Designing strategy for multi-agent system based large structural health monitoring

Density and different kinds of sensors have to be adopted to have a reliable monitoring of large scale engineering structures. To deal with the diverse, heterogeneous and distributed information obtained by these sensors at different sites on the structure, multi-agent system (MAS) based structural health monitoring (SHM) technology is researched in detail in this paper. A general framework to design the multi-agent system for large scale structural health monitoring is proposed. Compared to past work, this framework is typical and flexible, which is easy to be referenced by other researchers to develop their own multi-agent based SHM system. The design strategy of the MAS based SHM is discussed in detail, including function design, ontology design, individual agent design, facilitator design, society behavior and learning behavior design. A MAS based composite SHM system is researched as a case study to evaluate the efficiency of the proposed MAS based SHM system design strategy and the advantages of the MAS based SHM system. Very detail introduction of the implementation of this system is given in the paper.