Optimising real-time charging for e-micromobility devices
We are pleased to announce the publication of a new scientific article authored by researchers from the ETEC Department & MOBI Research Group at Vrije Universiteit Brussel (VUB), within the framework of the EU-funded NEMO project. The paper, titled “Optimization of charging process for electric micromobility devices with real-time operation”, has been published in the […]
We are pleased to announce the publication of a new scientific article authored by researchers from the ETEC Department & MOBI Research Group at Vrije Universiteit Brussel (VUB), within the framework of the EU-funded NEMO project.
The paper, titled “Optimization of charging process for electric micromobility devices with real-time operation”, has been published in the peer-reviewed journal Results in Engineering (Elsevier) and is now openly available via Zenodo.
Advancing fast charging for e-micromobility
Electric micromobility, including e-bikes, electric scooters, and light-weight electric motorcycles, is a rapidly expanding segment of sustainable urban transport. However, fast charging these devices presents key technical challenges, particularly when operating within low-cost processors and limited cooling systems. Traditional optimisation algorithms often require high computational resources, making them impractical for real-time charging control in micromobility.
To address these barriers, the VUB team developed a novel optimised charging algorithm that integrates offline training with real-time operation. The algorithm leverages a balance between charging speed and battery thermal behavior, producing optimized policy maps based on a 1D electrothermal model and dynamic programming. Once trained offline, these maps can be efficiently interpolated on low-cost microcontrollers during real-time use.
Key results and performance improvements
Experimental validation on a 43 Ah battery demonstrated that the proposed strategy significantly improves charging performance compared with conventional constant-current approaches:
Charging time reduced by 8.5 %, without compromising battery temperature or state-of-charge profiles.
Robust stability across varying initial states of charge and environmental conditions.
Minimal execution time of ~0.44 ms on an 8-bit microcontroller, highlighting suitability for real-world low-cost applications.
These findings offer a significant step toward efficient, safe, and scalable charging strategies in the micromobility landscape, a critical enabler for broader adoption of electric transportation in urban environments.
EU funding and acknowledgements
This research was supported by the European Commission under the NEMO project. This paper is openly accessible and available through the Zenodo open repository.
Read the paper: https://zenodo.org/records/18174820
