Adaptive ECMS with Gear Shift Control by Grey Wolf Optimization Algorithm and Neural Network for Plug-in Hybrid Electric Buses

The Plug-in hybrid electric bus (PHEB) is an important means of public transportation. For PHEB, this paper proposes an energy management strategy that considers both gear shift control and power splitting. For gear shift control, a large number of gear-switching data under different working conditi...

Full description

Saved in:
Bibliographic Details
Published inIEEE transactions on industrial electronics (1982) Vol. 71; no. 1; pp. 1 - 10
Main Authors Sun, Xiaodong, Jin, Zhijia, Xue, Mingzhou, Tian, Xiang
Format Journal Article
LanguageEnglish
Published New York IEEE 01.01.2024
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Subjects
Adaptive control
Algorithms
Batteries
Driving cycle division
Dynamic programming
Energy consumption
Energy management
Engines
Equivalence
Fuel consumption
Fuel economy
Gear shift control
Gears
Grey wolf optimizer algorithm
Hybrid electric vehicles
Neural network
Neural networks
Optimization
Optimization algorithms
Plug-in hybrid electric bus (PHEB)
Power split control
Public transportation
Switching
Torque
Online AccessGet full text

Cover

More Information
Summary:The Plug-in hybrid electric bus (PHEB) is an important means of public transportation. For PHEB, this paper proposes an energy management strategy that considers both gear shift control and power splitting. For gear shift control, a large number of gear-switching data under different working conditions are calculated by using the dynamic programming algorithm. The neural network is trained by these data in the offline part so that it can provide the appropriate gear-switching signal in time in the online model. For power-split control, this paper mainly selects an improved equivalent fuel consumption minimization strategy (ECMS) and uses the grey wolf optimization algorithm to iteratively solve the optimal equivalent factor. The proposed control strategy has not only been verified under various operating conditions but also verified by relevant experiments on the HIL platform. The results show that the proposed strategy has better fuel economy than ECMS and the rule-based strategy under the provided mixed operating conditions. Compared with the dynamic programming algorithm, the fuel consumption is only increased by 3.23%, but the problem of the curse of dimensionality is avoided.
ISSN:0278-0046
1557-9948
DOI:10.1109/TIE.2023.3243304