Some Research On Damage Localization Based On Lamb Wave

Steel is widely used in the structure, however, due to the external forces, the damage will be appeared gradually during the use of the product. If the damage is not found in time, it will affect the normal use of the structure and even the safety of life and property. Therefore, it should be found the damage early, and avoid the accidents.Steel detection based on Lamb wave is one of the prospect methods in the field of structural health monitoring. This paper uses the method of combination of the theory and experiment, and do some research on the damage detection, localization and optimization.According to the theory of elastic dynamics, deduce the Lamb wave dispersion equation based on the potential function. Solved the numerical solution by Matlab program, then draws the dispersion curve, and discussed the multi-modes and dispersion characteristics.Deducing the theory of displacement components and stress components,discussed the different defects sensitivity under the SO and AO mode. Proved the detection feasibility in the case of low-frequency thickness product through the experience, and use the variation of the peak damage factor proved the AO mode is suitable to detect the surface defects.This paper expounds the principle of the energy damping method, introduces the methods for calculate the energy of wave packet and points out the feasibility of using this method for damage localization.The paper obtains the principle of wave energy changes with the increase of damage depth based on the experimental analysis in the thin plate. Calculating the damage rate in different path by using the variation of wave energy, and using this method realize the preliminary localization of damage. It is proved the feasibility by the comparison of the actual damage location. It provides the reference of the monitoring for large area thin plate.The paper gives the necessity of the optimization for measurement points and introduces the method of number and position optimization in brief.By a four-points supported plate as an example, using a genetic algorithm calculate the available recognition unit number before and after the optimization. The optimization of sensor's number and position plays a significant role to improve the number of available recognition unit by contrast.