The acquisition method for structural synthesized damage identification decision based on information fusion

Many of the structural health monitoring systems for the tall buildings, long span bridges and large space structures have been established in recent years, which are used to detect the damage of the structure, assess the behavior of the structure, and give the estimation of the structural health. Nevertheless, to make good use of the information obtained from the various types of sensors with the aim of giving the synthesized damage identification decision is still one of the most important challenging issues for structural health monitoring. The acceleration sensors and strain sensors are the most frequently installed in the structure for acquiring the global and the local properties of the structure which can give the damage identification decision separately; however, there are some disadvantages in each damage identification method based on global properties or local properties. In the paper, an acquisition method for structural synthesized damage identification decision is proposed; firstly, the substructure damage index is introduced in applying to the damage identification of the large space structure; secondly, the damage identification method based on global properties or local properties using the artificial neural network is introduced and inputs and outputs of the networks are given; thirdly, the information fusion is introduced to give the structural synthesized damage decision; at last, the paper takes a space shell structure as an example to manifest the validity and stability of the method by comparing the three damage identification results mentioned above.