Three-coordinate-optical fusion blade detection device and detection method thereof

Abstract

The invention discloses a three-coordinate-optical fusion high-precision complex blade detection device and a detection method thereof, and belongs to the technical field of part detection methods. The blade detection device comprises a three-coordinate measuring machine, a four-axis driving device, an optical sensor, a blade clamp, an upper computer and a power supply module; based on the complex structure and curvature characteristics of the blade and the detection advantages of three-coordinate and optical sensors, the blade body section planning of the to-be-detected blade is divided into a blade basin, a blade back, a front edge and a rear edge, the front edge and the rear edge are measured by adopting a three-coordinate measuring machine with higher precision, and the blade basin and the blade back are measured by adopting the optical sensors with higher efficiency. Compared with a traditional detection method, the method comprehensively considers the detection precision and efficiency requirements, and achieves the high-precision and high-efficiency detection of the molded surface of the blade body through the cooperative measurement of the three-coordinate measuring machine and the optical sensor.

Description

Technical Field

[0001] The present invention belongs to the technical field of parts detection methods, and in particular relates to a three-coordinate-optical fusion blade detection device and a detection method thereof. Background Art

[0002] As the core parts of the engine, aircraft engine blades are characterized by a wide variety, large quantity, complex structure, high geometric precision requirements, and great processing difficulty. The manufacturing quality of the blades (dimensional accuracy, form and position tolerances, and surface quality of complex surfaces) directly affects the aerodynamic performance and service life of the engine. In addition, during the service process, the blades will be affected by various factors such as high temperature, high pressure, and corrosion, and are prone to deformation, cracks, wear, and corrosion, resulting in reduced performance of the blades and even serious failures. How to measure the complex spatial geometry of the blades with high precision and efficiency, evaluate the manufacturing quality, feedback control the processing deformation, and ensure the high-quality surface accuracy of the blades has become an important issue that needs to be urgently solved in the current aviation manufacturing industry.

[0003] At present, common blade inspection methods are divided into contact and non-contact methods. Contact measurement is represented by coordinate measuring machines (CMMs), which can realize multi-parameter and high-precision comprehensive measurement of blade profiles, but the measurement efficiency is low and the environmental adaptability is poor. Non-contact measurement mainly includes three-dimensional scanners and point laser profile measuring instruments, which have the advantages of high sampling rate and strong flexibility, but the detection accuracy of three-dimensional scanners is low. Point lasers always have the situation of missing points and distortion when detecting the leading and trailing edges of blades with small radius, especially unable to accurately measure large torsion angle and thin-walled aircraft engine compressor blades. Although the current non-contact optical detection is not as accurate as the data detected by CMM, the flexibility and high efficiency of non-contact measurement systems have led to its increasing application. How to give full play to the advantages of optical sensors (spectral confocal sensors, laser sensors and other non-contact optical measurement sensors) in measurement accuracy, sampling rate, high directivity and high brightness, while taking into account the detection accuracy of thin walls and complex structures, is a direction that needs to be broken through urgently. Therefore, it is necessary to carry out research on engine blade inspection technology based on the fusion of CMM and optical detection to achieve efficient and high-precision measurement and fusion of multi-source data. Summary of the invention

[0004] In view of the above problems, the present invention aims to provide a three-coordinate-optical fusion blade detection device and a detection method thereof, so as to solve the problem that the existing non-contact measurement technology cannot meet the needs of high-precision blade testing.

[0005] In order to achieve the above-mentioned object of the invention, the technical solution adopted by the present invention is as follows:

[0006] A three - coordinate and optical fusion blade detection device is provided, which includes a coordinate measuring machine, a four - axis driving device, an optical sensor, a blade fixture, a standard ball, a host computer and a power supply module;

[0007] The coordinate measuring machine includes a first horizontal platform;

[0008] The four - axis driving device includes a second horizontal platform arranged on the first horizontal platform. On the second horizontal platform, there are three linear axes X, Y and Z and a rotary axis C axis;

[0009] The bottom of the linear axis Z axis is installed on the linear axis X axis through a connecting plate; the optical sensor is installed on the linear axis Z axis through a sensor fixture; both the optical sensor and the four - axis driving device are electrically connected to the host computer and the power supply module;

[0010] The rotary axis C axis is installed on the linear axis Y axis, and a turntable is installed on the rotary axis C axis. On the turntable, there is a blade fixture and a standard ball; the optical sensor is located on one side of the blade fixture;

[0011] The optical sensor is a laser displacement sensor or a spectral confocal sensor.

[0012] Further, the linear axis X axis, the linear axis Y axis and the linear axis Z axis all include a motor, a grating scale, a ball screw module and a linear guide rail. The linear guide rail is arranged on the second horizontal platform, and the ball screw module and the grating scale are arranged on the linear guide rail; the motor is connected to the ball screw in the ball screw module; the ball screw nut seat in the ball screw module is connected to the connecting plate, the turntable or the sensor fixture. The grating scale is used to detect the position of the ball screw nut seat and the components connected thereto.

