A reverse turning method and device for machining an elongated shaft part

Abstract

The application provides a reverse turning method and device for machining an elongated shaft part, and relates to the technical field of turning machining. A turning gap boundary of the part in axial elastic clamping turning is generated according to turning response data and contact state characteristics, and a radial contact index between each supporting claw in the lathe chuck and the outer wall of the part is determined. When the radial contact index between any supporting claw and the outer wall of the part is lower than the turning gap boundary, it is determined that the corresponding supporting claw forms radial under-restraint to the part. When the radial contact index between any supporting claw and the outer wall of the part is higher than the turning gap boundary, it is determined that the corresponding supporting claw forms radial over-restraint to the part. The radial abutting amount of the corresponding supporting claw to the outer wall of the part in reverse turning is adjusted according to the radial under-restraint and the radial over-restraint. The application can realize self-adaptive adjustment of the radial restraint state in reverse turning, so as to improve the machining precision of the reverse turning of the elongated shaft part.

Description

Technical Field

[0001] This application relates to the field of turning technology, and more specifically, to a reverse turning method and apparatus for machining slender shaft parts. Background Technology

[0002] Turning is one of the most basic and widely used metal cutting methods in the field of mechanical manufacturing. Its core principle is that the spindle drives the workpiece to make a rotational motion on the lathe, while the cutting tool makes a linear feed motion along the axial or radial direction of the workpiece. The excess material is removed through the relative motion between the tool and the workpiece, so that the workpiece can obtain the dimensional accuracy, shape accuracy and surface quality that meet the design requirements. Turning can complete the machining of various features such as internal and external cylindrical surfaces, conical surfaces, end faces, threads, grooves and shaped rotating surfaces. It is suitable for most engineering materials such as steel, cast iron, non-ferrous metals and alloys, and is the main machining method for all rotating parts such as shafts and discs.

[0003] However, traditional reverse-feed turning of slender shaft parts lacks real-time dynamic perception and precise quantification of the radial constraint state during machining. It also fails to incorporate axial elastic clamping boundary conditions and chuck overall clamping force balance constraints for multi-dimensional correction and verification. This makes the radial under-constraint and over-constraint states susceptible to misjudgment due to instantaneous fluctuations. The adjustment process can easily lead to eccentric deformation and additional bending stress in the part. Furthermore, existing adaptive fixtures for thin-walled parts can only achieve static stiffness switching during clamping, failing to address the real-time changes in radial constraint state caused by dynamic factors such as cutting force fluctuations, thermal expansion, and vibration during the turning of slender shafts. This makes it difficult to meet the high-precision machining requirements of slender shaft parts. Therefore, how to achieve adaptive adjustment of the radial constraint state during reverse turning to improve the machining accuracy of slender shaft parts is a challenge facing the industry. Summary of the Invention

[0004] This application provides a reverse turning method and apparatus for machining slender shaft parts, which can adaptively adjust the radial constraint state during reverse turning to improve the machining accuracy of slender shaft parts.

[0005] In a first aspect, this application provides a reverse turning method for machining slender shaft parts, the reverse turning method comprising the following steps:

[0006] Collect axial turning response data and contact state characteristics of slender shaft parts during reverse feed turning;

[0007] Based on the turning response data and the contact state characteristics, the turning clearance boundary of the part during axial elastic clamping turning is generated, and the radial contact index between each support jaw in the lathe chuck and the outer wall of the part is determined.

[0008] When the radial contact index between any support claw and the outer wall of the part is detected to be lower than the turning clearance boundary, it is determined that the corresponding support claw forms a radial under-constraint on the part. When the radial contact index between any support claw and the outer wall of the part is detected to be higher than the turning clearance boundary, it is determined that the corresponding support claw forms a radial over-constraint on the part.

[0009] The radial clamping amount of the corresponding support claw against the outer wall of the part is adjusted according to the radial under-constraint and the radial over-constraint during reverse turning.

[0010] In this embodiment, the acquisition of axial turning response data and contact state characteristics of slender shaft-type parts during reverse feed turning specifically includes:

[0011] The turning start characteristics are determined based on the tool holder strain markings during reverse turning of slender shaft parts.

[0012] The turning state information of slender shaft parts during reverse feed turning is marked according to the turning start characteristics.

[0013] Extract the axial turning response data and contact state characteristics of slender shaft-type parts during reverse feed turning from the turning state information.

[0014] In this embodiment, generating the turning clearance boundary of the part during axial elastic clamping turning based on the turning response data and the contact state characteristics specifically includes:

[0015] Extract the time-domain synchronization feature set of cutting force fluctuation and workpiece elastic displacement from the turning response data;

[0016] Based on the time-domain synchronization feature set, the contact state identifier during axial elastic clamping turning is obtained, and the contact stiffness coefficient is determined.

[0017] The contact state identifier and the contact stiffness coefficient are mapped to the turning dynamics model to generate the turning clearance boundary during axial elastic clamping turning.

[0018] In this embodiment, determining the radial contact index between each support jaw in the lathe chuck and the outer wall of the part specifically includes:

[0019] Collect the time-domain signals of radial clamping force and micro-displacement between each support jaw in the lathe chuck and the outer wall of the part;

[0020] After determining the contact stiffness of each support claw based on the radial clamping force and the micro-displacement time-domain signal, the contact state feature spectrum is extracted.

[0021] By correlating and mapping the contact state characteristic spectrum and the elastic clamping boundary conditions, the radial contact index between each support jaw in the lathe chuck and the outer wall of the part is obtained.

[0022] In this embodiment, the extraction of the contact state feature spectrum after determining the contact stiffness of each support claw based on the radial clamping force and the micro-displacement time-domain signal specifically includes:

[0023] The dynamic matrix of contact stiffness of each support claw is determined based on the time-varying cross-correlation function of the radial clamping force and the micro-displacement time-domain signal.

[0024] Based on the dynamic matrix of contact stiffness, the energy distribution of each mode is extracted by variational mode decomposition and mapped to the contact state feature spectrum.

[0025] In this embodiment, when the radial contact index between any support claw and the outer wall of the part is detected to be lower than the turning clearance boundary, determining that the corresponding support claw forms a radial under-constraint on the part specifically includes:

[0026] Based on the real-time comparison between the radial contact index of each support claw and the turning clearance boundary, a radial constraint state residual sequence is constructed.

[0027] Based on the radial constraint state residual sequence, a residual cumulative energy threshold within a variable-scale time window is introduced to determine the radial contact degradation characteristic degree.

[0028] Extract the reduced-order modes that exceed the expected margin from the radial contact degradation eigenvalues ​​and generate the corresponding radial under-constraint decision events for the support claws.

