Error compensation method for multi-axis linkage machining of mold blank hole series

Abstract

The invention relates to the technical field of information, in particular to an error compensation method for multi-axis linkage machining of a mold blank hole system, which comprises the following steps: judging the time range of a tool changing window based on an error accumulation trend, if error accumulation reaches a preset threshold value, triggering a compensation action, otherwise, continuously monitoring error change; aiming at the complexity of a motion track, a nonlinear optimization algorithm is adopted, an execution path of a compensation action is optimized, and conflicts with a processing action are reduced; according to the optimized compensation path, the control parameters of the servo motor are adjusted in real time, and it is ensured that the compensation action is stably executed in the machining process; dynamically updating an error model through data of a main shaft temperature sensor and a tool wear monitoring device, and continuously optimizing a time point and an execution path of a compensation action; in the machining process, the error accumulation condition is monitored in real time, if the error exceeds the preset range, the compensation action is triggered again, and the machining precision is ensured.

Description

Technical Field

[0001] The present invention relates to the field of information technology, specifically to the processing technology of die blanks, and particularly to an error compensation method for multi-axis linkage machining of die blank hole systems. Background Art

[0002] In the fine machining method of die blank hole systems based on multi-axis linkage compensation, the core problem faced by the time guidance technology is how to achieve high-precision synchronization between the compensation action and the machining action during multi-axis linkage machining. In traditional methods, the compensation action is often triggered after detecting an error. However, due to the time delay between the machining action and the compensation action, the error has accumulated to a certain extent before compensation begins, making it difficult to effectively suppress the further expansion of the error. In addition, during multi-axis linkage machining, the motion trajectories of each axis are complex, and the timing relationship between the machining action and the compensation action is even more difficult to accurately control. If the compensation action is executed too early or too late, new errors may be introduced, or even conflicts with the machining action may occur, affecting the machining accuracy and efficiency.

[0003] In the actual machining scenario, the motion speeds, accelerations, and force states of the tool at different positions vary significantly, resulting in non-linear characteristics of the current fluctuations of the servo motors of each axis. This non-linear characteristic makes it difficult to predict the time point of the compensation action through a simple linear model. At the same time, the dynamic changes in the spindle temperature and tool wear further increase the complexity of time guidance. If the influence of these changes on the error cannot be accurately captured, it may lead to a deviation in the synchronization between the compensation action and the machining action, affecting the final machining accuracy. In addition, during multi-axis linkage machining, the time window for tool change clearance is limited. If the compensation verification action fails to make full use of this window, the compensation effect may not take effect in time, affecting the subsequent machining quality.

[0004] Therefore, how to achieve high-precision synchronization between the compensation action and the machining action within a limited time window has become a key problem that needs to be solved by the time guidance technology. Summary of the Invention

[0005] The present invention provides an error compensation method for multi-axis linkage machining of die blank hole systems, including the following steps:

[0006] S101. Obtain the motion trajectory data of each axis during multi-axis linkage machining, record the current fluctuation information of the servo motor in real time, and establish a dynamically changing error model in combination with the data of the spindle temperature sensor and the tool wear monitoring device;

[0007] S102. According to the error model, analyze the non-linear characteristics of the servo current fluctuation and the spindle temperature change, calculate the predicted value of the time point of the compensation action, and determine the trend of error accumulation;

[0008] S103. Based on the error accumulation trend, determine the time range of the tool change window. If the error accumulation reaches the preset threshold, trigger the compensation action; otherwise, continue to monitor the error change.

[0009] S104. Within the tool change window, through the analysis of the timing relationship, adjust the execution order of the compensation action and the machining action to ensure their precise synchronization in time and avoid introducing new errors.

[0010] S106. According to the optimized compensation path, adjust the control parameters of the servo motor in real time to ensure the smooth execution of the compensation action during the machining process.

[0011] S106. According to the optimized compensation path, adjust the control parameters of the servo motor in real time to ensure the smooth execution of the compensation action during the machining process.