[0013] Further, the resolution of the linear axis X axis, the linear axis Y axis and the linear axis Z axis is 0.1μm, and the repeat positioning accuracy is 1μm; the resolution of the rotary axis C axis is 70μrad, and the repeat positioning accuracy is 60μrad.

[0014] Further, the power supply module includes a 220V AC power supply and a 24V DC power supply; the 220V AC power supply supplies power to the four - axis driving device; the 24V DC power supply supplies power to the optical sensor.

[0015] The present invention also provides a detection method for a three - coordinate and optical fusion blade detection device, including the steps:

[0016] S1. Set the four-axis drive device on the first horizontal platform of the coordinate measuring machine, and place the blade to be detected in the blade fixture. Detect the three reference planes of the second horizontal platform through the coordinate measuring machine, so as to establish an optical detection reference coordinate system O1 in the coordinate measuring machine; detect the blade tenon through the coordinate measuring machine to construct a blade workpiece reference coordinate system O2; the origin of the optical detection reference coordinate system O1 is at the intersection of the axis of the rotary axis C and the installation surface of the second horizontal platform; the blade workpiece reference coordinate system O2 is the design reference coordinate system of the blade to be detected.

[0017] S2. Taking the optical detection reference coordinate system O1 as the reference, output the attitude of the blade workpiece reference coordinate system O2 relative to the optical detection reference coordinate system O1, and then correct the motion coordinate data of the four-axis drive device to ensure that the leading and trailing edges detected by the coordinate measuring machine and the blade bowl and blade back detected by the optical sensor are on the same cross-section of the blade.

[0018] S3. Establish an optical sensor measurement starting point coordinate system O3; the origin of the optical sensor measurement starting point coordinate system O3 is the measurement starting point of the optical sensor when the linear axis X and the linear axis Z are at the zero position, and the x, y, and z axes of the optical sensor measurement starting point coordinate system O3 are parallel to the x, y, and z axes of the optical detection reference coordinate system O1.

[0019] S4. According to the established optical detection reference coordinate system O1 and the blade workpiece reference coordinate system O2 above, install a standard ball in the space of the optical sensor measurement starting point coordinate system O3. Use the coordinate measuring machine to detect the standard ball to obtain the center coordinate and radius of the standard ball under the optical detection reference coordinate system O1, and use the optical sensor to detect the standard ball at the same position to obtain the center coordinate and radius of the standard ball under the optical sensor measurement starting point coordinate system O3; compare the detection data of the standard ball by the coordinate measuring machine and the optical sensor to find the relative position of the optical sensor measurement starting point coordinate system O3 relative to the optical detection reference coordinate system O1, and complete the calibration of the measurement starting point and the coordinate system fusion when the optical sensor detects the blade.

[0020] S5. Use the coordinate measuring machine to measure the blade profile point cloud data A of the leading and trailing edges of the blade to be detected in the blade workpiece reference coordinate system O2.

[0021] S6. Through the linkage of the linear axis X, the linear axis Y, the linear axis Z and the rotary axis C, use the optical sensor to measure the blade profile point cloud data B of the blade bowl and blade back in the optical detection reference coordinate system O1.

[0022] S7. Transform the point cloud data B to the blade workpiece reference coordinate system O2 through the space transformation matrix to obtain the point cloud data B`, and complete the rough registration.

[0023] S8. Taking the point cloud data A as a reference, the closest point matching between the point cloud data B` and A is performed through the ICP algorithm to obtain the point cloud data B. matched , and the data fusion fine registration is completed.

[0024] S9. Perform B-spline curve fitting on the finely registered point cloud data B matched . The curve after fitting is the cross-sectional contour curve of the measured blade.

[0025] Further, in step S2, the posture includes Δx, Δy, Δz, Δθx, Δθy, and Δθz. Among them, Δx is the X-axis coordinate of the origin of the blade workpiece reference coordinate system O2 in the optical detection reference coordinate system O1, Δy is the Y-axis coordinate of the origin of the blade workpiece reference coordinate system O2 in the optical detection reference coordinate system O1, Δz is the Z-axis coordinate of the origin of the blade workpiece reference coordinate system O2 in the optical detection reference coordinate system O1, Δθx is the rotation angle of the X-axis in the blade workpiece reference coordinate system O2 relative to the X-axis of the optical detection reference coordinate system O1, Δθy is the rotation angle of the Y-axis in the blade workpiece reference coordinate system O2 relative to the Y-axis in the optical detection reference coordinate system O1, and Δθz is the rotation angle of the Z-axis in the blade workpiece reference coordinate system O2 relative to the Z-axis in the optical detection reference coordinate system O1.

