Abstract:
The progressive development of electricity markets envisions the participation of smart community grids in energy trading, but it is still challenging to guarantee global...Show MoreMetadata
Abstract:
The progressive development of electricity markets envisions the participation of smart community grids in energy trading, but it is still challenging to guarantee global optimality for uncertain power management of multiple entities under a hierarchical framework. To tackle the cooperative strategic bidding problem with uncertainties, this paper proposes a hierarchical three-stage robust optimization (HTS-RO) method. With respect to the bilayer coordination framework involving an aggregator gathering multiple prosumers, a novel HTS-RO bidding model is firstly established relying on the canonical centralized two-stage RO model. In the first stage, the aggregator performs power bidding in the upper layer according to the derived power flow information of the shared buses from the lower-layer prosumers. Considering renewable-load uncertainties and privacy preservation, each prosumer conducts its own robust scheduling via the second and third stages based on the bidding prices from the aggregator and submits its optimal results of the shared bus back to the aggregator. For the robust scheduling model of each prosumer, the second stage identifies the basic plans of the nominal scenario, and the third stage checks the feasibility of the basic plans under uncertainties. To solve the resulting intractable multi-level mixed-integer nonlinear programming, a tri-loop decomposition-based iterative algorithm is tailored to ensure that the distributed computation converges to the optimal solution with the overall profit, thereby achieving the equivalent split of the centralized optimization into a hierarchical manner. Numerical tests on generic smart community grids with multiple prosumers validate the applicability and superiority of the proposed HTS-RO method for power bidding.
Published in: IEEE Transactions on Smart Grid ( Volume: 14, Issue: 4, July 2023)