Engineering in Action: My Projects

As part of my studies at IPSA, I created a complete 3D model of a radial engine using CATIA software. The goal of this project was to design a complex mechanical system, from the individual 3D modeling of each component (crankshaft, pistons, main connecting rod, auxiliary connecting rods, propeller blades) to their final assembly.

Beyond the 3D design itself, the real challenge was managing the kinematic assembly constraints to ensure the consistency of the entire system. I then produced the complete set of technical drawings for the system, including a detailed bill of materials. This project demonstrates my attention to detail in mechanical design and my proficiency in CAD—skills that are essential for a career in the aerospace and motorsports industries.

As part of a fluid mechanics project, I conducted a comprehensive 2D numerical study of a moving object using STAR-CCM+. The objective was to study the development of the boundary layer and propose geometric optimizations to reduce the drag on a raft that needed to cross a lake in a record time of 4 hours, with 4 people on board.

- Mesh & Modeling: I set up a quadrangular mesh with prism layers to accurately capture the high wall-normal velocity gradients, coupled with a k-epsilon turbulence model.

- Design Optimization: The image shown illustrates my optimized geometry. Adding chamfers to the ends helped delay boundary layer separation. This modification drastically reduces the turbulent wake and significantly decreases form drag.

This project demonstrates my ability to use computational fluid dynamics (CFD) not merely as an observational tool, but as a powerful lever to guide technical design decisions.

I use programming and statistics to transform raw data into reliable engineering models. This graph illustrates the development of a multiple regression algorithm designed to predict the behavior of a mechanical system.

- Complex Modeling: My code incorporates nonlinear variables (saturation effects) and time-dependent variables (cycles) to accurately reflect the physical reality of the system.

- Mathematical Rigor: I optimized the algorithm using the least squares method to minimize the overall error.

- Validated reliability: The model achieves a coefficient of determination (R²) of 0.9073, proving that it is capable of explaining and predicting more than 90% of the system’s variance.

For me, the ability to interpret data and design robust models is an essential asset for optimizing technical performance.