A hybrid GA-neurofuzzy damping control system for a unified power flow controller (UPFC) is proposed to enhance the transient and dynamic stability of a power system. The neurofuzzy damping controller is configured by employing a five layer feedforward fuzzy neural network. A novel "grandparenting" technique is employed for seeding the initial population to hasten the convergence speed of micro-GA. Also, a parallel micro-GA scheme is used to speed up the genetic algorithm search-space surfing process. The performance of the resulting neurofuzzy-based UPFC is validated in a multi-machine power system environment by means of digital simulation.