Method for adaptive correction of turning paths of a generator rotor

Abstract

The application is suitable for the technical field of generator rotor production, and particularly relates to a turning path adaptive correction method for a generator rotor, which comprises the following steps: after receiving an instruction, a plurality of groups of original detection coordinate data of a rotor end face are acquired; a spatial plane data of an actual rotor end face is solved by adopting a least square method spatial plane fitting algorithm on the plurality of groups of original detection coordinate data, and three coupled error parameters of a radial eccentricity, an axial displacement and a double inclination angle of the rotor clamping are synchronously separated; an actual machining reference coordinate system is reconstructed based on the three coupled error parameters, and a spatial pose transformation matrix is calculated; a double-dimension tool compensation parameter is generated according to the spatial pose transformation matrix; and the positioning coordinates and the tool path of the turning tool are corrected based on the double-dimension tool compensation parameter and the axial displacement. The method can comprehensively and accurately eliminate various coupled clamping errors generated in the rotor clamping process.

Description

Technical Field

[0001] This application belongs to the field of generator rotor manufacturing technology, and in particular relates to an adaptive correction method for generator rotor turning path. Background Technology

[0002] The generator rotor is the core rotating component of power generation equipment such as steam turbine generators and hydro turbine generators. The dimensional accuracy and geometric tolerances of its end face turning directly determine the rotor's assembly coaxiality, operational stability, and overall service life, making it a critical high-precision machining process. When turning the generator rotor on a CNC lathe, due to its large size and weight, the rotor is easily affected by multiple factors during clamping, including radial runout of the machine tool spindle, chuck clamping eccentricity, tailstock center coaxiality deviation, initial blank positioning error, and uneven distribution of clamping force. These factors simultaneously produce three types of clamping errors: radial eccentricity, axial movement, and end face spatial double tilt angle. Moreover, these three types of errors are coupled, superimposed, and propagated in three-dimensional space, making it impossible to achieve independent separation and precise quantification through conventional testing methods.

[0003] Currently, the industry's error correction methods for generator rotor turning paths mainly rely on manual alignment, single-point static measurement compensation, or fixed parameter compensation, which have many technical shortcomings: Manual alignment requires operators to manually check the runout of the rotor end face and outer diameter using measuring tools such as dial indicators and micrometers, and adjust the clamping position based on experience. This is not only inefficient and labor-intensive, but also highly susceptible to human factors, resulting in poor consistency and failing to meet the machining requirements of large, high-precision rotors. Single-point static measurement can only obtain offset data in a single direction and cannot identify the coupled error relationship of radial, axial, and end face tilt. The compensation dimension is singular, making it difficult to completely eliminate machining errors caused by clamping deviations. Fixed parameter compensation does not adaptively adjust to the actual clamping posture of the rotor and cannot adapt to error fluctuations under different clamping conditions. When the rotor clamping is eccentric, shifting, or the end face is tilted, the tool positioning coordinates and tool path cannot match the actual rotor posture, which can easily lead to out-of-tolerance issues in rotor end face flatness, perpendicularity, and axial dimensions. Summary of the Invention

[0004] This application provides an adaptive correction method for the turning path of a generator rotor, which can solve the problem of out-of-tolerance turning dimensions caused by generator rotor clamping errors.

[0005] In a first aspect, embodiments of this application provide an adaptive correction method for the turning path of a generator rotor, including: After detecting the receiving command, multiple sets of original detection coordinate data of the rotor end face are acquired; wherein, the receiving command is used to instruct the CNC lathe tool post to move along a preset safety path to the generator rotor end face to be processed, and to execute the command after multi-point touch position detection; The least squares spatial plane fitting algorithm is used to solve the spatial plane data of the actual end face of the rotor for multiple sets of original detection coordinate data, and the three coupled error parameters of the rotor clamping radial eccentricity, axial movement and end face spatial double tilt angle are obtained simultaneously. Based on the three coupling error parameters, a real machining reference coordinate system matching the actual rotor pose is reconstructed, and the spatial pose transformation matrix of the real machining reference coordinate system relative to the theoretical machining reference coordinate system is calculated. Based on the spatial pose transformation matrix, two-dimensional tool compensation parameters are generated, namely radial positioning translation compensation amount and end face tilt trajectory compensation slope. The positioning coordinates and tool path of the turning tool are corrected based on the dual-dimensional tool compensation parameters and the axial movement.

[0006] The technical solutions described in this application embodiment have at least the following technical effects: The adaptive correction method for the turning path of a generator rotor provided in this application acquires multiple sets of original detection coordinate data of the rotor end face by using a CNC lathe tool post carrying an electrical probe to complete multi-point contact detection along a safe path. It then uses a least-squares spatial plane fitting algorithm to accurately calculate the actual spatial plane data of the rotor end face and simultaneously separates three clamping coupling error parameters: radial eccentricity, axial movement, and end face spatial double tilt angle. Based on these three error parameters, it reconstructs an actual machining reference coordinate system matching the actual rotor posture and calculates the corresponding spatial posture transformation matrix. Based on this, it generates two-dimensional tool compensation parameters: radial positioning translation compensation and end face tilt trajectory compensation slope. Finally, it relies on this two-dimensional compensation... The parameters and axial movement are used to collaboratively correct the positioning coordinates and tool path of the turning tool, which can comprehensively and accurately eliminate various coupled clamping errors generated during rotor clamping. It can achieve adaptive matching between the machining datum and the actual position of the rotor, and complete the fully automatic and high-precision dynamic correction of the turning path without repeated manual alignment. This significantly improves the flatness, coaxiality, perpendicularity and axial dimensional accuracy of the generator rotor end face turning, effectively simplifies the machining process, improves machining efficiency and consistency, and fundamentally solves the technical problems of traditional correction methods, such as the difficulty in quantifying and separating clamping coupling errors, the inability to adaptively match the machining datum, low tool compensation accuracy and insufficient intelligence.

[0007] Secondly, embodiments of this application provide an adaptive correction device for the turning path of a generator rotor, comprising: The acquisition unit is used to acquire multiple sets of original detection coordinate data of the rotor end face after detecting the received command; wherein, the received command is used to instruct the CNC lathe tool post to move along a preset safety path to the generator rotor end face to be processed, and to execute the command after multi-point touch position detection. The calculation and separation unit is used to solve the spatial plane data of the actual end face of the rotor using the least squares spatial plane fitting algorithm on multiple sets of the original detection coordinate data, and simultaneously separates and obtains three coupled error parameters of the rotor clamping: radial eccentricity, axial movement, and end face spatial double tilt angle. The construction unit is used to reconstruct the actual machining reference coordinate system that matches the actual rotor pose based on the three coupling error parameters, and to calculate the spatial pose transformation matrix of the actual machining reference coordinate system relative to the theoretical machining reference coordinate system. The generation unit is used to generate two-dimensional tool compensation parameters based on the spatial pose transformation matrix, namely, radial positioning translation compensation amount and end face tilt trajectory compensation slope. The correction unit is used to correct the positioning coordinates and tool path of the turning tool based on the dual-dimensional tool compensation parameters and the axial movement.

[0008] Thirdly, embodiments of this application provide a turning device, which includes a turning apparatus and a control device. The turning apparatus is electrically connected to the control device. The control device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the method described in the first aspect above.

[0009] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in the first aspect above.

[0010] Fifthly, embodiments of this application provide a computer program product that, when run on a turning machine, causes the turning machine to execute the generator rotor turning path adaptive correction method described in the first aspect above.

