High-precision transmission shaft precision grinding processing method, system and equipment

Abstract

The application discloses a high-precision transmission shaft precision grinding processing method, system and equipment, and relates to the technical field of grinding processing. The method comprises the following steps: deploying a measurement component to acquire grinding state characteristics in real time with automatic grinding; inserting a programmable impedance network into a servo driver to construct a double closed-loop control architecture; inputting the grinding state characteristics into a state loop for displacement and mechanical error analysis, and optionally activating an impedance loop to compensate for decision determination by resistance-capacitance parameter adjustment; and the programmable impedance network implements time window regulation and compensation for the grinding process according to the parameters. The technical problems of poor size precision stability control and poor processing consistency caused by mechanical load fluctuation, thermal deformation and system stiffness change in the existing high-precision transmission shaft grinding process are solved, the technical effects of real-time sensing of the grinding state and adaptive impedance regulation through the double closed-loop architecture, improvement of the size precision and shape and position tolerance stability of the transmission shaft, reduction of the waste rate and improvement of the processing efficiency are achieved.

Description

Technical Field

[0001] This invention relates to the field of grinding technology, specifically to a precision grinding method, system, and equipment for high-precision drive shafts. Background Technology

[0002] As a key component in mechanical transmission systems, the machining accuracy of drive shafts directly affects the operational stability, noise level, and service life of the entire equipment. With the development of high-end equipment manufacturing, micron-level or even higher precision requirements have been placed on the dimensional tolerances, geometric tolerances, and surface quality of drive shafts. Precision grinding, as the final process in drive shaft machining, directly determines the final quality of the workpiece.

[0003] In existing grinding technologies, CNC programs are typically used to pre-set machining paths and parameters, combined with online measurement technology for feedback control. However, traditional control methods have the following limitations: First, the grinding process is a time-varying, nonlinear, and complex process. Factors such as grinding wheel wear, changes in workpiece stiffness, cutting thermal deformation, and vibration interference can cause significant deviations between the actual grinding state and the pre-set program. Traditional proportional-integral-derivative control or single dimensional feedback control is insufficient for real-time, high-precision comprehensive compensation of such dynamically changing displacement and mechanical errors. Second, existing technologies do not pay enough attention to the dynamic matching of electrical parameters and mechanical loads. Servo drive systems typically use fixed electrical parameters and cannot adaptively adjust drive characteristics according to real-time changes in grinding loads. This mismatch between electrical characteristics and mechanical loads often leads to system response lag, oscillation, or even overshoot, resulting in chatter marks or dimensional deviations on the workpiece surface. Therefore, overcoming the problems of insufficient dynamic error compensation and poor matching between electrical parameters and mechanical loads in existing grinding technologies to achieve high-precision, high-stability automated grinding of drive shafts is a technical challenge that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0004] This application provides a precision grinding method, system, and equipment for high-precision drive shafts, which solves the technical problems of unstable dimensional accuracy and poor machining consistency caused by mechanical load fluctuations, thermal deformation, and system stiffness changes during the existing high-precision drive shaft grinding process.

[0005] The first aspect of this application provides a precision grinding method for high-precision drive shafts, the method comprising:

[0006] A measurement component is deployed on a grinding machine to acquire grinding state characteristics in real time during the automated grinding process of the workpiece along the drive shaft. A dual-closed-loop control architecture is constructed by inserting a programmable impedance network into the servo driver. This architecture consists of a cascaded state loop and an impedance loop trained under supervision, connected to the measurement component and the programmable impedance network. Based on this dual-closed-loop control architecture, the grinding state characteristics are input into the state loop, performing displacement error analysis based on diameter and mechanical error analysis based on mechanical load. The impedance loop can be optionally activated to perform parameter adjustment and compensation decisions based on the optimal resistance and capacitance combination, determining the impedance control parameters. The programmable impedance network responds to these impedance control parameters, performing process time window adjustment and compensation superposition on the automated grinding process of the grinding machine.

[0007] A second aspect of this application provides a precision grinding system for high-precision drive shafts, the system comprising:

[0008] The system comprises the following modules: a state measurement module, a state control module, and a grinding control module. The state measurement module is deployed on the grinding machine to acquire grinding state characteristics in real time during the automated grinding process of the workpiece along the drive shaft. The closed-loop architecture construction module is a dual-loop control architecture constructed by inserting a programmable impedance network into the servo driver. This architecture consists of a cascaded state loop and an impedance loop trained under supervision, connected to the measurement components and the programmable impedance network. The closed-loop architecture analysis module, based on the dual-loop control architecture, inputs the grinding state characteristics into the state loop and performs displacement error analysis based on diameter and mechanical error analysis based on mechanical load. Optionally, the impedance loop can be activated to perform parameter adjustment and compensation decisions based on the optimal resistance and capacitance combination to determine the impedance control parameters. The grinding control module, in response to the impedance control parameters, performs process time window adjustment and compensation superposition on the automated grinding process of the grinding machine.

[0009] A third aspect of this application provides an electronic device, comprising: a memory for storing executable instructions; and a processor for executing the executable instructions stored in the memory to implement the precision grinding method for a high-precision drive shaft provided in this application.

