Research on Low-Carbon Optimization Scheduling of Microgrid Based on Improved Dream Optimization Algorithm

Abstract Energy optimization scheduling of microgrids is a key technology for achieving low-carbon operation of the system and enhancing the capacity to accommodate high proportions of intermittent renewable energy. Addressing the challenge of traditional optimization algorithms being prone to getting stuck in local optima and having insufficient convergence in the high-dimensional low-carbon economic scheduling problem of microgrids, this paper proposes an improved dream optimization algorithm (SCO-DOA) that integrates the sine and cosine operators. In the “forgetting-supplementing” mechanism of the dream optimization algorithm, this method introduces a probability-mixed positive and negative cosine oscillation update strategy to dynamically coordinate global exploration and local development, effectively enhancing the algorithm’s search ability and convergence accuracy in the high-dimensional solution space.To verify the algorithm’s performance, a microgrid model comprising photovoltaics, wind turbines, diesel generators, micro gas turbines, and energy storage is established and optimized by SCO-DOA for minimizing combined operational and carbon emission costs. Simulation results show SCO-DOA improves convergence speed by ∼15%, reduces total carbon emissions by 9.2%, and decreases grid dependence by 18.1% compared to traditional DOA, demonstrating superior efficiency, economy, and low-carbon performance.