Research On The Method Of Quantitative Evaluation Of Damage Based On Planar Electromagnetic Sensor Arrays

The damage to engine blade and bolt holes could lead to heavy flying accident. But if the damage can be previously tested, the accident could be entirely avoided. Traditional nondestructive testing methods have lots of problems, such as difficulty to test, low testing efficiency and uneasy to achieve quantitative evaluation of micro damage, etc. To overcome those problem, this thesis proposes the method of quantitative evaluation of damage based on planar electromagnetic sensor. At present, Meandering Winding Magnetometer(MWM) has been researched overseas while not researched at home.On the basis of MWM, the method of quantitative evaluation of damage based on planar electromagnetic sensor is studied in this thesis. The brief structure are as follows:Technologies for searching algorithm based on measurement grid are researched. This thesis studies the feature of measurement grid and present problems, proposes a nonlinear multivariate searching algorithm, which is a combination of rough search and fine search. Firstly, the initial grid cell is ascertained with dichotomy, and then the accurate goal grid cell is searched by the judgement of location relationship between grid cell and goal point. This algorithm can not only shorten the searching time by100times, but also reduce the failure ratio of searching by60times.Technologies for interpolation algorithm based on measurement grid are researched. This thesis studies previous interpolation algorithm and the distribution of conductivity and lift-off error. After the analysis of the relationship between features of measurement grid and error, this thesis presents a nonlinear multivariate interpolation algorithm to minish nonlinear error. Firstly, the correcting parameter of the grid center is calculated, and then the nonlinear error of the whole grid can be compensated according to the error distribution. Simulation results indicate that the algorithm could reduce the interpolation error of conductivity by80%and improve the quantitative evaluation accuracy of damage.Technologies for sensor response calibration are researched. Four types of calibration methods are researched, including air point calibration, air-short calibration, minimum variance calibration and air-minimum variance calibration. The last method could overcome the demerits of first three methods, such as low precision, hardness to implement and complex calculation.Finally, three parts experiments are established including messurement of lift-off, messurement of stress and the image of surface fatigue. The research results have been shown to be of validity.