[0029] In this embodiment, when the radial contact index between any support claw and the outer wall of the part is detected to be higher than the turning clearance boundary, determining that the corresponding support claw forms a radial over-constraint on the part specifically includes:

[0030] Based on the radial contact index of each support claw and the deviation sequence of the turning clearance boundary, a radial over-constraint feature comparison matrix is ​​constructed;

[0031] Perform amplitude-based hierarchical comparison of the radial over-constraint feature comparison matrix to determine the over-constraint contact margin;

[0032] The over-constraint contact margin is mapped to a preset radial clamping upper limit threshold to determine and generate the corresponding radial over-constraint event for the support claw.

[0033] In this embodiment, performing amplitude-based hierarchical comparison of the radial over-constraint feature comparison matrix to determine the over-constraint contact margin specifically includes:

[0034] Based on the contact stress fluctuation time history in the radial over-constraint feature comparison matrix, an amplitude-based comparison of contact stress fluctuation is performed to generate a graded identifier for the over-constraint contact stress amplitude.

[0035] Based on the graded identifier, a preset radial over-constraint contact margin mapping function is introduced to determine the over-constraint contact margin.

[0036] In this embodiment, adjusting the radial clamping amount of the corresponding support jaw against the outer wall of the part during reverse turning based on the radial under-constraint and the radial over-constraint specifically includes:

[0037] Based on the constraint deviation feature vectors of the radial under-constraint and the radial over-constraint, a clamping amount compensation mapping function is constructed;

[0038] Based on the aforementioned clamping amount compensation mapping function, the reverse turning axial elastic clamping boundary condition is introduced to determine the radial clamping correction amount of the corresponding support claw;

[0039] The radial clamping correction amount is output to the chuck servo control unit to adjust the radial clamping amount of the corresponding support claw against the outer wall of the part.

[0040] Secondly, this application provides a reverse turning apparatus for machining slender shaft parts, used to perform a reverse turning method for machining slender shaft parts, the reverse turning apparatus comprising:

[0041] The acquisition module is used to acquire axial turning response data and contact state characteristics of slender shaft parts during reverse feed turning;

[0042] The processing module is used to generate the turning clearance boundary of the part during axial elastic clamping turning based on the turning response data and the contact state characteristics, and to determine the radial contact index between each support jaw in the lathe chuck and the outer wall of the part.

[0043] The processing module is further configured to determine that the corresponding support claw forms a radial under-constraint on the part when the radial contact index between any support claw and the outer wall of the part is lower than the turning clearance boundary, and to determine that the corresponding support claw forms a radial over-constraint on the part when the radial contact index between any support claw and the outer wall of the part is higher than the turning clearance boundary.

[0044] The execution module is used to adjust the radial clamping amount of the corresponding support claw against the outer wall of the part during reverse turning based on the radial under-constraint and the radial over-constraint.

[0045] The technical solutions provided by the embodiments disclosed in this application have the following beneficial effects:

[0046] The axial turning response data and contact state characteristics of slender shaft-like parts are collected during reverse feed turning. Based on the turning response data and contact state characteristics, the turning clearance boundary of the part during axial elastic clamping turning is generated, and the radial contact index between each support jaw in the lathe chuck and the outer wall of the part is determined. When the radial contact index between any support jaw and the outer wall of the part is detected to be lower than the turning clearance boundary, it is determined that the corresponding support jaw forms a radial under-constraint on the part. When the radial contact index between any support jaw and the outer wall of the part is detected to be higher than the turning clearance boundary, it is determined that the corresponding support jaw forms a radial over-constraint on the part. The radial clamping amount of the corresponding support jaw on the outer wall of the part during reverse turning is adjusted according to the radial under-constraint and the radial over-constraint.

[0047] Therefore, in this application, the radial clamping amount of the corresponding support claw to the outer wall of the part is adjusted according to the radial under-constraint and the radial over-constraint during reverse turning. Among them, by determining the radial contact index, a dimensionless parameter that can quantitatively characterize the degree of radial contact between a single support claw and the outer wall of the part can be obtained, thereby transforming the originally ambiguous contact interface constraint state into a quantifiable and comparable standardized numerical index. This parameter integrates multi-dimensional mechanical characteristics such as contact stiffness, clamping force fluctuation, and micro-displacement change, while incorporating the coupling effect of axial elastic clamping boundary conditions. It eliminates the subjective error of traditional qualitative judgment, making the constraint state under different cutting conditions laterally comparable, and providing a precise input basis for the adaptive adjustment system. Determining radial over / under constraint allows for the identification of standardized judgment events that accurately distinguish between effective constraint anomalies and instantaneous process fluctuations. This enables precise identification and severity quantification of constraint state anomalies. Through time-series residual accumulation analysis, modal energy decomposition, and multi-feature hierarchical comparison, this process effectively filters out misjudgments caused by interference factors such as sensor noise and instantaneous fluctuations in cutting force. At the same time, it quantifies the duration and impact of the anomaly, allowing the system to output corresponding tightening correction values ​​based on the severity of the anomaly, achieving hierarchical and precise adjustment and avoiding additional deformations caused by over- or under-adjustment.

[0048] In summary, the technical solution adopted in this application can achieve adaptive adjustment of the radial constraint state during reverse turning, thereby improving the machining accuracy of reverse turning of slender shaft parts. Attached Figure Description

[0049] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, 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 for this embodiment of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0050] Figure 1This is an exemplary flowchart of a reverse turning method for machining slender shaft parts according to the present application;

[0051] Figure 2 This is a flowchart illustrating the determination of the radial contact index provided in this application;

[0052] Figure 3 It is based on the reverse feed turning control principle diagram provided in this application;

[0053] Figure 4 This is a modular structure diagram of a reverse turning device for machining slender shaft parts, provided in this application. Detailed Implementation

[0054] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0055] This application provides a reverse turning method and apparatus for machining slender shaft parts. The core of this method involves collecting axial turning response data and contact state characteristics of the slender shaft part during reverse feed turning; generating a turning clearance boundary for the part during axial elastic clamping turning based on the turning response data and contact state characteristics; and determining the radial contact index between each support jaw in the lathe chuck and the outer wall of the part; when the radial contact index between any support jaw and the outer wall of the part is detected to be lower than the turning clearance boundary, it is determined that the corresponding support jaw forms a radial under-constraint on the part; when the radial contact index between any support jaw and the outer wall of the part is detected to be higher than the turning clearance boundary, it is determined that the corresponding support jaw forms a radial over-constraint on the part; and adjusting the radial clamping amount of the corresponding support jaw against the outer wall of the part during reverse turning based on the radial under-constraint and the radial over-constraint.

[0056] Example 1: To better understand the above technical solution, the following will provide a detailed description of the technical solution in conjunction with the accompanying drawings and specific implementation methods. (Refer to...) Figure 1 As shown in the figure, this is an exemplary flowchart of a reverse turning method for machining slender shaft parts according to this embodiment of the present application. The reverse turning method includes the following steps:

[0057] In step S1, the axial turning response data and contact state characteristics of slender shaft-type parts are collected during reverse feed turning.

