To incentivize electric vehicles (EVs) to participate in demand-side response to reduce the peak-to-valley difference in grid load and enhance the economic viability of EV electricity usage, the Monte Carlo method is used to simulate the unordered charging load of EVs. EVs are then categorized into three types based on whether they are regulated by the grid or guided by price signals. Subsequently, a dynamic time-of-use (TOU) pricing demand response model is established on the basis of the electricity price demand elasticity matrix, with the objectives of minimizing the mean square value of the grid's peak-to-valley difference and minimizing user charging and discharging costs. Using historical load data from a region in Hunan and segmenting the electricity prices for a day, simulation analysis is conducted to verify that dynamic TOU pricing considering real-time load feedback can effectively manage load fluctuations, and it has a more pronounced effect on reducing the peak-to-valley difference and lowering user electricity costs.