Trustworthy and explainable deep reinforcement learning for safe and energy-efficient process control: A use case in industrial compressed air systems

Puntos clave

• Energy efficiency was enhanced by implementing a deep reinforcement learning algorithm in compressed air systems, leading to more sustainable operations.
• Optimal performance improvements resulted in up to 30% energy savings per cycle, indicating the potential for widespread application.
• Assessment used a simulation-based model to evaluate process control strategies under various operational scenarios in industrial environments.
• Highlights the necessity for safe and explainable AI systems in industrial settings, aiming to enhance reliability and user trust.