P1 precision thread grinding dynamic compensation device for planetary roller screw machining

Abstract

The invention discloses a P1 precision thread grinding dynamic compensation device for planetary roller screw machining, and belongs to the technical field of thread grinding. The device comprises a global temperature field sensing module which comprises a distributed fiber bragg grating sensing network and a high-speed infrared thermal imager and is used for synchronously collecting the internal temperature and the surface temperature of a grinding area; the thermal deformation prediction module is used for correcting the prediction model in real time and identifying the contribution degree of each heat source; the thermal deformation decoupling module is used for outputting a high-precision three-dimensional temperature field and predicting decomposition thermal deformation; the active cooling control module comprises a cooling system and a cooling strategy controller and is used for optimally distributing cooling parameters; and the real-time compensation control module is used for generating a compensation instruction based on the residual thermal deformation predicted value. Through closed-loop cooperative control of sensing, prediction, cooling and compensation, thermal deformation is restrained from the source, residual errors are compensated, and high-stability grinding machining of the planetary roller screw with the P1-level precision is achieved.

Description

Technical Field

[0001] This invention relates to the field of thread grinding technology, and specifically to a dynamic compensation device for P1 precision thread grinding in planetary roller screw machining. Background Technology

[0002] As a core transmission component of high-end electromechanical actuators, the thread grinding precision of planetary roller screws directly affects transmission efficiency, positioning accuracy, and service life. With the upgrading needs of equipment in fields such as humanoid robots, precision machine tools, and aerospace, the machining of planetary roller screws with P1-level precision (lead error ≤5μm / 25mm) has become a technical bottleneck in the industry.

[0003] In P1 precision thread grinding, thermal deformation is the main dynamic error source restricting the stability of machining accuracy. Studies have shown that thermally induced errors caused by cutting heat, frictional heat, and ambient temperature fluctuations can account for 40%-70% of the total machining error. Existing thermal error compensation technologies mostly use a limited number of point temperature sensors, which are difficult to reflect the real-time gradient distribution of the three-dimensional temperature field in the grinding zone; the prediction models are mostly "black box" or empirical models, and the prediction accuracy drops sharply when the machining conditions change; they are mostly post-compensation, lacking the ability to predict thermal deformation trends in advance; and they only compensate for thermal deformation that has already occurred through trajectory correction, without suppressing the generation and conduction of heat at the source.

[0004] Therefore, current thermal error compensation technologies suffer from problems such as limited sensing dimensions, lack of prediction mechanisms, compensation lag, and lack of source suppression. There is an urgent need for an active thermal error suppression device that can achieve closed-loop coordinated control of sensing, prediction, cooling, and compensation to meet the grinding requirements of P1 precision planetary roller screws. Summary of the Invention

[0005] The purpose of this invention is to provide a dynamic compensation device for P1 precision thread grinding in planetary roller screw machining, so as to solve the problems of single thermal error perception dimension, lack of prediction mechanism, compensation lag and lack of source suppression.

[0006] To address the aforementioned technical problems, this invention provides a dynamic compensation device for P1 precision thread grinding in planetary roller screw machining, comprising: a global temperature field sensing module, a thermal deformation prediction module, a thermal deformation decoupling module, an active cooling control module, and a real-time compensation control module.

