Abstract: In order to solve the problems of increased power grid loss, voltage and frequency fluctuation caused by disorderly integration of electric vehicles into multi-source microgrid, so as to reduce power quality, the disorderly access of electric vehicles can be changed into orderly access to the power grid for charging through optimal control, and the system network loss and voltage deviation can be improved by peak shaving and valley filling. The traditional optimal control algorithms, including iterative method, gradient method, Newton method and Lagrange method, are efficient and reliable, but they are highly dependent on the mathematical model. For example, they require the continuous derivation of the objective function, which can not be met in practical problems. This paper aims at the operation optimization of electric vehicles incorporated into multi-source microgrid. Microgrid considers electric vehicles, photovoltaic, wind power, fuel cells, steam turbines and energy storage systems, establishes the optimization model of collaborative scheduling between microgrid and electric vehicles, and optimizes the solution considering the two constraints of power generation operation cost and environmental governance cost. This paper improves the existing particle swarm optimization algorithm and proposes an improved hybrid particle swarm optimization algorithm to solve multiple constraints. Finally, an example is given to simulate the impact of electric vehicles on grid connected microgrid in different scenarios.