Aiming at the issues of multi-agent interest coordination and operational robustness under uncertainty in multi-microgrid integrated energy systems
an optimal scheduling strategy for multi-microgrid systems based on an attention-driven robust game is proposed. First
a multi-microgrid integrated energy system model with shared energy storage is constructed
and an integrated demand response mechanism along with source-load uncertainty sets is introduced. Second
a bi-level collaborative scheduling model combining Stackelberg game and robust optimization is established. The upper-level system operator provides guidance by setting dynamic prices
while the lower-level multi-microgrid alliance solves for the lowest-cost operation scheme under the worst-case scenario. An attention-driven particle swarm optimization algorithm is applied to solve the model. Finally
case study results show that
compared with the non-cooperative conventional scheduling mode
the proposed strategy increases the total system revenue by approximately 31%
effectively reduces the system's dependence on the external distribution grid
ensures supply reliability under severe source-load fluctuation scenarios
and achieves the collaborative optimization of multi-party interests.
SAHARI K S M,JALAL M F A,HOMOD R Z,et al. Dynamic indoor thermal comfort model identification based on neural computing PMV index [C ] //4th International Conference on Energy and Environment 2013.Putrajaya,Malaysia,2013:012113.