Physics-Informed Deep Transfer Reinforcement Learning Method for the Input-Series Output-Parallel Dual Active Bridge-Based Auxiliary Power Modules in Electrical Aircraft

Keyphrases

• Reinforcement Learning Algorithm 100%
• Dual Active Bridge 100%
• Input-series Output-parallel 100%
• Physics-informed 100%
• Transfer Reinforcement Learning 100%
• Electric Aircraft 100%
• Deep Transfer 100%
• Auxiliary Power Module 100%
• Dual Active Bridge Converter 75%
• Control Stage 25%
• Accurate Model 25%
• Multiple Agents 25%
• Knowledge Transfer 25%
• Phase Shift 25%
• Power Balance 25%
• Deep Reinforcement Learning (deep RL) 25%
• Uncertain Environment 25%
• Reinforcement Learning Approach 25%
• Current Stress 25%
• Stochastic Environment 25%
• Modulation Method 25%
• Optimized Modulation Strategy 25%
• Effective Knowledge 25%
• Reinforcement Learning Agent 25%
• Data Providing 25%
• Online Control 25%
• Triple Phase Shift Modulation 25%
• Model Information 25%
• Power Balance Control 25%
• Variable Duty Cycle 25%
• Centralized Training 25%
• Source Simulation 25%
• Cycle Shift 25%
• Simulation System 25%

Engineering

• Reinforcement Learning 100%
• Auxiliary Power Module 100%
• Phase Shift 50%
• Power Balance 50%
• Experimental Result 25%
• Duty Cycle 25%
• Learning Approach 25%
• Modulation Method 25%
• Simulation System 25%

Chemical Engineering

• Auxiliaries 100%
• Reinforcement Learning 100%