Abstract: Integrating electric vehicles into microgrid can better balance the supply and demand of power grid, improving the utilization of renewable energy. The randomness of electric vehicle charging significantly impacts the economic operation of microgrid. An improved dung beetle optimizer algorithm is proposed to optimize the economic dispatch of microgrid with electric vehicles. In response to the problems of uneven population distribution, weak global search capability, and easy falling into local optimum in dung beetle optimizer algorithm, quasi-oppositional-based learning is used to initialize the population, the rolling behavior of the dung beetle algorithm is replaced with the behavioral strategy of the artificial rabbit algorithm, and t-distribution perturbation mutation is applied to improve the stealing behavior. Case analysis shows that the maximum, minimum, and average values obtained by the improved dung beetle algorithm are better than those of the original algorithm, demonstrating its effectiveness and superiority in solving the economic model.