[0011] It is understood that the beneficial effects of the second to fifth aspects mentioned above can be found in the relevant descriptions in the first aspect mentioned above, and will not be repeated here. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 This is a flowchart illustrating the adaptive correction method for the turning path of a generator rotor provided in an embodiment of this application. Figure 2This is a schematic diagram of spatial plane fitting provided in an embodiment of this application; Figure 3 This is a schematic diagram of the reconstruction of the actual machining reference coordinate system and the spatial pose transformation provided in the embodiments of this application; Figure 4 This is a schematic diagram of the structure of the generator rotor turning path adaptive correction device provided in the embodiments of this application; Figure 5 This is a schematic diagram of the structure of the control device for the turning equipment provided in the embodiments of this application. Detailed Implementation

[0014] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0015] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0016] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0017] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determination" or "if the described condition or event is detected" may be interpreted, depending on the context, as "once determination," "in response to determination," "once the described condition or event is detected," or "in response to the detection of the described condition or event."

[0018] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0019] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0020] The generator rotor is the core rotating component of power generation equipment such as steam turbine generators and hydro turbine generators. The dimensional accuracy and geometric tolerances of its end face turning directly determine the rotor's assembly coaxiality, operational stability, and overall service life, making it a critical high-precision machining process. When turning the generator rotor on a CNC lathe, due to its large size and weight, the rotor is easily affected by multiple factors during clamping, including radial runout of the machine tool spindle, chuck clamping eccentricity, tailstock center coaxiality deviation, initial blank positioning error, and uneven distribution of clamping force. These factors simultaneously produce three types of clamping errors: radial eccentricity, axial movement, and end face spatial double tilt angle. Moreover, these three types of errors are coupled, superimposed, and propagated in three-dimensional space, making it impossible to achieve independent separation and precise quantification through conventional testing methods.

[0021] Currently, the industry's error correction methods for generator rotor turning paths mainly rely on manual alignment, single-point static measurement compensation, or fixed parameter compensation, which have many technical shortcomings: Manual alignment requires operators to manually check the runout of the rotor end face and outer diameter using measuring tools such as dial indicators and micrometers, and adjust the clamping position based on experience. This is not only inefficient and labor-intensive, but also highly susceptible to human factors, resulting in poor consistency and failing to meet the machining requirements of large, high-precision rotors. Single-point static measurement can only obtain offset data in a single direction and cannot identify the coupled error relationship of radial, axial, and end face tilt. The compensation dimension is singular, making it difficult to completely eliminate machining errors caused by clamping deviations. Fixed parameter compensation does not adaptively adjust to the actual clamping posture of the rotor and cannot adapt to error fluctuations under different clamping conditions. When the rotor clamping is eccentric, shifting, or the end face is tilted, the tool positioning coordinates and tool path cannot match the actual rotor posture, which can easily lead to out-of-tolerance issues in rotor end face flatness, perpendicularity, and axial dimensions.

[0022] To address the aforementioned issues, this application provides an adaptive correction method for the turning path of a generator rotor. In this method, a CNC lathe tool post carrying an electrical probe completes multi-point contact detection along a safe path to acquire multiple sets of original detection coordinate data for the rotor end face. A least-squares spatial plane fitting algorithm is used to accurately calculate the actual spatial plane data of the rotor end face and simultaneously separate three clamping coupling error parameters: radial eccentricity, axial movement, and end face spatial double tilt angle. Based on these three error parameters, an actual machining reference coordinate system matching the actual rotor posture is reconstructed, and the corresponding spatial posture transformation matrix is ​​calculated. Based on this, dual-dimensional tool compensation parameters—radial positioning translation compensation and end face tilt trajectory compensation slope—are generated. Finally, the dual-dimensional compensation parameters and axial movement are correlated... By correcting the positioning coordinates and tool path of the turning tool, it can comprehensively and accurately eliminate various coupled clamping errors generated during rotor clamping, realize adaptive matching between the machining datum and the actual position of the rotor, and complete the fully automatic, high-precision dynamic correction of the turning path without repeated manual alignment. This significantly improves the flatness, coaxiality, perpendicularity and axial dimensional accuracy of the generator rotor end face turning, effectively simplifies the machining process, improves machining efficiency and consistency, and fundamentally solves the technical problems of traditional correction methods, such as the difficulty in quantifying and separating clamping coupling errors, the inability to adaptively match the machining datum, low tool compensation accuracy and insufficient intelligence.

[0023] The adaptive correction method for the turning path of the generator rotor provided in this application embodiment can be applied to turning equipment. In this case, the turning equipment is the execution subject of the adaptive correction method for the turning path of the generator rotor provided in this application embodiment. This application embodiment does not impose any restrictions on the specific type of turning equipment.

[0024] For example, a turning machine may include a turning device and a control device, with the turning device and control device electrically connected. The turning device includes a bed, a spindle drive assembly, a rotor clamping and positioning assembly, a CNC tool post assembly, an electric probe detection assembly, a coordinate axis feed assembly, and a cooling and protection assembly. The bed, as the overall supporting foundation of the turning machine, possesses high rigidity and stability, supporting the other components and ensuring structural stability during the turning process. The spindle drive assembly includes a spindle motor, a spindle box, and a transmission mechanism, which is connected to the rotor clamping and positioning assembly to drive the generator rotor to rotate stably around its own axis, providing the main motion power for the turning process. The rotor clamping and positioning assembly includes a hydraulic chuck and a tailstock center; the hydraulic chuck is used for clamping and fixing. One end of the generator rotor has a tailstock center that can move axially along the bed and press against the other end of the rotor, achieving coaxial clamping and positioning of the rotor. It can also accommodate generator rotors of different lengths. The CNC tool post assembly is a servo-driven CNC tool post, capable of automatic indexing and tool changing, used to clamp turning tools and drive them to complete turning cutting operations. The electrical probe detection assembly is a high-precision contact-type electrical probe, detachably mounted on the CNC tool post. Driven by the CNC tool post, it can move along a preset safety path to perform multi-point contact-type position detection on the end face of the generator rotor to be machined, collecting the original detection coordinate data of the rotor end face. The coordinate axis feed assembly includes X... Both the radial feed mechanism and the Z-axis axial feed mechanism utilize servo motors paired with ball screws and linear guides to drive the CNC tool holder and electrical probe, achieving high-precision radial and axial displacement movements. They perform radial positioning and translation compensation, axial runout compensation, and dynamic trajectory correction. The cooling and protection components include cooling nozzles, a coolant circulation system, and protective baffles. The cooling nozzles spray coolant into the cutting area during turning, providing cooling and lubrication to the tool and rotor machining areas. The protective baffles prevent cutting debris from contacting the coolant. The control device controls or regulates the turning process of the turning apparatus.

[0025] For example, the control device can be a microcontroller, PLC, smart screen, smart TV and other terminal equipment, handheld device with wireless communication function, computing device or other processing device connected to a wireless modem, Internet of Things terminal, computer, laptop computer, customer premises equipment (CPE) and / or other devices for communication on a wireless system, as well as next-generation communication systems, such as mobile terminals in 5G networks or mobile terminals in future evolved Public Land Mobile Networks (PLMNs).

[0026] To better understand the adaptive correction method for generator rotor turning path provided in the embodiments of this application, the specific implementation process of the adaptive correction method for generator rotor turning path provided in the embodiments of this application will be described by way of example below.

[0027] Figure 1 A schematic flowchart of the adaptive correction method for generator rotor turning path provided in this application is shown. The adaptive correction method for generator rotor turning path includes: S100, after detecting the receiving command, acquires multiple sets of original detection coordinate data of the rotor end face; among them, the receiving command is used to instruct the CNC lathe tool post to move along the preset safety path to the generator rotor end face to be processed, and execute the command after multi-point touch position detection.

