Skip to main content

Spectroscopy Measurement

Optimizing the effect of battery relaxation on electrochemical impedance spectroscopy measurement for real-time SOC estimation using transfer learning


Conventional Electrochemical Impedance Spectroscopy (EIS) measurements require extended battery rest periods, restricting real-time use in battery management systems (BMS) and limiting generalization across chemistries. This work introduces transfer learning to significantly shorten rest time and improve model adaptability, enabling more practical and scalable EIS integration for real-world state of charge (SOC) estimations.
EIS experiments were performed on 52 Ah Lithium Iron Phosphate (LFP) and 3.6 Ah Nickel Cobalt Manganese (NCM) cells at varying SOCs, with and without rest periods. The transfer learning-based Deep Neural Network (DNN-TL) model achieved high SOC estimation accuracy for LFP cells with mean squared error (MSE) of 0.0063 and mean absolute error (MAE) of 0.0664, improving MSE by 77.58% and MAE by 50.92% compared to standard models. Additionally, only 30% of the original dataset size was needed for retraining. Applying the DNN-TL model trained on LFP data to NCM cells using unrested EIS data resulted in up to 82.08% reduction in MSE and 53.15% in MAE, requiring only 20% of the original data size for retraining.
Extensive EIS experiments were conducted on both 52 Ah LFP and 3.6 Ah NCM cells in a laboratory environment to obtain measurement data with and without rest time. For LFP cells, the proposed deep neural network with transfer learning achieves reliable SOC estimation using unrested EIS data, with an MSE of 0.0063 and MAE of 0.0664. This represents a significant improvement of up to 77.58% in MSE and 50.92% in MAE compared to conventional DNN models trained without transfer learning. 
Moreover, the burden of EIS data collection is substantially reduced—only 30% of the original source dataset is needed to retrain the DNN model while maintaining high accuracy. To further enhance the applicability of data-driven SOC estimation across different battery chemistries, the DNN-TL model trained on LFP cells is transferred to NCM cells. Using unrested EIS data, the transfer learning approach yields notable improvements in SOC estimation accuracy, achieving reductions of up to 82.08% in MSE and 53.15% in MAE relative to a standalone DNN trained solely on NCM data. 
Additionally, the retraining process requires only 20% of the original NCM dataset, significantly reducing data collection effort. These findings demonstrate the effectiveness of transfer learning in enabling real-time, cross-chemistry SOC estimation with reduced rest time and minimal data. Future investigations should address critical operating conditions encountered in EV applications, including variations in battery design, temperature-dependent EIS behavior, and the impact of battery aging and cycle life on EIS data distribution to support deployment in more dynamic and diverse scenarios.

Comments

Post a Comment

Popular posts from this blog

 How Network Polarization Shapes Our Politics! Network polarization amplifies political divisions by clustering like-minded individuals into echo chambers, where opposing views are rarely encountered. This reinforces biases, reduces dialogue, and deepens ideological rifts. Social media algorithms further intensify this divide, shaping public opinion and influencing political behavior in increasingly polarized and fragmented societies. Network polarization refers to the phenomenon where social networks—both offline and online—become ideologically homogenous, clustering individuals with similar political beliefs together. This segregation leads to the formation of echo chambers , where people are primarily exposed to information that reinforces their existing views and are shielded from opposing perspectives. In political contexts, such polarization has profound consequences: Reinforcement of Biases : When individuals only interact with like-minded peers, their existing beliefs bec...
Post a Comment
Read more

Quantum Network Nodes

An operating system for executing applications on quantum network nodes The goal of future quantum networks is to enable new internet applications that are impossible to achieve using only classical communication . Up to now, demonstrations of quantum network applications  and functionalities   on quantum processors have been performed in ad hoc software that was specific to the experimental setup, programmed to perform one single task (the application experiment) directly into low-level control devices using expertise in experimental physics.  Here we report on the design and implementation of an architecture capable of executing quantum network applications on quantum processors in platform-independent high-level software. We demonstrate the capability of the architecture to execute applications in high-level software by implementing it as a quantum network operating system-QNodeOS-and executing test programs, including a delegated computation from a client to a server ...
Post a Comment
Read more

Global Lighthouse Network

Smart, sustainable manufacturing: 3 lessons from the Global Lighthouse Network Launched in 2018, when more than 70% of factories struggled to scale digital transformation beyond isolated pilots, the Global Lighthouse Network set out to identify the world’s most advanced production sites and create a shared learning journey to up-level the global manufacturing community. In the past seven years, the network has grown from 16 to 201 industrial sites in more than 30 countries and 35 sectors, including the latest cohort of 13 new sites. This growing community of organizations is setting new standards for operational excellence, leveraging advanced technologies to drive growth, productivity, resilience and environmental sustainability. But what exactly is a Global Lighthouse and what has the network achieved? What is the Global Lighthouse Network? The Global Lighthouse Network is a community of operational facilities and value chains that harness digital technologies at scale to ac...
Post a Comment
Read more