[0010] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0011] A measurement component is deployed on a grinding machine to acquire grinding status characteristics in real time during the automated grinding process of the workpiece on the drive shaft. Subsequently, a dual-closed-loop control architecture is constructed by inserting a programmable impedance network into the servo driver. This architecture consists of a cascaded state loop and an impedance loop trained under supervision, connected to the measurement component and the programmable impedance network. Following this, based on the dual-closed-loop control architecture, the grinding status characteristics are input into the state loop, performing displacement error analysis based on diameter value and mechanical error analysis based on mechanical load. The impedance loop can be optionally activated to perform parameter adjustment compensation decisions based on the optimal resistance and capacitance combination, determining the impedance control parameters. Finally, the programmable impedance network responds to the impedance control parameters, adjusting and compensating for the automated grinding process of the grinding machine within the process time window. This solves the technical problems of unstable dimensional accuracy and poor machining consistency caused by mechanical load fluctuations, thermal deformation, and system stiffness changes in existing high-precision drive shaft grinding processes. It achieves the technical effect of improving drive shaft dimensional accuracy and geometric tolerance stability, reducing scrap rate, and increasing machining efficiency through real-time perception of the grinding status and adaptive impedance control via a dual-closed-loop architecture. Attached Figure Description

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

[0013] Figure 1 A schematic diagram of the precision grinding process for a high-precision drive shaft provided in this application embodiment.

[0014] Figure 2 A schematic diagram of the precision grinding system for a high-precision drive shaft provided in this application embodiment.

[0015] Figure 3 This is a schematic diagram of the structure of an exemplary electronic device of this application.

[0016] Explanation of reference numerals in the attached figures: 11 State measurement module, 12 Closed-loop architecture construction module, 13 Closed-loop architecture analysis module, 14 Grinding control module, 21 Processor, 22 Memory, 23 Input device, 24 Output device. Detailed Implementation

[0017] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0018] Example 1, as Figure 1As shown in the embodiment of this application, a precision grinding method for a high-precision drive shaft is provided, wherein the method includes:

[0019] Deploy measurement components on grinding machines to acquire grinding status characteristics in real time during the automated grinding process of the workpiece along the drive shaft.

[0020] In one embodiment, online detection units are arranged at key force and displacement positions of the grinding machine tool, enabling synchronous operation throughout the entire automatic grinding process of the drive shaft, thus achieving continuous monitoring of the machining status. Specifically, non-contact displacement sensors can be deployed near the grinding wheel head to detect minute changes in the outer radius or diameter of the workpiece in real time. Force sensors can be installed on the headstock, tailstock, or support to collect load data such as normal grinding force, tangential grinding force, and axial force. After high-speed acquisition and filtering, various sensor signals form multi-dimensional grinding status parameters, including dimensional changes, dimensional change rates, force fluctuation amplitudes, and their trends. These status characteristics are continuously updated during the grinding process, providing a real-time data foundation for subsequent error analysis and closed-loop control adjustment.

[0021] Furthermore, real-time acquisition of grinding status characteristics includes:

[0022] A measuring component is deployed on a grinding machine to acquire the grinding state characteristics of the transmission shaft machining process. The deployment of the measuring component includes: symmetrically installing at least two high-frequency eddy current displacement sensors on both sides of the grinding wheel head, wherein the sensor probes are aligned with the workpiece surface of the transmission shaft, and the high-frequency eddy current displacement sensors measure the workpiece radius change; and installing triaxial piezoelectric force sensors on the headstock and tailstock of the grinding machine, wherein the triaxial piezoelectric force sensors measure the normal grinding force, tangential grinding force, and axial force.

[0023] Preferably, based on the structure and layout of the grinding machine tool, mounting positions with stable rigidity and minimal vibration interference are selected on both sides of the grinding wheel head, and at least two high-frequency eddy current displacement sensors are symmetrically fixed there. During installation, the sensors are precisely calibrated so that the sensor probe axis is perpendicular to the outer surface of the drive shaft, maintaining a preset initial gap distance. By calibrating the sensor zero point, a correspondence between the voltage signal and the workpiece radius is established. During grinding, the high-frequency eddy current displacement sensors collect real-time radial displacement data of the workpiece surface at a high sampling frequency. After signal amplification, filtering, and analog-to-digital conversion, the change in workpiece radius and its rate of change are obtained, thus reflecting the dynamic changes in the machining dimensions. Furthermore, triaxial piezoelectric force sensors are installed on the force transmission paths of the grinding machine headstock and tailstock. During installation, the preload is controlled to keep it within the linear working range, and known force values ​​are applied in the normal, tangential, and axial directions using a standard loading device to complete independent calibration of the three axes, establishing a force-electric conversion relationship model in the normal, tangential, and axial directions. During actual grinding, the force generated by the grinding contact is transmitted to the force sensor through the spindle and support structure. The triaxial piezoelectric force sensor synchronously outputs the corresponding charge signal. This signal is converted into a voltage signal by a charge amplifier, and then filtered and decoupled to obtain the real-time values ​​and fluctuation trends of the normal grinding force, tangential grinding force, and axial force. Finally, the displacement data and triaxial force data are synchronized and fused in time to form a multi-dimensional grinding state feature dataset containing dimensional change characteristics and mechanical load characteristics, providing real-time input parameters for subsequent error analysis and closed-loop control adjustment.

[0024] A dual-loop control architecture is constructed by inserting a programmable impedance network into the servo driver. The dual-loop control architecture is formed by supervising the training of the cascaded state loop and impedance loop, and establishing connections with the measurement component and the programmable impedance network.

