Surface Features Analysis And Damage Identification Of Steel Structure Based On Electromagnetically Excitation Infrared Thermography

Cracks are the most common and harmful defects in steel structures.Electromagnetically excitation infrared thermography,as a new nondestructive testing technology,has lots of advantages.For example,it has large detection areas(high detection efficiency),high detection sensitivity and can locate crack damage directly.It has a good application potential in detecting surface cracks of steel structures.In the practical application,a rough surface state cannot be ignored.Using electromagnetically excitation infrared thermography,temperature distributions of surface cracks at different characteristics are analyzed by three parts: the theoretical analysis,simulation and experimental validation.Then,a crack identification algorithm,based on crack morphology and temperature gradient,is proposed to determine the location and size of the crack.It has great significances in surface crack detection.The main research content is as follows:(1)Analyze surface temperature distribution of steel structure by simulation.On the basic of surface crack mechanism analysis,different steel structure models can be built by COMSOL.Then we can get induced current and temperature distribution in the surface of steel structures under different surface features.Analysis results show that the rough surface can cause temperature anomaly area,not caused by cracks.It will affect the accuracy of tests.(2)Propose a crack identification algorithm based on crack morphology and temperature gradient.The algorithm restrains the influence of the steel structure surface features by thermal response differential normalization method.And then the abnormal temperature distribution are classified and marked by using the threshold segmentation and edge detection methods.Finally through calculation and comparison of the regional aspect ratio and edge temperature gradient,crack identification and extraction of crack size information can be realized.(3)Study surface crack detection of steel structure by using electromagnetically excitation infrared thermography.Surface temperature distributions are verified through lots of tests where use steel plate specimens at different rough surface characteristics.Then analyze thermal image graph by MATLAB,and the accuracy of crack identification algorithms based on crack morphology and temperature gradient gets verified.Results show that the algorithm can eliminate the influence of different surface roughness features,accurately identify the location and size of cracks,and have a high reduction degree of crack information.Through above research,validity of the crack identification algorithm based on the crack morphology and temperature gradient is verified,which lays foundation for the application of electromagnetically excitation infrared thermography in the field of steel structure surface crack detection.