Bin Zhang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Bin Zhang has a total of 74 co-authors. Here are the top co-authors by number of shared publications:
Xiaoyang Luan
8 shared publicationsShuya Hou
7 shared publicationsMingju Jing
5 shared publicationsYingnan Zhang
Nanjing Tech University
5 shared publicationsZhi Wang
5 shared publicationsHui Hu
4 shared publications
Latest Publications
- Open Access
Adaptive Hyperparameter-Tuned Transformer–LSTM for Lithium-Ion Battery State-of-Health Prediction
Accurate prediction of lithium-ion battery state of health (SOH) is crucial for improving the safety, reliability, and operational efficiency of battery management systems (BMSs). However, many data-driven methods still struggle to maintain robust forecasting performance when degradation trajectories differ across cells, especially in later-stage aging. To address this issue, this study developed a robustness-oriented SOH prediction framework, termed Ada-TL, by integrating a Transformer encoder, an LSTM regressor, and adaptive hyperparameter tuning. Cycle-level health indicators were extracted from the publicly available CALCE dataset and transformed into a compact representation for supervised learning. The Transformer module captures non-local dependencies within each input window, whereas the LSTM summarizes sequential degradation dynamics. The number of attention heads, the initial learning rate, and the L2 regularization coefficient are adaptively optimized to reduce manual trial-and-error in model configuration. Experimental results on four CS2 cells show that Ada-TL consistently outperformed BP, CNN–LSTM, and the fixed-hyperparameter baseline in overall SOH prediction accuracy, achieving RMSE values of 0.0210–0.0310, MAE values of 0.0163–0.0262, and MAPE values of 4.17–9.30%. Additional late-stage and cumulative-drift analyses further indicate that Ada-TL provided more stable post-knee tracking and better control of long-horizon bias accumulation, with late-stage RMSE reduced to 0.0169–0.0217 across the four cells. An ablation study also showed that the KPCA-based three-dimensional representation improved the overall test-set accuracy on most cells while reducing input dimensionality. These results suggest that the main value of Ada-TL lies in robustness-oriented SOH forecasting under cell-to-cell variability.
Open AccessAdaptive Hyperparameter-Tuned Transformer–LSTM for Lithium-Ion Battery State-of-Health Prediction
Accurate prediction of lithium-ion battery state of health (SOH) is crucial for improving the safety, reliability, and operational efficiency of battery management systems (BMSs). However, many data-driven methods still struggle to maintain robust forecasting performance when degradation trajectories differ across cells, especially in later-stage aging. To address this issue, this study developed a robustness-oriented SOH prediction framework, termed Ada-TL, by integrating a Transformer encoder, an LSTM regressor, and adaptive hyperparameter tuning. Cycle-level health indicators were extracted from the publicly available CALCE dataset and transformed into a compact representation for supervised learning. The Transformer module captures non-local dependencies within each input window, whereas the LSTM summarizes sequential degradation dynamics. The number of attention heads, the initial learning rate, and the L2 regularization coefficient are adaptively optimized to reduce manual trial-and-error in model configuration. Experimental results on four CS2 cells show that Ada-TL consistently outperformed BP, CNN–LSTM, and the fixed-hyperparameter baseline in overall SOH prediction accuracy, achieving RMSE values of 0.0210–0.0310, MAE values of 0.0163–0.0262, and MAPE values of 4.17–9.30%. Additional late-stage and cumulative-drift analyses further indicate that Ada-TL provided more stable post-knee tracking and better control of long-horizon bias accumulation, with late-stage RMSE reduced to 0.0169–0.0217 across the four cells. An ablation study also showed that the KPCA-based three-dimensional representation improved the overall test-set accuracy on most cells while reducing input dimensionality. These results suggest that the main value of Ada-TL lies in robustness-oriented SOH forecasting under cell-to-cell variability.
- 20 November 2025
- Published by Elsevier in Journal of Loss Prevention in the Process Industries
- Vol. 100
Effect of acidic conditions on the thermal hazard of 1-chloro-2,4-dinitrobenzene
- 20 November 2025
- Published by Elsevier in Journal of Loss Prevention in the Process Industries
- Vol. 100
Effect of acidic conditions on the thermal hazard of 1-chloro-2,4-dinitrobenzene
- 26 February 2025
- Published by Elsevier in International Journal of Hydrogen Energy
- Vol. 111, 667-680
Numerical study on mitigation effect of fence on the vapor dispersion of liquid hydrogen leak in a dike
- 26 February 2025
- Published by Elsevier in International Journal of Hydrogen Energy
- Vol. 111, 667-680
Numerical study on mitigation effect of fence on the vapor dispersion of liquid hydrogen leak in a dike
- Open Access
- 1 January 2025
- Published by National Library of Serbia in Journal of the Serbian Chemical Society
- Vol. 90 (2) , 233-245
Enhancing fire resistance in wood with high-water retention silica gel: A promising flame-retardant solution
This study aims to evaluate water retention and flame-retardant properties of silica gel prepared using anionic polyacrylamide (HPAM), gluconate- delta-lactone (GDL) and aluminum citrate (AlCit). Silica gel samples were synthesized with sodium silicate (8 wt. %), sodium bicarbonate (4 wt. %) and varying concentrations of HPAM (0.2?0.8 wt. %) and GDL (0.1?0.3 wt. %). The prepared gels were characterized using XPS, XRD, FTIR and TGA. Optimal water retention capacity was achieved with 0.4 wt. % HPAM and 0.2 wt. % GDL. Compared to traditional gels, silica gel has more surface water molecules due to the additional hydrophilic groups and the amorphous nature of silica. At high temperatures, silica forms a layer with the charcoal from treated wood combustion, inhibiting oxygen penetration and minimizing further combustion. After combustion at 500?C, the mass loss of wood treated with silica gel is 36?53 % less than untreated wood, indicating greater weight retention and demonstrating silica gel's effectiveness in preventing continued burning.
- 1 January 2025
- Published by National Library of Serbia in Journal of the Serbian Chemical Society
- Vol. 90 (2) , 233-245
Open AccessEnhancing fire resistance in wood with high-water retention silica gel: A promising flame-retardant solution
This study aims to evaluate water retention and flame-retardant properties of silica gel prepared using anionic polyacrylamide (HPAM), gluconate- delta-lactone (GDL) and aluminum citrate (AlCit). Silica gel samples were synthesized with sodium silicate (8 wt. %), sodium bicarbonate (4 wt. %) and varying concentrations of HPAM (0.2?0.8 wt. %) and GDL (0.1?0.3 wt. %). The prepared gels were characterized using XPS, XRD, FTIR and TGA. Optimal water retention capacity was achieved with 0.4 wt. % HPAM and 0.2 wt. % GDL. Compared to traditional gels, silica gel has more surface water molecules due to the additional hydrophilic groups and the amorphous nature of silica. At high temperatures, silica forms a layer with the charcoal from treated wood combustion, inhibiting oxygen penetration and minimizing further combustion. After combustion at 500?C, the mass loss of wood treated with silica gel is 36?53 % less than untreated wood, indicating greater weight retention and demonstrating silica gel's effectiveness in preventing continued burning.