Interactive 3D visualization of surface roughness predictions in additive manufacturing: A data-driven framework
arXiv:2603.09353v1 Announce Type: new Abstract: Surface roughness in Material Extrusion Additive Manufacturing varies across a part and is difficult to anticipate during process planning because it depends on both printing parameters and local surface inclination, which governs the staircase effect. A data-driven framework is presented to predict the arithmetic mean roughness (Ra) prior to fabrication using process parameters and surface angle. A structured experimental dataset was created using a three-level Box-Behnken design: 87 specimens were printed, each with multiple planar faces spanning different inclination angles, yielding 1566 Ra measurements acquired with a contact profilometer. A multilayer perceptron regressor was trained to capture nonlinear relationships between manufacturing conditions, inclination, and Ra. To mitigate limited experimental data, a conditional generative adversarial network was used to generate additional condition-specific tabular samples, thereby im
arXiv:2603.09353v1 Announce Type: new Abstract: Surface roughness in Material Extrusion Additive Manufacturing varies across a part and is difficult to anticipate during process planning because it depends on both printing parameters and local surface inclination, which governs the staircase effect. A data-driven framework is presented to predict the arithmetic mean roughness (Ra) prior to fabrication using process parameters and surface angle. A structured experimental dataset was created using a three-level Box-Behnken design: 87 specimens were printed, each with multiple planar faces spanning different inclination angles, yielding 1566 Ra measurements acquired with a contact profilometer. A multilayer perceptron regressor was trained to capture nonlinear relationships between manufacturing conditions, inclination, and Ra. To mitigate limited experimental data, a conditional generative adversarial network was used to generate additional condition-specific tabular samples, thereby improving predictive performance. Model performance was assessed on a hold-out test set. A web-based decision-support interface was also developed to enable interactive process planning by loading a 3D model, specifying printing parameters, and adjusting the part's orientation. The system computes face-wise inclination from the model geometry and visualizes predicted Ra as an interactive colormap over the surface, enabling rapid identification of regions prone to high roughness and immediate comparison of parameter and orientation choices.
Executive Summary
This article presents a data-driven framework for predicting surface roughness in additive manufacturing using interactive 3D visualization. The framework leverages a multilayer perceptron regressor and a conditional generative adversarial network to capture nonlinear relationships between manufacturing conditions, inclination, and arithmetic mean roughness (Ra). The model's performance was assessed on a hold-out test set, and the results were visualized using a web-based decision-support interface. This framework has the potential to improve process planning in additive manufacturing by enabling rapid identification of regions prone to high roughness and facilitating immediate comparison of parameter and orientation choices. The system's interactive nature allows for real-time exploration of the relationships between printing parameters, surface angle, and Ra.
Key Points
- ▸ The framework utilizes a structured experimental dataset created using a three-level Box-Behnken design
- ▸ A multilayer perceptron regressor was trained to capture nonlinear relationships between manufacturing conditions, inclination, and Ra
- ▸ A conditional generative adversarial network was used to generate additional condition-specific tabular samples to improve predictive performance
Merits
Strength in predictive performance
The framework demonstrates excellent predictive performance on a hold-out test set, suggesting its potential for real-world application in additive manufacturing.
Interactivity and visualization
The web-based decision-support interface enables interactive process planning and visualization of predicted Ra, facilitating rapid identification of regions prone to high roughness.
Demerits
Limitations of experimental data
The framework relies on a limited dataset, which may not be representative of all possible printing conditions and surface angles, potentially limiting its generalizability.
Dependence on nonlinear regression
The framework's performance is heavily reliant on the accuracy of the nonlinear regression model, which may be sensitive to overfitting or underfitting, particularly with limited experimental data.
Expert Commentary
The article presents a well-structured framework for predicting surface roughness in additive manufacturing. The use of a multilayer perceptron regressor and a conditional generative adversarial network is a promising approach to capturing nonlinear relationships between manufacturing conditions and Ra. However, the framework's performance is heavily reliant on the quality and representativeness of the experimental dataset. Future research could focus on expanding the dataset and exploring alternative machine learning techniques to improve predictive performance. Additionally, the framework's potential for real-world application in additive manufacturing is significant, and its development may inform policy decisions regarding the adoption and regulation of this technology.
Recommendations
- ✓ Future research should focus on expanding the experimental dataset to improve the framework's generalizability and predictive performance.
- ✓ Alternative machine learning techniques, such as gradient boosting or random forests, could be explored to improve the framework's robustness and accuracy.