- This event has passed.
Search for Assistant Professor, Process Intensification: Gabriel Patrón
The Department of Chemical Engineering & Applied Chemistry is pleased to host an interview for Assistant Professor, Process Intensification.
Please join us in welcoming Dr. Gabriel D. Patrón for his research seminar on February 26th and his teaching seminar on February 27th, as per the details below:
RESEARCH SEMINAR: Thursday February 26th from 11:00am – 12:00 PM
Title:
From Data to Decisions: ML-Aided Chemical Process Optimization Under Uncertainty
Abstract:
Chemical process systems must be optimized across multiple timescales to ensure profitability, safety, and sustainability. While reactor set points and control actions are determined in real time, the capacity and schedule of electricity-integrated processes are planned over much longer horizons. In practice, model-based optimizers are inevitably subject to uncertainty in real-world settings.
In this talk, I present recent work using machine learning (ML) to represent these uncertainties and embed them within optimization frameworks. First, I consider reactor-level real-time optimization in the presence of epistemic, endogenous model uncertainty. Leveraging the universal approximation properties of neural networks, gradients are learned from historical plant data to reconcile models with reality and recover plant-optimal solutions with theoretical guarantees. Second, I discuss how quantile neural networks can learn conditional probability distributions in stochastic optimization settings with exogenous market uncertainty. Optimizing over these learned distributions explicitly hedges against electricity price uncertainty in capacity planning and process scheduling for an electrolytic hydrogen system. These results demonstrate how combining ML with optimization enables data-driven and uncertainty-aware decision-making for chemical process systems, with direct implications for process intensification and the design and operation of future low-carbon infrastructure.
Join us Virtually!
Microsoft Teams meeting
Join: https://teams.microsoft.com/meet/27576045026927?p=SqF8SOCHsKEF16OCE8
Meeting ID: 275 760 450 269 27
Passcode: zJ2dK2S7
Dial in by phone
+1 647-794-1609,,249456086# Canada, Toronto
Phone conference ID: 249 456 086#
For organizers: Meeting options | Reset dial-in PIN
TEACHING SEMINAR: Friday February 27th from 10:00 – 11:00 AM
Teaching Seminar Title: In the Loop: A Hands-On Introduction to PID
Join us Virtually!
Microsoft Teams meeting
Join: https://teams.microsoft.com/meet/25186297758341?p=8RH8HwN6WmRyRbMGFD
Meeting ID: 251 862 977 583 41
Passcode: 6BU9Nx2Q
Dial in by phone
+1 647-794-1609,,588618099# Canada, Toronto
Phone conference ID: 588 618 099#
For organizers: Meeting options | Reset dial-in PIN
SPEAKER BIOGRPAHY:
Dr. Gabriel D. Patrón is a Research Associate in stochastic optimization at Imperial College London’s Department of Computing and the Sargent Centre for Process Systems Engineering. His current work, funded by BP, aims to bridge modern machine learning with process engineering to develop methods for the control and optimization of chemical and energy systems. His research interests include process sustainability and intensification, as well as interpretable AI to uncover the systemic incentives driving the green transition. Previously, Dr. Patrón was a Postdoctoral Fellow at the University of Waterloo, where he also completed his PhD. He holds an MSc from Imperial College London and a BASc from the University of Toronto.