[0058] In this embodiment, the following steps can be used to collect axial turning response data and contact state characteristics of slender shaft-type parts during reverse feed turning:

[0059] The turning start characteristics are determined based on the tool holder strain markings during reverse turning of slender shaft parts.

[0060] The turning state information of slender shaft parts during reverse feed turning is marked according to the turning start characteristics.

[0061] Extract the axial turning response data and contact state characteristics of slender shaft-type parts during reverse feed turning from the turning state information.

[0062] In practice, firstly, two resistance strain gauges are attached to the side of the tool holder near the tool tip to form a half-bridge measurement circuit. This circuit is connected to a multi-channel data acquisition system via shielded wires. Three no-load operation tests are performed beforehand, and the average strain signal of the tool holder under no cutting force is recorded as the baseline value. Then, five trial cutting tests with the same cutting parameters are performed, and the amplitude of the sudden change in strain signal when the tool just enters the workpiece surface is statistically analyzed. 80% of the minimum value is taken as the sudden change threshold. The strain signal is monitored in real time. When the strain value exceeds the baseline value and reaches the sudden change threshold and remains stable for more than 20ms, it is determined that a turning start characteristic has been detected. Then, a ring buffer with a depth of 200ms is pre-activated to continuously buffer all raw sensor sampling data. When the strain signal of the tool holder exceeds the pre-calibrated abrupt change threshold, the precise sampling point number corresponding to the trigger moment is immediately latched. All continuous sampling data from 20ms before the trigger moment to after the trigger moment are extracted from the annular buffer. An absolute timestamp with an accuracy of no less than 1 microsecond is added to each independent sampling point. The sampling point where the strain signal first exceeds the threshold is taken as the globally unified turning start time zero point, with a timestamp accuracy of no less than 1 microsecond, forming turning state information with continuous time sequence markings. Finally, the turning state information with precise timestamps is read from the high-speed buffer inside the system in chronological order. All axial displacement and axial acceleration raw data from the turning start time point to the end of the current acquisition cycle are extracted as axial turning response data. At the same time, the measurement data of all pressure sensors on each support claw within the same time period are extracted. The arithmetic mean of each support claw in each acquisition cycle is calculated. After removing outliers exceeding three times the standard deviation, it is taken as the contact state feature corresponding to that support claw.

[0063] It should be noted that, in this application, reverse feed turning refers to a machining method in which the cutting tool feeds along the lathe guideway from the tailstock end to the chuck end; tool holder strain indicator refers to the electrical signal characteristics output by the tool holder when it undergoes elastic deformation under the action of cutting force; turning start characteristics refer to the step change characteristics of the tool holder strain signal when the tool just enters the surface of the part; turning state information refers to a time series data set including timestamp, axial displacement, axial acceleration, and contact pressure of each support jaw; turning response data is a multi-channel synchronous time series data set including tool holder strain signal, axial displacement signal, and axial acceleration signal; contact state characteristics are a multi-dimensional mechanical characteristic set composed of contact stiffness, contact stress fluctuation amplitude, and micro-displacement change rate.

[0064] It should be noted that, in this application, before turning slender shaft-type parts, one end of the slender shaft-type parts is clamped in the lathe chuck, and the other end is axially elastically clamped by the tailstock elastic center. The cutting tool is driven to perform longitudinal feed turning along the axis of the part from the end near the lathe chuck to the end near the tailstock elastic center.

[0065] In step S2, the turning clearance boundary of the part during axial elastic clamping turning is generated based on the turning response data and the contact state characteristics, and the radial contact index between each support jaw in the lathe chuck and the outer wall of the part is determined.

[0066] In this embodiment, generating the turning clearance boundary of the part during axial elastic clamping turning based on the turning response data and the contact state characteristics can be achieved by the following steps:

[0067] Extract the time-domain synchronization feature set of cutting force fluctuation and workpiece elastic displacement from the turning response data;

[0068] Based on the time-domain synchronization feature set, the contact state identifier during axial elastic clamping turning is obtained, and the contact stiffness coefficient is determined.

[0069] The contact state identifier and the contact stiffness coefficient are mapped to the turning dynamics model to generate the turning clearance boundary during axial elastic clamping turning.

[0070] In practice, firstly, the cutting force signal converted from the tool holder strain signal and the workpiece axial elastic displacement signal collected by the axial displacement sensor are extracted from the axial turning response data according to a unified timestamp. The two sets of signals are divided into sliding windows, and the window length is set to a fixed value that includes multiple continuous acquisition cycles. The root mean square value and peak frequency of the cutting force in each window, as well as the maximum amplitude of the workpiece elastic displacement and the phase difference between the cutting force and the displacement, are calculated. The parameters of the cutting force in each window are combined in time order to form a time-domain synchronous feature set. Then, a correspondence table between time-domain synchronization characteristics and contact states was established in advance through multiple sets of process tests. During the tests, different support claw clamping amounts were set under the same cutting parameters. The corresponding characteristic parameters and actual contact states were recorded by pressure sensors and displacement sensors. All test data were stored in the database after being verified three times. In advance, independent contact stiffness coefficient tables were established for four contact states: separation, point contact, surface contact, and interference contact, through static calibration tests. Each contact stiffness coefficient table uses the radial clamping amount of the support claw as the independent variable and the contact stiffness coefficient as the dependent variable. After the contact state identifier is obtained through time-domain synchronization characteristic matching, the stiffness coefficient table of the corresponding state is selected first. Then, based on the actual clamping amount of the support claw fed back by the current servo encoder, linear interpolation is performed in the contact stiffness coefficient table to obtain the contact stiffness coefficient under the current working condition. Finally, the radial contact index is the ratio of the current contact stiffness to the reference contact stiffness multiplied by the clamping force normalization coefficient. This radial contact index changes continuously with the amount of clamping of the support jaws. In the simulation of the turning dynamics model, by scanning the range of values ​​of the clamping amount of the support jaws, the corresponding radial contact index and contact stiffness are calculated in sequence. Substituted into the turning dynamics model based on the Euler-Bernoulli beam theory, the maximum bending deformation and vibration amplitude of the part under different radial contact indices are obtained. When the maximum bending deformation and vibration amplitude simultaneously meet the machining accuracy requirements specified in the part design drawings, the corresponding radial contact index range is the turning clearance boundary.

