State of charge estimation of lithium-ion batteries based on PSO-TCN-Attention neural network
- Publisher:
- ELSEVIER
- Publication Type:
- Journal Article
- Citation:
- Journal of Energy Storage, 2024, 84
- Issue Date:
- 2024-04-15
Embargoed
| Filename | Description | Size | |||
|---|---|---|---|---|---|
| State of charge estimation of lithium-ion batteries based on PSO-TCN-Attention neural network.pdf | Accepted version | 5.82 MB |
Copyright Clearance Process
- Recently Added
- In Progress
- Open Access
This item is currently unavailable due to the publisher's embargo.
Lithium-ion batteries are acted as energy storage devices and widely used in many fields, such as mobile, electric vehicles, and renewable energy sources, etc. However, their reliability, performance and safety are limited by state of charge (SOC) estimation of Lithium-ion batteries. In order to address this issue, a PSO-TCN-Attention network model is proposed in this work. The particle swarm optimization (PSO) algorithm is utilized to optimize the structural parameters of temporal convolutional network (TCN), enabling the model automatically learn and adapt the characteristics of lithium-ion batteries under different temperature conditions. Then, the attention mechanism allows the network adaptively focus on key time steps, enhancing the capture of time dependency in the SOC estimation from the Lithium-ion battery dataset, and further improving the accuracy and robustness of the model. Moreover, as the dataset used for SOC prediction consists of battery data from all dynamic operating conditions at various temperatures, the model is validated against LSTM and TCN networks. Results demonstrate that the SOC estimation of PSO-TCN-Attention network model is the most optimal, whose RMSE and MAXE is less than 1 % and 5.75 %, respectively, and R2 coefficient of determination exceeds 99.88 %.
Please use this identifier to cite or link to this item: