four key process outputs: total energy consumption, yield stress, final workpiece temperature, and effective maximum plastic strain rate. A full-factorial design of experiments comprising 64 SIMUFACT Forming FE simulations was employed to generate a comprehensive process response dataset. A feedforward artificial neural network (ANN) surrogate model, implemented in MATLAB and trained on 80% of the dataset ( R 2 ≈ 0.99), was developed to capture the complex nonlinear relationships between process parameters and outputs, offering a computationally efficient alternative to repeated FE evaluation. Results demonstrate that friction coefficient is the dominant parameter governing energy consumption and yield stress, while initial temperature exerts the greatest influence on final workpiece temperature. The optimal process conditions _ μ = 0.05, v = 15 mm s −1 , T 0 = 100 °C_ are identified as achieving minimum energy consumption within the investigated design space. The proposed FE-ANN framework provides a practical and accurate predictive tool for cold forging process design and optimization.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Metallurgy and Material Forming · Metal Forming Simulation Techniques · Aluminum Alloy Microstructure Properties
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Jordan University of Science and Technology
The Hashemite University
References (17)
Show source-provided references
1. Finite element analysis of the axisymmetric upsetting process using the deformation theory of plasticity10.1115/1.2901789 ↗
2. Upsetting of a cylindrical specimen between elastic tools10.1016/S0924-0136(98)00236-2 ↗
3. An insight into ultrasonic vibration assisted conventional manufacturing processes: a comprehensive review10.1177/16878132221107812 ↗
4. Effect of design and process parameter to cold forging die design: a finite element analysis10.3923/jas.2007.777.784 ↗
5. Development of a CAD/CAM system for simulating closed forging process using finite-element method10.1108/02644400910943635 ↗
6. Experimental research and numerical simulation of the dynamic cylinder upsetting10.1016/j.msea.2007.11.128 ↗
7. Artificial intelligence in metal forming10.1016/j.cirp.2024.04.102 ↗
8. Investigation of new method of large ingots forging based on upsetting of workpieces with ledges10.1007/s00170-022-09989-1 ↗
9. Selection of forging process for compressor disc for aero engine using finite element analysis10.1016/j.matpr.2022.02.035 ↗
10. Assessment of process, parameters, residual stress mitigation, post treatments and finite element analysis simulations of wire arc additive manufacturing technique10.1007/s12540-021-01015-5 ↗
11. Integration of artificial neural network with finite element analysis for residual stress prediction of direct metal deposition process10.1016/j.mtcomm.2021.102197 ↗
12. Enhancement of upsetting technology for minimizing internal defects in heavy ingot casting: simulation, optimization, and experimental validation10.1007/s12289-025-01883-x ↗
13. Integrated finite element method, artificial intelligence, and digital twin technologies for drawing die design: a comprehensive review of advanced simulation and production efficiency optimization10.12913/22998624/213621 ↗
14. Simulation of upsetting process using kinematic schemes of metal flow10.1088/1757-899X/971/2/022041 ↗
15. Integrated FEM and neural network-based optimization of the drawing process for variable cross-section forgings10.46299/j.isjea.20250403.09 ↗
16. Simulation and optimization of an aluminum extrusion process using FEM and artificial intelligence10.15866/iremos.v16i3.22527 ↗
17. Estimating the state of health and remaining useful life of lithium-ion batteries using artificial neural networks and support vector machines10.15866/iremos.v18i5.26166 ↗