[0028] It is understood that the receiving command is a dedicated trigger command for the CNC lathe system. Its triggering prerequisite is that the CNC lathe tool post, equipped with a contact-type electrical probe, moves along a pre-set safety avoidance path of the CNC system, avoiding interference components such as the rotor blank, lathe chuck, and tailstock center throughout the movement. After moving to the area of ​​the generator rotor to be machined without collision risk, it performs single-point or multi-point circumferentially distributed contact-type position detection. After detection, the electrical probe sends a detection completion signal back to the CNC system. The system verifies the validity of the signal and generates this receiving command. The command also carries status indicators such as detection start, position completion, data feedback, and anomaly verification. The original detection coordinate data is the three-dimensional machine tool coordinate (X / Y / Z) collected by the electrical probe on the rotor end face. The collection points can be selected as one center detection point at the center of the rotor end face, and multiple edge detection points evenly distributed along the circumference of the end face. The electric probe can adopt the principle of high-precision contact displacement sensing. When the probe makes rigid contact with the rotor end face, it instantly triggers a sampling signal. The CNC system synchronously collects the machine tool coordinate values ​​of the current tool holder to form the original detection coordinate data.

[0029] S200 uses the least squares spatial plane fitting algorithm to solve the spatial plane data of the actual end face of the rotor from multiple sets of original detection coordinate data, and simultaneously separates and obtains three coupled error parameters: radial eccentricity, axial movement, and end face spatial double tilt angle of the rotor clamping.

[0030] It is understandable that when a generator rotor is clamped on a CNC lathe, it is affected by factors such as spindle radial runout, chuck clamping eccentricity, tailstock center coaxiality deviation, rotor blank positioning deviation, and uneven clamping force. These factors simultaneously generate three types of clamping errors: radial eccentricity, axial movement, and end-face spatial tilt. These three types of errors are coupled and superimposed, making them impossible to separate using a single measurement method. The least squares spatial plane fitting algorithm is a mathematical algorithm for solving the three-dimensional spatial plane pose. It is adapted to the machining characteristics of the circular plane of the rotor end face. Its core principle is to minimize the sum of the squares of the Euclidean distances from all original probe points to the fitting plane, maximally offsetting the random errors of the electrical probe and restoring the true spatial shape of the rotor end face. Before the algorithm is executed, multiple sets of original probe coordinate data can be preprocessed, including data denoising, outlier removal, and coordinate normalization, to ensure the validity of the input data. Then, the preprocessed three-dimensional coordinate data is substituted into the algorithm model to calculate the spatial plane equation and geometric parameters that best represent the actual end face of the rotor. The logic for synchronously separating the three coupled error parameters is as follows: First, the overall spatial pose of the actual rotor end face is obtained by fitting the plane. Then, the actual pose is compared with the theoretical design pose and decomposed into translational errors (radial eccentricity and axial movement) and rotational errors (end face spatial double tilt angle). The radial eccentricity represents the offset of the rotor center relative to the theoretical axis in the X / Y radial plane. The axial movement represents the clamping depth / movement deviation of the rotor in the Z axis. The end face spatial double tilt angle represents the pitch / yaw tilt deviation of the rotor end face around the X and Y axes.

[0031] For example, a least squares spatial plane fitting algorithm can be used on multiple sets of original detection coordinate data to calculate the general plane equation of the actual rotor end face. Based on the general plane equation, the normal vector of the actual end face can be extracted to obtain the spatial plane data of the actual rotor end face. Alternatively, a weighted least squares spatial plane fitting algorithm can be used. Based on the difference in detection accuracy between the center detection point and the edge detection point of the rotor end face, the corresponding coordinate data can be assigned differential weights. The weighted original coordinate data can be fitted with plane to obtain the general plane equation. Based on this equation, the weighted and optimized end face normal vector and geometric center coordinates can be extracted to obtain the spatial plane data that fits the true pose of the rotor end face, and so on, but not limited to these methods.

[0032] In one possible implementation, please refer to Figure 2 In step S200, the spatial plane data of the actual rotor end face is calculated using the least squares spatial plane fitting algorithm on multiple sets of original detection coordinate data, including: S210 uses the least squares spatial plane fitting algorithm to calculate the general plane equation of the actual end face of the rotor by applying the least squares method to multiple sets of original detection coordinate data.

[0033] It can be understood that the general standard form of the equation for a three-dimensional spatial plane is Ax + By + Cz + D = 0, where A, B, C, and D are the coefficients of the plane equation, and (x, y, z) are the coordinates of any point in space. This equation can completely describe the position, orientation, and spatial shape of the actual end face of the rotor in the three-dimensional coordinate system of the machine tool. The specific implementation process of the least squares spatial plane fitting algorithm is as follows: multiple sets of original probe coordinates are... Using the input samples, an error objective function is constructed with the plane equation coefficients A, B, C, and D as unknowns. The objective function is the sum of the squared distances from all probe points to the plane, i.e. Simultaneously, a constraint condition with a normal vector magnitude of 1 (A² + B² + C² = 1) is added. Then, a normal equation system is constructed through matrix operations, and the four coefficients A, B, C, and D are solved using matrix inversion and least squares estimation algorithms, ultimately obtaining the general planar equations of the actual rotor end face. This algorithm exhibits strong robustness to random errors in the original probe data; the more probe points there are, the closer the fitting result is to the true shape of the rotor end face. The plane equation coefficients are directly related to the end face pose; coefficients A and B correspond to the spatial tilt characteristics of the end face, and coefficient D corresponds to the axial position characteristics of the end face.

[0034] S220 extracts the normal vector of the actual end face based on the general equation of the plane, and obtains the spatial plane data of the actual end face of the rotor.

[0035] It can be understood that the spatial plane normal vector is a three-dimensional vector perpendicular to the fitted plane, and is the core geometric parameter characterizing the spatial attitude of the plane. Its direction follows the right-hand rule and directly corresponds to the axial reference of the rotor end face. The method of extracting the normal vector from the general plane equation Ax+By+Cz+D=0 is extremely straightforward, with the plane normal vector n=(A, B, C), requiring no additional complex calculations. After extraction, the CNC system normalizes the normal vector, setting its magnitude to 1 to eliminate the numerical influence of the plane equation coefficients, facilitating subsequent angle calculations with the theoretical reference normal vector. The spatial plane data of the actual rotor end face is a complete pose dataset, containing three core components: first, the general plane equation Ax+By+Cz+D=0; second, the normalized end face normal vector n=(A, B, C); and third, the geometric center coordinates of the fitted plane (obtained by weighted averaging of the original probe points combined with solving the plane equation). The accuracy of the normal vector fully inherits the accuracy of the least squares fitting plane, accurately reflecting the actual tilt attitude of the rotor end face and serving as the direct basis for calculating the spatial double tilt angle of the end face; the geometric center coordinates are the reference points for calculating radial eccentricity and axial movement. The spatial plane data completely and uniquely characterizes the true physical pose of the rotor end face in the machine tool coordinate system.

[0036] This setup, employing the least squares spatial plane fitting algorithm to solve the actual rotor end face, can maximally offset the random errors from the electrical probe and the interference from local burrs / defects on the end face, ensuring high precision and stability in the planar solution. This addresses the pain point of traditional manual and single-point measurements failing to reconstruct the overall end face pose. Directly extracting normal vectors based on the general plane equations simplifies the mathematical logic of spatial attitude representation, eliminating the need for additional geometric model construction, resulting in extremely high computational efficiency and adapting to the real-time computational requirements of CNC systems. This step simultaneously yields complete spatial plane data, accurately representing the spatial morphology of the rotor end face and providing a unique and precise geometric basis for the subsequent separation of the three coupled error parameters. This solves the industry problem of the inability to quantify and separate clamping error coupling in traditional machining. After normalization, the normal vectors can be directly used for angle calculations with the theoretical benchmark, significantly simplifying the algorithmic complexity of subsequent double tilt angle calculations. The plane geometric center coordinates provide a unified benchmark for radial and axial error calculations, reducing the error superposition caused by multiple benchmarks. This technical solution ensures accurate conversion from raw detection data to spatial plane data at the algorithm level, with minimal error propagation. It lays the core data foundation for the entire rotor turning path adaptive correction method. At the same time, the algorithm is highly versatile and adaptable to the end face fitting of generator rotors of different specifications and sizes.