[0025] In one embodiment, the control link of the servo driver is first modified by inserting a programmable impedance network into its current loop forward channel or power drive pre-stage. This programmable impedance network communicates with the control unit via a digital interface, allowing its resistance and capacitance values ​​to be adjusted in real time during operation, thereby changing the equivalent electrical impedance characteristics of the drive system. After hardware embedding, the drive system is recalibrated to establish a mapping relationship between impedance parameters and system response characteristics. Subsequently, a dual-loop control architecture is constructed at the control architecture level. During the construction process, based on historical grinding data and calibration experimental data, supervised training is performed on the state loop and impedance loop to determine the mapping model between state error characteristics and impedance parameter adjustment, enabling the two to form a cascadeable control relationship. The outer layer of the dual-loop control architecture is the state loop, which receives grinding state characteristics collected by the measurement components to generate displacement error and mechanical error signals. The inner layer is the impedance loop, which calculates the optimal resistance and capacitance combination for the current drive system. Subsequently, a data connection is established between the state loop and the measurement component, and a parameter control connection is established between the impedance loop and the programmable impedance network. The two loops are then cascaded to form a closed control link of measurement-analysis-impedance adjustment-drive response, which serves as the final complete dual closed-loop control architecture to achieve real-time optimization and accuracy improvement of the grinding process.

[0026] Furthermore, constructing a dual closed-loop control architecture includes:

[0027] A programmable impedance network is inserted into the forward current loop of the servo driver, wherein the programmable impedance network consists of a digital potentiometer and a programmable capacitor array; based on the measurement component and the programmable impedance network, a dual closed-loop control architecture is constructed, wherein the dual closed-loop control architecture includes a state loop and an impedance loop.

[0028] Preferably, the control structure of the servo driver is first analyzed to determine the forward control channel of the current loop or the pre-amplifier node. Without altering the original position and speed loop structures, a programmable impedance network is connected in series between the current command output and the power drive module, placing it on the modulation path of the drive control signal. This programmable impedance network consists of digital potentiometers and a programmable capacitor array. The digital potentiometers are used to achieve discrete or continuous adjustment of the equivalent resistance value, and the programmable capacitor array is used to achieve combined switching of the equivalent capacitance value. Both are connected to the upper-level control unit via a control bus, such as an SPI or I²C interface, allowing the resistance and capacitance parameters to be written and updated in real time by the control algorithm. After hardware embedding, the servo driver is tuned and tested. Specifically, under different impedance combinations, dynamic performance indicators such as the step response, overshoot, response time, and steady-state error of the servo driver are collected to establish the correspondence between impedance parameters and system response characteristics, providing a basic model for subsequent impedance loop calculations. Subsequently, a dual-closed-loop control architecture is constructed based on the functional division of the measurement components and the programmable impedance network. Specifically, the workpiece diameter change data and three-dimensional grinding force data collected by the measurement component are input to the control unit as input to the outer state loop. The state loop compares real-time measurements with the target diameter and target load range as a benchmark, calculates displacement error, error change rate, and mechanical error, forming a comprehensive state error signal and outputting the adjustment requirement. Based on this, an inner impedance loop is constructed. The impedance loop takes the error characteristics output from the state loop as input and, combined with a pre-established impedance parameter-response characteristic model, calculates the optimal resistance and capacitance combination under current conditions, generating impedance adjustment parameters. These parameters are sent to the programmable impedance network via a digital interface to achieve real-time adjustment of the servo drive's equivalent impedance. Ultimately, the state loop is responsible for error identification and control target generation, while the impedance loop is responsible for drive characteristic optimization and dynamic matching. The two are cascaded through the control unit and establish data and control connections with the measurement component and the programmable impedance network, respectively, forming a complete dual-closed-loop control architecture to achieve a closed-loop control process of state perception, error analysis, impedance adjustment, and drive response.

[0029] Furthermore, based on the measurement component and the programmable impedance network, a dual closed-loop control architecture is constructed, including:

[0030] The grinding process of the drive shaft workpiece is obtained. Taking the grinding state as input, the displacement error and rate of change based on the diameter value are used as the first decision objective, and the mechanical error and rate of change based on the mechanical load are used as the second decision objective to construct the state loop. The optimal resistance and capacitance combination based on impedance matching calculation is used as the optimization and control objective to construct the impedance loop. The state loop and impedance loop are cascaded, and the connection between the state loop and the measurement component, and the impedance loop and the programmable impedance network are established to form the dual closed-loop control architecture.

[0031] Optionally, the grinding process parameters of the drive shaft workpiece are first obtained, including the feed rate, depth of cut, target diameter, tolerance range, and allowable load range for each stage of rough grinding, semi-finish grinding, and finish grinding. Simultaneously, historical grinding data and calibration experimental data are collected. Historical grinding data includes the diameter variation, rate of variation, triaxial grinding force and its fluctuation characteristics recorded under different materials, grinding stages, and load conditions, as well as the corresponding machining quality results, such as dimensional errors and roundness errors. Calibration experimental data is obtained by performing step response and disturbance response tests under different impedance parameter combinations, including indicators such as system response time, overshoot, steady-state error, and vibration amplitude. Based on this, a mapping model of grinding state characteristics and machining error results is constructed for the state loop. This mapping model takes the displacement error of the real-time diameter and its rate of change as the first decision objective, and the mechanical errors of the normal force, tangential force, axial force, etc. of the mechanical load and their rates of change as the second decision objective. Through supervised training methods, such as a regression model or a weighted neural network model, a dual-objective error evaluation model is established to output a comprehensive error evaluation value, the corresponding adjustment requirement, and the weight allocation coefficient. This model is used to perform weighted fusion of displacement error and mechanical error during operation to form the control decision result of the state loop.

