Distributed modeling considering uncertainties for robust operation of integrated energy system

To improve the energy system flexibility and robustness, a distributed synergistic model with the min max-min robust optimization is proposed for a 3-block integrated energy system (IES). First, the uncertainties of distributed generation (DG) output, and electric vehicle (EV) aggregator's state of charge (SOC) in IES are thoroughly considered, and the uncertainty set considering the spatial-temporal correlation and symmetry is constructed. Second, a three-level min-max-min robust optimization model for IES is built to deal with the multiple uncertainties. Furthermore, to accelerate the solution of the third level with integer variables, this paper develops a column and constraint generation with an alternative optimization procedure (C&CG-AOP) algorithm in the robust optimization model. Finally, based on the prediction-correction-based alternating direction method with multipliers (PCB-ADMM) algorithm, a distributed model considering the multi-agent characteristics of IES is proposed to solve the 3-block IES distributed operation problem. The simulation results show that the constructed uncertainty set considering the spatial-temporal correlation and symmetry has lower operating cost than the traditional uncertainty set, which can eliminate some low probability scenarios and make the conservatism of robust optimization reduced in a certain degree. Meanwhile, the developed C&CG-AOP algorithm effectively reduces the solving time while keeping the results consistent, and has higher efficiency than the Nested-C&CG algorithm in solving complex multi-level models. Moreover, the 3-block IES distributed model based on the PCB-ADMM algorithm can converge reliably and has a faster convergence rate than the compared ADMM-based algorithm. (c) 2021 Elsevier Ltd. All rights reserved.

Automated summary

A distributed synergistic model with min max-min robust optimization is proposed for a 3-block integrated energy system, which effectively handles multiple uncertainties and accelerates the solution process with the developed C&CG-AOP algorithm. The simulation results show that the constructed uncertainty set considering spatial-temporal correlation and symmetry can reduce operating costs.