As the adoption of electric vehicles (EVs) grows, the need for more reliable and energy-efficient charging solutions becomes increasingly important. Recent research by the University of Michigan Transportation Research Institute (UMTRI) and startup Utilidata highlights how artificial intelligence (AI) could play a pivotal role in optimizing EV charging and ensuring the stability of the power grid.

Increasing EV Charging Efficiency with AI

The study conducted by UMTRI and Utilidata explores the potential of AI to analyze EV charging behaviour, offering insights that could enhance the driver’s experience and help utilities manage the increased electricity demand.Β 

One important finding is that EV charging often draws power inconsistently, leading to lower power quality. This inconsistency not only wastes energy but can also cause EV chargers to malfunction, which is a challenge as these chargers are not as efficient as Oil stations.

AI tools could provide utilities with real-time data, enabling them to anticipate and immediately detect issues in the power grid. By doing so, AI can help maximize the efficiency of electric energy usage. For instance, AI models could alert utilities about potential impacts on the power grid due to EV charging.

Additionally, these models could guide drivers on optimal charging times and locations, helping EV charging companies maintain their equipment more effectively.

Real-Time Data for Improved Power Management

In their study, UMTRI and Utilidata equipped six EV charging stations at the University of Michigan with electric meter adapters powered by Utilidata’s AI platform, Karman. This platform analyzed various dynamics such as current, voltage, and power. Moreover, the team installed devices on ten vehicles to monitor the drivers’ charging habits.

The study revealed a phenomenon known as short cycling, where vehicles would start and stop charging even after the battery was fully charged. This behaviour leads to inefficient energy consumption and could cause overheating of wires and transformers. 

Additionally, the researchers found that EV charging lowers power quality when electricity deviates from ideal voltage and frequency ranges, potentially causing wear and tear on the equipment.

Preparing for a Sustainable Future

Given these findings, the study authors believe that unmanaged EV charging in areas with high EV adoption could negatively impact the power grid. To handle these issues, UMTRI is collaborating with the North American Electric Reliability Council and planning a larger research project to further explore these ideas.

The key takeaway from the study is the recognition of numerous unknown behaviours associated with electric vehicles. As Utilidata’s Vice President of Product Solutions, Yingchen Zhang mentioned, there is a pressing need to open up all this data to better understand and manage EV charging.

Conclusion

AI has the potential to change the way we manage EV charging, making it more energy-efficient and reliable. By providing real-time data and insights, AI can help utilities and drivers alike optimise energy use, maintain equipment, and ensure the stability of the power grid. As research continues, the integration of AI in EV charging systems could play a crucial role in the transition to a more sustainable and energy-efficient future.