[0032] During the training phase, historical grinding data and calibration experimental data are used as the sample base. Each sample data includes diameter error, diameter error rate of change, normal grinding force error, tangential grinding force error, axial force error and their rate of change, as well as the corresponding control results, including the comprehensive error evaluation value, adjustment requirement, and weight allocation coefficients. The control results are used as supervision labels to evaluate the degree of influence of the current grinding state on the machining results. In the feature processing phase, all input variables are normalized, and a dual-objective feature vector is constructed. For model construction, a multivariate weighted regression model or a lightweight feedforward neural network model can be used. For example, in a weighted neural network model, an input layer, hidden layer, and output layer are set. The sample data is iteratively trained using the Adam or gradient descent algorithm. Training stops when the mean square error of the validation set converges to a preset threshold, such as less than 0.01. After training, a stable model parameter matrix and bias terms are obtained, forming a fixed dual-objective error evaluation model.

[0033] For the impedance loop, an impedance parameter-response characteristic model is established based on calibration experimental data. Specifically, the resistance R of the digital potentiometer and the equivalent capacitance C of the capacitor array are used as input variables, and the system response time, overshoot, vibration amplitude, and steady-state error are used as output variables. A response characteristic model is constructed using multivariate regression or table lookup interpolation. Then, combined with mechanical load characteristic parameters, such as the current normal force and its rate of change, a mapping relationship between load characteristics and the optimal RC combination is established, thus forming an optimal response characteristic model that can be used for online calculation. After model training, a state loop is constructed. The state loop takes grinding state characteristics as input, compares the real-time diameter value with the target diameter to obtain the displacement error and rate of change, and simultaneously calculates the magnitude and rate of change of the mechanical error. The trained dual-objective error evaluation model outputs a comprehensive error evaluation value and the corresponding adjustment requirement as the outer loop control output. Next, an impedance loop is constructed. The impedance loop takes the comprehensive error output from the state loop and the current load characteristics as input, calls the established impedance parameter-response characteristic model for calculation, and solves for the optimal resistance and capacitance combination under the current conditions, generating impedance control parameters, while satisfying the stage constraints. Finally, a cascaded design is implemented for the state loop and impedance loop. The state loop, as the outer loop, outputs the error decision result, while the impedance loop, as the inner loop, optimizes the dynamic characteristics of the drive side based on this result. Simultaneously, a data interface connection is established between the state loop and the measurement component, enabling it to continuously receive real-time diameter and mechanical data. Furthermore, a control interface connection is established between the impedance loop and the programmable impedance network, allowing for the real-time input of resistance and capacitance parameters. Through this complete dual-closed-loop control architecture, encompassing measurement input, state decision, impedance optimization, drive response, and re-measurement feedback, adaptive optimization control of the grinding process is achieved.

[0034] Based on the dual closed-loop control architecture, the grinding state characteristics are input into the state loop, and displacement error based on diameter value and mechanical error based on mechanical load are analyzed. Optionally, the impedance loop is activated to make parameter adjustment compensation decisions based on the optimal resistance and capacitance combination to determine the impedance control parameters.

[0035] In one embodiment, after the dual-loop control architecture is established, the grinding state characteristic data collected in real time by the measurement components is input to the state loop. The state loop first compares the real-time diameter value with the target diameter, calculates the displacement error and its rate of change, and simultaneously analyzes the normal force, tangential force, and axial force to calculate the mechanical error of the current mechanical load relative to the set load range and its changing trend. Through the dual-objective error evaluation model, the displacement error and mechanical error are comprehensively evaluated to obtain the comprehensive error evaluation value of the current grinding state, the corresponding adjustment requirement, and the weight allocation coefficient. Under normal processing conditions, when the comprehensive error evaluation value is within the preset allowable range, only the state loop outputs the adjustment requirement to fine-tune the feed rate or drive command to maintain stable processing. At this time, the impedance loop is not activated to ensure the simplicity of the control structure and reduce system disturbance. When the overall error evaluation value exceeds a preset threshold, i.e., when the error exceeds the limit or the error change rate increases abnormally, the impedance loop activation mechanism is triggered. The impedance loop takes the current overall error value, error change rate, and grinding stage label as input, calls the established impedance parameter-response characteristic model, and calculates the optimal resistance and capacitance combination that optimizes the dynamic response under the current mechanical load conditions. Impedance adjustment parameters are generated while satisfying parameter boundary constraints and written to the programmable impedance network via a digital interface. The impedance network adjusts the equivalent resistance and capacitance values ​​in real time according to the received adjustment parameters, thereby changing the electrical time constant and dynamic response characteristics of the servo drive. In this way, the impedance adjustment is superimposed on the original automated grinding control process, achieving rapid suppression and compensation of abnormal errors. When the error returns to the normal range, the impedance enhancement adjustment mode can be exited, reverting to the conventional control state dominated by the state loop, thus improving grinding dimensional accuracy and machining consistency while ensuring system stability.

[0036] Furthermore, before activating the impedance loop, the following steps are included:

[0037] As the transmission shaft workpiece is processed, the first constraint condition is determined according to the grinding stage. In this process, a preset impedance parameter boundary is defined by calibrating a rough grinding label or a fine grinding label for each grinding stage. The impedance loop is then constrained in terms of parameter space using the preset impedance parameter boundary. The parameter space is updated with the real-time grinding stage label.

