Abstract: The traditional charging load modeling for electric vehicles (EVs) usually adopts the method of sampling simulation for an individual EV, which fails to describe the macroscopic operating states formed by the interaction of the EV groups from the perspective of analysis mechanism. Therefore, this paper proposes a method to calculate the probability distribution of EVs' charging load based on the semi-dynamic traffic equilibrium model and the combined states of the charge (CSOC) probability calculation. Firstly, the traffic and charging characteristics of EVs are analyzed, and a feasible path set construction is proposed. Then, the traffic equilibrium theory is introduced to model the spatial distribution of the EVs, and a semi-dynamic traffic equilibrium model considering random utility is established to realize the balanced distribution of the macroscopic traffic flow. Furthermore, the variation of the states of charge of an EV group is analyzed theoretically, and the calculation model of the probability distribution of charging load base on the CSOC is established. Finally, the effectiveness of the proposed method is verified on the 13-node road network and the actual large road network respectively, and the influence of the penetration of the EV and road network structure on the probability distribution of charging load is analyzed.