An Evidential Reasoning-Enhanced African Vulture Optimization Algorithm for Two-Stage Optimization of Integrated Energy Systems Under Uncertainty

With the aim of mitigating the impact of wind power integration and source-load-side uncertainties on an integrated energy system, we initially employed the Monte Carlo simulation in this study to randomly generate multiple wind power output/load scenarios in accordance with probability distribution functions. Additionally, we proposed a two-stage optimization method. In the first stage of our study, an enhanced African vulture optimization algorithm was applied to perform multi-objective optimization targeting fuel cost and carbon emissions across various scenarios, thereby solving the Pareto frontier to obtain multiple candidate solutions. In the study’s second stage, comprehensively considering fuel cost, carbon emission, and wind power penetration rate, evidential reasoning was utilized to determine the optimal operation strategy among the candidates. Finally, a combined heat and power system composed of the IEEE 30-bus system and a 32-node heating network was simulated. The results demonstrate that this decision-making approach can effectively reflect the merits of candidate solutions, thus validating the feasibility of the designed research methodology.