Skip to Main Content
One of the main preoccupations of the next generation power network, i.e. the smart grid, is improving the reliability of the power network. In order to achieve that, monitoring and control applications play an important role, in particular at a substation's level. Using a wireless sensor and actuator network is an attractive option for these applications. However, what is missing so far is that information collection should be complemented with the ability to transform significant data into actions in a decentralized manner. In this paper, we address this issue by adopting an agent-based approach. Autonomous agents are embodied in sensor nodes in order to evaluate the importance of collected data and choose an appropriate communication policy. In order to evaluate each phase of the proposed algorithm, different variants are compared with the legacy client/server scheme. Simulation results have shown better performance in terms of network congestion, energy consumption, and transmission delay.