Search

This study presents a novel data-driven method for state-of-charge estimation in lithium-ion batteries. It integrates a temporal convolutional network with multi-verse optimization to enhance the accuracy of predicting the state of charge. The temporal convolutional network possesses advantages such as an extended memory window and efficient parallel computation, exhibiting exceptional performance in time-series tasks for state of charge estimation. Its hyperparameters are optimized by adopting multi-verse optimization to obtain better model performance. The driving model utilizes various measurable data as inputs, including battery terminal voltage, current, and surface temperature. To validate the effectiveness of the proposed method, extensive datasets from diverse dynamic working conditions at different ambient temperatures are employed for model training, validation, and testing. The numerical outcomes provide evidence of the proposed method's superior performance compared to the other two methods, providing a more robust and accurate solution for the state of charge estimation in lithium-ion batteries. [ABSTRACT FROM AUTHOR]

The accurate determination of electrochemical parameters in lithium-ion batteries is crucial for assessing battery health. This study conducted a comparative investigation utilizing 78 popular meta-heuristic algorithms for parameter identification in simulations. In the electrochemical identification framework proposed herein, the pseudo-two-dimensional model of a lithium-ion battery was solved using the finite element method, and the electrochemical parameters were identified using meta-heuristic algorithms in a one-step strategy. Parameter identification was conducted under high-rate discharge/charge conditions with a loading current of 5C. The discussion encompassed the accuracy, convergence speed, and robustness of the 78 different meta-heuristic algorithms. Notably, the teaching learning-based optimization algorithm exhibited the highest accuracy, albeit with a moderate computational burden. With the exception of the search and rescue optimization algorithm, other algorithms with mean absolute percentage errors of less than 15% demonstrated relatively high robustness. Furthermore, a piecewise C-rates working condition was employed to validate the previous conclusions. Ultimately, this study proposed a modified teaching learning-based optimization algorithm to enhance the precision and computational efficiency of electrochemical parameter identification. This comparative analysis contributed novel insights into electrochemical parameter identification methods employing meta-heuristic algorithms. • 78 meta-heuristic algorithms were used for parameter identification of P2D model. • A one-step electrochemical parameter identification framework was proposed. • High-rate discharge/charge loading current was used for parameter identification. • Accuracy, convergence speed, and robustness of 78 algorithms were discussed. • A modified teaching learning-based optimization algorithm was proposed. [ABSTRACT FROM AUTHOR]