[0071] It should be noted that, in this application, axial elastic clamping turning refers to a process method in the turning of slender shafts in which a tailstock center with axial elastic extension capability is used to axially clamp the part; cutting force fluctuation refers to the dynamic change of the cutting force acting on the tool over time during the turning process; workpiece elastic displacement refers to the recoverable elastic deformation displacement of the workpiece under the action of cutting force; time-domain synchronous feature set refers to the associated feature group composed of the dynamic change characteristics of cutting force and the dynamic response characteristics of workpiece axial elastic displacement under the same time reference; contact state identifier is a classification mark that characterizes the macroscopic contact state of the contact interface between the chuck support jaw and the outer wall of the part; contact stiffness coefficient refers to a physical parameter that quantitatively describes the ability of the contact interface to resist elastic deformation; turning dynamics model refers to a dynamic simulation model of the turning process of slender shafts based on the classical Euler-Bernoulli beam theory; turning clearance boundary refers to the critical range of the radial contact index that ensures the stability of the turning process and that the machining error of the part is within the allowable range.

[0072] Preferably, in this embodiment, the radial contact index between each support jaw in the lathe chuck and the outer wall of the part is determined by reference. Figure 2 As shown in the figure, this is a flowchart illustrating the determination of the radial contact index in some embodiments of this application. In this embodiment, the radial contact index can be determined using the following steps:

[0073] In step S21, the radial clamping force and micro-displacement time-domain signals between each support jaw in the lathe chuck and the outer wall of the part are acquired;

[0074] In step S22, the contact stiffness of each support claw is determined based on the radial clamping force and the micro-displacement time-domain signal, and then the contact state feature spectrum is extracted.

[0075] In step S23, the contact state characteristic spectrum and the elastic clamping boundary condition are correlated and mapped to obtain the radial contact index between each support claw in the lathe chuck and the outer wall of the part.

[0076] In practice, firstly, a miniature piezoresistive force sensor is embedded on the working surface of each support claw in contact with the part. The force-bearing surface of the sensor is flush with the working surface of the support claw. A high-precision capacitive micro-displacement sensor is installed on the side of each support claw, with the sensor measuring head facing the outer wall of the part and the measurement direction consistent with the radial direction. A multi-channel synchronous data acquisition system is used to simultaneously acquire the radial clamping force from all force sensors and the micro-displacement time-domain signal from the micro-displacement sensor at a uniform sampling frequency. The acquisition process is synchronized with the turning process. Then, the synchronously acquired radial clamping force and micro-displacement time-domain signals are divided into sliding windows. The window length is set to a fixed value containing multiple continuous acquisition cycles. Within each window, a force-displacement curve is plotted with the radial clamping force as the abscissa and the micro-displacement as the ordinate. The least squares method is used to fit the linear segment of the curve. The slope of the fitted line is the contact stiffness within that window. The average value, standard deviation, rate of change of the contact stiffness within each window, as well as the root mean square value of the radial clamping force, are extracted to form a contact state characteristic spectrum. Finally, an association mapping table between the contact state characteristic spectrum, elastic clamping boundary conditions, and radial contact index was established in advance through orthogonal experiments. During the experiment, the corresponding contact state characteristic spectrum and actual processing effect were measured under different axial clamping forces and different support claw clamping amounts. The true value label of the radial contact index can be calculated for each test point using the above formula. The radial contact index corresponding to the best processing effect is calibrated as the benchmark value. The contact state characteristic spectrum and the current elastic clamping boundary conditions are input into the association mapping table in real time, and the current radial contact index is calculated using the weighted average method.

[0077] It should be noted that in this application, radial clamping force refers to the normal pressure applied by the support claw to the outer wall of the part in the radial direction; micro-displacement time-domain signal refers to the continuous signal of the small relative displacement in the radial direction at the contact interface between the support claw and the outer wall of the part changing with time; contact stiffness refers to the increase in radial clamping force required to produce a unit radial elastic deformation at the contact interface; contact state characteristic spectrum refers to an ordered set of features composed of multiple characteristic parameters that can reflect the state of the contact interface from different angles; elastic clamping boundary condition refers to the axial constraint boundary condition jointly determined by the axial elastic clamping turning process parameters and the mechanical properties of the part; correlation mapping refers to the process of mapping the contact state characteristic spectrum and elastic clamping boundary condition to the radial contact index; the radial contact index refers to the dimensionless parameter of the radial contact tightness between a single support claw and the outer wall of the part, which can be directly calculated by the following formula: ,in, Radial contact index, dimensionless. The measured contact stiffness is expressed in N / μm. This refers to the reference contact stiffness (rated contact stiffness under surface contact conditions), expressed in N / μm. This represents the current radial clamping force, in N. The reference clamping force is expressed in N.

[0078] In addition, in this embodiment, the extraction of the contact state feature spectrum after determining the contact stiffness of each support claw based on the radial clamping force and the micro-displacement time-domain signal can be achieved by the following steps:

[0079] The dynamic matrix of contact stiffness of each support claw is determined based on the time-varying cross-correlation function of the radial clamping force and the micro-displacement time-domain signal.

[0080] Based on the dynamic matrix of contact stiffness, the energy distribution of each mode is extracted by variational mode decomposition and mapped to the contact state feature spectrum.

[0081] In practice, firstly, the synchronously acquired radial clamping force time-domain signal and micro-displacement time-domain signal are preprocessed for detrending and noise reduction, retaining the effective temporal fluctuation components. Then, the time-varying cross-correlation function of the two sets of signals is calculated at each time node to characterize the linkage and correlation characteristics between force and micro-displacement over time. Based on Hertzian cylindrical contact theory, the relationship between the contact stiffness and contact force when two cylindrical surfaces are in contact is as follows: ,in For contact stiffness, These are constants related to the material's elastic modulus, Poisson's ratio, and contact geometry. For radial contact force, the peak value of the time-varying cross-correlation function output. With contact force Proportional to each other, the proportionality coefficient k is determined through static calibration tests; therefore, the contact stiffness can be expressed by the formula... The calculated stiffness values, arranged in a time series, are regularized to form the dynamic contact stiffness matrix of the corresponding support claw. Then, the dynamic contact stiffness matrix is ​​used as the raw data to be decomposed and input into the variational modal decomposition process. A reasonable modal decomposition order and convergence constraints are pre-set. The algorithm iteratively solves for each order of intrinsic modal components, calculates the signal energy of each modal component, and statistically analyzes the proportion of each energy level in the total energy, forming a complete distribution of modal energy. The modal order, energy amplitude, energy proportion, and other parameters are integrated and normalized according to the standard dimension to complete the parameter space mapping, ultimately generating a standardized contact state characteristic spectrum.

[0082] It should be noted that in this application, the time-varying cross-correlation function characterizes the correlation between the time-domain signal of the radial clamping force and the time-domain signal of the micro-displacement corresponding to the same support claw as time progresses; the contact stiffness dynamic matrix refers to storing the contact stiffness values ​​of the support claw and the part at different times in matrix form; variational mode decomposition refers to the decomposition method of decomposing the complex fluctuation components contained in the contact stiffness dynamic matrix into several stable natural mode components; and the energy distribution of each mode refers to the energy proportion and amplitude distribution corresponding to each mode component after variational mode decomposition.

