Thesis Work - Electric Vehicle Smart Charging based on Predictive Models

About this opportunity - Background

The transition towards a more sustainable society is driving a trend of increased vehicle electrification and an increasing share of renewable energy production. In order to optimize operational cost and system efficiency, it is crucial to schedule charging of electrified vehicles in an intelligent manner. 

Currently, such functions are typically developed to minimize the cost to charge an individual battery electric vehicle (BEV) up to a specified battery energy level, at a known departure time. This can reduce charging costs significantly for customers who have electricity contracts with time-of-use pricing. A drawback with current state-of-the-art charging cost optimization systems is that they often only optimize the charging on a day-to-day basis. In general, there is rarely an absolute need to charge a BEV daily, and for most vehicles, it should be sufficient to charge at most a few times per week. However, longer time horizons may necessitate forecasting of energy demand and/or electricity prices. To ensure that the optimized charging schedules are of high quality, it is important that such forecasts are accurate.

Scope of the thesis work

•    Develop prediction models and optimization methods for Smart Charging applications where the time horizon stretches over several days. 
•    The prediction models should predict upcoming vehicle energy demand and/or trends for the hourly electricity prices, based on historical vehicle usage, market conditions, weather, and weather forecasts.
•    Integrate the developed prediction models with state-of-the-art Smart Charging optimization methods. 
•    Evaluate the performance on realistic data and assess the sensitivity towards disturbances.

What you'll bring

•    MSc. in Engineering, within Computer Science, Data Science, Mathematics, Control, or similar. 
•    Relevant courses in machine learning, algorithms, data science, statistics, control, optimization, and programming. 
•    Residence permit in the EU/Schengen, i.e., a valid visa for Sweden. 
•    Willingness to work on site at the Volvo Cars office in Torslanda-Göteborg, in pair with another student. 

Duration

•    The work will start in January 2026. 
•    The duration for this thesis work is 20 weeks.
•    30 ECTS in agreement with your Thesis Advisor at the University
•    This thesis is to be conducted by two students working in pair.