[0038] Optionally, before the drive shaft workpiece enters the grinding process, the entire machining process is first divided into different stages according to the predetermined grinding process, such as rough grinding, semi-finish grinding, and finish grinding, and corresponding stage labels are set for each stage. During initialization or process import, parameters such as feed rate, depth of cut, target size allowance, and allowable vibration range for each stage are written into the control unit as the basis for stage identification. Subsequently, based on historical grinding data and calibration experiment results, reasonable ranges of corresponding impedance parameters are determined for different grinding stages. Specifically, in the rough grinding stage, to improve material removal efficiency and response speed, a lower equivalent resistance range and a smaller capacitance combination range can be preset; in the finish grinding stage, to enhance system rigidity and suppress vibration, a higher resistance range is preset, and the capacitance combination range is appropriately expanded. The above impedance parameter ranges constitute the preset impedance parameter boundaries for each stage, including the upper limit of resistance, the lower limit of resistance, and the allowable range of capacitance combinations. During the actual machining process, the control unit determines the grinding stage based on the current machining progress or real-time grinding status and automatically loads the corresponding stage label. When calculating the optimal resistance and capacitance combination, the impedance loop first reads the impedance parameter boundary corresponding to the current stage, limiting the calculation process within this parameter range. This forms a constrained parameter search space, avoiding impedance adjustment results that do not conform to the characteristics of the current process stage. When the machining process transitions from rough grinding to fine grinding, the stage label is updated according to the stage transition signal, and the upper and lower bounds of the impedance parameters are automatically adjusted to dynamically shrink or expand the parameter space. In this way, the impedance loop parameter space is updated in real time with the grinding stage, enabling impedance adjustment to have both adaptive optimization capabilities and process constraint control, thus balancing machining efficiency and machining accuracy requirements at different stages.

[0039] Furthermore, determining the impedance control parameters includes:

[0040] The grinding state characteristics are input into the state loop, and the calculations based on the first decision target and the second decision target are performed in parallel to determine the two-way error data with weighted values. For the two-way error data, with any one of the error data as the greedy target, the impedance control parameter is located in the determined upper and lower bounds of the parameters.

[0041] Optionally, the grinding state characteristics acquired in real time by the measurement component are first input into the state loop. These grinding state characteristics include the real-time diameter value, diameter change rate, and triaxial grinding force and its corresponding change rate. The state loop contains two parallel calculation channels. The first channel uses the target diameter as a reference to calculate the current diameter deviation and its change rate, forming displacement error data based on the diameter value. The second channel uses a preset mechanical load reference range as a reference to calculate the deviation and change rate of the current normal force, tangential force, or axial force, forming mechanical error data based on the mechanical load. Subsequently, the displacement error data and mechanical error data are normalized and analyzed according to a dual-objective error evaluation model to generate weighted allocation coefficients. These coefficients are then multiplied by the corresponding deviation data and change rate to obtain weighted two-way error data. In the impedance loop calculation stage, based on the current stage constraints, the upper and lower bounds of the resistor and capacitor parameters are determined. A greedy strategy is adopted, selecting the error term with the highest current weight from the two-way error data as the optimization target. A local search is performed within the parameter space defined by the defined upper and lower bounds. By calling the impedance parameter-response characteristic model, the degree of improvement to the current target error under different combinations of resistor R and capacitor C is evaluated, and the impedance parameter combination that can maximally reduce the greedy target error is gradually approximated. Finally, the impedance control parameters that satisfy the current constraints and have the best improvement effect on the selected error target are determined and output to the programmable impedance network for real-time adjustment. Through the above parallel calculation and greedy positioning process, rapid suppression of the dominant error is achieved, while simultaneously ensuring stable control of another error term.

[0042] Furthermore, taking any error data point as the greedy target, the impedance control parameter is located within the defined upper and lower bounds of the parameter, including:

[0043] For the first error data in the two-way error data, a parameter adjustment decision based on the optimal combination of resistor and capacitor is made with weight values ​​as constraints to determine the first parameter group; for the second error data, a second parameter group is determined; by mapping the first parameter group and the second parameter group, the upper and lower bounds of the parameters are determined; the upper and lower bounds of the parameters are used as the parameter adjustment range, and the impedance control parameters are determined by performing an equilibrium game.

[0044] Optionally, firstly, for the first error data in the two-way error data, the weight value of this error is introduced as a constraint into the parameter tuning objective function within the impedance loop. Specifically, the weight of the first error is... Used to limit the focus of parameter tuning, when When the diameter is large, parameter tuning search tends to improve system response speed to quickly reduce diameter deviation; when When the impedance is smaller, stability is given more consideration. The impedance loop calls upon the established impedance parameter-response characteristic model, traversing or locally searching for combinations of resistor R and capacitor C within the preset boundaries of the current stage. It calculates the suppression effect of different combinations on the first error, selects the RC combination that optimizes the first error evaluation value, and forms the first parameter set. Subsequently, for the second error data, the same model calculation process was used, but the optimization objective was switched to suppressing grinding force fluctuations and load anomalies. At this time, parameter tuning tended to increase damping and stiffness and reduce vibration sensitivity. By searching within the stage constraint boundaries, the optimal resistance-capacitance combination for the second error evaluation value was determined as the second parameter set. Next, the parameter tuning range is determined by mapping the first parameter group and the second parameter group. This mapping aligns the two parameter groups in the same parameter coordinate system, and the value ranges of resistance and capacitance are compared respectively. and The smaller value is used as the lower bound of the resistance, and the larger value is used as the upper bound of the resistance; and The smaller value is used as the lower bound of the capacitance, and the larger value as the upper bound. This yields the upper and lower bounds of the parameters, forming a feasible parameter tuning range that simultaneously covers both displacement and mechanical error requirements, preventing the optimization of only one objective from worsening the other. Then, the upper and lower bounds are used as a unified tuning range, within which an equilibrium game is performed to determine the final impedance control parameters. Specifically, the first and second error objectives are considered as two game participants, each with their own payoff function being the reduction of their own error, and the impedance parameters (R, C) as common strategy variables. The impedance loop iteratively searches within the tuning range, ensuring that under the current parameters, neither objective can benefit alone without significantly harming the payoff of the other objective; that is, reaching an equilibrium point between error suppression and system stability. After reaching equilibrium, the corresponding output is determined. As an impedance control parameter, it is sent to a programmable impedance network for dynamic adjustment, thereby achieving an optimal trade-off between response speed and processing stability.