[0037] In one possible implementation, in step S200, multiple sets of original detection coordinate data include at least one of the coordinate data of the rotor end face center detection point and the coordinate data of the rotor end face edge detection point. Three coupled error parameters are then obtained: the radial eccentricity of the rotor clamping, the axial movement, and the end face spatial double tilt angle. S230 uses the coordinate data of the center detection point on the rotor end face as a reference to calculate the difference between the center detection point and the theoretical shaft center coordinates, thus obtaining the radial eccentricity and axial movement.

[0038] It can be understood that the rotor end face center detection point is a pre-set end face center detection point, which is the core reference point characterizing the rotor clamping center position; the theoretical axis coordinates are the ideal machining center coordinates preset by the CNC system according to the rotor design drawings, which are fixed standard values ​​representing the theoretical center position of the rotor when there is no clamping error. The core logic for calculating radial eccentricity and axial movement is to find the coordinate difference between the actual center and the theoretical center. This can be calculated using the machine tool's three-dimensional coordinate system to reduce the additional errors caused by coordinate system transformation: the radial eccentricity is the coordinate difference of the center detection point in the X and Y axes, i.e., ΔX = X actual center. Theoretical axis, actual center of ΔY=Y The theoretical shaft center, ΔX and ΔY are vector components that characterize the magnitude and direction of the eccentricity, reflecting the degree of rotor clamping eccentricity in the radial plane; the axial movement is the coordinate difference of the center detection point in the Z-axis direction, i.e., ΔZ = Z actual center. The theoretical shaft center is a scalar parameter that reflects the axial clamping depth of the rotor and the deviation of the spindle movement.

[0039] S240, based on the angle between the normal vector of the actual end face and the normal vector of the theoretical end face reference, calculate the spatial double tilt angle of the rotor end face around the X-axis and Y-axis respectively.

[0040] It is understandable that the theoretical end face datum is the ideal end face specified in the rotor design drawings, whose normal vector completely coincides with the theoretical shaft center, and the theoretical normal vector is fixed at... The end face orientation serves as the standard reference. The actual end face normal vector n=(A, B, C) is the unit vector extracted from the previous fitting. The spatial angle between the two directly reflects the overall tilt of the rotor end face. The tilt of the generator rotor clamping end face can be decomposed into the pitch angle α around the X-axis and the yaw angle β around the Y-axis, i.e., spatial double tilt angles. The two tilt angles are orthogonal to each other, perfectly matching the motion characteristics of the CNC lathe's X / Y / Z axes. The calculation process uses the three-dimensional spatial vector angle formula: ,because All are unit vectors with a magnitude of 1. The formula can be simplified to: The spatial angle θ is then decomposed into inclination angles in two orthogonal planes, X and Y. The inclination angles α=arctan(B / C) around the X-axis and β=arctan(A / C) around the Y-axis are calculated using the arctangent function to obtain the spatial double inclination angles of the end face.

[0041] This setup uses the rotor end face center detection point as the reference to calculate radial eccentricity and axial movement. The reference is unified and the logic is simple, allowing direct quantification of translational errors caused by clamping. The error values ​​directly correspond to the lathe's X / Z axis motion, providing directly usable parameters for subsequent tool positioning compensation and reducing complex coordinate transformations. By decomposing the spatial double tilt angle through the angle between the actual end face normal vector and the theoretical reference normal vector, the rotational errors caused by clamping are accurately characterized. The three-dimensional spatial tilt is decomposed into the tilt angles of two orthogonal axes, adapting to the orthogonal motion axis characteristics of CNC lathes, resulting in extremely strong compensation adaptability. The originally coupled and complex clamping errors are decomposed into two independent parameters: translational errors (radial eccentricity, axial movement) and rotational errors (spatial double tilt angles), completely solving the industry pain point of inaccurate correction of multiple coupled errors. The error calculation results are unique, accurate, and unambiguous.

[0042] S300 reconstructs the actual machining reference coordinate system that matches the actual rotor pose based on three coupling error parameters, and calculates the spatial pose transformation matrix of the actual machining reference coordinate system relative to the theoretical machining reference coordinate system.

[0043] It is understandable that the theoretical machining reference coordinate system is an ideal workpiece coordinate system constructed based on the rotor design drawings, with the origin at the theoretical axis, the Z-axis coinciding with the theoretical axis, and the X / Y axes being radially orthogonal axes. However, there is a deviation between the actual and theoretical positions of the rotor after clamping. If machining is still performed using the theoretical reference, it will lead to deviations in end-face flatness, coaxiality, and perpendicularity. Therefore, the actual machining reference coordinate system can be reconstructed based on three coupled error parameters, allowing the machining reference to perfectly match the actual rotor position, achieving adaptive machining based on the actual workpiece, and eliminating the influence of clamping errors. The core of reconstructing the actual machining reference coordinate system is to correct the origin position of the coordinate system using radial eccentricity and axial runout, and to correct the axial attitude of the coordinate system using spatial double tilt angles, so that the coordinate system perfectly matches the actual clamping position of the rotor. The spatial pose transformation matrix is ​​a mathematical matrix describing all pose changes from the theoretical machining reference coordinate system to the actual machining reference coordinate system, containing translation and rotation transformation components.

[0044] For example, the actual machining reference coordinate system can be constructed with the geometric center of the rotor's actual end face as the origin of the coordinate system, the normal vector of the actual end face as the Z-axis, and the theoretical radial direction of the rotor as the X and Y axes; or the projection point of the rotor's theoretical axis center on the actual end face can be used as the origin of the coordinate system, and the origin can be translated and corrected by combining the radial eccentricity and axial movement, with the end face normal after spatial double tilt angle correction as the Z-axis, and the orthogonal radial direction of the machine tool as the X and Y axes, etc., but not limited to these.

[0045] In one possible implementation, step S300 involves reconstructing an actual machining reference coordinate system that matches the actual rotor pose based on three coupling error parameters, including: A coordinate system for actual machining is constructed with the geometric center of the actual end face of the rotor as the origin, the normal vector of the actual end face as the Z-axis, and the theoretical radial direction of the rotor as the X and Y axes.

[0046] It is understandable that setting the geometric center of the actual end face of the rotor as the origin of the coordinate system can directly offset the translation error of the origin caused by radial eccentricity and axial movement, so that the machining origin coincides completely with the actual center of the rotor, eliminating the clamping translation deviation from the datum level; using the unit normal vector of the actual end face as the Z-axis can directly offset the axial attitude error caused by the double tilt angle of the end face space, so that the Z-axis of the coordinate system is perpendicular to the actual end face of the rotor, ensuring the flatness of the machined end face; the X and Y axes follow the theoretical radial direction of the rotor, keeping the radial machining datum consistent with the design drawings, ensuring that the dimensional accuracy and geometric tolerance of the rotor outer circle and end face meet the design requirements.

[0047] This setup, using the actual geometric center of the end face as the origin and the normal vector as the Z-axis, reconstructs the actual machining reference coordinate system. This eliminates translational and rotational errors caused by rotor clamping at the reference level, ensuring a perfect match between the machining reference and the actual rotor posture. This solves the machining deviation problem caused by the mismatch between the theoretical reference and the actual workpiece posture in traditional machining. The X / Y axes follow the theoretical radial direction, ensuring the consistency of the radial reference for turning, meeting the design requirements of the drawings, and reducing dimensional deviations caused by changes in the radial reference. This coordinate system directly maps to the X / Z motion axes of the CNC lathe, providing a standardized and directly calculable mathematical reference for subsequent spatial posture transformations and tool compensation, without requiring additional format conversion. Through reference reconstruction, subsequent tool path calculations are all based on the actual rotor posture, avoiding all machining deviations caused by clamping errors from the source, and providing a fundamental guarantee for the geometrical accuracy of rotor end face turning.

