Study On The Fiber Structure Monitoring System Based On Multi-agent Collaboration And Rubber Curing Research

In modern engineering fields such as aviation aircraft and so on, function and structure of large complex machinery equipment and use environment has become increasingly diversif ied and complicated. The number of sensors that structural health monitoring needed is increasing, and conventional structural health monitor ing system is difficult to meet the requirements of the precision and reliability. As the new research achievement in the field of artificial intelligence, main idea of multi-agent collaboration technology is to deal with complex problems which are beyond single agent's capability and knowledge through multi-agents collaboration. This article was done from two aspects, theory and experiment, and applied multi-agent technology to the field of structural health monitoring. Using large-sized aviation aluminum plate as the monitoring object, fiber intelligent structural health monitoring system was constructed based on multi-agent technology, and the feasibility and reliability of the multi-agent system were studied and discussed which were verified by experiment. The main research work of this article includes the following several parts:1) Research status of the multi-agent technology and its application in structural health monitoring research at home and abroad was analyzed and introduced. And the research background and progress of rubber curing monitoring were briefly summarized.2) The relevant knowledge and the working principle of the agent and the multi-agent theory were expounded, and the functional definitions of every agent in the multi-agent structural health monitoring system were made. On this basis, the frame of fiber impact monitoring system based on the multi-agent collaboration was constructed.3) Aiming at accurate identif ication problem for impact load of large structure, two positioning monitoring methods, the maximum inner product matching and the energy amplitude ratio, were proposed. Using the data fusion method, the advantages of two positioning methods were fused effectively to locate impact through blackboard collaboration in the multi-agent technology with epoxy plate and aluminum plate as experimental subject. The results of the experiments showed that the method could improve the accuracy and reliability of positioning impact load of the structure.4) To solve the problem that boundary area location accuracy was not high, by introducing dependence graph, the corresponding collaboration models of sensing agents were established in boundary area of two subregions and four subregions. Using large aviation aluminum plate as experimental object, the impact load monitor ing experiment showed these collaboration models could effectively improve the accuracy of impact load location, which could also improve the reliability of the whole monitoring system.5) Using composite material plate/aluminum alloy plate bonding structure as the experimental object, fiber Bragg grating sensors were proposed to be placed in the rubber to monitor changes in the solidif ication process. The corresponding encapsulated form of temperature compensation sensor was designed, and the experimental results show ed that fiber Bragg grating sensors could realize online monitoring of rubber curing, and were of great significance to the rubber quality evaluation.