The model of a 60 kW permanent magnet synchronous motor for electric vehicles was established. The optimal number of flow channels and the range of flow channel parameters were acquired by theoretical analysis. The cooling channel width aand height bwere taken as design variables, and the motor temperature rise and the inlet outlet pressure difference were the objective functions. Through the Isight integrated mesh deformation software Sculptor and CFD calculation software Fluent, the sample points were created by using the Latin hypercube design method, and the data were automatically submitted for simulation calculation. Based on the obtained data, a response surface model was established, and the multi island genetic algorithm was used to optimize the approximate model. After optimization, the motor temperature rise and inlet outlet pressure difference were decreased by 6.01% and 6.81%, respectively. The performance and safe operation of the motor were guaranteed.