[0083] In step S3, when the radial contact index between any support claw and the outer wall of the part is detected to be lower than the turning clearance boundary, it is determined that the corresponding support claw forms a radial under-constraint on the part. When the radial contact index between any support claw and the outer wall of the part is detected to be higher than the turning clearance boundary, it is determined that the corresponding support claw forms a radial over-constraint on the part.

[0084] In this embodiment, when the radial contact index between any support claw and the outer wall of the part is detected to be lower than the turning clearance boundary, the determination that the corresponding support claw forms a radial under-constraint on the part can be achieved by the following steps:

[0085] Based on the real-time comparison between the radial contact index of each support claw and the turning clearance boundary, a radial constraint state residual sequence is constructed.

[0086] Based on the radial constraint state residual sequence, a residual cumulative energy threshold within a variable-scale time window is introduced to determine the radial contact degradation characteristic degree.

[0087] Extract the reduced-order modes that exceed the expected margin from the radial contact degradation eigenvalues ​​and generate the corresponding radial under-constraint decision events for the support claws.

[0088] In practice, firstly, the real-time radial contact index of each support claw is extracted cycle by cycle. This index is then compared point by point with the pre-generated lower limit of the turning clearance boundary. The difference between the contact index and the lower limit at each moment is calculated. When the difference is negative, the original value is retained; when it is positive, it is uniformly recorded as 0. Here, 0 represents no under-constraint deviation. The calculated difference sequence is processed by moving average filtering to remove high-frequency instantaneous interference, and then arranged in timestamp order to form a regular radial constraint state residual sequence. Then, according to the current turning speed and feed rate, the time window length is dynamically adjusted according to a preset mapping relationship. The higher the speed and the larger the feed rate, the shorter the window length. Within each time window, the sum of squares of all non-zero differences in the residual sequence is calculated. This sum of squares is used as the residual cumulative energy, and it is compared with the residual cumulative energy threshold calibrated in advance through multiple sets of under-constraint tests. The ratio of "cumulative energy / energy threshold" is used as the contact degradation parameter, and combined with the duration of the residual within the time window, a weighted calculation is performed to obtain the radial contact degradation characteristic degree. Finally, the one-dimensional time series of radial contact degradation characteristics is used as the input of the variational mode decomposition (VMD) algorithm. Based on the frequency characteristics of the slender shaft turning process, the mode decomposition order is set to 4, and the convergence accuracy is 1e-7. The VMD algorithm decomposes the original sequence into four intrinsic mode components with different center frequencies by solving a variational optimization problem. The energy proportion of each mode component is calculated, and the top two modes with a cumulative energy proportion of 85% are selected as reduced-order modes. The amplitude of each reduced-order mode is compared with the preset expected margin. When the amplitude of any reduced-order mode exceeds the expected margin and lasts for two time windows, it is determined that there is effective underconstrained degradation. The system automatically records the corresponding support claw number, degradation start time, and current degradation characteristic value, generates a standardized radial underconstrained judgment event, and sends it to the control controller.

[0089] It should be noted that, in this application, the radial constraint state residual sequence refers to the discrete data sequence composed of the difference between the real-time radial contact index of each support claw and the lower limit of the turning clearance boundary in chronological order; the lower limit of the turning clearance boundary is the minimum critical value of the turning clearance boundary; the variable scale time window refers to the time analysis window that dynamically adjusts the length according to the turning speed and feed rate; the residual cumulative energy threshold refers to the energy critical value used to distinguish between normal fluctuations and abnormal degradation, calibrated based on multiple sets of process tests; the radial contact degradation characteristic degree refers to the physical quantity of the degree and trend of degradation when the radial contact index is continuously lower than the lower limit of the boundary; the expected margin refers to the maximum allowable fluctuation range of the radial contact degradation characteristic degree; the reduced-order mode refers to the few key modal components that contribute the most to the underconstraint, extracted from the degradation characteristic degree sequence through modal decomposition; and the radial underconstraint refers to the standardized judgment signal containing the support claw number, degradation start time, and degradation degree.

[0090] In this embodiment, when the radial contact index between any support claw and the outer wall of the part is detected to be higher than the turning clearance boundary, the determination that the corresponding support claw forms a radial over-constraint on the part can be achieved by the following steps:

[0091] Based on the radial contact index of each support claw and the deviation sequence of the turning clearance boundary, a radial over-constraint feature comparison matrix is ​​constructed;

[0092] Perform amplitude-based hierarchical comparison of the radial over-constraint feature comparison matrix to determine the over-constraint contact margin;

[0093] The over-constraint contact margin is mapped to a preset radial clamping upper limit threshold to determine and generate the corresponding radial over-constraint event for the support claw.

[0094] In practice, firstly, the real-time radial contact index of each support claw is extracted cycle by cycle. The difference between this index and the upper limit of the turning clearance boundary is calculated. When the difference is positive, the original value is retained; when it is negative, it is recorded as 0. Simultaneously, the contact stress fluctuation amplitude and fluctuation frequency data at the corresponding time are extracted and aligned with the deviation sequence by timestamp. The deviation value, contact stress fluctuation amplitude, and fluctuation frequency at the same time are used as a row of a matrix. The feature data of all times are arranged in chronological order to construct a radial over-constraint feature comparison matrix. Then, through multiple sets of over-constraint process tests, several levels are divided according to the magnitude of the contact stress fluctuation amplitude. At the same time, the corresponding deviation value range and actual over-constraint effect under each level are recorded to form a graded comparison standard library. The contact stress fluctuation amplitude column in the feature comparison matrix is ​​graded and matched to determine the stress level at each time. Combined with the deviation value at the corresponding time, the over-constraint quantization value at each time is calculated by weighting the weight coefficients in the graded comparison standard library. The average value of the quantization values ​​at multiple consecutive times is taken to obtain the over-constraint contact margin. Finally, a preset radial clamping upper limit threshold is calibrated in advance through orthogonal experiments. During the experiments, the over-constraint contact margin at which the part produces the maximum elastic deformation within the allowable range is measured under different materials and cutting parameters. The average value of three repeated experiments is taken as the final threshold. The over-constraint contact margin calculated in real time is compared with this threshold through boundary mapping. When the margin value exceeds the threshold and continues for two consecutive matrix rows, it is determined to be a valid over-constraint. The system automatically records the corresponding support claw number, the over-constraint start time, and the current margin value, generates a standardized radial over-constraint event, and sends it to the adjustment controller.

