Abstract
Fused deposition modelling (FDM) is a versatile additive manufacturing technique involving extruding thermoplastic polymers layer by layer. FDM enables the fabrication of complex geometries, however prediction and minimization of warpage deformation is critical for high-precision applications such as medical implants. Simulations based on finite element analysis (FEA) have been shown to accurately predict warpage deformation in FDM; however, this approach is computationally intensive and has limitations for process optimisation and monitoring. In this study, a feed-forward neural network (FFNN) model was trained using deformation values from FEA simulations. Input features included nozzle temperature, material deposition rate, bed temperature, layer height, layer number, and spatial coordinates, while warpage deformation served as the output. The trained FFNN was then combined with the particle swarm optimization (PSO) algorithm to find the optimal FDM parameters to minimize the warpage deformation. The combination of a trained FFNN based on FEA with PSO identified optimal FDM parameters, producing a minimum warpage deformation of 0.062 mm. The proposed PSO-FFNN approach demonstrates the feasibility of using machine learning and metaheuristic algorithms to minimize warpage defects in additive manufacturing while reducing computational cost and computation time.
| Original language | English |
|---|---|
| Title of host publication | 12th 2026 International Conference on Control, Decision and Information Technologies, CoDIT 2026 |
| Publisher | Institute of Electrical and Electronics Engineers Inc. |
| Pages | 622-627 |
| Number of pages | 6 |
| ISBN (Electronic) | 9798319520777 |
| DOIs | |
| Publication status | Published - 2026 |
| Event | 12th International Conference on Control, Decision and Information Technologies, CoDIT 2026 - Bari, Italy Duration: 13 Jul 2026 → 16 Jul 2026 |
Publication series
| Name | 12th 2026 International Conference on Control, Decision and Information Technologies, CoDIT 2026 |
|---|
Conference
| Conference | 12th International Conference on Control, Decision and Information Technologies, CoDIT 2026 |
|---|---|
| Country/Territory | Italy |
| City | Bari |
| Period | 13/07/26 → 16/07/26 |
UN SDGs
This output contributes to the following UN Sustainable Development Goals (SDGs)
-
SDG 2 Zero Hunger
Keywords
- Additive manufacturing
- Feed forward neural network
- Fused deposition modeling
- Particle swarm optimization algorithm
- Warpage deformation
Fingerprint
Dive into the research topics of 'Warpage Deformation Prediction and Optimization in Fused Deposition Modeling Using a Hybrid FFNN and PSO Approach'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver