In order to fully tap the flexible regulation potential of electric vehicles in residential areas and relieve the pressure caused by disorderly charging on power grid operation and new energy consumption, a response model was improved and constructed for electric vehicle charging behavior in response to the shortcomings of traditional time of use electricity pricing period division and pricing optimization. For the first time, a quantitative model for the spatiotemporal coupling degree of new energy charging load was established, and a multi-objective time of use electricity pricing optimization model with improved fuzzy C-means (FCM) clustering algorithm and improved particle swarm optimization (PSO) algorithm was proposed. The optimization mechanism of time of use electricity pricing for residents’ slow charging scenarios was studied, which took into account power grid safety, new energy consumption, user costs, and charging pile operator benefits. The results show that the improved fuzzy C-means clustering method can realize the reasonable division of peak, average and valley periods guided by new energy consumption. The time of use pricing scheme formed on this basis can effectively guide the transfer of charging load from the peak period of power grid to the new energy friendly period, reduce the peak valley difference of power grid by 28.18 MW, and improve the overall coupling degree of new energy charging load from -0.274 to 0.457. At the same time, under the premise of controllable increase in user charging costs, the revenue of charging pile operators increases significantly. It can be seen that the proposed method for optimizing the time of use electricity price of residential charging pile can effectively stimulate the flexible adjustment ability of electric vehicles, and realize the collaborative optimization of new energy consumption, stable operation of power grid and multi-agent economic benefits.