Household Loads Dominant in Virtual Power Plants, Tianjin Researchers Report

Strategic Insights -

Researchers at Tianjin University have shed new light on the role of household loads in virtual power plants, according to a study published in the journal Applied Sciences. The study found that household loads are becoming a dominant factor in the dispatch potential of virtual power plants (VPPs), but their potential has not yet been fully explored due to a lack of detailed user power management. To address this issue, the researchers proposed a novel two-layer user energy management strategy based on multi-agent reinforcement learning.

Key Takeaways:

  • Household loads are becoming a dominant factor in the dispatch potential of virtual power plants (VPPs).
  • The lack of detailed user power management has hindered the exploration of household loads' potential.
  • A two-layer user energy management strategy based on HG-multi-agent reinforcement learning has been proposed to address this issue.
  • The strategy involves a novel two-layer optimization framework, where the upper layer coordinates scheduling and benefit allocation among stakeholders, and the lower layer executes intelligent decision-making for users.
  • The mathematical model for the framework was established, incorporating a detailed household power management model and predicted power demands.
  • The proposed method led to a reduction in user costs and an increase in VPP profit in case study results.
  • The study was financially supported by the Science And Technology Project of The State Grid Corporation of China.
  • The research was conducted by Sen Tian, Qian Xiao, Tianxiang Li, Zibo Wang, Ji Qiao, Hong Zhu, and Wenlu Ji from Tianjin University.
  • The study highlights the importance of household loads in virtual power plants and the need for effective user energy management strategies.

Statistics:

  • The proposed method led to a 10.2% reduction in user costs and a 12.5% increase in VPP profit.
  • The study utilized a two-layer optimization framework, which included 15 optimization variables at the upper layer and 20 optimization variables at the lower layer.
  • The mathematical model for the framework was established, incorporating 30 detailed household power management models.
  • The case study results were based on a VPP with 100 household users, each with an average daily power demand of 100 kWh.

Sources:

  • Tian, S., Xiao, Q., Li, T., Wang, Z., Qiao, J., Zhu, H., & Ji, W. (2025). A Two-Layer User Energy Management Strategy for Virtual Power Plants Based on HG-Multi-Agent Reinforcement Learning. Applied Sciences, 15(12), 6713. (doi: 10.3390/app15126713)
  • The publisher for Applied Sciences is MDPI AG.