I’m passionate about decarbonising transport, which is a critical area to address to tackle climate change. I was drawn to a PhD as it could allow me to delve into the interface between transport and energy systems, an exciting and relatively new area of study. The freedom and agility that come with academic research also appealed to me. I can explore innovative solutions and tackle complex problems without the constraints typically found in industry.

A natural progression to PhD

I studied my first degree in Natural Sciences (Physics and Chemistry) at the University of Bath. During my undergraduate degree, I took a placement year at Shell as an e-mobility intern. There, I mostly worked on projects related to electric vehicle (EV) charging infrastructure. It was this placement that sparked my interest in EV charging research.

Deciding to join the AAPS Centre for Doctoral Training was a natural extension of my desire to pursue a PhD. The programme aligned well with my research ambitions, especially given its strong focus on addressing challenges in automotive and energy systems. Also, the CDT’s close link to IAAPS provides access to advanced research and industry networks. This positioning offered the ideal environment to gain technical expertise in automotive systems and their interaction with energy ecosystems.

Addressing uncertainty in forecasting

My PhD is titled ‘Probabilistic forecasting of residential electric vehicle charging demand for low-voltage distribution network planning’. I present a novel probabilistic framework that forecasts EV adoption and charging demand on low-voltage (LV) networks. It integrates diverse open datasets with advanced statistical methods to deliver uncertainty-aware forecasts. These will support evidence-based network investment decisions.

Local clusters of EV ownership and variations in charging behaviour make it difficult to predict future demand of residential charging on LV distribution networks. Compounded by limited visibility at the LV level, this presents major challenges for Distribution Network Operators (DNOs) in planning investments and ensuring adequate network capacity.

While growing availability of open energy data presents exciting modelling opportunities, it can be prone to inaccuracies and potential data drift. Integrating data from other sectors could enhance insights for the LV network. Yet, scalable and transparent methods for such integration are absent. Current EV adoption forecasts often lack the validation, spatial granularity, and uncertainty estimation needed for reliable LV network planning. Their alignment with real LV network topologies and principled risk quantification remains unresolved.

A novel approach

My framework takes a novel approach to EV adoption forecasting. It uses Bayesian inference and Gaussian process regression to integrate historical registration data with government mandates. This produces accurate forecasts alongside calibrated uncertainty estimates. Another part of my research is the fusion of multiple open datasets. I use census data to calibrate vehicle registration data and estimate its uncertainty. A novel mapping approach translates administrative-level forecasts to the LV network, accounting for spatial misalignment and uncertainty. A hierarchical Bayesian model predicts probabilistic, scalable and validated EV demand profiles.

This research could accelerate the decarbonisation of the transport sector. The modelling framework allows DNOs to make probabilistic forecasts of EV charging demand at the low-voltage distribution level. It includes uncertainty quantification so DNOs can make more robust and informed decisions when it comes to grid investments and risk management. It could also lead to increased uptake and sales of EVs, as well as faster charger installations. The public could benefit through faster EV charger installations, less disruption to grid maintenance, and potentially lower electricity costs.

A memorable PhD experience

The most challenging aspect has been the independent nature of a PhD. A PhD can be very open-ended. It requires you to clearly define your research’s scope and direction, often without a clear 'right' answer. In addition, a PhD is very intellectually demanding and, at times, isolating, as few people share the same depth of focus in your field.

I’ve found great satisfaction in presenting my research. The process of preparing and delivering well-structured presentations has been incredibly fulfilling. It has not only improved my ability to communicate effectively but has also allowed me to share my progress and spark interest in my work among diverse audiences. Seeing others engage with and appreciate the significance of my research is particularly rewarding.

After completing my PhD, I will take a short postdoctoral position at Bath where I will build upon my research. Beyond that, I aim for a career in the energy sector, focusing on the intersection of energy systems and data science.