[0048] In one possible implementation, please refer to Figure 3 In step S300, the spatial pose transformation matrix of the actual machining reference coordinate system relative to the theoretical machining reference coordinate system is calculated, including: S310, based on the radial eccentricity and axial movement, construct the translation transformation matrix of the actual machining reference coordinate system relative to the theoretical machining reference coordinate system.

[0049] It can be understood that the translation transformation matrix is ​​a 4×4 three-dimensional homogeneous coordinate matrix describing the translation between two coordinate systems only at the origin, without any attitude rotation. It is the core sub-matrix of spatial pose transformation, and its construction logic directly corresponds to the translational errors in rotor clamping. The standard form of the translation transformation matrix in three-dimensional homogeneous coordinates is: all elements on the main diagonal are 1, the first three elements of the last column are the translation amounts of the X-axis, Y-axis, and Z-axis, respectively, and the last row is 0001. During the construction process, the radial eccentricity ΔX and ΔY are directly assigned as the translation components in the X and Y directions of the matrix, and the axial runout ΔZ is directly assigned as the translation component in the Z direction of the matrix. That is, the translation components of the translation matrix are completely determined by the translation error among the three error parameters, without any additional calculations or error introduction. This matrix only represents the offset of the origin position of the coordinate system and does not change the axial attitude of the coordinate system. It is independent of the rotation transformation matrix, which facilitates subsequent fusion calculations.

[0050] S320, based on the spatial double tilt angle, constructs the rotation transformation matrix of the actual machining reference coordinate system relative to the theoretical machining reference coordinate system.

[0051] The rotation transformation matrix is ​​a 4×4 three-dimensional homogeneous coordinate matrix describing the attitude rotation between two coordinate systems without origin translation. It corresponds to the rotational errors in rotor clamping and is constructed based on a standard mathematical model of three-dimensional spatial rotation transformation. The rotor end face has two spatial tilt angles: α around the X-axis and β around the Y-axis. During construction, it follows the spatial rotation synthesis rule of first rotating around the X-axis and then around the Y-axis, constructing rotation sub-matrices around the X and Y axes respectively, and then fusing them into the total rotation transformation matrix through matrix multiplication. The elements of the rotation matrix for rotating around the X-axis by angle α and around the Y-axis by angle β are calculated using cosine and sinine trigonometric functions. During construction, the matrix is ​​normalized to ensure no scaling distortion or attitude distortion during rotation. The rotation transformation matrix only represents the axial tilt attitude of the coordinate system and does not change the origin position. It is independent of the translation transformation matrix, reducing the coupling interference between translation and rotation errors.

[0052] S330 merges the translation and rotation transformation matrices to obtain the final spatial pose transformation matrix.

[0053] It is understandable that the rotor clamping pose error involves both translation and rotation transformations. Therefore, it is necessary to fuse the independently constructed translation and rotation transformation matrices to obtain a complete spatial pose transformation matrix that includes both translational offset and rotational tilt, thus achieving a one-stop conversion from the theoretical coordinate system to the actual coordinate system. The core mathematical basis of this fusion is that the generator rotor is a rigid workpiece, and the spatial pose transformation only includes translation and rotation, without scaling deformation, which conforms to the rules of three-dimensional rigid body transformation.

[0054] For example, the three-dimensional homogeneous coordinate transformation rule can be used to perform matrix multiplication operations on the rotation transformation matrix and the translation transformation matrix in sequence, and then the matrix multiplication results can be used to construct a homogeneous spatial pose transformation matrix that simultaneously contains translation offset and rotation tilt, as the final spatial pose transformation matrix; alternatively, the rotation transformation matrix can be first converted into a unit quaternion to represent the spatial rotational attitude of the rotor end face, and the translation transformation matrix can be extracted into a three-dimensional translation vector to represent radial eccentricity and axial movement, and then the quaternion and translation vector can be combined according to the rigid body transformation rule and converted into a standard 4×4 homogeneous matrix to obtain the spatial pose transformation matrix, and so on, but not limited to these.

[0055] This configuration, by constructing a translation transformation matrix based on radial eccentricity and axial axial movement, and a rotation transformation matrix based on spatial double tilt angles, and then fusing the two types of transformation matrices to obtain the final spatial pose transformation matrix, can accurately quantify the translational and rotational deviations generated by rotor clamping. It decomposes the coupled pose error into independent and computable mathematical matrices, and then forms a complete and unified coordinate transformation mathematical model through matrix fusion. This achieves accurate and efficient conversion from the theoretical machining reference coordinate system to the actual machining reference coordinate system, ensuring the mathematical rigor and accuracy stability of the pose transformation, reducing the coordinate transformation distortion problem caused by the coupling of translational and rotational errors, simplifying the coordinate calculation logic of the CNC system, and improving the real-time performance and reliability of the pose transformation. It provides a standardized and high-precision core mathematical basis for subsequent tool compensation parameter generation and dynamic correction of turning trajectory, effectively ensuring the overall accuracy and stability of adaptive correction of generator rotor turning path.

[0056] In one possible implementation, in step S330, the translation transformation matrix and the rotation transformation matrix are fused to obtain the final spatial pose transformation matrix, including: S331 uses the three-dimensional homogeneous coordinate transformation rule to perform matrix multiplication operations on the rotation transformation matrix and the translation transformation matrix in sequence.

[0057] It is understandable that the three-dimensional homogeneous coordinate transformation rule is a universally accepted standard for three-dimensional coordinate transformation in the CNC machining field. Extending three-dimensional rectangular coordinates to four-dimensional homogeneous coordinates unifies all spatial transformations such as translation, rotation, and scaling into matrix multiplication operations, simplifying the algorithmic logic of complex coordinate transformations. This is the core rule for achieving coordinate system fusion transformation. The order of matrix multiplication operations strictly follows the spatial rigid body transformation rule of rotation transformation first, followed by translation transformation. This order closely matches the actual impact mechanism of rotor clamping errors: first, the axial tilt of the coordinate system is corrected through rotation transformation, and then the origin position of the coordinate system is corrected through translation transformation. Reversing this order will lead to coordinate transformation distortion and error amplification. The 4×4 matrix multiplication operation follows the standard rule of row multiplication and column summation, uses a double-precision floating-point arithmetic unit to perform the operation, retaining an accuracy to 0.0001mm, with no rounding errors and no calculation deviations.

[0058] S332 constructs a homogeneous spatial pose transformation matrix that simultaneously includes translational offset and rotational tilt from the matrix multiplication result, which serves as the final spatial pose transformation matrix.

[0059] It can be understood that the result of matrix multiplication directly contains the attitude components of rotation transformation and the position components of translation transformation. Regularizing these components into a 4×4 homogeneous matrix yields the homogeneous spatial pose transformation matrix, which serves as the core parameter for the final pose transformation. Each component of the matrix has a clear physical meaning: the rotation component represents the double-tilt attitude of the actual coordinate system relative to the theoretical coordinate system; the translation component represents the origin offset due to radial eccentricity and axial axial movement. The matrix as a whole completely and uniquely represents the entire pose transformation relationship from the theoretical machining datum to the actual machining datum.

[0060] This configuration, by constructing translation and rotation transformation matrices separately, decomposes the complex spatial pose transformation into two independent and simple sub-operations, significantly reducing algorithm complexity, alleviating the computational load on the CNC system, and improving computational efficiency. The matrix fusion using 3D homogeneous coordinate transformation rules unifies the operational forms of translation and rotation transformations. The matrix multiplication order of rotation first, then translation, aligns with the actual impact logic of rotor clamping errors, ensuring the accuracy of the pose transformation results and reducing coordinate distortion caused by incorrect transformation order. The fused homogeneous spatial pose transformation matrix integrates the pose influence of all clamping errors in a one-stop manner, without omissions or redundancy, achieving accurate and rapid conversion from theoretical coordinates to actual coordinates.

[0061] S400 generates two-dimensional tool compensation parameters, namely radial positioning translation compensation amount and end face tilt trajectory compensation slope, based on the spatial pose transformation matrix.

