This paper addresses a robust fuzzy control problem for an uncertain large-scale nonlinear system using decentralized static output-feedback scheme for both continuous-time and discrete-time cases. In both cases, sufficient design conditions are derived for robust asymptotic stabilization in terms of linear matrix inequalities (LMIs), and are therefore easily tractable by convex optimization. An illustrative example, a two-area power system with parametric uncertainties and the valve-position limit nonlinearity is provided to verify the effectiveness of the proposed technique.