The invention discloses a relay protection setting calculating method of a power plant electric power system. When a computer algorithm is adopted, power network models (the primary wiring relations of power networks) need to be converted into mathematic models of the power networks, a conventionally adopted method involves converting all the power network models, yet the method provided by the invention only involves converting a part of the power network models. The method provided by the invention greatly reduces the size of the loaded mathematic models, reduces the dimension number of a node admittance matrix, shortens the matrix inversion time and greatly accelerates the operation speed. According to the relay protection setting calculating method of the power plant electric power system, the power network models are loaded towards a last load layer according to the position of a fault branch, the power network models are not loaded for devices at the same layer or at a lower load layer, and the power network models are all loaded for a primary wiring portion.