In order to solve the requirements of scheduling automation system for access control and the defects of access control model, an automatic access control model of power information system based on intelligent semantic network was designed. The overall structure of access control model of power information system is given. The detector is designed by artificial intelligence technology, combining artificial neural network and artificial immune algorithm, which provides the basis for checking access request module. By checking the access request module to find out whether there is a detector matching the access request, to judge whether the access request is legitimate. The verification results show that, the model can effectively support the access and modification of legitimate users and prevent the access of illegal users with high control precision.