Abstract: To enhance the accuracy and dynamic response capability of distribution network state estimation, this paper takes a 10 kV medium-voltage distribution network in a certain city as the research object, analyzing the shortcomings of traditional static estimation methods in data asynchronous processing and dynamic response. By introducing the wide-area measurement system to construct a dynamic state estimation model, an improved unscented Kalman filtering algorithm and model correction measures considering topological changes are proposed. The research results show that this method can effectively integrate high-precision synchronous measurement data, significantly reduce the estimation error, and solve the problems of high pseudo-measurement dependency and lagging dynamic tracking.