[0062] It can be understood that the dual-dimensional tool compensation parameters are physical compensation commands that can be directly executed by the CNC lathe. These are divided into radial positioning translation compensation and end-face tilt trajectory compensation slope, corresponding to the translational and rotational errors of the rotor clamping, respectively. This covers all the effects of clamping errors, achieving dual-dimensional coordinated correction of positioning and trajectory compensation. The spatial pose transformation matrix completely contains all translational and rotational transformation components. The core logic for generating compensation parameters is to extract the corresponding transformation components from the matrix and convert the matrix parameters into displacement and slope parameters executable by the lathe's X / Z axes through mathematical transformation. The radial positioning translation compensation is used to correct the radial position of the tool's cutting start point, offsetting radial eccentricity errors; the end-face tilt trajectory compensation slope is used to dynamically correct the radial following displacement of the tool's axial movement, offsetting end-face double tilt angle errors.

[0063] For example, the radial offset component and the rotational tilt component can be extracted from the spatial pose transformation matrix, and then the radial eccentricity can be vector-synthesized to obtain the radial positioning translation compensation amount. The rotational tilt component can be vector-synthesized to obtain the end face comprehensive tilt angle. Then, the trajectory compensation slope of the tool axial movement can be calculated based on the end face comprehensive tilt angle. Alternatively, the homogeneous coordinate inverse solution can be directly performed on the spatial pose transformation matrix to separate the translation compensation vector and rotational tilt parameter from the inverse matrix. The radial positioning translation compensation amount can be directly calculated through coordinate projection, and the end face tilt trajectory compensation slope can be directly solved through spatial angle linear mapping, etc., but not limited to these methods.

[0064] In one possible implementation, step S400, generating the radial positioning translation compensation amount based on the spatial pose transformation matrix, includes: S410 extracts the radial offset component from the spatial pose change matrix.

[0065] It can be understood that the first two elements of the last column of the spatial pose transformation matrix represent the radial offset components of the actual coordinate system relative to the theoretical coordinate system along the X and Y axes. These directly correspond to the radial eccentricity ΔX and ΔY of the rotor clamping, and are the core input data for generating the radial positioning translation compensation. The extraction process can directly index the elements at the corresponding positions in the matrix. The extracted radial offset components are vector parameters, containing both the magnitude and direction of the offset, fully characterizing all features of the rotor's radial eccentricity; the component accuracy is double-precision floating-point, retained to 0.001mm.

[0066] S420, the radial eccentricity is vector-combined to obtain the radial positioning translation compensation.

[0067] It can be understood that the rotor radial eccentricity is a vector component in two orthogonal directions, the X and Y axes. This two-dimensional planar eccentricity can be converted into a single radial vector compensation amount through vector synthesis calculation. The vector synthesis calculation consists of two steps: first, calculating the total radial offset (eccentricity modulus), and then performing directional calibration in conjunction with the turning radial feed direction, ultimately obtaining the complete radial positioning translation compensation amount, including both magnitude and direction. Since the radial eccentricity is a planar vector, its synthesis follows vector operation rules, accurately representing the total degree of eccentricity of the rotor center relative to the theoretical shaft center; directional calibration ensures that the compensation direction is completely opposite to the eccentricity direction.

[0068] This setup, by accurately extracting the radial offset component from the spatial pose transformation matrix and performing vector synthesis calculations on the radial eccentricity to obtain the radial positioning translation compensation, can transform the two-dimensional radial plane eccentricity deviation generated by rotor clamping into a one-dimensional radial compensation parameter that can be directly executed by the CNC lathe. It accurately quantifies the magnitude and direction of the rotor's radial eccentricity, achieving efficient and distortion-free extraction and synthesis of clamping radial eccentricity errors. The generated radial positioning translation compensation can be directly used for accurate correction of the tool's radial cutting start point, offsetting the machining deviation caused by radial eccentricity from the source, effectively improving the radial positioning accuracy and coaxiality accuracy of generator rotor turning. At the same time, the extraction and vector synthesis operation logic is simple and has a small computational load, adapting to the real-time operation requirements of the CNC system, and providing an accurate and reliable radial compensation basis for the adaptive correction of the rotor turning path.

[0069] In one possible implementation, in step S420, the radial eccentricity is vector-synthesized to obtain the radial positioning translation compensation amount, including: S421, calculate the vector magnitude of the offset components of the radial eccentricity on the X and Y axes to obtain the total radial offset.

[0070] It is understandable that the vector magnitude calculation uses the Pythagorean theorem, with the mathematical formula L=√(ΔX²+ΔY²), where L is the total radial offset, and ΔX and ΔY are the radial offset components along the X and Y axes, respectively. This calculation can accurately obtain the absolute distance of the rotor's radial eccentricity, characterizing the severity of the eccentricity. The total radial offset is a scalar parameter, representing only the magnitude of the eccentricity and not including directional information.

[0071] S422, combined with the radial feed direction of rotor turning, the direction of the total radial offset is calibrated to obtain the final radial positioning translation compensation amount.

[0072] It is understandable that the radial feed direction in rotor turning is a fixed process parameter, typically feed from the outer circle of the rotor towards the center or from the center towards the outer circle. The core of direction calibration is to assign a positive or negative sign to the total radial offset, ensuring that the tool compensation direction is completely opposite to the radial eccentricity direction, thus accurately offsetting the eccentricity error. If the eccentricity direction is in the same direction as the feed direction, the compensation amount is negative; if they are opposite, the compensation amount is positive. The calibrated radial positioning translation compensation amount is a vector parameter, containing both the compensation magnitude and direction, and can be directly written into the lathe tool compensation register for real-time effect.

[0073] This configuration, by decomposing the radial eccentricity into X / Y components and calculating the vector modulus, accurately yields the total radial eccentricity of the rotor, reducing the limitations of single-direction compensation and comprehensively covering all eccentric deviations in the radial plane. Combined with radial feed direction calibration, this ensures the tool compensation direction is completely opposite to the eccentricity deviation, achieving precise cancellation of radial eccentricity errors. This solves the problem of rotor coaxiality exceeding tolerances caused by inaccurate radial eccentricity compensation and incorrect compensation direction in traditional machining. The vector synthesis and direction calibration algorithms are simple, efficient, and robust, and can be executed in real-time in the CNC system, meeting the speed requirements of online dynamic compensation. The compensation parameters are vector parameters directly executable by the lathe, requiring no secondary conversion and achieving extremely high execution accuracy. The generated radial positioning translation compensation adaptively matches clamping scenarios with different eccentricities and directions.

[0074] In one possible implementation, step S400, generating the end-face tilt trajectory compensation slope based on the spatial pose transformation matrix, includes: S430 extracts the rotation tilt component from the spatial pose transformation matrix, and performs vector synthesis of the rotation tilt component to obtain the end face composite tilt angle.

[0075] It can be understood that the first three rows and first three columns of the spatial pose transformation matrix are rotational tilt components, which directly correspond to the spatial double tilt angles α and β of the rotor end face around the X and Y axes. The extraction process involves directly indexing the corresponding elements of the matrix. The end face double tilt angles are rotational vectors in two orthogonal directions. They are converted into a comprehensive end face tilt angle through spatial vector synthesis, which characterizes the total tilt degree of the rotor end face relative to the theoretical reference.

[0076] S440 calculates the trajectory compensation slope of the tool's axial movement based on the end face comprehensive tilt angle.

[0077] It can be understood that the end face tilt trajectory compensation slope is a real-time proportionality coefficient between the tool's axial feed displacement and radial compensation displacement, mathematically expressed as k = tanθ (where θ is the overall end face tilt angle), converting the angular error into a linear slope parameter. The physical meaning of the slope k is: for every 1mm axial movement of the tool, it moves radially by kmm synchronously, dynamically following the end face tilt posture. The slope calculation uses trigonometric tangent operations, executed in real-time by the CNC system, with an accuracy retained to 0.0001.

