Increasing penetration of distributed generation (DG) has brought more uncertainty to the operation of active distribution networks (ADNs). With the reformation of the power system, increasingly more flexible loads access to distribution network through load aggregators (LAs), which becomes an effective way to solve these issues. Since LAs and ADNs are separate entities with different interests, the traditional centralized and deterministic optimization methods fail to meet the actual operational requirements of ADNs. Based on the linear power flow model, a robust optimal dispatching model of ADNs considering the influence of renewable DG’s uncertain output on voltage security constraints is established. Then, an independent optimal scheduling model for LAs is modeled based on the analysis of the composition and characteristics of flexible load in LAs. LAs and ADNs, as two different stakeholders, use a distributed modeling method to establish different economic optimization goals. The optimization problem is solved by decoupling the coupling exchanging power between LAs and ADNs into virtual controllable loads and virtual DGs. Finally, with the case study of a modified IEEE 33-bus system, the correctness and effectiveness of the proposed method are verified. The effects of the robust level and demand response incentive on the results are also analyzed.

No abstract

To promote the energy accommodation of both electrical and heating power while considering the source-load uncertainties, this paper proposes a peer-to-peer (P2P) energy trading model among prosumers in the integrated electric-thermal system, considering the energy trading agent (ETA) with self-build energy system. The Solar Heat-Pump Hybrid Thermal Water System (SPTS) and the transferring loss of heating power is considered and modelled based on the principle of steady-state thermal transfer. Since the variations of load and PV cannot be described by any single common distribution, the chance-constraints programming based on the Gaussian mixture model (GMM) is proposed to handle the uncertainty of the net load. As the proposed power trading problem is a non-convex problem with binary variables, an improved distributed alternating direction method of multiplier (ADMM) based on the predictor-corrector and two-stage cycle iterations is proposed to enhance the convergence performance of standard ADMM. Finally, an example simulation verifies the effectiveness, results show that the proposed model can decrease the total cost by 26.7% and enhance local energy balance by 61.8% compared to other cases.