With the high penetration of wind power and photovoltaic power generation, bilateral transactions between multiple microgrid operators (MGO) and the distribution system operator (DSO) lead to higher challenges to the distribution network. Considering the advantages of flexible distribution devices for fast and flexible regulation of power flow, this paper proposes a collaborative planning method that considers bilateral transactions between multiple MGOs and the DSO in a flexible network structure. First, a generalized Nash bargaining model for collaborative planning is constructed on the basis of the soft open point (SOP) planning. Then, the alternating optimization procedure (AOP) is used to transform the nonconvex and discontinuous problem into a bi-level iterative negotiation process. Each investor individually optimizes the planning scheme in the upper level, and achieves bilateral transaction in the lower level. Next, the lower-level problem is further equivalently transformed into two subproblems of power clearing and cost settlement, and then they are solved by alternating direction multiplier method respectively. Finally, the IEEE33-bus calculation analysis shows that the proposed planning scheme can effectively reduce the total social cost and the overall annual cost of each investor.