[0007] The global temperature field sensing module is used to simultaneously acquire dynamic images of the temperature and surface temperature fields at key points inside the grinding zone. The global temperature field sensing module includes a distributed fiber Bragg grating sensor network and a high-speed infrared thermal imager. The distributed fiber Bragg grating sensor network includes multiple Bragg gratings serially fabricated on a single optical fiber. These gratings are embedded at least in the grinding machine's grinding wheel head spindle bearing housing, workpiece headstock spindle bearing, workpiece tailstock, bed guide rails, coolant nozzles, and ambient temperature monitoring points, and are connected to the high-speed fiber Bragg grating demodulator. The high-speed infrared thermal imager is aligned with the grinding zone. The thermal deformation prediction module includes a thermal-structural coupled finite element model, a parameter adaptive updater, and a thermal deformation source tracing analysis unit. The thermal-structural coupled finite element model is used to solve for simulated temperature data based on boundary conditions. The parameter adaptive updater receives measured temperature data from the global temperature field sensing module and corrects the thermal boundary conditions in the thermal-structural coupled finite element model in real time by comparing the deviation between the simulated temperature data and the measured temperature data. The thermal deformation source tracing analysis unit calculates the transfer function matrix of the unit heat change of each heat source on the thermal deformation of key positions of the workpiece based on the sensitivity analysis of the thermal-structural coupled finite element model, and identifies the contribution of each heat source to the thermal deformation of the workpiece. The thermal deformation decoupling module includes a Kalman filter fusion algorithm and a thermal deformation decoupling unit. The state equation of the Kalman filter fusion algorithm is constructed based on the transient heat conduction finite element discrete equation. The observation equation is associated with the measured temperature data of the distributed fiber grating sensor network and the high-speed infrared thermal imager. The optimal fusion of internal and external temperature data is achieved through Kalman filter recursion, and a high-precision three-dimensional temperature field is output. The thermal deformation decoupling unit predicts the thermal deformation trend within a preset time period based on the high-precision three-dimensional temperature field and decomposes it into axial thermal deformation of the workpiece, radial thermal deformation of the workpiece, and additional error of guide rail bending. The active cooling control module includes a cooling system and a cooling strategy controller. The cooling system includes at least a spindle bearing cooling circuit, a grinding zone jet cooling circuit, and a bed guide constant temperature circulation circuit. Each circuit is equipped with an independently controlled electronically controlled proportional valve, a flow meter, and a temperature sensor. The cooling strategy controller receives the contribution of each heat source from the thermal deformation source analysis unit and the predicted thermal deformation from the thermal deformation decoupling unit. Based on the minimum thermal deformation control algorithm, with minimizing the total thermal deformation as the objective function and cooling energy consumption as the constraint, it optimizes and solves the flow rate setpoint and temperature setpoint of each cooling circuit in the next control cycle. The real-time compensation control module includes a feedforward-feedback composite controller, which receives the predicted value of residual thermal deformation after active cooling suppression output by the thermal deformation decoupling module, and generates compensation instructions that are superimposed on the original interpolation instructions of the CNC system of the grinding machine. The compensation instructions include at least a Z-axis position correction instruction for the axial thermal deformation of the workpiece, an X-axis position correction instruction for the radial thermal deformation of the workpiece, and a multi-axis linkage correction instruction for the additional error of the guide rail bending.

[0008] Furthermore, the high-speed fiber Bragg grating demodulator and the high-speed infrared thermal imager are both connected to the same synchronous acquisition card. The synchronous acquisition card outputs a unified TTL synchronous trigger signal with a frequency higher than a preset value to trigger the synchronous acquisition of the high-speed fiber Bragg grating demodulator and the high-speed infrared thermal imager.

[0009] Furthermore, the thermal deformation decoupling module adopts a first control cycle, and the active cooling control module adopts a second control cycle, wherein the first control cycle is longer than the second control cycle.

[0010] Furthermore, the total thermal deformation in the minimum thermal deformation control algorithm is a weighted sum of the workpiece's axial thermal deformation, the workpiece's radial thermal deformation, and the additional error caused by the guide rail bending.

[0011] Furthermore, the thermal boundary conditions corrected by the parameter adaptive updater include at least: heat source intensity, convective heat transfer coefficient, and contact thermal resistance.

[0012] The beneficial effects of this invention are as follows: By combining a distributed fiber optic grating sensor network with a high-speed infrared thermal imager, a complete reconstruction of the three-dimensional temperature field of the grinding zone is achieved, providing a comprehensive data foundation for accurate thermal deformation analysis; the finite element mechanism model and measured data are deeply integrated, and the thermal boundary conditions are corrected in real time through a parameter adaptive updater, ensuring both the physical interpretability of thermal deformation prediction and improving the model's adaptability to different working conditions; the optimal fusion of internal and external temperature data is achieved based on the Kalman filter fusion algorithm, improving the accuracy of the three-dimensional temperature field; the error is decomposed into three independent components: axial, radial, and guide rail additional error, realizing the ability to predict and decouple thermal deformation in advance; active cooling suppresses the generation of heat and thermal deformation from the source, and dynamic compensation corrects residual errors. The synergistic effect of the two is better than that of a single method in suppressing thermal errors; active cooling and compensation operate at different cycles, forming a cascaded control, with the inner loop rapidly cooling to suppress the main thermal deformation and the outer loop compensating for residual errors, reducing the requirements for the servo system's response speed and enhancing system stability; through closed-loop collaborative control of sensing, prediction, cooling, and compensation, the high-precision requirements of planetary roller screw thread grinding can be achieved. Attached Figure Description

[0013] The accompanying drawings, which form part of this application, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings: Figure 1 This is a schematic diagram of the dynamic compensation device according to an embodiment of the present invention.

