Research on Electricity Scheduling Strategies for University Dormitory Areas Based on the Sine-Triggered PSO Algorithm

The global energy structure is undergoing a transition toward green and low-carbon development. To address the challenges of large-scale energy consumption management and insufficient low-carbon potential exploitation in university buildings, this study first develops models of photovoltaic (PV) generation and battery energy storage systems. Subsequently, an optimal scheduling model is established with the objectives of minimizing economic operating costs and carbon tax, while considering power balance constraints and equipment output constraints. Furthermore, to overcome the problems of premature convergence and local optima in the traditional particle swarm optimization (PSO) algorithm, a Sine-triggered Particle Swarm Optimization (STPSO) algorithm is proposed. By dynamically adjusting the particle search range using a sine function, the proposed algorithm enhances global exploration capability and effectively avoids premature convergence. Experimental results demonstrate that, compared with MEDDPG, MA, MOPSO, and PSO, the proposed STPSO algorithm achieves the lowest relative percentage growth value and the shortest computation time, requiring only 46.5 s, which verifies the effectiveness of the proposed improvements. Compared with the conventional PSO algorithm, STPSO achieves the most significant reduction in daily operating costs under rainy conditions, with a maximum saving of 45%. Moreover, it reduces carbon dioxide emissions by 119.82 kg on sunny days. The proposed method provides an effective solution for green energy management and low-carbon operation of university dormitory building clusters.