Research On Electromagnetic Nondestructive Testing Of Defects In The Engine Blades Of Aircraft

Along with the development of Chinese aviation industry, and the updating cycle ofaircraft shortened continually, more and more strict performance requirements are putforward on aircraft engine. As one of the most important parts of the aircraft engine, thequality and performance of blade directly relate to the normal operation of the aircraft.Accurate detection of the structural integrity and state of blades, timely detection andprevention of hazards, and improvement of engine safety has become a common focus in thefield of non-destructive testing. The blade at work is affected by a variety of factors, includingalternating load, hot corrosion, high temperature oxidation and mechanical wear, and mostprone to fatigue crack damage. Therefore, this paper combines the "twelfth five-year"national defense fund programme "****blade micro defects NDT technology" aiming atdetection of the micro crack defects on the surface and subsurface of the blade.On the otherhand, in contrast to the advantages and disadvantages of common methods for detectingsurface cracks on the blade, the paper also focuses on sensors designation, the blade cracksignals procession, and the crack defect recognition and classification, based on ACFM andeddy current, the two electromagnetic techniques, as they are more suitable for the surfacecrack of aircraft engine blade. The main contents are as follows,(1) According to the characteristics and detection principles of ACFM electromagneticnondestructive testing technology, the ACFM magnetic fields of the aero engine bladeare analyzed, the interaction of alternating magnetic field and micro defects of aviationturbine blades is discussed, and the rotating incentive method for randomly distributeddefects is put forward, then the rotating magnetic field model for defecting cracks of theengine blade, combined with finite element numerical analysis method, is established byAnsoft simulation software. From the simulation results, it can be seen that the method canrealize the rotating magnetic field theoretically, which lays a foundation for experiments.(2) The ACFM detecting system of the aeroengine turbine blade has been developed.Based on the analysis of the present situation of the detection system, covering its basicproblems presently, the electromagnetic rotational excitation device, suitable for the defectsof the turbine blade, has been researched. The dedicated system circuit and the dataacquisition system have been researched.. It can be seen that the detection system developedin this paper is effective and reliable.(3) To satisfy the demand for detecting micro cracks on the blades surface, thispaper designs a kind of miniaturized differential sensor, namely the difference incentive andmeasurement eddy current sensor, used for blades with micro cracks on the smallcurvature surface. Experiments show that, the sensor has advantages of high sensitivity, goodanti-interference, and better suitability for detecting surface with micro cracks. On the basisof the sensor, scanning defect system has been designed with the single probe.(4) By the finite element simulation software, ANSOFT, a3D simulation model,belonging to the blade specimen with surface micro-crack damage, has been established, toachieve the simulation of magnetic field distribution state with influence parameters, such aslift-off height, changing and different sizes and positions of the micro-crack defect. According to the simulation results, adjustment and optimization of the sensor parameterhave been carried out.(5) Research on the eddy current array sensor of blade defects has been carried out.The manufacture cost of flexible array sensor is so high, which weighs againstthe development of basic experiments, so a kind of embedded planar coil array sensor is putforward. Its characteristic is that the planar coil made of PCB board is closely combined withrubber flexible substrate, which not only reduces the production cost, but makes the sensorsimilar with flexible arrays in principle and strictly. Then a method for the early basicexperiment of array sensor is provided(6) Theories and methods of extracting characteristics information from ACFM andeddy current testing signal have been researched. The detection software of signal dataacquisition and signal processing system of the aeroengine turbine blade based on GUI, havebeen developed. Realize data acquisition and processing of the blade detection signals withmulti-channel. Mirrorizing extension EMD/EEMD-the wavelet singularity analysis method,for the blade crack signals has been put forward. Blade surface defect imaging by embeddedelectromagnetic array sensor has been achieved.(7) Considering the shortcomings of neural network algorithm and the quantity limit ofcrack samples on the single blade specimen, the method for evaluating and classifying crackdefects of engine blade has been researched based on the support vector machine SVM theory.For the micro-cracks under different positions and sizes, different angles of the same size anddifferent positions of the same size, three kinds of classifiers have been established on thebasis of "one against one" strategy, to realize the effective classification of blade cracks.