The study, published in the Journal of Energy Storage, details the work of the team in the School of Engineering, Technology and Design at Canterbury Christ Church University that has leveraged AI to reduce the environmental impact of batteries.
Lithium-ion batteries are integral to a wide range of industries, powering numerous modern technologies, from consumer electronics in smartphones, tablets and wearable technology, electric vehicles, energy storage, aerospace, medical devices, power tools, marine and military applications, renewable energy storage, and smart grids.
Despite their versatility, one of the challenges in battery technology is accurately estimating a battery’s State of Charge (SOC). Inaccurate SOC estimations can lead to overcharging, faster degradation, and early disposal of batteries, posing environmental and financial concerns.
To address this issue, the team led by PhD student Mohammed Khalifa Al-Alawi, Dr Hany Hassanin Reader in Engineering, Ali Jaddoa Senior Lecturer in Computing and Cybersecurity, and James Cugley Senior Lecturer in Applied Engineering, developed an AI model known as the Cluster-Based Learning Model (CBLM) that can accurately estimate the State of Charge by analysing real-time data to predict battery health levels more precisely than traditional methods. This can enhance battery management systems, allowing for more precise monitoring and preventing overcharging or discharging.
Improving efficiency and effectiveness
Hany explained the importance of the study: “This research is not just about making batteries last longer; it’s about making them smarter, safer, and more reliable. By improving SOC estimation, we can extend the life of second-life batteries, reduce waste, and make energy storage systems more efficient and cost-effective, and in turn reduce our reliance on fossil fuels.
“The research team are keen to collaborate with industry partners to bring this technology from the lab to real-world applications.”
The practical impact of the improved SOC estimation was explored in various real-world scenarios. There were many benefits of this study for sustainable energy solutions including prolonging battery health and cost savings.
How did the AI model perform?
- Tested on Tesla model lithium-ion battery: The new approach achieved lower estimation errors compared to state-of-the-art methods
- Performance: The CBLM model outperformed existing methods by nearly 60%, making it highly reliable for real-world applications
- Grid services: The new method preserved 5.4% more battery capacity, slowing battery degradation and extending operational life
- Solar energy storage: The AI model saved approximately £11,278 and maintained battery health for a longer period, improving energy storage efficiency
- Large-scale energy systems: The method saved up to £170,000 by enhancing battery lifespan and efficiency, increasing economic value and supporting sustainable energy practices.
In the upcoming Feb/Mar issue of Renewable Energy Installer magazine we consider the latest advancements in battery technology and take at a look at the drivers behind the emergence of sodium-ion as an alternative battery technology. If you are not subscribed, sign up here to receive your copy.
