Process Parameter Control And Optimization Strategy Of Aircraft Blade Defect Detection Based On Hot-air Active Infrared

Most of the aero-engine blades are made of high-performance titanium alloys,which are the core components of aero-engines that provide high propulsion.Long-term service in harsh working conditions may cause damage such as cracks and coating peeling.However,the existing conventional non-destructive testing technology is affected by the experience of technicians,and is prone to missed detection and false detection,which increases the difficulty of detection and maintenance.Therefore,efficient and accurate detection of blade near-surface defects is of great significance to ensure the flight safety of civil aircraft.In this thesis,an active infrared detection method using hot air as the excitation source is proposed to detect the near-surface defects of aero-engine blades.Based on the infrared feature identification test system of hot air excitation source aero-engine blade defects designed by the research group,the excitation temperature and excitation wind speed of the hot air excitation source are calibrated,and the infrared detection of the defect area of the specimen after excitation by different excitation parameters,defect characteristics and material properties is studied.Effect;build a hot air forced convection aero-engine blade model,analyze the coupling mechanism of the blade surface and the internal transient temperature distribution after hot air is applied,and simulate the excitation mode,excitation time,excitation parameters,defect characteristics and the change law of the surface temperature of the specimen With the help of GA-BP neural network,a comprehensive scoring model of hot air infrared detection parameters and detection effect was constructed,and a self-defined defect effect evaluation function was used to evaluate the significance and integrity of defects.Leaf inspection verification.The results show that the R2 of the GA-BP neural network is 0.99772,which has high accuracy;the average error of the actual and predicted comprehensive scores is 3.35%,which is within a reasonable error range,and it also shows that this hot air active infrared detection parameter optimization method is feasible.