Intelligent Structural State Retrieval Model Inspired from Human Memory

Long-term SHM systems need to control the scale of monitoring data storage and reduce the maintenance cost of sensor failures. Inspired by human memory, an intelligent structural state retrieval model is presented. First, the mental model of human memory is introduced and then the structural state recognition model and recall model are established. Methods of calculating familiarity values for recognition and similarity values for recall are discussed for the model. An experiment on strain monitoring was carried out to show the implementation process while demonstrating the effectiveness of the model. It is shown that the structural state recognition model can discard a large amount of redundant information and the structural state recall model can restore the missing data of failed sensors.