Present paper proposes a solution in the form of a Distributed Model Predictive Control (DMPC) strategy for excitation control in large scale networked applications, such as Multi-machine Power Systems (MMPS). Each MPC solves its own optimization problem using local decomposed model of the overall system, as it is more convenient to design control laws in distributed manner, based on only local measurements and reduced order dynamical model of the system. The coordination amongst these controllers is achieved through information exchange to obtain performance close to that of centralized MPC scheme. The effectiveness of the proposed DMPC is tested on a 3rd order model of 2-machine system with lossy transmission lines and loads.