Research 

Amanchukuw Lab

University of Chicago Pritzker School of Molecular Engineering

Project Title: Machine Learning for Battery Electrolyte Discovery

Duration: 10 weeks

Project Description: 

The advancement of next-generation batteries is hindered by several challenges, encompassing issues like low energy density, safety hazards, and sluggish charge-discharge rates. To address these challenges, this project focused on the discovery of innovative electrolytes with outstanding ionic conductivity—a vital property that enhances both energy density and safety in batteries. We achieved this by training a Chemprop model using an extensive dataset featuring over ten thousand unique electrolytes. Rigorous model performance assessments were conducted, including comparisons with alternative models, to ensure its suitability for our project's objectives.

_______________________

Bioresources Valorization Laboratory

University of Benin

Project Title: Data-Driven Intelligent Modeling, Optimization, and Global Sensitivity Analysis of a Xanthan Gum Biosynthesis Process 

Duration: 6 months

Project Description:

Xanthan gum, a commonly used thickening agent in various fast-moving consumer goods (FMCG) products, was efficiently synthesized from agricultural waste materials. Machine learning models were designed to consider three crucial input factors for xanthan gum production: fermentation time, potassium phosphate concentration, and ammonium nitrate concentration.

To identify the most effective predictive model,  a comprehensive evaluation of several machine learning techniques, including extreme learning machines, extreme gradient boosting, kernel ridge regression, random forest, support vector machines, and artificial neural networks was conducted. Additionally, assessment of the optimization capabilities of two metaheuristic algorithms: genetic algorithm and particle swarm optimization was done.

This research aimed to streamline and enhance xanthan gum production from agricultural waste, contributing to both sustainability and cost-efficiency in the FMCG industry.

_______________________

Central Research Laboratory 

University of Benin

Project Title: Optimization of Lipase Yield from Agricultural Waste Using Machine Learning

Duration: 

Project Description (final year thesis):

Lipase, a vital enzyme utilized in the transesterification process for biodiesel production, was synthesized using a ternary substrate mixture comprising both solid and liquid feedstock components. To achieve the most favorable fermentation medium formulation, a constrained mixture design approach was employed. 

In addition to optimizing the feedstock blend, experiments to enhance lipase yield by harnessing the inductive effects of surfactants were conducted. This comprehensive research contributes to the efficient and sustainable production of biodiesel, offering promising solutions for the biofuel industry.