[0026] Further, in step S6, the method for using the optical sensor to measure the blade surface point cloud data B of the blade's suction side and pressure side in the optical detection reference coordinate system O1 through the linkage of the linear axis X-axis, linear axis Y-axis, linear axis Z-axis, and rotary axis C-axis is as follows:

[0027] When measuring the blade, considering that the linearity of spectral confocal, laser and other displacement sensors is the best at the midpoint of the measurement range, and taking into account the fluctuation of the distance during measurement, the ranging reference value during the measurement of the optical sensor is set to the midpoint M0 of the measurement range. When performing blade detection, after recording the displacement values X0, Y0, Z0, C0 of each moving axis relative to the measurement starting point and the distance M from the optical sensor to the blade surface, the three-dimensional point cloud data B of the blade surface is obtained according to the following formula:

[0028]

[0029] z = Z0.

[0030] Further, in step S7, the calculation formula of the point cloud data B` is:

[0031] B` = R·B(x, y, z)+T0;

[0032]

[0033] Where, Δx is the X-axis coordinate of the origin of the reference coordinate system O2 of the blade workpiece in the optical detection reference coordinate system O1, Δy is the Y-axis coordinate of the origin of the reference coordinate system O2 of the blade workpiece in the optical detection reference coordinate system O1, Δz is the Z-axis coordinate of the origin of the reference coordinate system O2 of the blade workpiece in the optical detection reference coordinate system O1, Δθx is the rotation angle of the X-axis in the reference coordinate system O2 of the blade workpiece relative to the X-axis of the optical detection reference coordinate system O1, Δθy is the rotation angle of the Y-axis in the reference coordinate system O2 of the blade workpiece relative to the Y-axis in the optical detection reference coordinate system O1, and Δθz is the rotation angle of the Z-axis in the reference coordinate system O2 of the blade workpiece relative to the Z-axis in the optical detection reference coordinate system O1.

[0034] Further, in step S8, the point cloud data B matched is obtained by the following method:

[0035] For the matched point cloud data A and B matched , in order to solve for R and T such that RB matched + T - A is minimized, the function is defined as:

[0036]

[0037] Based on singular value decomposition to solve for the spatial coordinate transformation matrices R and T of the matched point cloud data A and B when J is minimized; the process is as follows: matched The process is as follows:

[0038] To minimize J, let:

[0039]

[0040] Denote the weighted centroids of the point cloud data A and B as:

[0041]

[0042] Substitute T into J to get:

[0043]

[0044] Denote Then:

[0045]

[0046] At this time,

[0047] , and

[0048] Define the covariance matrix S = XWY T , perform singular value decomposition on S to get S = UΣV T Then:

[0049] tr(RXWY T ) = tr(RS) = tr(RU∑V T ) = tr(∑V T RU);

[0050] Since U, R, and V are all orthogonal matrices, let M = V T RU, then M is also an orthogonal matrix. According to the properties of orthogonal matrices, each element |m ij | in matrix M is ≤ 1. Therefore, when and only when M = V T RU = E, tr(ΣV T RU) reaches its maximum value, that is reaches its maximum value, and at this time R = VU T ;

[0051] For the obtained R and T, by comparing and to determine whether R and T converge, thereby controlling the output of R and T.

[0052] The detection method of a three - coordinate - optical fusion blade detection device in the present invention is mainly applied to a three - coordinate - optical collaborative measurement system with a characteristic structure like blades, etc., which present complex geometric contours and uneven curvature radii at different positions. The three - coordinate is used to detect the standard ball to obtain the center coordinates and radius of the standard ball relative to the optical detection reference coordinate system O1, and the optical sensor is used to detect the standard ball at the same position to obtain the center position and radius of the standard ball relative to the optical sensor measurement starting coordinate system O3 with the origin at the optical sensor and each coordinate axis parallel to the optical detection reference coordinate system O1. By comparing the detection data of the standard ball by the coordinate measuring machine and the optical sensor, the relative position of the optical sensor measurement starting coordinate system O3 relative to the optical detection reference coordinate system O1 can be found, and the measurement starting point calibration and coordinate system fusion for the optical sensor to detect the blade can be completed; the more accurate coordinate measuring machine is used to measure the leading and trailing edge parts with smaller curvature radii and finer structures in the blade; the more efficient optical sensor is used to measure the blade basin and blade back parts with larger curvature radii. Based on the complex structure and curvature characteristics of the blade and the detection advantages of the three - coordinate and the optical sensor, the blade body cross - section to be detected is divided into the blade basin, blade back, leading edge, and trailing edge. The more accurate coordinate measuring machine is used to measure the leading and trailing edge parts, and the more efficient optical sensor is used to measure the blade basin and blade back. The high - precision and high - efficiency detection of the blade body profile is realized through the collaborative measurement of the coordinate measuring machine and the optical sensor.

