Oluwamayowa Amusat, Ph.D.

Computing Sciences Research Area
Lawrence Berkeley National Laboratory
1 Cyclotron Road
50B-2270A
Berkeley, CA 94720
United States
ooamusat@lbl.gov
Google Scholar | LinkedIn | ORCID

About me

I am an experienced Computational Research Scientist in the Scientific Data Division at Berkeley Lab.

Previously, I was a post-doctoral research scholar in the Computational Research Division(CRD) where I worked on a number of energy and water-related projects under the supervision of Dan Gunter at Berkeley Lab. I have a Masters degree in Advanced Chemical Engineering from the University of Leeds, and a PhD. in Chemical Engineering from University College London (UCL) , where I worked under the supervision of Professor Eric S. Fraga and Professor Paul Shearing in the Product and Process Systems Engineering (PPSE) research group.

Among other things, I currently work on developing equation-oriented surrogate modelling tools to aid in the design and optimization of advanced energy and water desalination systems. This builds on my previous (PhD) research which focused on the design and optimization of off-grid hybrid renewable energy systems under renewables variability.

Research Interests

My interests generally centre around in the topic of accelerating first principles models, scientific simulation, and optimization with AI/ML. More specifically, my research interests lie in the area of computer-aided process engineering: the incorporation of the latest advances in modelling, optimization, machine learning and decision support techniques to the improvement and enhancement of process systems. I am especially passionate about leveraging data-driven solutions and optimization to advance sustainability goals in the energy and water sectors. My current efforts in this space are as part of the DOE-funded IDAES and NAWI projects and explore the interface between machine learning, operations research and process systems engineering.

Ongoing and Previous Projects

I am (or have previously been) involved a number of DOE-funded research projects:

Rapid Decarbonization of Energy Systems by Platform-Based Design: Focusing on Industrial Systems (Ongoing)

I am the technical lead of computational sciences team on research project focused on demonstrating the platform-based design (PBD) approach for the design of affordable, uninterruptible energy systemd for industrial applications (e.g., chemical plants, AI data centers).

PrOMMiS (Ongoing)

PrOMMiS (Process Optimization and Modeling for Minerals Sustainability) is a multi-institutional collaboration building a framework and modeling platform to enable design choices with costing, to perform process optimization, and to accelerate development and deployment of extraction and purification processes for critical minerals/rare earth elements at reduced risk.

As part of the project, I am collaborating with researchers at Carnegie Mellon University to develop a superstructure-based multi-objective optimization framework for the techno-economic assessment of rare earth element (REE) recovery processes from end-of-life hard disk drives (HDDs) and electric vehicle (EV) batteries. We are also working on developing optimization-based predictive models for the chemical precipitation of are earths from solution.

IDAES

I am part of a team of national lab researchers working on the IDAES Process Systems Engineering (IDAES PSE) Framework, a next generation multi-Scale modeling and optimization Framework to support the US power industry. The project is funded by the US Department of Energy’s Office of Fossil Energy through the Simulation-Based Engineering, Crosscutting Research Program.

I am developing equation-oriented surrogate modelling tools to aid in the design and optimization of advanced energy systems. The tools provide a way for important legacy simulations and pilot-stage energy generation technologies to be integrated into the IDAES PSE framework for techno-economic analysis and performance optimization.

The highlight of my contributions is the development of PySMO , an open-source surrogate modelling software providing users with a set of surrogate modeling tools which support flowsheeting and direct integration into a Pyomo and IDAES. The library has been adopted by the PSE community and has been instrumental in various PSE applications, including water treatment systems design, carbon capture optimization, and programming with Bayesian hybrid models .

IDAES was a winner of the prestigious 2020 R&D 100 award.

NAWI (Ongoing)

NAWI is a DOE-funded Energy-Water Desalination Hub headquartered at Berkeley Lab.

As part of NAWI, I am an active contributor to the development of the Water treatment Technoeconomic Assessment Platform (WaterTAP), an open-source Python-based software package that supports the technoeconomic assessment of full water treatment trains and advanced water desalination systems. Details about the WaterTAP platform may be found on GitHub here.

I am also heavily involved in research into how information from complex high-fidelity first principles and black-box models can be integrated into open-source EO-optimization frameworks such as WaterTAP using surrogate and AI/ML models.

ScienceSearch (Ongoing)

ScienceSearch is a platform that enables search across different types of scientific data. ScienceSearch is funded out of the Office of Advanced Scientific Computing Research (ASCR), U.S. Department of Energy.

I am supervising of a team of researchers developing machine learning techniques for text labeling and automated metadata generation that will help improve indexing and scientific search.

Professional Activities