PhD Thesis at Western University (Defended: May 2021)

Summary

My Ph.D. research was part of a collaborative project to solve a problem for one of the major Oil & Gas companies in Canada. I applied data analysis, mathematical modelling, Machine Learning and programming Python as tools to optimize the separation process.

You can find a copy of my thesis here

Objective

To develop a mathematical model able to predict the behaviour of hydrocarbon/water mixtures using traditional methodologies in combination with machine learning techniques

Business Problem

Hydrocarbon/water mixtures are not completely miscible. To optimize the separation improving process economy, it is needed to understand hydrocarbon/water mixtures behavior

Business opportunities

The opportunity is to:

Applied statistics and machine learning for data analysis of experimental results
Identified opportunities for process optimization using mathematical modelling, statistics and data analysis
Regression, classification and predictive analysis using machine learning and traditional techniques
Continuous communication of research outputs to team and industry stakeholders
Programing using Python (Pandas, Scikit-Learn, Matplotlib) and Matlab

Tools

Python
Pandas
Scikit-learn
Mathematical modelling
Statistical analysis

Achievements

I was the presenter for the results of our research at Canadian Chemical Engineering Conference 2020

I was the leading author of the following papers:
- Lopez-Zamora, S., et al. (2021). "Vapour-Liquid-Liquid and Vapour-Liquid Equilibrium of Paraffinic Aromatic Synthetic Naphtha/WaterBlends: Prediction of The Number of Phases" Link
- Lopez-Zamora, S., et al. (2021). "Thermodynamics and Machine Learning Based Approaches for Vapor-Liquid-Liquid Phase Equilibria in NOctane/Water Blends, as a Naphtha-Water Surrogate in Water Streams." Link

I was co-author of the following papers:
- Escobedo, S., et al. (2021). "Synthetic Naphtha Recovery from Water Streams: Vapor-Liquid-Liquid Equilibrium (VLLE) Studies in a Dynamic VL-Cell Unit with High Intensity Mixing."
  My contribution to this paper was related to the data analysis and thermodynamic modelling of the blends. Link
- Kong, J.; et al. (2021). "Phase Equilibrium in N-Octane/Water Separation Units: Vapor Pressures, Vapor and Liquid Molar Fractions."
  My contribution to this paper was related to the data analysis and thermodynamic modelling of the blends. Link

Additional publications after Ph.D. defense:
- Lopez-Ramirez, E., et al. (2023). "Application of Artificial Neural Networks to Predict Phase Behavior of n-Octane/Water Blends." Link
- Lopez-Zamora, S., et al. (2022). "A Machine Learning Approach for Phase-Split Calculations in n-Octane/Water and PASN/Water Systems." Link