Hybrid modeling-based reinforcement learning for boiler control in data-scarce buildings using a simulation-informed digital twin

Key Points

• Boiler control efficiency improved with hybrid reinforcement learning, adapting to data-scarce building conditions.
• Key evidence shows a 20% increase in performance metrics measured during simulated environments.
• Hybrid modeling approach utilizes both simulation and digital twin technologies to optimize control strategies for building operations.
• Significance highlights the potential benefits for energy efficiency in real-world applications, especially in data-limited situations.