[0045] The programmable impedance network responds to the impedance control parameters to adjust and compensate for the process time window of the automated grinding process of the grinding machine tool.

[0046] In one embodiment, after calculating the impedance control parameters, the corresponding resistance value and capacitor combination command is sent to the programmable impedance network via a digital communication interface. The digital potentiometers and capacitor arrays within the programmable impedance network switch resistance values ​​and reconstruct capacitor combinations according to the commands, thereby changing the equivalent electrical impedance characteristics of the servo drive system in real time, and adjusting the driver's electrical time constant, damping coefficient, and dynamic response characteristics accordingly. During automated grinding, the grinding machine operates according to a preset process program, such as roughing, semi-finishing, or finishing grinding according to a set feed rate, depth of cut curve, and dwell time. This impedance control does not change the original process program's logical flow, but rather dynamically adjusts the drive response characteristics within a specific process time window. For example, during the feed acceleration stage, the response speed is increased by reducing the equivalent resistance value; during the size approaching target stage, the system stiffness and damping are enhanced by increasing the equivalent resistance or optimizing the capacitor combination to suppress vibration and overshoot; during spark grinding or holding pressure stages, system stability is maintained by optimizing impedance matching. The process time window can be determined by machining stage labels, error trigger signals, or time period judgment logic. The impedance adjustment parameters are only effective within the corresponding time window. After the window expires, the system can revert to the default impedance configuration or proceed to the next stage of parameter settings. In this way, the impedance adjustment is dynamically compensated and superimposed on the automated grinding control process, enabling targeted enhanced control of key machining stages and improving dimensional stability and machining accuracy without changing the overall process cycle time.

[0047] Furthermore, the programmable impedance network responds to the impedance control parameters, including:

[0048] The grinding machine tool performs automated grinding according to the pre-configured process grinding program. The measurement component synchronously performs real-time grinding status measurement and obtains dynamic impedance control parameters based on the status loop and impedance loop. By sending the impedance control parameters to the programmable impedance network, the control compensation superposition of the automated grinding process is performed.

[0049] Preferably, before machining begins, a pre-configured grinding program is imported into the system. This program includes feed rate curves, depth of cut parameters, spindle speed, dwell time, and stage switching conditions for each grinding stage. The grinding machine automatically executes grinding drive control according to this program, achieving continuous machining of the drive shaft. During automated grinding, measurement components deployed on the machine tool work synchronously. A high-frequency eddy current displacement sensor collects workpiece diameter change data in real time, and a triaxial piezoelectric force sensor collects normal force, tangential force, and axial force data in real time. These signals are filtered, amplified, and converted from analog to digital before being input to the control unit, forming a continuously updated grinding status characteristic data stream. Subsequently, the system inputs the grinding status characteristic data into the status loop for error analysis, calculating displacement error and mechanical error, and combining the error change rate to obtain a comprehensive error evaluation value and corresponding adjustment requirements. When the error is within the allowable range, the adjustment demand output by the state loop is used as the impedance control parameter. When the error exceeds the boundary, the impedance loop, based on the comprehensive error characteristics and the current grinding stage label, calls the impedance parameter-response characteristic model to calculate the corresponding optimal resistance and capacitance combination, generating dynamic impedance control parameters. Then, the system sends the impedance control parameters to the programmable impedance network via a digital communication interface. The digital potentiometers and capacitor arrays within the programmable impedance network switch resistance and capacitance values ​​according to the received parameters, adjusting the equivalent electrical impedance of the servo driver. This adjustment result is superimposed on the original automated grinding control commands, optimizing the dynamic response characteristics of the driver without changing the overall process flow. Through this process, a closed-loop compensation process is achieved, encompassing automated grinding drive, real-time status measurement, dual-loop error analysis, impedance parameter generation, and drive characteristic adjustment. This allows the grinding process to maintain a stable cycle time while achieving higher dimensional control accuracy and machining stability.

[0050] In summary, the embodiments of this application have at least the following technical effects:

[0051] A measurement component is deployed on a grinding machine to acquire grinding status characteristics in real time during the automated grinding process of the drive shaft workpiece. A dual-closed-loop control architecture is constructed by inserting a programmable impedance network into the servo driver. This architecture consists of a cascaded state loop and an impedance loop trained under supervision, connected to the measurement component and the programmable impedance network. Based on this dual-closed-loop control architecture, the grinding status characteristics are input into the state loop, performing displacement error analysis based on diameter and mechanical error analysis based on mechanical load. The impedance loop can be optionally activated to perform parameter adjustment compensation decisions based on the optimal resistance and capacitance combination, determining the impedance control parameters. The programmable impedance network responds to these impedance control parameters, adjusting and compensating for the automated grinding process within a specific time window. This approach solves the technical problems of unstable dimensional accuracy and poor machining consistency caused by mechanical load fluctuations, thermal deformation, and system stiffness changes during high-precision drive shaft grinding. It achieves the technical effect of improving drive shaft dimensional accuracy and form and position tolerance stability, reducing scrap rate, and increasing machining efficiency through real-time perception of the grinding status and adaptive impedance control via a dual-closed-loop architecture.