[0014] Figure 2 This is a control logic block diagram of the dynamic compensation device according to an embodiment of the present invention.

[0015] Figure 3 This is a flowchart of the control steps in an embodiment of the present invention.

[0016] In the picture: 1. Global temperature field sensing module; 11. Distributed fiber optic grating sensor network; 111. Bragg grating; 112. High-speed fiber optic grating demodulator; 12. High-speed infrared thermal imager. 2. Grinding machine; 21. Grinding wheel head spindle bearing housing; 22. Workpiece headstock spindle bearing; 23. Workpiece tailstock; 24. Bed guide rail; 25. Coolant nozzle. 3. Thermal deformation prediction module; 31. Thermal-structure coupled finite element model; 32. Parameter adaptive updater; 33. Thermal deformation source analysis unit. 4. Thermal deformation decoupling module; 41. Kalman filter fusion algorithm; 42. Thermal deformation decoupling unit; 5. Active cooling control module; 51. Cooling system; 511. Spindle bearing cooling circuit; 512. Grinding zone jet cooling circuit; 513. Bed guide rail constant temperature circulation circuit; 52. Cooling strategy controller. 6. Real-time compensation control module. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments implemented by those skilled in the art without creative effort are within the protection scope of the present invention.

[0018] Example 1 This embodiment provides a dynamic compensation device for P1 precision thread grinding in planetary roller screw machining, and its structural schematic diagram is shown below. Figure 1 As shown, its control logic block diagram is as follows: Figure 2 As shown, it includes a global temperature field sensing module 1, a thermal deformation prediction module 3, a thermal deformation decoupling module 4, an active cooling control module 5, and a real-time compensation control module 6, and is connected to the grinding machine body 2.

[0019] The global temperature field sensing module 1 is used to simultaneously acquire dynamic images of the temperature and surface temperature fields at key points inside the grinding zone, providing a complete data foundation for thermal deformation analysis.

[0020] The global temperature field sensing module 1 includes a distributed fiber optic grating sensor network 11 and a high-speed infrared thermal imager 12. The distributed fiber optic grating sensor network 11 comprises multiple Bragg gratings 111 serially fabricated on a single optical fiber. These gratings are embedded at least in the grinding machine 2's grinding wheel head spindle bearing housing 21, workpiece headstock spindle bearing 22, workpiece tailstock 23, bed guide rail 24, coolant nozzle 25, and ambient temperature monitoring points, and are connected to the high-speed fiber optic grating demodulator 112. The high-speed infrared thermal imager 12 is aligned with the grinding area. The distributed fiber optic grating sensor network 11 is constructed by serially fabricating multiple Bragg gratings 111 on a single optical fiber, each grating having a different center wavelength to achieve wavelength division multiplexing. The gratings are embedded in key heat source and heat-sensitive locations on the grinding machine, including the grinding wheel head spindle bearing housing, workpiece headstock spindle bearing, workpiece tailstock, bed guide rail, coolant nozzle, and ambient temperature monitoring points. All optical fibers are led out of the machine tool and connected to the high-speed fiber optic grating demodulator. The high-speed infrared thermal imager 12 can be mounted above the grinding wheel holder, with its lens aimed at the grinding wheel-workpiece grinding area. It is used for non-contact acquisition of the surface temperature field distribution in the grinding area. To prevent splashing, an air curtain can be used for protection. Through the internal and external integration of the distributed fiber optic grating sensor network 11 and the high-speed infrared thermal imager 12, a complete reconstruction of the three-dimensional temperature field of the grinding area is achieved. The fiber optic grating provides high-precision internal point temperature data, while the infrared thermal imager provides the full-field surface temperature distribution. The fusion of these two technologies overcomes the limitations of a single sensor, providing a complete data foundation for accurate thermal deformation analysis.

