| The aero-casting turbines/gas turbines blades are the cores of aero-engines.The manufacturing accuracy and profile quality determine the performance and service life of aero-engines.However,the walls of these parts are thin and the rigidity is weak.The features make the parts easy to deform during the processing.The traditional measurement mainly relies on the offline CMM(Coordinate Measuring Machine)and the contact OMM(On Machine Measurement).The offline CMM is expensive and the parts need to be transported and re-clamped repeatedly;The contact OMM is complicated and easy to interfere.To improve the efficiency of measurement,this paper focuses on the construction of on-site inspection system based on laser displacement sensor,the direction calibration of laser beam,the on-site inspection path generation method based on ant colony algorithm,data collection,data processing algorithms,etc.The main contents include:(1)Study the calibration method of laser beam direction and the integration method of multi-angle points.Considering the relative positional relationship between the machine coordinate system,the workpiece and the sensor coordinate system,the mathematical model of the beam direction calibration and the integration of multi-angle points method are established.A group of nonlinear equations are constructed by scanning the fixed ball.The nonlinear equations are solved by the Levenberg-Marquardt algorithm to get the unit direction vector of the laser beam and the center position of the standard ball.The calibration of laser beam direction is realized in high-precision.As for the integration,several sets of beam direction vectors and several sets of sphere center position are solved.The coordinate transformation parameters of the sensor coordinate system are calculated.The scanned points from different coordinate systems are transformed into the same coordinate system by the coordinate transformation.(2)Study the generation method of on-site inspection path based on triangular mesh.A grid algorithm method of points extraction from the STL triangle grid is proposed.The generation of theoretical path points based on triangular meshes is realized.A voxel filtering algorithm is used to homogenize and simplify the extracted points.Considering the constraint conditions of sensor,the position coordinate relationship between the spherical points and the spindle of the machine is analyzed.The path planning model for on-site inspection with a laser sensor is established and the optimization objective function is constructed.A path generation method based on ant colony algorithm is proposed.The scanning sequence of points is optimized.The composition and structure of the FANUC system macro program are analyzed.Then the optimized points are post-processed and derived to the NC code.(3)The on-site inspection system with a laser displacement sensor is established and the software is developed.The calibration of the laser beam and integration of multi-angle points experiments are carried out to verify the method proposed in this paper.The error analysis experiment based on a standard ball is carried out to optimize the scan depth and the incident angle.The diameter of single ball and center distance of the double ball are measured with the on-site inspection system.The results showed that the accuracy of diameter inspection can reach ±0.03 mm,and the accuracy of the center distance can reach ±0.01 mm.Then the measurement of a plane,a blade and a skin is carried out.The results show that this on-site inspection system is suitable for a variety of complex aerospace parts,and the inspection efficiency is improved. |
PDF Full Text Request