水准点(测量)
模型预测控制
功率(物理)
汽车工程
过程(计算)
工程类
控制(管理)
荷电状态
动力传动系统
计算机科学
电动汽车
燃料效率
控制工程
电池(电)
人工智能
扭矩
物理
量子力学
大地测量学
热力学
地理
操作系统
作者
Ruihu Chen,Chao Yang,Yue Ma,Weida Wang,Muyao Wang,Xuelong Du
出处
期刊:Applied Energy
[Elsevier]
日期:2022-10-01
卷期号:323: 119592-119592
被引量:12
标识
DOI:10.1016/j.apenergy.2022.119592
摘要
For off-road hybrid electric vehicles (HEV), due to the variability of road environments and the dynamic and deferred response characteristic of the engine generator set (EGS) for off-road HEVs, it is difficult to coordinate the power output of multi-energy sources to meet the demand power of the vehicle. Therefore, designing an efficient power control strategy for off-road HEVs remains a major challenge. Motivated by this issue, an online learning predictive power coordinated control strategy for off-road HEVs is proposed in this study. Firstly, the online sequential extreme learning machine is used for short-term power prediction for the first time. With the online learning capability, the precision of power prediction under irregular road conditions is significantly improved. Secondly, to determine the optimal control behavior of power distribution between two energy sources, a novel predictive adaptive equivalent consumption minimization strategy is designed. The equivalent factor is rolling optimized in the prediction horizon to maintain battery state of charge and ensure fuel economy. Thirdly, considering the actual response process of EGS, a one-step-ahead coordinated control is presented to guarantee adequate electric power output. Finally, the performance of the proposed strategy is verified by simulation and hardware-in-loop test. The results show that fuel consumption using the proposed strategy is reduced by 6.72% and 8.63% over benchmark method under the two test driving cycles, respectively. Meanwhile, the setting time of EGS power is decreased by 61.12% and 64.63% to ensure the dynamic performance of vehicle.
科研通智能强力驱动
Strongly Powered by AbleSci AI