Scaling Up Bioreactors via Dimensionless Latent Space: Translating Optimal Lab Performance to Industry
| Director | Álvaro Ibáñez, Salvador Izquierdo |
|---|
| Titulaciones | |
|---|---|
|
Master Universitario en Inteligencia Artificial e Ingeniería Computacional Máster Universitario en Ingeniería Industrial Máster Universitario en Ingeniería Informática Máster Universitario en Ingeniería Mecánica Máster Universitario en Ingeniería Química |
| Duracion | 4-5 months |
|---|---|
| Lugar | Online / Fluid Mechanics Area |
| Fecha Alta | 2026-09-15 |
| Fecha Baja | 2027-09-15 |
| Resumen | DESCRIPTION. Bioreactors are highly sensitive systems on which small operating conditions variations lead to huge changes in performance. Data-driven techniques allow for an easier visualization and analysis of how variables are related. Instead of using brute-force analysis or blind calculations, this study proposes leveraging the information hidden in the results of reactor simulations to find optimal points, crucial for a safe design and scale-up of these devices. OBJECTIVE. Using several data-driven algorithms (PCA, U-Net among others) the latent space of a bioreactor productivity will be explored in order to find optimals as well as gaining insights of the system that can be applied in the reactor itself. FRAMEWORK. The data used for the latent space building stems from CFD simulations. The different techniques to be used as well as the analysis will be implemented using Python. Possibility to assess the results performing additional simulations. IMPACT. This study is considered ground-breaking in the field of bioreactors since it opens a new path in the traditional scale-up approaches, showing a new workflow and analysis in this area. |
| Detalles/Contacto | If you are interested and want to know more about this and other available projects, reach to: salvador.izquierdo@unizar.es / a.ibanez@unizar.es |