Abstract: To address the problems of large load peak and high operating costs brought by the large-scale charging of electric trucks to the power grid,a two-layer coordinated optimal dispatch method is proposed for peak shaving and valley filling.Firstly,load prediction of electric trucks is carried out by considering logistic characteristics,acquiring the driving behaviour of electric trucks,and obtaining charging load curves.Fuzzy C-mean clustering is used to divide the charging loads into dispatchable and non-dispatchable categories,and the schedulable potential is measured.Secondly,the minimum variance of total load is selected as the objective function.The upper layer optimization model is constructed,and the daily charging scheme of electric trucks is formulated.Then,the optimal daily charging scheme of the upper layer is transferred to the lower layer,and the lower layer optimization model with the lowest operating cost of the electric trucks as the objective function is constructed by considering the flexible time window constraints.The two-layer optimization dispatch is implemented with the variance of charging loads and the operating cost.Finally,the two-layer optimization model is solved by an adaptive particle swarm algorithm under the master-slave game framework,and the simulation verification is carried out using actual electric truck data in a logistics park.The results show that the proposed two-layer optimization dispatch method can effectively achieve smoothing the load profile of the grid while reducing the operating costs of electric trucks.