[0095] It should be noted that, in this application, the deviation sequence refers to a discrete data sequence composed of the difference between the real-time radial contact index of each support claw and the upper limit of the turning clearance boundary in chronological order; the radial over-constraint feature comparison matrix refers to a two-dimensional data matrix with the deviation sequence as the core, integrating multi-dimensional features such as contact stress fluctuation amplitude and fluctuation frequency; contact stress fluctuation refers to the dynamic change of contact stress at the interface between the support claw and the part over time; amplitude classification comparison refers to an analysis method that divides the contact stress fluctuation amplitude into different levels and compares it with the deviation sequence; over-constraint contact margin refers to a quantitative index of the degree of over-constraint exceeding the normal allowable range; the radial clamping upper limit threshold refers to the over-constraint critical value calibrated based on the allowable stress of the part material and the machining accuracy requirements; boundary mapping refers to the process of correspondingly associating the over-constraint contact margin with the preset threshold to determine whether it exceeds the allowable range; radial over-constraint refers to a standardized judgment signal containing the support claw number, over-constraint start time, and over-constraint contact margin.

[0096] In addition, in this embodiment, the determination of the over-constraint contact margin by performing amplitude-level comparison based on contact stress fluctuation on the radial over-constraint feature comparison matrix can be achieved by the following steps:

[0097] Based on the contact stress fluctuation time history in the radial over-constraint feature comparison matrix, an amplitude-based comparison of contact stress fluctuation is performed to generate a graded identifier for the over-constraint contact stress amplitude.

[0098] Based on the graded identifier, a preset radial over-constraint contact margin mapping function is introduced to determine the over-constraint contact margin.

[0099] In practice, firstly, contact stress column data is extracted from the radial over-constraint feature comparison matrix and organized into a contact stress fluctuation time history according to timestamp order. An amplitude grading standard is established beforehand through multiple sets of process experiments. During the experiments, the contact stress amplitude and corresponding elastic deformation of the part are measured under different degrees of over-constraint. Based on the influence of deformation on machining accuracy, several levels are defined. The stress amplitude at each moment in the time history is compared with the grading standard to determine the corresponding level and generate a grading identifier for the over-constraint contact stress amplitude. Then, when fitting the mapping function, the measured radial elastic deformation of the part is first standardized to obtain the over-constraint contact margin.

[0100]

[0101] in To provide over-constrained contact margin, The radial elastic deformation of the part is measured. To design the maximum allowable elastic deformation, The ultimate elastic deformation measured in the experiment is used as the independent variable, and the standardized over-constraint contact margin is used as the dependent variable. Least square fitting is performed to obtain the preset radial over-constraint contact margin mapping function. The classification label at each time moment is substituted into the function to calculate the instantaneous over-constraint contact margin. The instantaneous margin values ​​at multiple consecutive time moments are processed by moving average to remove high-frequency fluctuation interference and obtain the final over-constraint contact margin.

[0102] It should be noted that, in this application, the contact stress fluctuation time history refers to the continuous record of contact stress amplitude arranged in chronological order in the radial over-constraint feature comparison matrix; the classification label of the over-constraint contact stress amplitude refers to the standardized classification label of contact stress fluctuations in different amplitude ranges; the preset radial over-constraint contact margin mapping function refers to the mathematical function that describes the quantitative relationship between the classification label and the over-constraint contact margin, which is established in advance by fitting experimental data.

[0103] In step S4, the radial clamping amount of the corresponding support claw against the outer wall of the part is adjusted according to the radial under-constraint and the radial over-constraint during reverse turning.

[0104] In this embodiment, adjusting the radial clamping amount of the corresponding support jaw against the outer wall of the part during reverse turning based on the radial under-constraint and the radial over-constraint can be achieved by the following steps:

[0105] Based on the constraint deviation feature vectors of the radial under-constraint and the radial over-constraint, a clamping amount compensation mapping function is constructed;

[0106] Based on the aforementioned clamping amount compensation mapping function, the reverse turning axial elastic clamping boundary condition is introduced to determine the radial clamping correction amount of the corresponding support claw;

[0107] The radial clamping correction amount is output to the chuck servo control unit to adjust the radial clamping amount of the corresponding support claw against the outer wall of the part.

[0108] In practice, firstly, four core parameters—constraint type, deviation amplitude, duration, and contact stiffness change rate—are extracted from the radial under-constraint and radial over-constraint judgment events. These parameters are then arranged in a fixed order to form a constraint deviation feature vector. The contact stiffness change rate is defined as the change in contact stiffness per unit time, which can be directly extracted from the contact stiffness dynamic matrix. The contact stiffness dynamic matrix stores the contact stiffness values ​​at each sampling moment in a time series. The contact stiffness change rate can be obtained by calculating the difference in contact stiffness between two adjacent moments and dividing it by the sampling time interval. Multiple sets of orthogonal process tests are conducted, dividing the cutting parameter range into three intervals—low speed, medium speed, and high speed—based on the spindle speed. Within each interval, the adjustment effect of different compensation values ​​under different combinations of deviation features is tested. The quantitative criterion for "restoring the constraint state to normal" is clearly defined as follows: the radial contact index returns to the boundary range of the turning clearance and lasts for 5 consecutive acquisition cycles. The compensation value that meets this criterion and has the shortest adjustment time is recorded as the optimal compensation value. Using the feature vector as the independent variable and the optimal compensation value as the dependent variable, multiple linear regression is performed in each interval to obtain the piecewise clamping amount compensation mapping function. Through 100 sets of closed-loop adjustment tests under different working conditions, the adjustment error of the piecewise regression function is less than 5%, and the adjustment convergence time is less than 3 acquisition cycles, which meets the real-time control requirements. Then, the current axial elastic clamping force, elastic tip compression, and part axial elongation are collected in real time to form the axial elastic clamping boundary conditions. The constraint deviation feature vector is substituted into the clamping amount compensation mapping function to calculate the initial correction amount. The initial correction amount is weighted according to the axial clamping boundary conditions. The larger the axial clamping force, the smaller the correction amount weight; the larger the axial elongation, the larger the correction amount weight. The correction result is limited, and the maximum single correction amount does not exceed the preset value to obtain the final radial clamping correction amount. Finally, the radial clamping correction amount is converted into a pulse command for the corresponding servo motor. The conversion coefficient is pre-calibrated using the servo motor parameters and the lead screw. After receiving the pulse command, the chuck servo control unit drives the servo motor of the corresponding support claw to rotate. The rotational motion is converted into radial linear motion through the ball screw mechanism, causing the support claw to move the corresponding distance. After the adjustment is completed, the encoder feedback value of the servo motor is read to confirm that the actual displacement of the support claw is consistent with the correction amount, thus completing one adjustment cycle.

