A neuro-fuzzy method is used to estimate the departure from nucleate boiling (DNB) protection limit using the measured average temperature and pressure of a reactor core. The neuro-fuzzy system parameters are optimized by two learning methods. A genetic algorithm is used to optimize the antecedent parameters of the neuro-fuzzy inference system, and a least-squares algorithm is used to solve the consequent parameters. Two neuro-fuzzy inference systems are used according to the pressure and temperature regions. The proposed method, which is applied to the Yonggwang 3 and 4 nuclear power plants, has a 6.09% larger thermal margin than the conventional Westinghouse OTT DNB protection logic. This simple algorithm can provide good information for nuclear power plant operation and diagnosis by estimating the DNB protection limit each time step.