The smart electricity meter-based measurement can provide a new model for the management of power grid in the future. This paper presents a new state estimation method, which uses smart electricity meter measurement to monitor the distribution network. The analytical expressions required to calculate the uncertainty of voltage and power are derived, which is a function of the number of smart electricity meters and their accuracy characteristics. The transformer is modeled in order to fully apply the state evaluation results of low-voltage power grid to the state evaluation of medium-voltage power grid. The uncertainty propagation theory is used to coordinate the state evaluation level to maximize the accuracy performance. The accuracy performance of different levels of state estimation of distribution system is given. The cloud architecture for interconnecting different voltage level evaluation models is also introduced. The proposed method is tested in the simulation. The results show that the proposed method can not only ensure the evaluation accuracy, but also significantly reduce the calculation time. It is suitable for real-time state evaluation and monitoring of distribution network.