The invention discloses a natural language processing-based electrical secondary equipment modeling method and system. The method comprises the steps of defining different secondary equipment types and name rules of relay protection elements; identifying primary equipment type and name, secondary equipment types and names and signal types from signal point names; at the moment, analyzing the relayprotection signal point names one by one, adding the identified secondary equipment to a model, and adding information of association with primary equipment and relay protection signal points; and after the relay protection signal points are maintained and modified, updating a secondary equipment model. The modeling method and system has the beneficial effects that existing power grid model resources can be effectively utilized; the problem of electrical secondary equipment automatic modeling is solved; the working efficiency is improved; automatic association between the secondary equipmentand the primary device and between the secondary equipment and the relay protection signal points is realized; and possible errors during manual association are avoided.