Research On Registration Technology In Adaptive Machining Of Titanium Alloy Fan Blade

Manufacturing precision of titanium alloy fan blade has important effect on the performance of turbofan engine.The blade blank is made by superplastic forming/diffusion connection process,and the problems of uneven distribution of machining allowance and large size difference between blanks exist,so it’s needed to research on adaptive machining method aiming at characteristics of blade blank.Based on the adaptive machining requirements of titanium alloy fan blades,a registration algorithm is proposed to meet the blade profile characteristics.Furthermore,the constraint condition of machining allowance is incorporated into the registration model,and the constraint registration algorithm is proposed so that the automatic location and allowance optimization of blade blank are realized.Based on the results of registration,the blade machining path is non-rigidly transformed to eliminate the step of blade inlet/exhaust side.This paper mainly completed the following work:(1)Research on basic algorithm for fast and accurate registration of rigid bodyThe accuracy and efficiency of the existing registration algorithms are studied,and the registration algorithm is proposed to adapt to the fan blade shape features.The principal component analysis method is modified,and the main direction of the possible reverse is corrected to solve the defects of principal component analysis registration.The iterative closest point algorithm is improved,and the point cloud topology is reconstructed based on kd-tree to improve search efficiency.The improved principal component analysis method is used for rough registration,and the rough registration result is used as the input of the improved iterative closest point algorithm.(2)Optimization of blade machining allowance based on constrained registrationThe basic constraint registration algorithm is used to analyze the machinability of blanks.Aiming at the problem of large torsion error of blade blank,the torsional curve of the blank along the stacking axis is calculated,and the blade theoretical model is modified based on the torsional curve.Furthermore,the complex constraint registration algorithm is proposed to realize the automatic positioning of the blank,and meet the processing margin and tolerance requirements.(3)Optimization of blade machining trajectory based on non-rigid variationBy using the properties of Bezier curves and surfaces,twice continuously Bezier curves and surfaces with the cutter location and measured data of blank are constructed.Based on the constructed curve,cutter location is adjusted to be smoothly connected with the blank,and the step of milled exhaust side is eliminated.Normal vector of cutter location is calculated and tool offset is applied to cutter location to get the cutter head center position,finally generate the tool path.(4)Verification experiment of milling inlet and exhaust side of flat plate test pieceThe flat plate test piece is measured in machine and the data and theoretical model are restricted registered and processing allowance is allocated.Cutter location is adjusted to be smoothly connected with the blank to eliminate the step of milled exhaust side.Tool offset is applied to cutter location to get the cutter head center position,finally generate the tool path.