[0109] It should be noted that, in this application, the constraint deviation feature vector refers to a multi-dimensional feature set composed of parameters such as constraint type, deviation amplitude, duration, and contact stiffness change rate; the clamping amount compensation mapping function refers to a mathematical model that establishes a quantitative relationship between constraint deviation features and optimal clamping amount compensation value; the reverse turning axial elastic clamping boundary condition refers to a set of axial constraint parameters composed of the current axial clamping force, elastic center stiffness, and axial elongation of the part; the radial clamping correction amount refers to the radial displacement that the support jaw needs to adjust, with positive values ​​indicating an increase in clamping amount and negative values ​​indicating a decrease in clamping amount; the chuck servo control unit refers to a servo drive system that independently controls the radial movement of each support jaw; the radial clamping amount refers to the degree of radial clamping applied by the support jaw to the outer wall of the part, which is determined by the radial extension length of the support jaw relative to the chuck body.

[0110] In addition, in this embodiment, based on the clamping amount compensation mapping function, the axial elastic clamping boundary condition of reverse turning is introduced, and the radial clamping correction amount of the corresponding support claw can be determined by the following steps:

[0111] Based on the clamping amount compensation mapping function and the reverse turning axial elastic clamping boundary condition, determine the radial clamping compensation expectation value of the corresponding support claw;

[0112] Based on the radial clamping compensation expectation value, a dynamic balance constraint of the chuck clamping force is introduced to determine the radial clamping correction amount of the corresponding support claw.

[0113] In practical implementation, firstly, the current constraint deviation feature vector is substituted into the pre-fitted clamping amount compensation mapping function to calculate the initial compensation value of a single support claw. The axial elastic clamping force, elastic tip compression, and axial elongation of the part are collected in real time. Using a pre-calibrated axial constraint correction coefficient table, the correction coefficient under the current working condition is interpolated. The initial compensation value is multiplied by the correction coefficient to obtain the radial clamping compensation expectation value considering the axial boundary effect. The correction coefficient table is calibrated through clamping stiffness tests under different axial clamping forces, with each set of parameters repeated three times and the average value taken. Then, the radial clamping compensation expectation values ​​of all support claws that need adjustment are extracted. With the optimization objective of "minimizing the sum of squares of the deviations between the actual correction amount and the compensation expectation value of each support claw" and the equality constraint condition of "the vector sum of the radial clamping forces of all support claws is zero," a constrained least squares optimization model is established: Optimization objective:

[0114] Constraints:

[0115] in For the first The actual correction amount for each support claw. For the first The expected compensation value for each support claw. For the first The radial clamping force of each support claw, For the first The radial unit vector corresponding to each support claw is used to solve the optimization model using the Lagrange multiplier method to obtain the optimal correction amount that simultaneously satisfies the compensation requirements of a single support claw and the overall force balance of the chuck. The preliminary correction results are then subjected to amplitude limiting processing, ensuring that the maximum correction amount in a single instance does not exceed a preset value, and finally, the radial clamping correction amount is obtained.

[0116] It should be noted that, in this application, the radial clamping compensation expectation value refers to the ideal radial displacement that a single support claw needs to adjust to eliminate the current constraint anomaly when the overall clamping force balance of the chuck is not considered; the chuck clamping force dynamic balance constraint refers to the mechanical constraint condition established based on the principle of static balance, which requires the sum of the radial clamping force vectors of all support claws on the part to be zero.

[0117] In this embodiment, reference Figure 3 As shown in the figure, this is a schematic diagram of the reverse feed turning control principle. In this diagram, a slender shaft-like part is used as the machining object. The left end of the part is held by an electric chuck support jaw with integrated force and displacement sensors, and the right end is axially clamped by an elastic center, forming the basic clamping structure for axial elastic clamping turning. During the machining process, the sensors on each support jaw of the chuck collect the radial clamping force and micro-displacement time-domain signals in real time, and the tailstock elastic center synchronously collects the axial turning response data. All signals are preprocessed by the data acquisition unit and then transmitted to the controller. The controller calculates the radial contact index of each support jaw according to a preset algorithm, and dynamically generates the turning clearance boundary in combination with the axial working conditions. It accurately determines the radial under-constraint or over-constraint state of each support jaw. Finally, the controller outputs adjustment commands to the chuck servo unit to independently adjust the radial clamping amount of the corresponding support jaw, so that the part is always in the optimal radial constraint state, effectively suppressing bending deformation and vibration, and significantly improving the machining accuracy and surface quality of the slender shaft.

[0118] Therefore, in this application, the radial clamping amount of the corresponding support claw to the outer wall of the part is adjusted according to the radial under-constraint and the radial over-constraint during reverse turning. Among them, by determining the radial contact index, a dimensionless parameter that can quantitatively characterize the degree of radial contact between a single support claw and the outer wall of the part can be obtained, thereby transforming the originally ambiguous contact interface constraint state into a quantifiable and comparable standardized numerical index. This parameter integrates multi-dimensional mechanical characteristics such as contact stiffness, clamping force fluctuation, and micro-displacement change, while incorporating the coupling effect of axial elastic clamping boundary conditions. It eliminates the subjective error of traditional qualitative judgment, making the constraint state under different cutting conditions laterally comparable, and providing a precise input basis for the adaptive adjustment system. Determining radial over / under constraint allows for the identification of standardized judgment events that accurately distinguish between effective constraint anomalies and instantaneous process fluctuations. This enables precise identification and severity quantification of constraint state anomalies. Through time-series residual accumulation analysis, modal energy decomposition, and multi-feature hierarchical comparison, this process effectively filters out misjudgments caused by interference factors such as sensor noise and instantaneous fluctuations in cutting force. At the same time, it quantifies the duration and impact of the anomaly, allowing the system to output corresponding tightening correction values ​​based on the severity of the anomaly, achieving hierarchical and precise adjustment and avoiding additional deformations caused by over- or under-adjustment.

[0119] In summary, the technical solution adopted in this application can achieve adaptive adjustment of the radial constraint state during reverse turning, thereby improving the machining accuracy of slender shaft parts during reverse turning.

[0120] Example 2: This application provides a reverse turning apparatus for machining slender shaft parts, referencing... Figure 4 As shown in the figure, this is a modular structure diagram of a reverse turning device for machining slender shaft parts according to this embodiment of the present application. The reverse turning device includes:

[0121] The acquisition module 100 is used to acquire axial turning response data and contact state characteristics of slender shaft parts during reverse feed turning.

[0122] The processing module 200 is used to generate the turning clearance boundary of the part during axial elastic clamping turning based on the turning response data and the contact state characteristics, and to determine the radial contact index between each support jaw in the lathe chuck and the outer wall of the part.

[0123] The processing module 200 is further configured to determine that the corresponding support claw forms a radial under-constraint on the part when the radial contact index between any support claw and the outer wall of the part is lower than the turning clearance boundary, and to determine that the corresponding support claw forms a radial over-constraint on the part when the radial contact index between any support claw and the outer wall of the part is higher than the turning clearance boundary.