[0078] This configuration extracts the rotational tilt component from the spatial pose transformation matrix and synthesizes the overall tilt angle of the end face, accurately quantifying the overall tilt attitude of the rotor end face, eliminating dual tilt angle coupling interference, and providing a precise angular basis for trajectory compensation. The overall tilt angle is converted into the end face tilt trajectory compensation slope, transforming the angular error into a linear displacement mapping relationship that the CNC system can directly calculate. This perfectly adapts to the axial + radial linkage feed characteristics of the lathe, resulting in more precise and smoother compensation execution. The end face tilt trajectory compensation slope is a dynamically following compensation parameter, which can correct the tool path in real time throughout the turning process, always conforming to the actual tilt attitude of the rotor end face.

[0079] S500 corrects the positioning coordinates and tool path of turning tools based on two-dimensional tool compensation parameters and axial movement.

[0080] It is understandable that correcting the positioning coordinates and tool path of the turning tool is the final step in the adaptive correction method. The core objective is to use the synergistic effect of two-dimensional tool compensation parameters and axial movement to ensure that the tool positioning coordinates and tool path perfectly match the actual clamping posture of the rotor, eliminating all machining deviations caused by clamping errors. The correction process is divided into two stages: cutting start point positioning correction and full-course tool path correction. The starting point correction corrects the X / Z coordinates of the tool's cutting start point through radial positioning translation compensation and axial movement, eliminating clamping translation errors. The full-course path correction dynamically corrects the radial following displacement of the tool's axial movement through end face tilt trajectory compensation slope, eliminating clamping rotation errors.

[0081] For example, the radial cutting start coordinates of the turning tool can be translated and corrected according to the radial positioning translation compensation amount, and the axial cutting start coordinates of the turning tool can be compensated and corrected according to the axial movement amount. The axial feed displacement of the lathe and the radial compensation displacement can be established in real time according to the end face tilt trajectory compensation slope, and the axial tool path can be dynamically corrected. Alternatively, the theoretical tool path can be transformed point by point according to the spatial pose transformation matrix, and all theoretical trajectory points can be converted into target coordinate points that are adapted to the actual clamping pose to form a continuously corrected tool path. At the same time, the transformed first point coordinates can be directly used as the tool cutting positioning coordinates, etc., but not limited to these.

[0082] In one possible implementation, step S500 involves correcting the positioning coordinates and tool path of the turning tool based on the dual-dimensional tool compensation parameters and axial movement, including: Based on the radial positioning translation compensation amount, the radial cutting start coordinates of the turning tool are translated and corrected; based on the axial movement amount, the axial cutting start coordinates of the turning tool are compensated and corrected; based on the end face tilt trajectory compensation slope, a real-time mapping relationship is established between the lathe axial feed displacement and the radial compensation displacement, and the axial tool path is dynamically corrected.

[0083] It can be understood that the radial cutting start coordinate of the turning tool is the theoretical X-axis coordinate. The logic of translation correction is Xactual = Xtheoretical + radial positioning translation compensation amount. The radial compensation amount directly corrects the initial radial position of the tool, offsetting the rotor's radial eccentricity error. The axial cutting start coordinate of the turning tool is the theoretical Z-axis coordinate. The mathematical logic of axial movement compensation correction is Zactual = Ztheoretical + axial movement amount. The axial movement amount directly corrects the initial axial position of the tool, offsetting the rotor's axial movement / clamping depth error. The end face tilt trajectory compensation slope is a linear compensation coefficient obtained in the early stage through spatial double tilt angle vector synthesis and trigonometric function calculation. Its physical meaning is the proportion of radial displacement that needs to be synchronously compensated for every unit displacement of the tool's axial movement. Its core function is to offset the spatial double tilt angle error of the rotor end face around the X and Y axes, ensuring that the tool trajectory always conforms to the actual end face shape of the rotor. When turning the rotor end face on a CNC lathe, the tool's main motion is axial (Z-axis) feed, and the radial (X-axis) motion is an auxiliary linkage motion. The core mathematical model for establishing the real-time mapping relationship is: radial real-time compensation displacement ΔX = end face tilt trajectory compensation slope k × axial real-time feed displacement ΔZ. This mapping relationship is embedded in the interpolation calculation module of the CNC system and is executed synchronously within each interpolation cycle (microsecond level) of the lathe, realizing dynamic real-time correction of the tool trajectory. During the turning process, the CNC system collects the feed displacement data of the Z-axis in real time, calculates the corresponding compensation displacement of the X-axis in real time according to the mapping relationship, and superimposes the compensation displacement into the radial motion command of the tool. This allows the tool to feed along the Z-axis while the X-axis synchronously follows the end face tilt posture to make adaptive offset, forming a dynamically corrected tool trajectory that perfectly matches the actual end face of the rotor. This dynamic correction method is not fixed-point compensation, but a continuous and uninterrupted trajectory adaptation throughout the entire process. It can cover the tilt error of the entire rotor end face from the center to the edge, and can accurately follow and correct regardless of the size or direction of the end face tilt angle.

[0084] This setup, through radial positioning translation compensation, corrects the radial cutting start coordinates of the turning tool, and combines axial movement compensation to correct the axial cutting start coordinates. Then, based on the end face tilt trajectory compensation slope, a real-time mapping relationship between the lathe's axial feed displacement and radial compensation displacement is established to dynamically correct the tool's axial path. This comprehensively offsets the radial eccentricity, axial movement, and coupled clamping errors of the double tilt angles of the end face space generated by the generator rotor clamping from two dimensions: cutting start positioning and the entire tool path. This allows the positioning coordinates of the turning tool to accurately match the actual clamping posture of the rotor, significantly improving the coaxiality, flatness, and axial dimensional accuracy of the rotor end face turning.

[0085] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0086] Corresponding to the adaptive correction method for generator rotor turning path described in the above embodiments, this application also provides an adaptive correction device for generator rotor turning path, the various modules of which can implement the various steps of the adaptive correction method for generator rotor turning path. Figure 4 The diagram shows a structural block diagram of the generator rotor turning path adaptive correction device provided in the embodiments of this application. For ease of explanation, only the parts related to the embodiments of this application are shown.

[0087] Reference Figure 4 The adaptive correction device for the generator rotor turning path includes: The acquisition unit is used to acquire multiple sets of original detection coordinate data of the rotor end face after detecting the received command; wherein, the received command is used to instruct the CNC lathe tool post to move along a preset safety path to the generator rotor end face to be processed, and to execute the command after multi-point touch position detection.

[0088] The calculation and separation unit is used to solve the spatial plane data of the actual end face of the rotor using the least squares spatial plane fitting algorithm on multiple sets of original detection coordinate data, and simultaneously separates the three coupled error parameters of the rotor clamping: radial eccentricity, axial movement, and end face spatial double tilt angle.

[0089] The construction unit is used to reconstruct the actual machining reference coordinate system that matches the actual rotor pose based on three coupling error parameters, and to calculate the spatial pose transformation matrix of the actual machining reference coordinate system relative to the theoretical machining reference coordinate system.

[0090] The generation unit is used to generate two-dimensional tool compensation parameters based on the spatial pose transformation matrix, namely the radial positioning translation compensation amount and the end face tilt trajectory compensation slope.

[0091] The correction unit is used to correct the positioning coordinates and tool path of the turning tool based on two-dimensional tool compensation parameters and axial movement.

