Feature Recognition and Intelligent Energy Management Based on Multidimensional Cyclic Concatenation
Abstract
features describing time, route, charging and energy use are cyclically concatenated over a 14-day window. A shared sequence encoder and
four decoders predict route, charging probability, congestion and energy demand for the next seven days. These outputs adjust engine startstop thresholds and generate a target state-of-charge (SOC) trajectory. The model was trained with data from five drivers and deployed on a
BAIC DHT vehicle controller. Validation accuracy was 90.2% for route recognition and 85.3% for charging recognition; energy-consumption
MAPE was 7.8%. Four-week vehicle tests reduced combined fuel consumption from 7.2 to 6.8 L/100 km, corresponding to a 5.5% improvement, while also reducing unnecessary engine starts. The method is suitable for embedded predictive energy management.
Keywords
Full Text:
PDFReferences
[1] Sciarretta A, Serrao L, Contino F, et al. Optimal control of PHEVs using preview information[J]. IEEE Transactions on Control Systems
Technology, 2019, 27(3): 1262–1269.
[2] Zhang J Z, et al. Recent advances in energy management strategies for plug-in hybrid electric vehicles[J]. Journal of Mechanical Engineering, 2019, 55(14): 1–14.
[3] Liu T, Chen H, Wang J, et al. High-definition map-based speed prediction for eco-driving[J]. Transportation Research Part C, 2022, 134:
103478.
[4] Wang J, Zhang Y, Liu X, et al. Navigation usage patterns in daily commutes: a large-scale study[J]. Transportation Research Record,
2022, 2676(8): 345–359.
[5] Zhang H, Wang L, Chen G. Trajectory replication for speed prediction in navigation-disabled scenarios[J]. SAE International Journal of
Alternative Powertrains, 2021, 10(2): 178–189.
[6] Liu Y, Li S, Zhang W, et al. Machine-learning-based speed prediction for PHEV energy management[J]. IEEE Transactions on Vehicular
Technology, 2021, 70(5): 4287–4300.
[7] Wang H, Chen L, Zhang X. V2X-enhanced speed prediction for predictive vehicle control[J]. SAE International Journal of Connected
and Automated Vehicles, 2022, 5(2): 145–157.
[8] Chen W J, Wang J X, Zhang H. Research progress on predictive energy management for hybrid electric vehicles[J]. Automotive Engineering, 2023, 45(2): 201–212.
DOI: http://dx.doi.org/10.70711/frim.v4i9.10231
Refbacks
- There are currently no refbacks.