[0053] The beneficial effects of the present invention are:

[0054] 1. A three - coordinate - optical fusion blade detection device in the present invention has the advantages of flexibility, easy disassembly, and high precision. When high - precision metrological detection of complex and delicate parts such as blades is required, the optical detection device can be placed on the detection platform of the coordinate measuring machine to complete the fusion measurement. When the precision requirement is not high and the structure of the detected part is simple, the optical detection system can be used alone to complete the detection.

[0055] 2. The detection method of a three - coordinate - optical fusion blade detection device in the present invention is proposed according to the complex structure and curvature characteristics of the blade. Compared with the traditional detection method, it comprehensively considers the requirements of detection precision and efficiency. In the fusion of three - coordinate and optical detection data, the relative positions of each reference are measured with high precision by the three - coordinate to complete the rough registration of the data, and the ICP algorithm is used to complete the fine registration and fusion of the data, which can meet the high - precision blade detection requirements. Brief Description of the Drawings

[0056] Figure 1 It is a structural diagram of a three - coordinate - optical fusion blade detection system.

[0057] Figure 2 It is a structural diagram of the optical detection device.

[0058] Figure 3 It is a schematic diagram of the reference coordinate systems O1 - O3.

[0059] Among them, 1. Blade to be detected; 2. Blade fixture; 21. Standard ball; 3. Rotary table; 4. Linear axis Y; 5. Second horizontal platform; 6. Linear axis X; 7. Connecting plate; 8. Linear axis Z; 9. Optical sensor; 91. Laser displacement sensor; 92. Spectral confocal sensor; 10. Sensor fixture; 11. Coordinate measuring machine; 111. Coordinate measuring machine moving axis; 112. Coordinate measuring machine probe; 12. Motor; 13. Grating scale. Detailed Embodiments

[0060] The following describes the detailed embodiments of the present invention to facilitate those skilled in the art of this technology to understand the present invention. However, it should be clear that the present invention is not limited to the scope of the detailed embodiments. For those of ordinary skill in the art of this technology, as long as various changes are within the spirit and scope of the present invention defined and determined by the appended claims, these changes are obvious, and all inventions created using the concept of the present invention are within the scope of protection.

[0061] Such as Figures 1 - 2As shown in the figure, the present invention provides a high-precision complex blade detection device integrating coordinate measuring machine and optics, which includes a coordinate measuring machine 11, a four-axis driving device, an optical sensor 9, a blade fixture 2, a standard ball 21, a host computer and a power supply module; the coordinate measuring machine 11 includes a first horizontal platform, a coordinate measuring machine moving axis 111 and a coordinate measuring machine probe 112.

[0062] The four-axis driving device is arranged on a second horizontal platform 5 provided on the first horizontal platform. The second horizontal platform 5 is provided with three linear axes X, Y and Z and a rotary axis C axis; the bottom of the linear axis Z axis 8 is installed on the linear axis X axis 6 through a connecting plate 7; the optical sensor 9 is installed on the linear axis Z axis 8 through a sensor fixture 10; both the optical sensor 9 and the four-axis driving device are electrically connected to the host computer and the power supply module.

[0063] The rotary axis C axis is installed on the linear axis Y axis 4, and a turntable 3 is installed on the rotary axis C axis. A blade fixture 2 and the standard ball 21 are arranged on the turntable 3; the optical sensor 9 is located on one side of the blade fixture 2; the optical sensor 9 is a laser displacement sensor 91 or a spectral confocal sensor 92. The optical sensor 9 is used to measure the distance from the starting point to the measured point on the blade body cross-section, so as to obtain the three-dimensional coordinates of the measured point on the blade body cross-section through space coordinate transformation in combination with the axis information.

[0064] Specifically, the linear axis X axis 6, the linear axis Y axis 4 and the linear axis Z axis 8 all include a motor 12, a grating scale 13, a ball screw module and a linear guide rail. The linear guide rail is arranged on the second horizontal platform 5, and the ball screw module and the grating scale 13 are arranged on the linear guide rail; the motor 12 is connected to the ball screw in the ball screw module; the ball screw nut seat in the ball screw module is connected to the connecting plate 7, the turntable 3 or the sensor fixture 10. The ball screw module converts the rotational motion of the motor 12 into a linear reciprocating motion. The grating scale 13 is used to detect the position of the ball screw nut seat and the components connected thereto; the structures and motion principles of the linear axis X axis 6, the linear axis Y axis 4 and the linear axis Z axis 8 are prior arts and will not be elaborated here.

[0065] Further, the resolutions of the linear axis X axis 6, the linear axis Y axis 4 and the linear axis Z axis 8 are 0.1 μm, and the repeat positioning accuracies are 1 μm; the resolution of the rotary axis C axis is 70 μrad, and the repeat positioning accuracy is 60 μrad.

