Giorgia Capobianchi
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Giorgia Capobianchi has a total of 20 co-authors. Here are the top co-authors by number of shared publications:
- Luana Georgiana Stoica
University of Roma Tre
4 shared publications - Alessandro Di Marco
University Roma Tre
3 shared publications - Elisa de Paola
Department of Civil, Computer Science and Aeronautical Technologies Engineering, University of Roma Tre, Roma, Italy
3 shared publications R. Camussi
2 shared publications- Roberto Camussi
Department of Civil, Computer Science and Aeronautical Technologies Engineering, University of Roma Tre, Roma, Italy
2 shared publications A. Di Marco
1 shared publication
Latest Publications
- 17 August 2026
- Published by Elsevier in Aerospace Science and Technology
- Vol. 179
Numerical–experimental investigation of the aeroacoustic environment of a launch vehicle in transonic descent
- 17 August 2026
- Published by Elsevier in Aerospace Science and Technology
- Vol. 179
Numerical–experimental investigation of the aeroacoustic environment of a launch vehicle in transonic descent
- Open Access
- 15 August 2025
- Published by Elsevier in Aerospace Science and Technology
- Vol. 168
Overview of the VEGA-E space launcher model aerodynamic and aeroacoustic environment
- 15 August 2025
- Published by Elsevier in Aerospace Science and Technology
- Vol. 168
Open AccessOverview of the VEGA-E space launcher model aerodynamic and aeroacoustic environment
- Open Access
Predicting Wall Pressure Fluctuations on Aerospace Launchers Through Machine Learning Approaches
Artificial intelligence (AI) can be used to optimize the prediction of pressure fluctuations over the external surfaces of aerospace launchers and minimize the number of wind tunnel tests. In the present research, various machine learning (ML) techniques capable of predicting the acoustic load were tested and validated. The methods included decision trees, Gaussian Process Regression (GPR), Support Vector Machines (SVMs), artificial neural networks (ANNs), linear regression, and ensemble methods such as bagged and boosted trees. These algorithms were trained using experimental data from an extensive wind tunnel test campaign conducted to support the design of a VEGA (Advanced Generation European Vehicle) launcher vehicle and provide wall pressure fluctuations in many configurations. The main objective of this study was to identify, among several algorithms, the most suitable method able to process such complex databases efficiently and to provide reliable predictions. Different statistical indices, including the root mean square error (RMSE), the mean square error (MSE), and a correlation coefficient (R-squared), were employed to evaluate the performance of the ML methods. Among all the methods, the bagged tree algorithm outperformed the others, providing the most accurate predictions, with low RMSE and high R-squared values across all test cases. Other methods, such as the ANNs and GPR, exhibited higher errors, indicating their reduced suitability for this dataset. The results demonstrate that ensemble decision tree methods are highly effective in predicting acoustic loads, offering reliable predictions, even for configurations outside the training database. These findings support the application of ML-based models to optimize experimental campaigns and enhance the design of aerospace launch vehicles.
Open AccessPredicting Wall Pressure Fluctuations on Aerospace Launchers Through Machine Learning Approaches
Artificial intelligence (AI) can be used to optimize the prediction of pressure fluctuations over the external surfaces of aerospace launchers and minimize the number of wind tunnel tests. In the present research, various machine learning (ML) techniques capable of predicting the acoustic load were tested and validated. The methods included decision trees, Gaussian Process Regression (GPR), Support Vector Machines (SVMs), artificial neural networks (ANNs), linear regression, and ensemble methods such as bagged and boosted trees. These algorithms were trained using experimental data from an extensive wind tunnel test campaign conducted to support the design of a VEGA (Advanced Generation European Vehicle) launcher vehicle and provide wall pressure fluctuations in many configurations. The main objective of this study was to identify, among several algorithms, the most suitable method able to process such complex databases efficiently and to provide reliable predictions. Different statistical indices, including the root mean square error (RMSE), the mean square error (MSE), and a correlation coefficient (R-squared), were employed to evaluate the performance of the ML methods. Among all the methods, the bagged tree algorithm outperformed the others, providing the most accurate predictions, with low RMSE and high R-squared values across all test cases. Other methods, such as the ANNs and GPR, exhibited higher errors, indicating their reduced suitability for this dataset. The results demonstrate that ensemble decision tree methods are highly effective in predicting acoustic loads, offering reliable predictions, even for configurations outside the training database. These findings support the application of ML-based models to optimize experimental campaigns and enhance the design of aerospace launch vehicles.
- Open Access
- 24 August 2024
- Published by Elsevier in Aerospace Science and Technology
- Vol. 154
Aerodynamic and aeroacoustic experimental investigation of a three propellers DEP configuration
- 24 August 2024
- Published by Elsevier in Aerospace Science and Technology
- Vol. 154
Open AccessAerodynamic and aeroacoustic experimental investigation of a three propellers DEP configuration