[0124] The execution module 300 is used to adjust the radial clamping amount of the corresponding support claw against the outer wall of the part during reverse turning based on the radial under-constraint and the radial over-constraint.

[0125] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0126] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, including read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-Erasable Programmable Read-Only Memory (EEPROM), compactdisc read-only memory (CD-ROM) or other optical disc storage, disk storage, magnetic tape storage, or any other computer-readable medium capable of carrying or storing data.

[0127] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

Claims

1. A reverse turning method for machining slender shaft parts, characterized in that, The reverse turning method includes the following steps: Collect axial turning response data and contact state characteristics of slender shaft parts during reverse feed turning; Based on the turning response data and the contact state characteristics, the turning clearance boundary of the part during axial elastic clamping turning is generated, and the radial contact index between each support jaw in the lathe chuck and the outer wall of the part is determined. When the radial contact index between any support claw and the outer wall of the part is detected to be lower than the turning clearance boundary, it is determined that the corresponding support claw forms a radial under-constraint on the part. When the radial contact index between any support claw and the outer wall of the part is detected to be higher than the turning clearance boundary, it is determined that the corresponding support claw forms a radial over-constraint on the part. Adjust the radial clamping amount of the corresponding support claw against the outer wall of the part during reverse turning based on the radial under-constraint and the radial over-constraint. Specifically, the data collected on the axial turning response and contact state characteristics of slender shaft-type parts during reverse feed turning includes: The turning start characteristics are determined based on the tool holder strain marking of the slender shaft part during reverse turning; the axial turning state information of the slender shaft part during reverse feed turning is marked based on the turning start characteristics; and the axial turning response data and contact state characteristics of the slender shaft part during reverse feed turning are extracted from the turning state information. Specifically, generating the turning clearance boundary of the part during axial elastic clamping turning based on the turning response data and the contact state characteristics includes: Extract the time-domain synchronous feature set of cutting force fluctuation and workpiece elastic displacement from the turning response data; obtain the contact state identifier during axial elastic clamping turning based on the time-domain synchronous feature set, and determine the contact stiffness coefficient; map the contact state identifier and the contact stiffness coefficient to the turning dynamics model to generate the turning clearance boundary during axial elastic clamping turning. Specifically, determining the radial contact index between each support jaw in the lathe chuck and the outer wall of the part includes: The radial clamping force and micro-displacement time-domain signals between each support jaw in the lathe chuck and the outer wall of the part are collected; the contact stiffness of each support jaw is determined based on the radial clamping force and the micro-displacement time-domain signals, and then the contact state feature spectrum is extracted; the contact state feature spectrum and the elastic clamping boundary condition are correlated and mapped to obtain the radial contact index between each support jaw in the lathe chuck and the outer wall of the part.

2. The reverse turning method for machining slender shaft parts as described in claim 1, characterized in that, After determining the contact stiffness of each support claw based on the radial clamping force and the micro-displacement time-domain signal, the contact state feature spectrum is extracted, specifically including: The dynamic matrix of contact stiffness of each support claw is determined based on the time-varying cross-correlation function of the radial clamping force and the micro-displacement time-domain signal. Based on the dynamic matrix of contact stiffness, the energy distribution of each mode is extracted by variational mode decomposition and mapped to the contact state feature spectrum.

3. The reverse turning method for machining slender shaft parts as described in claim 1, characterized in that, When the radial contact index between any support claw and the outer wall of the part is detected to be lower than the turning clearance boundary, it is determined that the corresponding support claw forms a radial under-constraint on the part, specifically including: Based on the real-time comparison between the radial contact index of each support claw and the turning clearance boundary, a radial constraint state residual sequence is constructed. Based on the radial constraint state residual sequence, a residual cumulative energy threshold within a variable-scale time window is introduced to determine the radial contact degradation characteristic degree. Extract the reduced-order modes that exceed the expected margin from the radial contact degradation eigenvalues ​​and generate the corresponding radial under-constraint decision events for the support claws.

4. The reverse turning method for machining slender shaft parts as described in claim 1, characterized in that, When the radial contact index between any support claw and the outer wall of the part is detected to be higher than the turning clearance boundary, it is determined that the corresponding support claw forms a radial over-constraint on the part. Specifically, this includes: Based on the radial contact index of each support claw and the deviation sequence of the turning clearance boundary, a radial over-constraint feature comparison matrix is ​​constructed; Perform amplitude-based hierarchical comparison of the radial over-constraint feature comparison matrix to determine the over-constraint contact margin; The over-constraint contact margin is mapped to a preset radial clamping upper limit threshold to determine and generate the corresponding radial over-constraint event for the support claw.

5. The reverse turning method for machining slender shaft parts as described in claim 4, characterized in that, Performing amplitude-based hierarchical comparisons of the radial over-constraint feature comparison matrix to determine the over-constraint contact margin specifically includes: Based on the contact stress fluctuation time history in the radial over-constraint feature comparison matrix, an amplitude-based comparison of contact stress fluctuation is performed to generate a graded identifier for the over-constraint contact stress amplitude. Based on the graded identifier, a preset radial over-constraint contact margin mapping function is introduced to determine the over-constraint contact margin.

6. The reverse turning method for machining slender shaft parts as described in claim 1, characterized in that, Adjusting the radial clamping amount of the corresponding support jaw against the outer wall of the part during reverse turning, based on the aforementioned radial under-constraint and radial over-constraint, specifically includes: Based on the constraint deviation feature vectors of the radial under-constraint and the radial over-constraint, a clamping amount compensation mapping function is constructed; Based on the aforementioned clamping amount compensation mapping function, the reverse turning axial elastic clamping boundary condition is introduced to determine the radial clamping correction amount of the corresponding support claw; The radial clamping correction amount is output to the chuck servo control unit to adjust the radial clamping amount of the corresponding support claw against the outer wall of the part.

7. A reverse turning apparatus for machining slender shaft parts, used to perform a reverse turning method for machining slender shaft parts as described in any one of claims 1 to 6, characterized in that, The reverse turning device includes: The acquisition module is used to acquire axial turning response data and contact state characteristics of slender shaft parts during reverse feed turning; The processing module is used to generate the turning clearance boundary of the part during axial elastic clamping turning based on the turning response data and the contact state characteristics, and to determine the radial contact index between each support jaw in the lathe chuck and the outer wall of the part. The processing module is further configured to determine that the corresponding support claw forms a radial under-constraint on the part when the radial contact index between any support claw and the outer wall of the part is lower than the turning clearance boundary, and to determine that the corresponding support claw forms a radial over-constraint on the part when the radial contact index between any support claw and the outer wall of the part is higher than the turning clearance boundary. The execution module is used to adjust the radial clamping amount of the corresponding support claw against the outer wall of the part during reverse turning based on the radial under-constraint and the radial over-constraint.

Images