[0066] Further, the power supply module includes a 220V AC power supply and a 24V DC power supply; the 220V AC power supply supplies power to the four-axis driving device; the 24V DC power supply supplies power to the optical sensor 9.

[0067] A three - coordinate - optical fusion blade detection device in the present invention has the advantages of flexibility, easy disassembly, and high precision. When high - precision metrological detection of complex and delicate parts such as blades is required, the optical detection device can be placed on the detection platform of the coordinate measuring machine 11 to complete the fusion measurement. When the precision requirement is not high and the structure of the detected part is simple, the optical detection system can be used alone to complete the detection.

[0068] The present invention also provides a detection method for a three - coordinate - optical fusion blade detection device, including the steps:

[0069] S1. As shown in Figures 1 - 3 , the four - axis drive device is set on the first horizontal platform of the coordinate measuring machine 11, and the blade 1 to be detected is placed in the blade fixture 2. The three - coordinate measuring machine 11 detects the three reference planes of the second horizontal platform 5, thereby establishing an optical detection reference coordinate system O1 in the coordinate measuring machine 11; the blade workpiece reference coordinate system O2 is constructed by detecting the blade tenon by the coordinate measuring machine 11; the origin of the optical detection reference coordinate system O1 is at the intersection of the axis of the rotary axis C and the installation surface of the second horizontal platform 5; the blade workpiece reference coordinate system O2 is the design reference coordinate system of the blade 1 to be detected.

[0070] S2. Taking the optical detection reference coordinate system O1 as the reference, output the attitude of the blade workpiece reference coordinate system O2 relative to the optical detection reference coordinate system O1, and then correct the motion coordinate data of the four - axis drive device to ensure that the leading and trailing edges detected by the coordinate measuring machine 11 and the blade basin and blade back detected by the optical sensor 9 are on the same cross - section of the blade.

[0071] S3. Establish an optical sensor 9 measurement starting point coordinate system O3; the origin of the optical sensor 9 measurement starting point coordinate system O3 is the measurement starting point of the optical sensor 9 when the linear axis X - axis 6 and the linear axis Z - axis 8 are at the zero position. The x, y, and z axes of the optical sensor 9 measurement starting point coordinate system O3 are parallel to the x, y, and z axes of the optical detection reference coordinate system O1.

[0072] S4. According to the established optical detection reference coordinate system O1 and blade workpiece reference coordinate system O2 above, install a standard sphere 21 in the space of the optical sensor 9 measurement starting point coordinate system O3. Use the coordinate measuring machine 11 to detect the standard sphere 21 to obtain the center coordinate and radius of the standard sphere 21 under the optical detection reference coordinate system O1, and use the optical sensor 9 to detect the standard sphere 21 at the same position to obtain the center coordinate and radius of the standard sphere 21 under the optical sensor 9 measurement starting point coordinate system O3; compare the detection data of the standard sphere 21 by the coordinate measuring machine 11 and the optical sensor 9 to find the relative position of the coordinate system O3 relative to the coordinate system O1, and complete the calibration of the measurement starting point and the coordinate system fusion when the optical sensor detects the blade.

[0073] S5. Use a coordinate measuring machine 11 to measure the blade profile point cloud data A of the leading and trailing edges of the blade 1 to be detected in the blade workpiece reference coordinate system O2;

[0074] S6. Through the linkage of the linear axis X-axis, linear axis Y-axis, linear axis Z-axis, and rotary axis C-axis, use the optical sensor 9 to measure the blade profile point cloud data B of the blade basin and blade back of the blade 1 to be detected in the optical detection reference coordinate system O1;

[0075] S7. Transform the point cloud data B to the blade workpiece reference coordinate system O2 through a spatial transformation matrix to obtain the point cloud data B`, and complete the rough registration;

[0076] S8. Based on the point cloud data A, perform the nearest point matching of the point cloud data B` and A through the ICP algorithm to obtain the point cloud data B matched and complete the data fusion and fine registration;

[0077] S9. Perform B-spline curve fitting on the finely registered point cloud data B matched The fitted curve is the cross-sectional contour curve of the measured blade.

[0078] Further, in step S2, the posture includes Δx, Δy, Δz, Δθx, Δθy, and Δθz, where Δx is the X-axis coordinate of the origin of the blade workpiece reference coordinate system O2 in the optical detection reference coordinate system O1, Δy is the Y-axis coordinate of the origin of the blade workpiece reference coordinate system O2 in the optical detection reference coordinate system O1, Δz is the Z-axis coordinate of the origin of the blade workpiece reference coordinate system O2 in the optical detection reference coordinate system O1, Δθx is the rotation angle of the X-axis in the blade workpiece reference coordinate system O2 relative to the X-axis in the optical detection reference coordinate system O1, Δθy is the rotation angle of the Y-axis in the blade workpiece reference coordinate system O2 relative to the Y-axis in the optical detection reference coordinate system O1, and Δθz is the rotation angle of the Z-axis in the blade workpiece reference coordinate system O2 relative to the Z-axis in the optical detection reference coordinate system O1.