[0092] It should be noted that the information interaction and execution process between the above modules are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0093] Those skilled in the art will understand that, for the sake of convenience and brevity, the above-described module division is merely an example. In practical applications, the functions described above can be assigned to different modules as needed, that is, the internal structure of the device can be divided into different modules to complete all or part of the functions described above. The modules in the embodiments can be integrated into one processing unit, or each module can exist physically separately, or two or more modules can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the modules in the above device can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0094] This application also provides a control device. Figure 5 This is a schematic diagram of the structure of a control device 6 provided in an embodiment of this application. Figure 5 As shown, the control device 6 in this embodiment includes: at least one processor 60 ( Figure 5 Only one is shown in the image), at least one memory 61 ( Figure 5 (Only one is shown in the image) and a computer program 62 stored in the at least one memory 61 and executable on the at least one processor 60, wherein when the processor 60 executes the computer program 62, it causes the control device 6 to implement the steps in any of the above embodiments of the adaptive correction method for the turning path of the generator rotor, or causes the control device 6 to implement the functions of each module in the above embodiments of the device.

[0095] For example, the computer program 62 may be divided into one or more modules / units, which are stored in the memory 61 and executed by the processor 60 to complete this application. The one or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program 62 in the control device 6.

[0096] The control device 6 can be a computing device such as a PLC or a microcontroller. This control device may include, but is not limited to, a processor 60 and a memory 61. Those skilled in the art will understand that... Figure 5 This is merely an example of control device 6 and does not constitute a limitation on control device 6. It may include more or fewer components than shown, or combine certain components, or different components, such as input / output devices, network access devices, buses, etc.

[0097] The processor 60 can be a Central Processing Unit (CPU), or it can be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0098] In some embodiments, the memory 61 may be an internal storage unit of the control device 6, such as a hard disk or memory of the control device 6. In other embodiments, the memory 61 may be an external storage device of the control device 6, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the control device 6. Furthermore, the memory 61 may include both internal storage units and external storage devices of the control device 6. The memory 61 is used to store operating systems, applications, bootloaders, data, and other programs, such as the program code of computer programs. The memory 61 can also be used to temporarily store data that has been output or will be output.

[0099] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps in any of the above method embodiments.

[0100] This application provides a computer program product that, when run on a control device 6, causes the control device 6 to perform the steps in any of the above-described method embodiments.

[0101] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying the computer program code to a control device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks.

[0102] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0103] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0104] In the embodiments provided in this application, it should be understood that the disclosed control devices and apparatuses can be implemented in other ways. For example, the above-described embodiments of the generator rotor turning path adaptive correction device are merely illustrative. For instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual couplings, direct couplings, or communication connections may be indirect couplings or communication connections through some interfaces, devices, or modules, and may be electrical, mechanical, or other forms.

[0105] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0106] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. An adaptive correction method for the turning path of a generator rotor, characterized in that, include: After detecting the receiving command, multiple sets of original detection coordinate data of the rotor end face are acquired; wherein, the receiving command is used to instruct the CNC lathe tool post to move along a preset safety path to the generator rotor end face to be processed, and to execute the command after multi-point touch position detection; The least squares spatial plane fitting algorithm is used to solve the spatial plane data of the actual end face of the rotor for multiple sets of original detection coordinate data, and the three coupled error parameters of the rotor clamping radial eccentricity, axial movement and end face spatial double tilt angle are obtained simultaneously. Based on the three coupling error parameters, a real machining reference coordinate system matching the actual rotor pose is reconstructed, and the spatial pose transformation matrix of the real machining reference coordinate system relative to the theoretical machining reference coordinate system is calculated. Based on the spatial pose transformation matrix, two-dimensional tool compensation parameters are generated, namely radial positioning translation compensation amount and end face tilt trajectory compensation slope. The positioning coordinates and tool path of the turning tool are corrected based on the dual-dimensional tool compensation parameters and the axial movement.

2. The adaptive correction method for generator rotor turning path as described in claim 1, characterized in that, The process of using a least-squares spatial plane fitting algorithm to solve for the spatial plane data of the actual rotor end face from multiple sets of the original detection coordinate data includes: The least squares spatial plane fitting algorithm is used to calculate the general plane equation of the actual end face of the rotor by applying the least squares method to multiple sets of the original detection coordinate data. Based on the general equation of the plane, the normal vector of the actual end face is extracted to obtain the spatial plane data of the actual end face of the rotor.

3. The adaptive correction method for generator rotor turning path as described in claim 2, characterized in that, The multiple sets of original detection coordinate data include at least one of the coordinate data of the rotor end face center detection point and the coordinate data of the rotor end face edge detection point. The separation of the three coupled error parameters of the rotor clamping radial eccentricity, axial movement, and end face spatial double tilt angle includes: Using the coordinate data of the center detection point on the rotor end face as a reference, the difference between the center detection point and the theoretical shaft center coordinates is calculated to obtain the radial eccentricity and axial movement. Based on the angle between the normal vector of the actual end face and the normal vector of the theoretical end face reference, the spatial double tilt angles of the rotor end face around the X-axis and Y-axis are calculated respectively.

4. The adaptive correction method for generator rotor turning path as described in claim 3, characterized in that, The reconstruction of the actual machining reference coordinate system based on the three coupling error parameters and matching the actual rotor pose includes: A coordinate system for actual machining is constructed with the geometric center of the actual end face of the rotor as the origin of the coordinate system, the normal vector of the actual end face as the Z-axis, and the theoretical radial direction of the rotor as the X and Y axes.

5. The adaptive correction method for generator rotor turning path as described in claim 4, characterized in that, The calculation of the spatial pose transformation matrix of the actual machining reference coordinate system relative to the theoretical machining reference coordinate system includes: Based on the radial eccentricity and the axial axial displacement, construct the translation transformation matrix of the actual machining reference coordinate system relative to the theoretical machining reference coordinate system; Based on the aforementioned spatial double tilt angle, construct the rotation transformation matrix of the actual machining reference coordinate system relative to the theoretical machining reference coordinate system; The translation transformation matrix and the rotation transformation matrix are fused to obtain the final spatial pose transformation matrix.

6. The adaptive correction method for generator rotor turning path as described in claim 5, characterized in that, The step of fusing the translation transformation matrix and the rotation transformation matrix to obtain the final spatial pose transformation matrix includes: Using the three-dimensional homogeneous coordinate transformation rule, the rotation transformation matrix and the translation transformation matrix are multiplied sequentially. The matrix multiplication result is used to construct a homogeneous spatial pose transformation matrix that simultaneously includes translational offset and rotational tilt, which serves as the final spatial pose transformation matrix.

7. The adaptive correction method for generator rotor turning path as described in claim 1, characterized in that, Generate radial positioning translation compensation based on the spatial pose transformation matrix, including: Extract the radial offset component from the spatial pose change matrix; The radial eccentricity is vector-synthesized to obtain the radial positioning translation compensation.

8. The adaptive correction method for generator rotor turning path as described in claim 7, characterized in that, The step of performing vector synthesis calculation on the radial eccentricity to obtain the radial positioning translation compensation includes: The total radial offset is obtained by calculating the vector magnitude of the offset components of the radial eccentricity on the X and Y axes. By combining the radial feed direction of rotor turning, the total radial offset is calibrated to obtain the final radial positioning translation compensation amount.

9. The adaptive correction method for generator rotor turning path as described in claim 1, characterized in that, Generating the end face tilt trajectory compensation slope based on the spatial pose transformation matrix includes: The rotation tilt component is extracted from the spatial pose transformation matrix, and the rotation tilt component is vector synthesized to obtain the end face comprehensive tilt angle; The trajectory compensation slope of the tool's axial movement is calculated based on the comprehensive tilt angle of the end face.

10. The adaptive correction method for generator rotor turning path as described in claim 1, characterized in that, The positioning coordinates and tool path of the turning tool corrected based on the dual-dimensional tool compensation parameters and the axial movement amount include: Based on the radial positioning translation compensation amount, the radial cutting start coordinates of the turning tool are translated and corrected; based on the axial movement amount, the axial cutting start coordinates of the turning tool are compensated and corrected; based on the end face tilt trajectory compensation slope, a real-time mapping relationship is established between the lathe axial feed displacement and the radial compensation displacement, and the axial tool path is dynamically corrected.

Images