[0021] Furthermore, both the high-speed fiber Bragg grating demodulator 112 and the high-speed infrared thermal imager 12 are connected to the same synchronous acquisition card. The synchronous acquisition card outputs a unified TTL synchronous trigger signal with a frequency higher than a preset value, triggering the synchronous acquisition of the high-speed fiber Bragg grating demodulator 112 and the high-speed infrared thermal imager 12. Synchronous acquisition ensures that the timestamps of the two sensor data are accurately aligned, laying the foundation for subsequent data fusion; the higher the output frequency, the faster the acquisition and the higher the timestamp accuracy.

[0022] The thermal deformation prediction module 3 includes a thermal-structural coupled finite element model 31, a parameter adaptive updater 32, and a thermal deformation tracing analysis unit 33. The thermal deformation prediction module can be deployed on an edge computing server.

[0023] The thermal-structural coupled finite element model 31 is used to solve for the simulated temperature data based on boundary conditions. The thermal-structural coupled finite element model 31 is constructed based on the CAD geometric models of the grinding machine and the workpiece, and undergoes model reduction processing to meet real-time simulation requirements. This model comprehensively considers heat conduction, heat convection, and thermal-structural coupling effects, and can solve for the simulated temperature and thermal deformation data based on preset thermal boundary conditions.

[0024] The parameter adaptive updater 32 receives measured temperature data from the global temperature field sensing module 1 and corrects the thermal boundary conditions in the thermal-structural coupled finite element model 31 in real time by comparing the deviation between the simulated temperature data and the measured temperature data. Furthermore, the corrected thermal boundary conditions include at least the heat source intensity, convective heat transfer coefficient, and contact thermal resistance. The correction mechanism is as follows: using the measured temperature data as a reference, optimization algorithms such as gradient descent or Kalman filtering are employed to solve in reverse for the thermal boundary condition parameters that minimize the simulation error, thereby achieving dynamic matching between the model and the physical reality.

[0025] The thermal deformation source analysis unit 33, based on the sensitivity analysis of the thermal-structural coupled finite element model 31, calculates the transfer function matrix of the effect of unit heat change of each heat source on the thermal deformation of the workpiece at key locations, and identifies the contribution of each heat source to the thermal deformation of the workpiece. Its principle is as follows: a unit perturbation is applied at the input of each heat source in the model, and the thermal deformation response at key locations of the workpiece is calculated through finite element simulation. This establishes the mapping relationship between input and output, and then quantitatively evaluates the influence weight of each heat source.

[0026] The thermal deformation decoupling module 4 includes a Kalman filter fusion algorithm 41 and a thermal deformation decoupling unit 42; The state equation of the Kalman filter fusion algorithm 41 is constructed based on the transient heat conduction finite element discrete equation. The observation equation is associated with the measured temperature data of the distributed fiber optic grating sensor network 11 and the high-speed infrared thermal imager 12. Optimal fusion of internal and external temperature data is achieved through Kalman filtering recursion, outputting a high-precision three-dimensional temperature field. Its state transition matrix is ​​determined by the material's thermal properties and the spatial grid size, reflecting the physical laws of heat conduction. The observation matrix is ​​determined by shape function interpolation based on the position of each sensor measurement point in the finite element grid, realizing the mapping from the continuous temperature field to discrete measurement point observations. Through the prediction-correction recursive process of Kalman filtering, the advantages of the two types of sensors are fused: the fiber optic grating provides high-precision internal point temperature, and the infrared thermal imager provides full-field distribution information, ultimately outputting a high-precision three-dimensional temperature field with spatial resolution reaching the finite element grid density and accuracy superior to that of a single sensor.

[0027] The thermal deformation decoupling unit 42, based on a high-precision three-dimensional temperature field, predicts the thermal deformation trend within a preset time period and decomposes it into axial thermal deformation of the workpiece, radial thermal deformation of the workpiece, and additional errors caused by guide rail bending. Its prediction mechanism is as follows: the three-dimensional temperature field is used as a load input to the thermo-structure coupling model to solve the thermoelastic deformation of the workpiece and machine tool structure; combined with time series analysis or model extrapolation methods, the evolution trend of thermal deformation is predicted in advance. Its decoupling principle is as follows: based on deformation modal analysis or coordinate transformation, the comprehensive thermal deformation is decomposed into axial thermal deformation along the workpiece axis (affecting thread lead), radial thermal deformation along the workpiece radial direction (affecting thread pitch diameter), and additional errors caused by machine tool guide rail bending (affecting motion straightness).

