We are please to announce that Zohreh Moradinia has won the Best Poster Award at SIMULTECH 2025 for her work on “Machine Learning-Driven Framework for Identifying Parameter-Driven Anomalies in Multiphysics Simulations”. This work investigates whether errors in scientific simulations can be detected using machine learning. Zohreh assumes errors resulting from incorrect configuration of the simulation, such as time steps which are too large. She has trained several models that can identify when and where the simulations have gone wrong. This is useful as a means to check validity of simulation results, especially when the simulation is configured with liberal parameter settings that aim to result in high simulation speed.

Zohreh performed this work during her PhD in DIPSA. She is currently a Research Fellow at Imperial College London.