|Funding for:||UK Students, EU Students, International Students|
|Funding amount:||Not Specified|
|Placed On:||14th May 2021|
|Closes:||14th August 2021|
4 year PhD studentship covering fees and stipend (£15,609 in 20201 ‐22).
Available to applicants Worldwide. Start dates available September 2021 or January 2022.
Dr Stefan Güttel
Professor Igor Larrosa
Professor Kody Law
Dr Matthew Thorpe
Project Title: Learning chemical reactions from simulated and measured data
Applications are invited for a fully funded 4‐year PhD project within the Manchester Mathematical Modelling in Science & Industry initiative (MMMSI, https://tinyurl.com/mmmsi). This is a collaborative project between the departments of Mathematics and Chemistry, with advisory involvement of an industry partner.
The aim of this project is to develop new machine learning approaches to learn parameters of chemical reactions from simulated and measured data. The outcome of this work has potential for high impact in the optimization of industrial chemical processes, the design of new synthetic tools and new improved catalysts, and for the understanding of biochemical systems.
Experimentally, for a given chemical reaction the concentration of reagents and products can be measured over time from different initial concentrations. Mathematically, the mechanisms of catalytic chemical reactions are governed by ordinary differential equations (ODEs). Inferring the mechanism’s defining ODEs from the measured concentrations using currently available approaches is a nontrivial task, generally requiring numerous simplifications and assumptions, and leading to imperfect models.
This project will develop new mathematical methods that can recover sparse ODE formulations from simulated and measured data, incorporating important constraints which are naturally satisfied by chemical reaction equations (such as conservation of matter). The project will build on recent progress in the sparse recovery of nonlinear dynamical systems and will significantly develop them further.
Academic background of candidates.
We are looking for an enthusiastic and motivated graduate with the following:
Contact for further Information
Interested candidates should email both Dr Stefan Güttel email@example.com and Prof. Igor Larrosa firstname.lastname@example.org in the first instance with a CV, cover letter, transcripts and contact details for 2 referees.
Type / Role: