Sign in Subscribe
Batteries

Optimized Graphical Neural Network for Diagnosing Voltage Faults in Lithium-Ion Batteries

Strategic Insights

— 2 min read

Researchers at Guangdong Polytechnic Normal University have developed an optimized Graphical Neural Network (GNN) model to diagnose voltage faults in lithium-ion batteries, a critical function in battery management systems. The proposed method combines the physical coupling between batteries and the entanglement of measurement results with the strong nonlinear processing capability of neural networks to improve the effectiveness of fault localization. Experimental results show that the proposed method outperforms baseline methods in terms of Accuracy, Precision, Recall, and F1-score, verifying its effectiveness and accuracy in fault localization of voltage data.

Key Takeaways:

  • The proposed optimized GNN model extracts the relationships between various batteries by learning the topology of the batteries, combining physical coupling and measurement results with neural network processing capabilities.
  • Experimental results on three publicly available datasets show that the proposed method outperforms baseline methods such as GraphConv, GCNConv, ChebConv, SGConv, CNN, DBN, LSTM, and CNN-LSTM in terms of Accuracy, Precision, Recall, and F1-score.
  • The proposed method achieves a maximum improvement of 4.31% and 3.68% in the accuracy of abrupt fault and gradual fault localization respectively, compared to the highest-performing baseline method.
  • The optimized GNN model demonstrates satisfactory accuracy and stability in diagnosing voltage faults, which is of remarkable significance for the development of electric vehicles.
  • The researchers used three publicly available datasets to evaluate the performance of the proposed method, showing its effectiveness in real-world scenarios.
  • The model's ability to learn the topology of batteries and combine physical coupling with measurement results improves its effectiveness in fault localization.
  • The proposed method has the potential to be applied in various energy storage applications and electric vehicles, ensuring their safety and reliability.

Statistics:

  • The proposed method achieved an accuracy improvement of 4.31% and 3.68% in abrupt fault and gradual fault localization respectively, compared to the highest-performing baseline method.
  • The model demonstrated satisfactory accuracy and stability in diagnosing voltage faults, with a maximum improvement of 4.31% and 3.68% in accuracy.
  • The proposed GNN model outperformed baseline methods such as GraphConv, GCNConv, ChebConv, SGConv, CNN, DBN, LSTM, and CNN-LSTM in terms of Accuracy, Precision, Recall, and F1-score.

Sources:

  • Voltage faults diagnosis for lithium-ion batteries in electric vehicles using optimized graphical neural network. Scientific Reports, 2025;15(1):27328. Nature Publishing Group - www.nature.com/; Scientific Reports - www.nature.com/srep/
  • Guangdong Polytechnic Normal University. School of Automation. Guangzhou, 510665, Guangdong, People's Republic of China.
  • Nature Portfolio. Heidelberger Platz 3, Berlin, 14197, Germany.

Read more