Aiming at the problem of large-scale distributed photovoltaic and electric vehicle access distribution network impacting spatial load forecasting, this paper proposes a spatial load forecasting method for urban distribution network with large-scale electric vehicles participating in V2G considering the long-term roof distributed photovoltaic saturation installation and large-scale electric vehicles. It differentiates different communities, calculates the saturated installation area of the roof photovoltaic according to the corresponding floor area ratio and availability coefficient, and calculates the output of the photovoltaic by combining the historical radiation value data. Based on the improved parking generation rate model to predict the parking demand, and combined with the daily driving mileage, parking characteristics and charging and discharging strategy, the V2G load forecasting model of electric vehicle is established, and the Monte Carlo simulation is used to obtain the spatial and temporal distribution of V2G load. The improved load density index method is used to predict the traditional daily load of the distribution network considering the timing. Taking a planning area as an example, the prediction results show that the roof distributed PV and electric vehicle V2G have a great influence on the spatial load forecasting results of the distribution network, and the degree of impact on the load of different communities is different. This work is supported by National Natural Science Foundation of China (No. 51577058).