A multi-objective optimisation scheduling method for microgrids based on improved sparrow algorithm

This study investigates the multi-objective scheduling optimisation of microgrid clusters to enhance power operation, energy management, and cost efficiency. An improved sparrow search algorithm is developed by integrating Bernoulli chaotic mapping, Cauchy mutation, dynamic adaptive weights, and reverse learning, enabling faster convergence and stronger global search capability. Simulation results show that for the F1 test function, the algorithm converges around 88 iterations with accuracy above 92.05%, and deviations remain within 2%. In a practical 24-hour grid-connected scheduling case, the original operating cost of the regional microgrid cluster was 31.82 × 104 yuan, while the optimised cost decreased to 25.36 × 104 yuan. Overall efficiency improved by approximately 25.46%. These findings demonstrate that the proposed method significantly enhances energy utilisation and provides a reliable theoretical basis for improving the economic performance and operational sustainability of microgrid clusters.