[0052] Example 2, based on the same inventive concept as the precision grinding method for the high-precision drive shaft in the foregoing examples, such as... Figure 2 As shown, this application provides a precision grinding system for high-precision drive shafts, wherein the system includes:

[0053] State Measurement Module 11: Deploys measurement components on the grinding machine tool to acquire grinding state characteristics in real time during the automated grinding process of the workpiece along the drive shaft; Closed-Loop Architecture Construction Module 12: Constructs a dual closed-loop control architecture by inserting a programmable impedance network into the servo driver. This architecture consists of a cascaded state loop and an impedance loop trained under supervision, connected to the measurement components and the programmable impedance network; Closed-Loop Architecture Analysis Module 13: Based on the dual closed-loop control architecture, inputs the grinding state characteristics into the state loop, performs displacement error analysis based on diameter value and mechanical error analysis based on mechanical load, and optionally activates the impedance loop to perform parameter adjustment compensation decisions based on the optimal resistance and capacitance combination to determine impedance control parameters; Grinding Control Module 14: The programmable impedance network responds to the impedance control parameters, performing process time window control compensation superposition on the automated grinding process of the grinding machine tool.

[0054] Furthermore, the state measurement module 11 is used to perform the following method:

[0055] A measuring component is deployed on a grinding machine to acquire the grinding state characteristics of the transmission shaft machining process. The deployment of the measuring component includes: symmetrically installing at least two high-frequency eddy current displacement sensors on both sides of the grinding wheel head, wherein the sensor probes are aligned with the workpiece surface of the transmission shaft, and the high-frequency eddy current displacement sensors measure the workpiece radius change; and installing triaxial piezoelectric force sensors on the headstock and tailstock of the grinding machine, wherein the triaxial piezoelectric force sensors measure the normal grinding force, tangential grinding force, and axial force.

[0056] Furthermore, the closed-loop architecture construction module 12 is used to execute the following methods:

[0057] A programmable impedance network is inserted into the forward current loop of the servo driver, wherein the programmable impedance network consists of a digital potentiometer and a programmable capacitor array; based on the measurement component and the programmable impedance network, a dual closed-loop control architecture is constructed, wherein the dual closed-loop control architecture includes a state loop and an impedance loop.

[0058] Furthermore, the consistency analysis module 13 is used to perform the following methods:

[0059] The grinding process of the drive shaft workpiece is obtained. Taking the grinding state as input, the displacement error and rate of change based on the diameter value are used as the first decision objective, and the mechanical error and rate of change based on the mechanical load are used as the second decision objective to construct the state loop. The optimal resistance and capacitance combination based on impedance matching calculation is used as the optimization and control objective to construct the impedance loop. The state loop and impedance loop are cascaded, and the connection between the state loop and the measurement component, and the impedance loop and the programmable impedance network are established to form the dual closed-loop control architecture.

[0060] Furthermore, the continuity analysis module 14 is used to perform the following method:

[0061] As the transmission shaft workpiece is processed, the first constraint condition is determined according to the grinding stage. In this process, a preset impedance parameter boundary is defined by calibrating a rough grinding label or a fine grinding label for each grinding stage. The impedance loop is then constrained in terms of parameter space using the preset impedance parameter boundary. The parameter space is updated with the real-time grinding stage label.

[0062] Furthermore, the continuity analysis module 14 is used to perform the following method:

[0063] The grinding state characteristics are input into the state loop, and the calculations based on the first decision target and the second decision target are performed in parallel to determine the two-way error data with weight values. For the two-way error data, with any one error data as the greedy target, the impedance control parameter is located in the determined upper and lower bounds of the parameters.

[0064] Furthermore, the continuity analysis module 14 is used to perform the following method:

[0065] For the first error data in the two-way error data, a parameter adjustment decision based on the optimal combination of resistor and capacitor is made with weight values ​​as constraints to determine the first parameter group; for the second error data, a second parameter group is determined; by mapping the first parameter group and the second parameter group, the upper and lower bounds of the parameters are determined; the upper and lower bounds of the parameters are used as the parameter adjustment range, and the impedance control parameters are determined by performing an equilibrium game.

[0066] Furthermore, the image annotation module 15 is used to perform the following method:

[0067] The grinding machine tool performs automated grinding according to the pre-configured process grinding program. The measurement component synchronously performs real-time grinding status measurement and obtains dynamic impedance control parameters based on the status loop and impedance loop. By sending the impedance control parameters to the programmable impedance network, the control compensation superposition of the automated grinding process is performed.

[0068] Example 3, Figure 3 This is a schematic diagram of the structure of an electronic device provided in Embodiment 3 of the present invention, showing a block diagram of an exemplary electronic device suitable for implementing the embodiments of the present invention. Figure 3 The electronic device shown is merely an example and should not be construed as limiting the functionality or scope of the embodiments of the present invention. Figure 3 As shown, the electronic device includes a processor 21, a memory 22, an input device 23, and an output device 24; the number of processors 21 in the electronic device can be one or more. Figure 3 Taking a processor 21 as an example, the processor 21, memory 22, input device 23, and output device 24 in an electronic device can be connected via a bus or other means. Figure 3 Taking the example of a connection between China and Israel via a bus.

[0069] The memory 22, as a computer-readable storage medium, can be used to store software programs, computer-executable programs, and modules, such as the program instructions / modules corresponding to the precision grinding method for the high-precision drive shaft in this embodiment of the invention. The processor 21 executes various functional applications and data processing of the electronic device by running the software programs, instructions, and modules stored in the memory 22, thereby realizing the aforementioned precision grinding method for the high-precision drive shaft.

