Energy Management Strategy for Fuel Cell Vehicles Based on Online Driving Condition Recognition Using Dual-Model...

Area of Science:

• Sustainable Transportation
• Artificial Intelligence in Automotive Engineering
• Control Systems

Background:

• Fuel cell electric vehicles (FCEVs) are crucial for addressing climate change and energy crises.
• Effective energy management strategies (EMSs) are vital for optimizing FCEV powertrain efficiency under varied driving conditions.

Purpose of the Study:

• To propose a novel dual-model predictive control energy management strategy (DMPC EMS) for FCEVs.
• To enhance FCEV economic performance and robustness using LSTM-based driving condition recognition.
• To improve the accuracy and adaptability of driving condition recognition through Wavelet Transform (WT).

Main Methods:

• Integration of Wavelet Transform (WT) with Long Short-Term Memory (LSTM) for enhanced driving condition recognition.
• Development of a dual Model Predictive Control (MPC) and explicit MPC (eMPC) framework with unified objectives and constraints.
• Implementation of a dual MPC switching logic based on predicted driving conditions for improved adaptability.

Main Results:

• The proposed WTL-DMPC EMS demonstrates significant improvements in FCEV economic performance.
• The strategy achieves remarkable energy-saving capabilities compared to existing baseline methods.
• Enhanced accuracy and generalization in driving condition recognition were observed.

Conclusions:

• The WTL-DMPC EMS offers a robust and efficient solution for optimizing FCEV energy management.
• The integration of WT-LSTM and dual MPC enhances adaptability across diverse driving scenarios.
• This approach shows considerable promise for advancing the efficiency and sustainability of FCEVs.