Networks are evolving continuously in terms of number of users and services offered. This evolution makes them more complex and consequently more vulnerable to various kinds of attacks. Therefore, security management of these networks requires more intelligent and sophisticated models. Multi-agents systems and intelligent agent technology provide a powerful paradigm for the modeling and development of complex systems. So, to adapt to networks' evolution and complexity, we propose an intelligent agent-based security management model. In this paper, we describe our model of security management of networks, which is based on security policies, security events and BDI concepts.