[0028] The thermal-structural coupled finite element model 31 ensures the physical interpretability of the prediction, the parameter adaptive updater 32 realizes the dynamic matching between the model and the measured data, and the Kalman filter fusion algorithm 41 optimally combines the physical model and the measured data. The synergy of these three elements enables the thermal deformation prediction to possess both mechanistic reliability and operational adaptability. The comprehensive thermal deformation is decomposed into three independent components: axial, radial, and guideway additional errors. This decoupling makes subsequent cooling control and compensation corrections more targeted: axial thermal deformation mainly affects lead accuracy, radial thermal deformation mainly affects pitch diameter accuracy, and guideway additional errors affect motion straightness. Differential control for different error components improves the overall control effect.

[0029] The active cooling control module 5 includes a cooling system 51 and a cooling strategy controller 52.

[0030] The cooling system 51 includes at least a spindle bearing cooling circuit 511, a grinding zone jet cooling circuit 512, and a bed guideway constant temperature circulation circuit 513. Each circuit is equipped with an independently controlled electronically controlled proportional valve, flow meter, and temperature sensor to achieve precise control of cooling parameters (flow rate, temperature, and pressure). Furthermore, the grinding zone jet cooling circuit 512 includes multiple micro-cooling nozzles arranged around the circumference of the grinding wheel. The angle between the spray direction of each nozzle and the tangential direction of the grinding wheel is adjustable, forming a uniform cooling curtain covering the grinding zone.

[0031] The cooling strategy controller 52 receives the contribution values ​​of each heat source from the thermal deformation tracing analysis unit 33 and the predicted thermal deformation values ​​from the thermal deformation decoupling unit 42. Based on the minimum thermal deformation control algorithm, it optimizes the flow rate and temperature setpoints of each cooling loop in the next control cycle by minimizing the total thermal deformation as the objective function and cooling energy consumption as the constraint. Its control mechanism is as follows: based on the contribution value of each heat source, it determines the optimal allocation strategy for cooling resources—prioritizing the allocation of cooling capacity to heat sources with high contribution values; it uses the predicted thermal deformation value as feedback to evaluate the effectiveness of the current cooling strategy; and it constructs and solves an optimization problem by minimizing the total thermal deformation as the objective function and cooling energy consumption and temperature constraints as the constraints. The total thermal deformation is a weighted sum of each thermal deformation component, with the weighting coefficients determined according to the machining accuracy requirements. Further, the total thermal deformation in the minimum thermal deformation control algorithm is a weighted sum of the workpiece axial thermal deformation, the workpiece radial thermal deformation, and the additional error of the guide rail bending. Alternatively, other weighting methods such as exponential weighted averages can also be used.

[0032] Furthermore, the thermal deformation decoupling module 4 employs a first control cycle, while the active cooling control module 5 employs a second control cycle, with the first control cycle being longer than the second. The active cooling control module uses a faster control cycle to achieve rapid suppression of thermal deformation; the thermal deformation decoupling module uses a slower control cycle to achieve accurate prediction and decoupling of thermal deformation; the inner loop response speed is higher than the outer loop, forming a collaborative control mechanism of "rapid cooling to suppress major deformation and slow compensation to correct residual errors".

[0033] The real-time compensation control module 6 receives the predicted residual thermal deformation value after active cooling suppression from the thermal deformation decoupling module 4, and generates compensation commands that are superimposed on the original interpolation commands of the CNC system of the grinding machine 2. The compensation commands include at least a Z-axis position correction command for axial thermal deformation of the workpiece, an X-axis position correction command for radial thermal deformation of the workpiece, and a multi-axis linkage correction command for additional errors caused by guideway bending. Its compensation mechanism is a feedforward-feedback controller: the negative value of the predicted residual thermal deformation is used as a feedforward quantity and superimposed on the servo position command in advance, causing the tool path to deviate in the opposite direction to the workpiece's thermal deformation direction, thereby offsetting the effect of thermal deformation. Simultaneously, the feedback controller adjusts the compensation parameters online according to the actual machining error, forming a feedforward-feedback composite control to improve compensation accuracy and robustness.

