A data-driven robust optimization framework for CCHP-P2G system considering the correlation of RES output

The interconnection between the primary grid and the combined cooling, heating, and power (CCHP) system promotes the emerging distribution of power-to-gas (P2G) facilities. The P2G system is effectively controlled to improve the electrical–gas coupling in the P2G in the CCHP system. The growing demand for distributed renewable energy (RES) generation, such as solar (PV) and wind turbine (WT), has introduced significant challenges in the economic management of integrated CCHP systems. Integrating these factors not only enhances system efficiency but also supports environmental sustainability. This research employs a data-driven two-stage robust optimization (TSRO) approach to optimize the dispatch model to define the electrical, heating, and cooling load. The big M approach is employed to relax the nonconvex constraints using second-order cone (SOC) relaxation. The PV-WT output uncertainty set is constructed through a maximum entropy ellipsoid (MEE) set to mitigate conservatism estimates and obtain extreme scenarios. The data-driven approach captures simultaneous, spatial, and temporal correlations within PV-WT power over several periods. A column and constraint generation method (C&CG) is used to solve the TSRO model, which utilizes extreme scenarios. The results of the simulations show that the proposed data-driven TSRO technique effectively minimizes the cost of energy conversion scheduling in real time and the day-ahead through the analysis of the IEEE 33-node bus system. In addition to improving the economic efficiency of the dispatching system CCHP, the approach ensures stability by sustaining a balanced supply and demand of cooling, heat, and electricity.