Large structural impact localization based on multi-agent system

In practical applications of structural health monitoring, a huge amount of distributed sensors are usually used to monitor structures of large dimensions. In order to obtain fast and accurate evaluation of a structure, a multi-agent system is introduced to manage different sensor sets and to fuse distributed information. In this paper, a multi-agent system based on impact location is presented to deal with the impact load localization problem for large-scale structures. The monitoring system firstly detects whether an impact event happens in the monitored subregion, and focuses on the impact source on the sub-region boundary to obtain the sensor network data with blackboard systems. Then the collaborative evaluation of both the acoustic emission and the inverse analysis localization method is employed to obtain precise and fast localization result. Finally, a reliable assessment for the whole structure is provided by fusing evaluation results from the sub-regions. The performance of the proposed multi-agent system is illustrated by means of experimental on a large aerospace aluminum plate structure. Extensive testing of the proposed system demonstrated its effectiveness for the impact load localization in each sub-region, particularly for impacts lying next to the borders of the sub-regions.