[0079] Further, in step S6, the method of using the optical sensor to measure the blade profile point cloud data B of the blade basin and blade back of the blade to be detected in the optical detection reference coordinate system O1 through the linkage of the linear axis X-axis, linear axis Y-axis, linear axis Z-axis, and rotary axis C-axis is as follows:

[0080] When measuring the blade, considering that the linearity of spectral confocal, laser and other displacement sensors is the best at the midpoint of the measurement range, and taking into account the fluctuation of the distance during measurement, the ranging reference value during the measurement of the optical sensor is set to the midpoint M0 of the measurement range. When performing blade detection, record the displacement values X0, Y0, Z0, C0 of each moving axis relative to the measurement starting point and the distance M measured by the optical sensor to the blade surface. Then, according to the following formula, obtain the three-dimensional point cloud data B of the blade profile:

[0081]

[0082] z = Z0.

[0083] Further, in step S7, the calculation formula for the point cloud data B` is:

[0084] B` = R·B(x, y, z) + T0;

[0085]

[0086] where Δx is the X-axis coordinate of the origin of the reference coordinate system O2 of the blade workpiece in the optical detection reference coordinate system O1, Δy is the Y-axis coordinate of the origin of the reference coordinate system O2 of the blade workpiece in the optical detection reference coordinate system O1, Δz is the Z-axis coordinate of the origin of the reference coordinate system O2 of the blade workpiece in the optical detection reference coordinate system O1, Δθx is the rotation angle of the X-axis in the reference coordinate system O2 of the blade workpiece relative to the X-axis of the optical detection reference coordinate system O1, Δθy is the rotation angle of the Y-axis in the reference coordinate system O2 of the blade workpiece relative to the Y-axis in the optical detection reference coordinate system O1, and Δθz is the rotation angle of the Z-axis in the reference coordinate system O2 of the blade workpiece relative to the Z-axis in the optical detection reference coordinate system O1.

[0087] Further, in step S8, the method to obtain the point cloud data B matched is:

[0088] For the matched point cloud data A and B matched , in order to solve for R and T such that RB matched + T - A is minimized, the function is defined as:

[0089]

[0090] Based on singular value decomposition to solve for the spatial coordinate transformation matrices R and T of the matched point cloud data A and B when J is minimized; the process is as follows: matched The spatial coordinate transformation matrices R and T; the process is as follows:

[0091] To minimize J, let:

[0092]

[0093] Denote the weighted centroids of the point cloud data A and B as:

[0094]

[0095] Substitute T into J to get:

[0096]

[0097] Denote Then:

[0098]

[0099] At this time,

[0100]

[0101] , while

[0102] Define the covariance matrix S = XWY T , perform singular value decomposition on S to get S = UΣV T Then:

[0103] tr(RXWY T ) = tr(RS) = tr(RUΣV T ) = tr(∑V T RU);

[0104] Since U, R, and V are all orthogonal matrices, let M = V T RU, then M is also an orthogonal matrix. According to the properties of orthogonal matrices, each element |m ij | in matrix M is ≤ 1. So when and only when M = V T RU = E, tr(ΣV T RU) reaches the maximum value, that is reaches the maximum value. At this time, R = VU T ;

[0105] For the obtained R and T, by judging and to determine whether R and T converge, so as to control the output of R and T.

[0106] The detection method of a three-coordinate and optical fusion blade detection device in the present invention is mainly applied to a three-coordinate and optical collaborative measurement system with a characteristic structure such as blades that present complex geometric contours and uneven curvature radii at different positions. The three-coordinate is used to detect the standard sphere 21 to obtain the center coordinates and radius of the standard sphere 21 relative to the optical detection reference coordinate system O1, and the optical sensor 9 is used to detect the standard sphere 21 at the same position to obtain the center position and radius of the standard sphere 21 relative to the measurement starting point coordinate system O3 of the optical sensor 9 with the origin at the optical sensor 9 and each coordinate axis parallel to the optical detection reference coordinate system O1 respectively. By comparing the detection data of the standard sphere 21 by the coordinate measuring machine 11 and the optical sensor 9, the relative position of the measurement starting point coordinate system O3 of the optical sensor relative to the optical detection reference coordinate system O1 is found, and the calibration of the measurement starting point and the coordinate system fusion when the optical sensor detects the blade are completed; the coordinate measuring machine 11 with higher precision is used to measure the leading and trailing edge parts with smaller curvature radii and finer structures in the blade; the optical sensor 9 with higher efficiency is used to measure the blade basin and blade back parts with larger curvature radii in the blade. Based on the complex structure and curvature characteristics of the blade and the detection advantages of the coordinate measuring machine 11 and the optical sensor 9, the blade body cross-section of the blade to be detected 1 is planned and divided into the blade basin, blade back, leading edge, and trailing edge. The coordinate measuring machine 11 with higher precision is used to measure the leading and trailing edge parts, and the optical sensor 9 with higher efficiency is used to measure the blade basin and blade back. The high-precision and high-efficiency detection of the blade body profile is realized through the collaborative measurement of the coordinate measuring machine 11 and the optical sensor 9.

