The uncertainty of measurement makes the estimation accuracy of traditional methods difficult to meet the requirements of scheduling. The interval analysis is used to describe the measurement uncertainty. The dynamic interval optimization model of state variables of active distribution network is established, and the dynamic feasible region of state variables is obtained by using linear programming method. Based on the uncertainty of measurement and taking the highest accuracy of measurement point as the objective function, the robust state estimation model is established and the state estimation value is solved by modern interior point method. Taking the dynamic feasible region of the state variable as the constraint and the median value of the interval as the initial value, compared with the actual physical constraint and the maximum value of the historical data as the constraint, the range of solving the state variable is reduced and the amount of calculation is reduced. The feasibility and effectiveness of this method are verified by IEEE 30, IEEE 118 and actual distribution network systems. Combined with the evaluation function, a new evaluation index is proposed in the test of tolerance. Under three different constraints, the results of state estimation are compared, and the feasibility and accuracy of the proposed method are proved. At the same time, it can effectively track and estimate the voltage of the distributed generation access node.