Forecast-enhanced bilevel real-time pricing for microgrids via hybrid-action reinforcement learning

Puntos clave

• Real-time pricing strategies were enhanced by applying hybrid-action reinforcement learning methods.
• Forecasting models showed significant improvements, leading to reduced operational costs by 15% compared to traditional methods.
• The approach combined reinforcement learning with bilevel optimization, effectively managing distributed energy resources.
• Findings highlight potential for better demand response strategies, but further validation in diverse scenarios is needed.