Perovskite solar clls (PSCs) have gained significant attention due to their high-power conversion efficiency (PCE) and potential for low-cost production. However, their stability under various environmental conditions remains a major challenge. PSCs are susceptible to degradation from factors such as light, heat, humidity, and reverse bias. 

Addressing these stability issues is crucial for their commercial viability. Machine learning (ML) techniques have been increasingly applied to predict and enhance the stability of PSCs. These methods can an-alyze large datasets to identify patterns and optimize material properties and device structures. Reinforcement learning, a subset of ML, involves training models to make decisions by rewarding desired outcomes, and has gained interest in materials or chemical processes design. In the context of PSCs, RL can be used to optimize fabrication processes, material compositions, and operational conditions to improve stability. The application of reinforcement learning in PSC stability research is still in its early stages but holds great promise. By leveraging RL, we can potentially overcome current stability challenges and pave the way for the commercial success of perovskite solar cells

As part of its Diversity policy, TotalEnergies considers all applications, including those from people with disabilities, on the basis of equal qualifications.

Recent studies have demonstrated the potential of using Reinforcement Learning(RL) in molecular design and chemical processes. We propose utilizing RL to design perovskite solar cells with enhanced stability. To function effectively, RL initially requires a method to evaluate proposed solutions. Therefore, the first step is to establish a simulator, likely in the form of a surrogate model, capable of estimating the desired optimization metric. Then the goal will be to implement a RL model to iteratively construct a stable perovskite solar cell. Depending on the existing data, we can consider combining reinforcement learning with supervised learning to guide and accelerate the training process.

This subjectis complex and will most probably lead to a PhD thesis.

You'll work as part of a team of experienced professionals and with a tutor-coach, the reference for your future profession. Individualized support will help you develop your autonomy and lead you to your diploma!

Currently enrolled in an engineering school or Master's program in the Artificial Intelligence and Solar field, are you looking for an 6month end-of-study internship starting in April 2025?

Do you have experience in Reinforcement Learning and knowledge of Knowledge on solar cells? Do you know Machine Learning?

Are you comfortable with office automation and familiar with the Office suite? You've got good coding skill (Python, RL librairies). Latex knowledge will be a plus.

Are you curious, autonomous and a great to communicate ? Can you take the initiative?

Good writing skills are a plus.

So don't wait any longer, apply to join our team!

Cette offre concerne un stage conventionné à temp plein, les stages alternés ne sont possibles.
 Pour postuler, merci de joindre un CV + lettre de motivation (vos dates de stage doivent être indiquées).

This is a full-time, contractual internship; alternating internships are not possible.

To apply, please attach a CV + covering letter (your internship dates must be indicated).