[0034] The dynamic compensation device in this embodiment achieves closed-loop coordinated control of "sensing-prediction-cooling-compensation" through the following steps: Figure 3 As shown: S1 Multi-Source Temperature Sensing: Activates the global temperature field sensing module. A distributed fiber optic grating sensor network collects temperatures at key points inside the grinding machine, including spindle bearing temperature rise, guide rail temperature gradient, and coolant temperature changes. A high-speed infrared thermal imager simultaneously acquires the surface temperature field distribution of the grinding zone, capturing localized high temperatures caused by grinding heat. The synchronous acquisition card outputs a unified trigger signal to ensure precise temporal alignment of the two types of temperature data.

[0035] S2 Model Online Calibration and Thermal Deformation Source Tracing: Measured temperature data is input into the thermal deformation prediction module. The parameter adaptive updater compares the deviation between the simulated temperature and the measured temperature, and corrects the heat source intensity, convective heat transfer coefficient, and contact thermal resistance in the thermo-structural coupled finite element model in real time, dynamically matching the model with physical reality. The thermal deformation source tracing analysis unit performs sensitivity analysis based on the calibrated model, calculates the contribution of each heat source to the workpiece's thermal deformation, and identifies the main heat sources.

[0036] S3 Thermal Deformation Fusion Prediction and Decoupling: Measured temperature data is input into the thermal deformation decoupling module. The Kalman filter fusion algorithm, based on the state equation constructed from the physical equation of heat conduction and the observation equations relating the two types of sensors, achieves optimal fusion of internal and external temperature data through recursive filtering, outputting a high-precision three-dimensional temperature field. Based on this three-dimensional temperature field, the thermal deformation decoupling unit predicts the thermal deformation evolution trend over a future period and decomposes it into axial thermal deformation of the workpiece, radial thermal deformation of the workpiece, and additional error due to guide rail bending.

[0037] S4 Active Cooling Co-control: The heat source contribution and predicted thermal deformation are input into the active cooling control module. The cooling strategy controller, aiming to minimize total thermal deformation and constrained by cooling energy consumption, optimizes the flow rate and temperature setpoints of each cooling loop. The optimization results are sent to the cooling system: the spindle bearing cooling loop controls bearing temperature rise, the grinding zone jet cooling loop reduces the temperature in the grinding arc zone, and the bed guideway constant temperature circulation loop maintains the stability of the basic structure. Active cooling suppresses heat generation and conduction at the source, reducing major thermal deformation.

[0038] S5 Residual Thermal Deformation Compensation: The predicted value of residual thermal deformation after active cooling suppression is input into the real-time compensation control module. The feedforward-feedback composite controller generates compensation commands to correct the Z-axis position for axial thermal deformation, the X-axis position for radial thermal deformation, and perform multi-axis linkage correction for additional guideway errors. The compensation commands are superimposed on the original interpolation commands of the CNC system to drive the servo axes to achieve micron-level or even submicron-level trajectory correction.

[0039] S6 Iterative Cycle: Repeat steps one through five above to form a continuous closed-loop control. As the processing progresses, the model is continuously corrected, predictions are constantly updated, cooling is dynamically adjusted, and compensation is executed in real time, ensuring that thermal errors remain under control.