Claims

1. A high-precision complex blade detection device integrating three-coordinate and optical technologies, characterized in that, It includes a coordinate measuring machine, a four-axis drive device, an optical sensor, a blade fixture, a standard ball, a host computer, and a power supply module; The coordinate measuring machine includes a first horizontal platform; The four-axis drive device includes a second horizontal platform disposed on the first horizontal platform, and three linear axes X, Y, and Z and a rotary axis C axis are provided on the second horizontal platform; The bottom of the linear axis Z axis is mounted on the linear axis X axis through a connecting plate; the optical sensor is mounted on the linear axis Z axis through a sensor fixture; both the optical sensor and the four-axis drive device are electrically connected to the host computer and the power supply module; The rotary axis C axis is mounted on the linear axis Y axis, and a turntable is mounted on the rotary axis C axis, and a blade fixture and the standard ball are provided on the turntable; the optical sensor is located on one side of the blade fixture; The optical sensor is a laser displacement sensor or a spectral confocal sensor.

2. The three-coordinate and optical fusion blade detection device according to claim 1, wherein The linear axis X axis, the linear axis Y axis, and the linear axis Z axis all include a motor, a grating scale, a ball screw module, and a linear guide rail. The linear guide rail is disposed on the second horizontal platform, and the ball screw module and the grating scale are disposed on the linear guide rail; the motor is connected to the ball screw in the ball screw module; the ball screw nut seat in the ball screw module is connected to the connecting plate, the turntable, or the sensor fixture.

3. The three-coordinate and optical fusion blade detection device according to claim 2, characterized in that, The resolution of the linear axis X axis, the linear axis Y axis, and the linear axis Z axis is 0.1 μm, and the repeat positioning accuracy is 1 μm; the resolution of the rotary axis C axis is 70 μrad, and the repeat positioning accuracy is 60 μrad.

4. The three-coordinate and optical fusion blade detection device according to claim 1, wherein, The power supply module includes a 220V AC power supply and a 24V DC power supply; the 220V AC power supply supplies power to the four-axis drive device; the 24V DC power supply supplies power to the optical sensor.

5. A detection method for a blade detection system based on the three - coordinate and optical fusion according to any one of claims 1 to 4, characterized in that, It includes the steps: S1. Set the four-axis drive device on the first horizontal platform of the coordinate measuring machine, and place the blade to be detected in the blade fixture. Detect the three reference planes of the second horizontal platform through the coordinate measuring machine, so as to establish an optical detection reference coordinate system O1 in the coordinate measuring machine; Detect the blade tenon through the coordinate measuring machine to construct a blade workpiece reference coordinate system O2; the origin of the optical detection reference coordinate system O1 is at the intersection of the axis center of the rotary axis C and the installation surface of the second horizontal platform; the blade workpiece reference coordinate system O2 is the design reference coordinate system of the blade to be detected; S2. Taking the optical detection reference coordinate system O1 as a reference, output the attitude of the blade workpiece reference coordinate system O2 relative to the optical detection reference coordinate system O1, and then correct the motion coordinate data of the four-axis drive device to ensure that the leading and trailing edges detected by the coordinate measuring machine and the blade basin and blade back detected by the optical sensor are in the same cross-section of the blade; S3. Establish an optical sensor measurement starting point coordinate system O3; the origin of the optical sensor measurement starting point coordinate system O3 is the measurement starting point of the optical sensor when the linear axis X axis and the linear axis Z axis are at the zero position, and the x, y, and z axes of the optical sensor measurement starting point coordinate system O3 are parallel to the x, y, and z axes of the optical detection reference coordinate system O1; S4. According to the established optical detection reference coordinate system O1 and the blade workpiece reference coordinate system O2, install a standard sphere within the space of the optical sensor measurement starting point coordinate system O3. Use a coordinate measuring machine to detect the standard sphere to obtain the center coordinates and radius of the standard sphere under the optical detection reference coordinate system O1, and use the optical sensor to detect the standard sphere at the same position to obtain the center coordinates and radius of the standard sphere under the optical sensor measurement starting point coordinate system O3. Compare the detection data of the standard sphere by the coordinate measuring machine and the optical sensor to find the relative position of the optical sensor measurement starting point coordinate system O3 relative to the optical detection reference coordinate system O1, and complete the measurement starting point calibration and coordinate system fusion when the optical sensor detects the blade. S5. Use a coordinate measuring machine to measure the blade profile point cloud data A of the leading and trailing edges of the blade to be detected in the blade workpiece reference coordinate system O2. S6. Through the linkage of the linear axis X, linear axis Y, linear axis Z, and rotary axis C, use the optical sensor to measure the blade profile point cloud data B of the blade basin and blade back of the blade to be detected in the optical detection reference coordinate system O1. S7. Transform the point cloud data B to the blade workpiece reference coordinate system O2 through a spatial transformation matrix to obtain the point cloud data B`, and complete the rough registration. S8. Taking the point cloud data A as a reference, the closest point matching between the point cloud data B` and A is performed through the ICP algorithm to obtain the point cloud data B matched , and the data fusion fine registration is completed; S9. Perform B-spline curve fitting on the accurately registered point cloud data B matched The curve obtained after fitting is the cross-sectional contour curve of the measured blade.

