Suboptimal Online Energy Management for a Fuel Cell/Supercapacitor/ Battery Electric Vehicle Using Artificial Neural Network Approach
- Delivery
- Available on this site
- Format
- Price
- Non-members (tax incl.)：￥1,100 Members (tax incl.)：￥880
- Publication code
- 20229094
- Paper/Info type
- SETC
No.2022-32-0094
- Pages
- 1-9(Total 9 p)
- Date of publication
- Oct 2022
- Publisher
- JSAE
- Language
- English
- Event
- SETC2022
Detailed Information
| Author(E) | 1) CHIEN-LIANG CHEN, 2) YI-HSUAN HUNG, 3) ZHU-YANG QIU |
|---|---|
| Affiliation(E) | 1) Department of Industrial Education, National Taiwan Normal University, Taipei, 106, Taiwan, 2) Undergraduate Program of Vehicle and Energy Engineering, National Taiwan Normal University, Taipei, 106, Taiwan, 3) Department of Industrial Education, National Taiwan Normal University, Taipei, 106, Taiwan |
| Abstract(E) | This research aims at developing the suboptimal energy management strategy by using artificial neural network (ANN) for a triple-electrical-energy electric vehicle (EV). The controller hardware designs will be implemented in the future. Firstly, we constructed a low-order dynamic equations that abstracted the characteristics of the vehicle, including energy sources (the fuel cell, lithium battery, and supercapacitor), driver’s model, traction motor, transmission, and longitudinal vehicle dynamics, etc.. The key parameters were mostly retrieved from the commercialization software-Advanced Vehicle Simulator (ADVISOR). Base on the vehicle structure of the Toyota Mirai, we built the range-extended EV. The powertrain system included an 110kW fuel cell set, a 40Ah lithium-ion battery set, and a 165F/48V supercapacitor and a 150kW AC motor. The ECMS control strategy included a six-layer for-loop: the battery state-of-health (SOH), power demand, the battery state-of-charge (SOC𝑏), the supercapacitor state-of-charge (SOC𝑆𝐶), the power ratio of battery to power demand(α) and the power ratio of the supercapacitor to power demand(β). The ECMS data for ANN training was divided to two parts, the input part is four for-loop and the output part is α and β. To evaluated the benefit of the ANN, a rule-based (RB) control was designed as well. A standard driving cycle, New European Drive Cycle (NEDC), was chosen for the energy improvement evaluation. The energy consumption for RB and ECMS is [17.6258, 8.9141kWh], respectively, in two-time NEDC cycles. The energy improvement for ECMS is approximately 50% and the ANN accuracy is higher than 90%. The hybrid system can be scaled down to a small mobility in the near future. |
