Abstract:

Under the "dual carbon" goals, the park-level integrated energy system (PIES), as an important carrier for achieving multi-energy complementarity and low-carbon transition, has attracted extensive attention. However, its operation is affected by multiple sources of uncertainty, such as wind power output fluctuations and indirect carbon emission intensity of the power grid, which poses challenges to both economic efficiency and carbon performance of the system. To this end, this paper proposes a distributionally robust optimization method for PIES considering the uncertainty of power generation and carbon emissions. A hybrid fuzzy set is constructed based on the Wasserstein distance and moment information, while chance constraints are employed to handle wind power uncertainty. Meanwhile, a polyhedral uncertainty set is used to characterize the fluctuations in carbon emission intensity, and user-side demand response is incorporated to enhance system flexibility. The proposed model is transformed into a solvable mixed-integer linear programming (MILP) problem through the column-and-constraint generation (C&CG) algorithm and Karush–Kuhn–Tucker (KKT) conditions. Case study results demonstrate that the proposed method enhances economic efficiency while ensuring system robustness, and effectively coordinates the relationship between renewable energy accommodation, carbon emission constraints, and economic operation.

Key words: park-level integrated energy system, distributionally robust chance constraint, uncertainty of carbon emission intensity, column-and-constraint generation algorithm, KKT