Distributed generations (DGs) have been increasingly connected on the distribution systems, which cause stability concerns. This paper presents a novel adaptive generator control based on step-ahead model predictive methodology. This control, named model prediction adaptive control (MPAC), is built upon optimization of a specific performance index that is defined as a weighted combination of generator voltage deviation, mechanical and electrical torques mismatch, and incremental generator speed. This paper demonstrates the capability of MPAC that can improve significantly the power system stability. This paper offers a unique study and stability control of power distribution systems even with the operations of DGs, whereas most studies in the literatures were focused on the transmission-level power systems.Index Terms -AC generator excitation, adaptive control, power system stability, prediction methods. L. Wang (lwang@ee.ryerson.ca) and R. Cheung are with Ryerson University. 978-1-4244-1583-0/07/$25.00 ©2007 IEEE Lin Wang received her B.Eng., M.Eng., and Ph.D. degrees from Huazhong