[0040] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A dynamic compensation device for P1 precision thread grinding in planetary roller screw machining, characterized in that, include: The global temperature field sensing module (1) is used to synchronously acquire dynamic images of the temperature and surface temperature fields of key points inside the grinding zone; the global temperature field sensing module (1) includes a distributed fiber optic grating sensor network (11) and a high-speed infrared thermal imager (12). The distributed fiber grating sensing network (11) includes multiple Bragg gratings (111) fabricated in series on a single optical fiber. The gratings are at least embedded in the grinding wheel head spindle bearing seat (21), the workpiece headstock spindle bearing (22), the workpiece tailstock (23), the bed guide rail (24), the coolant nozzle (25), and the ambient temperature monitoring point of the grinding machine (2), and are connected to a high-speed fiber grating demodulator (112). The high-speed infrared thermal imager (12) is aligned with the grinding area; The thermal deformation prediction module (3) includes a thermal-structure coupled finite element model (31), a parameter adaptive updater (32), and a thermal deformation source analysis unit (33). The thermal-structural coupled finite element model (31) is used to solve the simulation temperature data based on the boundary conditions; The parameter adaptive updater (32) receives the measured temperature data from the global temperature field sensing module (1) and corrects the thermal boundary conditions in the thermal-structure coupled finite element model (31) in real time by comparing the deviation between the simulated temperature data and the measured temperature data. The thermal deformation source analysis unit (33) is based on the sensitivity analysis of the thermal-structure coupled finite element model (31), calculates the transfer function matrix of the unit heat change of each heat source on the thermal deformation of the key position of the workpiece, and identifies the contribution of each heat source to the thermal deformation of the workpiece. The thermal deformation decoupling module (4) includes a Kalman filter fusion algorithm (41) and a thermal deformation decoupling unit (42). The state equation of the Kalman filter fusion algorithm (41) is constructed based on the transient heat conduction finite element discrete equation. The observation equation is associated with the measured temperature data of the distributed fiber grating sensor network (11) and the high-speed infrared thermal imager (12). The optimal fusion of internal and external temperature data is achieved through Kalman filter recursion, and a high-precision three-dimensional temperature field is output. The thermal deformation decoupling unit (42) predicts the thermal deformation trend within a preset time based on the high-precision three-dimensional temperature field and decomposes it into workpiece axial thermal deformation, workpiece radial thermal deformation and guide rail bending additional error. The active cooling control module (5) includes a cooling system (51) and a cooling strategy controller (52); The cooling system (51) includes at least a spindle bearing cooling circuit (511), a grinding zone jet cooling circuit (512), and a bed guide rail constant temperature circulation circuit (513). Each circuit is equipped with an independently controlled electronically controlled proportional valve, a flow meter, and a temperature sensor. The cooling strategy controller (52) receives the contribution of each heat source output by the thermal deformation source analysis unit (33) and the thermal deformation prediction value output by the thermal deformation decoupling unit (42). Based on the minimum thermal deformation control algorithm, with the goal of minimizing the total thermal deformation and with the cooling energy consumption as the constraint, it optimizes and solves the flow rate setpoint and temperature setpoint of each cooling loop in the next control cycle. The real-time compensation control module (6) receives the predicted value of residual thermal deformation after active cooling suppression output by the thermal deformation decoupling module (4), and generates compensation instructions that are superimposed on the original interpolation instructions of the CNC system of the grinding machine (2). The compensation instructions include at least the Z-axis position correction instructions for the axial thermal deformation of the workpiece, the X-axis position correction instructions for the radial thermal deformation of the workpiece, and the multi-axis linkage correction instructions for the additional error of the guide rail bending.

2. The dynamic compensation device for P1 precision thread grinding in planetary roller screw machining according to claim 1, characterized in that, The high-speed fiber optic demodulator (112) and the high-speed infrared thermal imager (12) are both connected to the same synchronous acquisition card. The synchronous acquisition card outputs a unified TTL synchronous trigger signal with a frequency higher than a preset value to trigger the synchronous acquisition of the high-speed fiber optic demodulator (112) and the high-speed infrared thermal imager (12).

3. The dynamic compensation device for P1 precision thread grinding in planetary roller screw machining according to claim 2, characterized in that, The thermal deformation decoupling module (4) adopts a first control cycle, and the active cooling control module (5) adopts a second control cycle, wherein the first control cycle is longer than the second control cycle.

4. The dynamic compensation device for P1 precision thread grinding in planetary roller screw machining according to claim 3, characterized in that, The total thermal deformation in the minimum thermal deformation control algorithm is a weighted sum of the workpiece's axial thermal deformation, the workpiece's radial thermal deformation, and the additional error caused by the guide rail bending.

5. The dynamic compensation device for P1 precision thread grinding in planetary roller screw machining according to claim 4, characterized in that, The thermal boundary conditions corrected by the parameter adaptive updater (32) include at least: heat source intensity, convective heat transfer coefficient and contact thermal resistance.

6. The dynamic compensation device for P1 precision thread grinding in planetary roller screw machining according to claim 5, characterized in that, The grinding zone jet cooling circuit (512) includes multiple micro-cooling nozzles arranged around the circumference of the grinding wheel, and the angle between the jet direction of each nozzle and the tangential direction of the grinding wheel is adjustable.

Images