[0012] S107. Through the data of the spindle temperature sensor and the tool wear monitoring device, dynamically update the error model and continuously optimize the time point and execution path of the compensation action.

[0013] S108. During the machining process, monitor the error accumulation in real time. If the error exceeds the preset range, trigger the compensation action again to ensure the machining accuracy.

[0014] The technical solution provided by the embodiment of the present invention may include the following beneficial effects:

[0015] The present invention discloses an error compensation method for multi-axis linkage machining of die blank hole systems. The method collects data such as the motion trajectories of each axis, servo current fluctuations, spindle temperature, and tool wear in real time to establish a dynamic error model. Based on this model, the present invention analyzes the non-linear characteristics, predicts the compensation time point, and determines the tool change window. When the error accumulation reaches the threshold, the present invention triggers the compensation action and ensures the precise synchronization of the compensation and machining actions through timing analysis. For complex trajectories, the present invention uses a non-linear optimization algorithm to optimize the compensation path and reduce conflicts. At the same time, the present invention dynamically adjusts the control parameters of the servo motor to ensure the smooth execution of the compensation. Through continuous updating of the error model and real-time monitoring, the present invention can trigger compensation in time and effectively guarantee the machining accuracy. This method significantly improves the accuracy and stability of multi-axis linkage machining and provides an innovative solution for high-precision manufacturing. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a flowchart of an error compensation method for multi-axis linkage machining of die blank hole systems according to the present invention.

[0017] Figure 2 It is a schematic diagram of an error compensation method for multi-axis linkage machining of die blank hole systems according to the present invention.

[0018] Figure 3Another schematic diagram of the error compensation method for multi-axis linkage machining of die blank hole systems according to the present invention. Specific implementation manner

[0019] To further understand the content of the present invention, the present invention will be described in detail in combination with the accompanying drawings and embodiments. The following further describes the present application in detail with reference to the drawings and embodiments. It can be understood that the specific embodiments described herein are only used to explain the related invention, rather than limiting the invention. Additionally, it should be noted that for the sake of description, only the parts related to the invention are shown in the drawings.

[0020] As Figures 1-3 , a method for compensating errors in multi-axis linkage machining of die blank hole systems in this embodiment may specifically include:

[0021] Step S101, obtain the motion trajectory data of each axis during multi-axis linkage machining, record the current fluctuation information of the servo motor in real time, and establish a dynamically changing error model in combination with the data of the spindle temperature sensor and the tool wear monitoring device.

[0022] Obtain the motion trajectory data of each axis during multi-axis linkage machining, and use a high-speed data acquisition module to record the real-time position information of the moving axis; extract the current fluctuation data from the control system of the servo motor, and perform synchronous timestamp alignment in combination with the motion trajectory data; collect temperature data through the spindle temperature sensor, and perform fusion processing with the data of the tool wear monitoring device to generate temperature-wear correlation information; according to the motion trajectory data, the current fluctuation data, and the temperature-wear correlation information, establish an initial parameter set of the dynamic error model; if the deviation between the motion trajectory data and the current fluctuation data exceeds a preset threshold, adjust the weight coefficient of the dynamic error model; use a regression algorithm to optimize the dynamic error model to obtain a corrected parameter set of the error model; according to the corrected dynamic error model, generate real-time error compensation values for each axis during the machining process.