6. The detection method of the three-coordinate and optical fusion blade detection device according to claim 5, characterized in that In step S2, the posture includes Δx, Δy, Δz, Δθx, Δθy, and Δθz. Among them, Δx is the X-axis coordinate of the origin of the blade workpiece reference coordinate system O2 under the optical detection reference coordinate system O1, Δy is the Y-axis coordinate of the origin of the blade workpiece reference coordinate system O2 under the optical detection reference coordinate system O1, Δz is the Z-axis coordinate of the origin of the blade workpiece reference coordinate system O2 under the optical detection reference coordinate system O1, Δθx is the rotation angle of the X-axis in the blade workpiece reference coordinate system O2 relative to the X-axis of the optical detection reference coordinate system O1, Δθy is the rotation angle of the Y-axis in the blade workpiece reference coordinate system O2 relative to the Y-axis in the optical detection reference coordinate system O1, and Δθz is the rotation angle of the Z-axis in the blade workpiece reference coordinate system O2 relative to the Z-axis in the optical detection reference coordinate system O1.

7. The detection method of the three-coordinate and optical fusion blade detection device according to claim 5, characterized in that, In step S6, the method of using the optical sensor to measure the blade profile point cloud data B of the blade basin and blade back of the blade to be detected in the optical detection reference coordinate system O1 through the linkage of the linear axis X, linear axis Y, linear axis Z, and rotary axis C is as follows: When measuring the blade, set the ranging reference value during the measurement of the optical sensor to the midpoint M0 of the measurement range. When performing blade detection, record the displacement values X0, Y0, Z0, C0 of each moving axis relative to the measurement starting point and the distance M from the optical sensor to the blade surface. Then, obtain the three-dimensional point cloud data B of the blade profile according to the following formula: z = Z0.

8. The detection method of the three-coordinate and optical fusion blade detection device according to claim 5, characterized in that In step S7, the calculation formula for the point cloud data B` is: B` = R·B(x,y,z) + T0; Where, Δx is the X-axis coordinate of the origin of the reference coordinate system O2 of the blade workpiece in the optical detection reference coordinate system O1, Δy is the Y-axis coordinate of the origin of the reference coordinate system O2 of the blade workpiece in the optical detection reference coordinate system O1, Δz is the Z-axis coordinate of the origin of the reference coordinate system O2 of the blade workpiece in the optical detection reference coordinate system O1, Δθx is the rotation angle of the X-axis in the reference coordinate system O2 of the blade workpiece relative to the X-axis in the optical detection reference coordinate system O1, Δθy is the rotation angle of the Y-axis in the reference coordinate system O2 of the blade workpiece relative to the Y-axis in the optical detection reference coordinate system O1, and Δθz is the rotation angle of the Z-axis in the reference coordinate system O2 of the blade workpiece relative to the Z-axis in the optical detection reference coordinate system O1.

9. The detection method of the three-coordinate and optical fusion blade detection device according to claim 5, characterized in that, In step S8, the point cloud data B is obtained. matched The method is as follows: For the point cloud data A and B after matching matched , in order to solve for R and T such that RB matched + T - A is minimized, the function is defined as: When solving for the minimum J based on singular value decomposition, the spatially transformed matrices R and T of the point cloud data A and B after matching; the process is as follows: matched ​ To minimize J, let: Denote the weighted centroids of the point cloud data A and B as: Substitute T into J to get: Record Then: At this time, ， while Define the covariance matrix S = XWY T , perform singular value decomposition on S to obtain S = UΣV T , then: tr(RXWY T ) = tr(RS) = tr(RUV T ) = tr(ΣV T RU); Since U, R, and V are all orthogonal matrices, let M = V T RU. Then M is also an orthogonal matrix. According to the properties of orthogonal matrices, each element |m ij | in matrix M is ≤ 1. Therefore, when and only when M = V T RU = E, tr(ΣV T RU) reaches its maximum value, that is reaches its maximum value, and at this time R = VU T ; For the obtained R and T, by and to determine whether R and T converge, thereby controlling the output of R and T.

Images