Dr Brendan Wright | UNSW Research

Keywords

photovoltaics

machine learning

computational modelling

Fields of Research (FoR)

Condensed matter physics, Statistical data science, Machine learning, Photovoltaic devices (solar cells)

Biography

My Grants

Australian Renewable Energy Agency (ARENA) R&D Project Grant, "Machine learning applications for utility-scale...view more

My Grants

Australian Renewable Energy Agency (ARENA) R&D Project Grant, "Machine learning applications for utility-scale PV", RG220681, Dec 2022
Australian Centre for Advanced Photovoltaics (ACAP) Industrial Collaboration Grant, "End-of-Life Classification of Solar Panels", Mar 2022

My Qualifications

Ph.D. Physics (2018)
Intelligent Polymer Research Institute, University of Wollongong
Major: Physics of Excited-States in Condensed-Matter
Supervision: Prof. Attila Mozer & Assoc. Prof. Tracey Clarke
Thesis: ”The driving force dependence of charge carrier dynamics in donor-acceptor organic photovoltaic systems using optical and electronic techniques”
B. Nanotechnology Honours (2011)
School of Chemistry and Molecular Bioscience, University of Wollongong
Major: Computational Chemistry
Supervision: Assoc. Prof. Haibo Yu
Thesis: ”Development of a polarisable force field for Nafion”

My Awards

Winning Entry, 1st International Eurovision AI Song Contest, hosted by VPRO, Netherlands, May 2020; ”Beautiful the World” available at youtu.be/sAzULywAHUM; Australian team as collaboration between academics at RMIT and UNSW, and industry partner Uncanny Valley; personal contribution included developing raw audio and song structure generation capabilities with learning algorithms.

My Research Activities

Current research at UNSW focuses on the study of electrochemical defect mechanisms in silicon photovoltaics; modelling state-space dynamics using generative representation learning to investigate underlying physical mechanisms; more broadly, utilising recent developments in machine learning to model complex dynamic systems.

My Research Supervision

Areas of supervision

Development of generative representation learning systems to model complex dynamics of physical systems from experimental characterisation data, focusing on excited-state physics within photovoltaic systems (charge-carrier generation, transport, recombination