Join us! This event features Selen Cremaschi, who will talk about "Optimizing Processes with Hybrid Models: Energy and Biomedical Case Studies" as part of the Lehigh University Chemical and Biomolecular Engineering's Colloquium Seminar Series.

Abstract: Leveraging the growing capabilities of artificial intelligence (AI) and machine learning (ML)
holds great potential to advance the fundamental understanding of the underlying phenomena for
processing and biomanufacturing systems and to optimally design and operate them. These systems,
unlike many of the native application domains of AI/ML algorithms, typically produce relatively
structured data sets that may have poor information content and may not be abundant. Additionally,
there is a wealth of accumulated knowledge about some of these systems. Therefore, the application of
AI/ML techniques requires careful selection and customization to incorporate existing knowledge for
processing and biomanufacturing systems.

Hybrid modeling, which combines first-principles models with data-driven models based on AI/ML
algorithms, offers a promising approach. In this framework, the first-principles model captures known
process knowledge, while the data-driven model addresses discrepancies due to incomplete
understanding of the process mechanisms, thereby enhancing model accuracy. This talk will explore the
strengths and challenges of building hybrid models for engineering applications using three seemingly
disparate examples from energy and biomedical domains.