The invention discloses a
steel box girder evaluation method and
system based on
monitoring data, and the method comprises the steps of firstly carrying out the preprocessing of the
stress time history and temperature information of different monitoring points of a monitored
steel box girder, and carrying out the separation through an empirical mode
decomposition method, and obtaining a fast-change component and a slow-change component; secondly, extracting stress data according to a multiple
linear regression method, and carrying out mutual correction fusion on the stress data, the fast-change component and the slow-change component; then calculating
fatigue damage and fatigue life of the
steel box girder based on
monitoring data, and evaluating the service state of the steel
box girder with the fatigue life as an index; and meanwhile, building a steel
box girder stress evaluation system, so that the functions of data import, preprocessing, state evaluation and the like are realized. According to the method, the service performance of the main beam can be dynamically evaluated by using the existing
monitoring data, damage and diseases of the structure can be found in time, and a theoretical basis can be provided for intelligent early warning of the service state of the key component of the large-span cable-stayed bridge.