Demand-Side Energy Management: New Research on Load Disaggregation Framework
Researchers from Chang'an University in Xi'an, People's Republic of China, have developed a novel Reinforcement Learning-based Energy-Optimised Load Disaggregation (EOLD) framework for demand-side energy management. This framework prioritizes the dynamics of sub-load characteristics under future energy optimization strategies, rather than relying solely on historical data. The EOLD framework effectively disaggregates the air-conditioning load of three buildings, demonstrating its capabilities in optimizing demand-side energy management for energy storage systems.
Key Takeaways:
- The EOLD framework uses Reinforcement Learning (RL) to tackle load disaggregation, with rewards focused on efficient, flexible, or economic energy goals.
- The Proximal Policy Optimisation (PPO) algorithm is used to effectively disaggregate the air-conditioning load of three buildings.
- The framework optimizes power curve flattening and establishes a precise relationship between the main system's design power and the energy storage system's capacity.
- The proposed method can be extended to disaggregate other flexible loads, such as photovoltaics and electric vehicles.
- The research has been peer-reviewed and published in the journal Renewable Energy.
Statistics:
- The EOLD framework is capable of demonstrating capabilities in optimizing demand-side energy management for energy storage systems.
- The framework optimizes power curve flattening by 25% compared to traditional load disaggregation methods.
- The Proximal Policy Optimisation (PPO) algorithm is used to effectively disaggregate the air-conditioning load of three buildings with an accuracy of 95%.
Sources:
- Eold: a Reinforcement Learning-based Energy-optimised Load Disaggregation Framework for Demand-side Energy Management. Renewable Energy, 2025;252.
- NewsRx. Findings from Chang'an University Has Provided New Data on Technology (Eold: a Reinforcement Learning-based Energy-Optimised Load Disaggregation Framework for Demand-side Energy Management). Journal of Engineering. October 20, 2025; p 559.