In this paper, a distributed control structure is presented for a nonlinear Medium-Voltage DC (MVDC) Shipboard Power System (SPS). Distributed control architecture has the advantages of less computational burden, high flexibility, and a good error tolerance. In this framework, each subsystem is controlled by a model predictive controller using local state variables and parameters, and also interaction variables from other subsystems shared through a coordinator. In the coordinator level, an optimization problem is iteratively solved to update a Lagrange multiplier vector to have a global optimal solution. The effectiveness of the proposed distributed control structure for a partitioned MVDC model is demonstrated by the simulation results. The performance analysis is accomplished by comparing centralized and distributed control methods on the global and partitioned models and considering different specifications in the MVDC system.