The controllable series injected voltage can be used to damp low frequency power and rotor angle oscillations. Conventional linear and NeuroFuzzy control schemes perform well only for a specific operating condition, or in the vicinity of the tuned operating point of highly nonlinear power system, due to their fixed parameters architecture. To improve the performance of the damping control, nonlinear behavior of power system must be incorporated via some nonlinear control scheme. This work presents an online adaptive nonlinear control paradigm by incorporating Legendre polynomial NNs in the consequent part of the conventional TSK structure. The proposed control scheme is successfully applied to damp local and inter-area modes of oscillations for different contingencies and operating conditions. The robustness of the proposed control scheme is validated using comparative analysis based on nonlinear time domain simulations and different performance indices.