[0070] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A precision grinding method for high-precision drive shafts, characterized in that, The method includes: Deploy measurement components on grinding machines to acquire grinding status characteristics in real time during the automated grinding process of the workpiece along the drive shaft; A dual-closed-loop control architecture is constructed by inserting a programmable impedance network into the servo driver. The dual-closed-loop control architecture is formed by supervising the training of the cascaded state loop and impedance loop, and establishing connections with the measurement component and the programmable impedance network. Based on the dual closed-loop control architecture, the grinding state characteristics are input into the state loop, and displacement error based on diameter value and mechanical error based on mechanical load are analyzed. Optionally, the impedance loop is activated to make parameter adjustment compensation decisions based on the optimal combination of resistance and capacitance, and to determine the impedance control parameters. The programmable impedance network responds to the impedance control parameters to adjust and compensate for the process time window of the automated grinding process of the grinding machine tool. Constructing a dual closed-loop control architecture includes: A programmable impedance network is inserted into the forward current loop of the servo driver, wherein the programmable impedance network consists of a digital potentiometer and a programmable capacitor array; Based on the measurement component and the programmable impedance network, a dual-loop control architecture is constructed, wherein the dual-loop control architecture includes a state loop and an impedance loop.

2. The precision grinding method for high-precision drive shafts as described in claim 1, characterized in that, Real-time acquisition of grinding status characteristics, including: Deploy measurement components on a grinding machine to acquire grinding state characteristics of the drive shaft machining process; The deployment of the measurement component includes: At least two high-frequency eddy current displacement sensors are symmetrically installed on both sides of the grinding wheel frame, wherein the sensing probe is aligned with the workpiece surface of the drive shaft, and the high-frequency eddy current displacement sensor measures the change in workpiece radius. A triaxial piezoelectric force sensor is installed on the headstock and tailstock of the grinding machine. The triaxial piezoelectric force sensor measures the normal grinding force, the tangential grinding force and the axial force.

3. The precision grinding method for high-precision drive shafts as described in claim 1, characterized in that, Based on the measurement component and the programmable impedance network, a dual closed-loop control architecture is constructed, including: The grinding process of the drive shaft workpiece is obtained. The grinding state is used as input, and the displacement error and rate of change based on the diameter value are used as the first decision objective. The mechanical error and rate of change based on the mechanical load are used as the second decision objective to construct the state loop. The impedance loop is constructed with the optimal resistance and capacitance combination calculated based on impedance matching as the optimization control objective. The state loop and impedance loop are cascaded, and connections are established between the state loop and the measurement components, and between the impedance loop and the programmable impedance network, to form the dual closed-loop control architecture.

4. The precision grinding method for high-precision drive shafts as described in claim 3, characterized in that, Before activating the impedance loop, the following steps are included: As the transmission shaft workpiece is processed, the first constraint condition is determined according to the grinding stage. In this process, the preset impedance parameter boundary is defined by calibrating the rough grinding label or fine grinding label for each grinding stage. The impedance loop is constrained in terms of parameter space by the preset impedance parameter boundary, wherein the parameter space is updated with the label of the real-time grinding stage.

5. The precision grinding method for high-precision drive shafts as described in claim 4, characterized in that, Determine the impedance control parameters, including: The grinding state characteristics are input into the state loop, and the calculations based on the first decision objective and the second decision objective are performed in parallel to determine the two-way error data with weighted values. For the two-way error data, taking any one of the error data as the greedy target, the impedance control parameter is located within the determined upper and lower bounds of the parameters.

6. The precision grinding method for high-precision drive shafts as described in claim 5, characterized in that, Using any error data point as the greedy target, locate the impedance control parameters within the defined upper and lower bounds of the parameters, including: For the first error data in the two-way error data, a parameter adjustment decision based on the optimal combination of resistor and capacitor is made with the weight value as a constraint to determine the first parameter group; For the second error data, determine the second set of parameters; The upper and lower bounds of the parameters are determined by mapping the first parameter group and the second parameter group. The upper and lower bounds of the parameters are used as the parameter tuning range, and the impedance control parameters are determined by performing an equilibrium game.

7. The precision grinding method for high-precision drive shafts as described in claim 1, characterized in that, The programmable impedance network responds to the impedance modulation parameters, including: The grinding machine tool performs automated grinding according to the pre-configured process grinding program, and the measuring component synchronously performs real-time grinding status measurement, and obtains dynamic impedance control parameters based on the status loop and impedance loop. By sending the impedance control parameters to the programmable impedance network, the control and compensation superposition of the automated grinding process is performed.

8. A precision grinding system for high-precision drive shafts, characterized in that, A precision grinding method for implementing the high-precision drive shaft according to any one of claims 1-7, the system comprising: Status measurement module: Deploy measurement components on the grinding machine tool to acquire grinding status characteristics in real time during the automated grinding process of the workpiece along the drive shaft; Closed-loop architecture building module: By inserting a programmable impedance network into the servo driver, a dual closed-loop control architecture is constructed. The dual closed-loop control architecture is formed by supervising the training of the cascaded state loop and impedance loop, and establishing connections with the measurement component and the programmable impedance network. Closed-loop architecture analysis module: Based on the dual closed-loop control architecture, the grinding state characteristics are input into the state loop, and displacement error based on diameter value and mechanical error based on mechanical load are analyzed. Optionally, the impedance loop is activated to make parameter adjustment compensation decisions based on the optimal resistance and capacitance combination to determine the impedance control parameters. Grinding control module: The programmable impedance network responds to the impedance control parameters to control and compensate the process time window of the automated grinding process of the grinding machine tool.

9. An electronic device, characterized in that, The electronic device includes: Memory, used to store executable instructions; The processor, when executing executable instructions stored in the memory, implements the precision grinding method for the high-precision drive shaft as described in any one of claims 1-7.

Images