Abstract: The large-scale interconnection of distributed generators（DGs） makes the medium-and low-voltage distribution network have the ability to participate in optimal dispatching, but the randomness of DG output and the complexity of power flow distribution bring new challenges to the reactive power optimization of medium-and low-voltage distribution networks. Considering the practical problem of incomplete station area topology information and the limitations of traditional cooperative optimization, a distributed cooperative reactive power optimization strategy for medium-and low-voltage distribution networks is proposed in this paper. Firstly, the feeder physical model and station area fitting model are established for the feeder and station area, respectively.Then, the master-slave splitting method is used to decompose the global reactive power optimization model, and the power flow fitting model of uncontrollable station area is introduced into the feeder reactive power optimization model, so as to make full use of feeder resources, improve the voltage quality of each node in the uncontrollable station area, solve the limitations of cooperative optimization based on physical model, and improve the level of cooperative reactive power optimization. The simulation results of an example verify the effectiveness and superiority of the proposed strategy.