Optimising Smart Charging Plans for EVs: Maximising Savings and Efficiency
Through 206 experiments conducted over the three-month data set, Senior Research Developer at Volue, Johannes Phillip Maree, PhD, quantified the benefits of forecasted charging plans for EVs.
The research utilised a "Friends of Spark" platform where EV users voluntarily shared access to their vehicles' data. This data included charging behaviour, location, and state of charge. The analysis focused on 14 users across three energy pricing areas in Norway: Oslo, Tromsø, and an intermediate location. Random segments of seven days were chosen for each user, considering the respective price areas.
Smart charging increased weekly energy savings by 23%
Results revealed that users adopting smart charging plans based on granular price data could increase their weekly energy savings by an average of 23,2% during this period. While the potential is much larger, it varies depending on the location and price fluctuations. For example, Oslo's volatile energy prices offered greater potential savings than areas with more stable prices like Tromsø. In an extreme price scenario, we saw that a user would have saved up to 80,4%, which translates to 457,4 NOK/week.
Significant energy efficiency improvements are possible even with the existing price prediction accuracy. The results highlight the potential for collective efficiency gains and the importance of building confidence in smarter energy management.
While the research provides valuable insights, it acknowledges certain assumptions that can be challenged. Factors like varying user behaviour, deviations from weekly energy needs, and already optimised charging plans in some vehicles warrant further investigation. Additionally, the value of data and insights for better forecasting and planning should be emphasised. More precise predictions may enhance the benefits of smart charging, but even relative price signals can drive substantial behaviour shifts and savings.
Optimising smart charging plans through price segmentation and week-long forecasting holds immense potential for EV fleets, charging operators, and OEMs. This simple concept allows stakeholders to improve the planning, scheduling, and orchestration of distributed energy resources. It is a low-hanging fruit that can yield better results by gaining more insight into price signals and user behaviour.
These price signals will activate and influence consumer behaviour, helping to shape future incentives and initiatives in the renewable industry.
APIs used in the test can be accessed here.