Artificial immune systems (AIS) is a complicated system with the ability of self-adapting, self-learning, self-organizing, parallel processing and distributed coordinating, and it also has the basic function to distinguish self and non-self and clean non-self. One significant feature of the theory immunology is the ability to adapt to changing environments and dynamically learning continuously. Inspired by the theory of artificial immune systems, a novel model of Agents of Network Danger Evaluation is presented. The concepts and formal definitions of immune cells are given, and dynamically evaluative equations for self, antigen, immune tolerance, mature-lymphocyte lifecycle and immune memory are presented, and the hierarchical and distributed management framework of the proposed model are built. Furthermore, the idea of dynamic immunological surveillance period is applied for enhancing the self-learning ability to adapt continuously variety environments. The experimental results show that the proposed model has the features of real-time processing that provide a good solution for network surveillance.