In the ever-evolving landscape of energy storage, a groundbreaking study published by Joelton Deonei Gotz, a researcher from the Graduate Program in Electrical and Computer Engineering at the Federal University of Technology – Paraná, Brazil, is set to revolutionize the way we manage second-life lithium-ion batteries. This innovative research, published in IEEE Access, focuses on the development of a sophisticated system that can accurately estimate the state of charge (SOC) of second-life batteries, paving the way for more efficient and cost-effective energy storage solutions.
Lithium-ion batteries (LIBs) have long been the gold standard for energy storage in various industrial applications, from electric vehicles to renewable energy systems. However, as these batteries age, their state of health (SOH) declines, making them less suitable for high-demand applications. Traditionally, batteries with an SOH below 80% are either recycled or repurposed for less demanding tasks. Recycling, however, is an expensive and complex process, while repurposing requires intelligent battery management systems (BMS) to handle the diverse characteristics of aged cells.
Gotz’s research addresses this challenge head-on. “The idea is to create a system that can intelligently manage second-life batteries, making them viable for a wider range of applications,” Gotz explains. The system he developed consists of two layers of machine learning models, specifically Random Forest (RF) algorithms, which work together to estimate the SOC of second-life batteries with remarkable accuracy.
In the first phase, the system uses a RF model to discover the capacity curves of different second-life batteries, regardless of their age or initial capacity. Once the capacity curve is selected, a second RF model is trained to make SOC inferences for each specific capacity curve. This dual-layer approach ensures that the system can adapt to the unique characteristics of each battery, providing precise SOC estimates in real-time.
The results of Gotz’s study are nothing short of impressive. The system achieved a root square mean error (RSME) below 45 mAh for capacity estimation and below 0.87% for SOC estimation during the development phase. When integrated into a Raspberry Pi system for real-time measurements, the errors remained below 100 mAh for the first layer and below 1% for the second layer. These results demonstrate the system’s potential to significantly enhance the efficiency and reliability of second-life batteries.
The implications of this research are far-reaching. As the demand for renewable energy storage solutions continues to grow, the ability to repurpose aged lithium-ion batteries could lead to substantial cost savings and reduced environmental impact. By extending the useful life of these batteries, we can also reduce the need for raw materials, contributing to a more sustainable energy ecosystem.
Gotz’s work, published in IEEE Access, represents a significant step forward in the field of energy storage. As we continue to explore new ways to harness and store energy, innovations like this will be crucial in shaping a more sustainable and efficient future. The energy sector stands on the brink of a new era, where second-life batteries could play a pivotal role in meeting our growing energy demands.