[0023] Specifically, multi-axis linkage machining involves precise control of the real-time positions of each axis. Through a high-speed data acquisition module, the position coordinates of each axis during machining can be obtained. For example, for the X, Y, and Z axes of a three-axis vertical machining center, when the acquisition frequency is set to 1000 Hz, the coordinate positions of each axis can be recorded every millisecond, forming a motion trajectory database. The current fluctuation data in the servo motor control system reflects the changes in the machining load. Taking milling machining as an example, when the tool cuts into the workpiece, the motor current will generate obvious fluctuations. By setting the sampling period to 0.1 ms, the real-time current values of the motors of each axis can be recorded. Aligning the current data with the position data through timestamps can analyze the load distribution during the machining process. The spindle temperature is closely related to tool wear. Temperature sensors are arranged at the spindle bearings to monitor the temperature rise changes in real time. The tool wear monitoring device uses a photoelectric sensor to measure the changes in the tool length and diameter. The fusion of temperature and wear data can establish a wear prediction model. When the temperature reaches 65 degrees Celsius, the tool wear value usually increases by 0.02 mm. The initial parameters of the dynamic error model include the thermal deformation coefficient, geometric error coefficient, and load deformation coefficient. Taking a three-axis machining center as an example, when the X axis moves to the 500 mm position, if the sudden change in the motor current exceeds 5 A and the deviation from the standard trajectory exceeds 0.01 mm, the load deformation coefficient needs to be adjusted. At the same time, combined with the temperature change, the thermal deformation compensation value is corrected. During the optimization of the error model by the regression algorithm, the least squares method is used to fit the deviation between the actual trajectory and the theoretical trajectory. The correction values of various parameters are obtained through iterative calculations. For example, the thermal deformation coefficient is adjusted from the initial value of 0.005 to 0.008. The finally generated compensation value can achieve real-time error compensation for each axis, ensuring the machining accuracy. Taking the lifting motion of the Z axis as an example, when it is detected that the bearing temperature rises by 10 degrees Celsius, a thermal deformation amount of 0.015 mm is calculated through the temperature compensation model. At the same time, considering that the tool wear reaches 0.03 mm, the wear compensation amount is superimposed on the original compensation value to achieve precision control. In circular interpolation motion, the coordinated cooperation between the transverse feed axis and the longitudinal feed axis is particularly important. When machining a circular arc with a radius of 100 mm, the speed ratio of the two axes needs to change with the position. Through the dynamic error model, the real-time compensation values of each point can be calculated to ensure that the circular arc profile accuracy is controlled within 0.01 mm.

[0024] Step S102: According to the error model, analyze the non-linear characteristics of the servo current fluctuation and the spindle temperature change, calculate the predicted value of the time point of the compensation action, and determine the trend of error accumulation.

[0025] Obtain the real-time current data and spindle temperature data of the servo system; perform filtering processing on the real-time current data and spindle temperature data to obtain a smoothed current value and a temperature value; input the smoothed current value and temperature value into a pre-established error model to calculate the non-linear relationship characteristic parameters between the smoothed current value and the temperature value; according to the non-linear relationship characteristic parameters, use the time series analysis method to predict the time point of the compensation action to obtain a sequence of time point prediction values; for the sequence of time point prediction values, determine the error accumulation trend curve through the cumulative error calculation method; if the error accumulation trend curve exceeds a preset threshold, trigger the compensation mechanism to generate a compensation action instruction; according to the compensation action instruction, adjust the servo system control parameters to achieve the dynamic balance of the real-time current data and the spindle temperature data; record the adjusted control parameters and the system operation state, and update the relationship characteristic parameters in the error model.

[0026] Specifically, the real-time current data acquisition of the servo system usually manifests as the current fluctuation value during the operation of the motor in industrial production. For example, during the machining process of a numerically controlled machine tool, when the spindle servo motor operates, there will be a fluctuation range of the reference current of ±5 amperes. The original current data collected has high-frequency noise, which can be processed by a Butterworth low-pass filter. The cut-off frequency is set to 100 Hz to make the current data smoother. The acquisition of the spindle temperature data is achieved through a temperature sensor. Under common working conditions, the spindle temperature varies between 20 and 60 degrees Celsius. The temperature signal also needs filtering treatment, and a moving average filtering method can be used. The sampling window is set to 10 data points, which can effectively eliminate the random fluctuations during the temperature acquisition process. The calculation of the non-linear relationship characteristic parameters involves the coupling relationship between current and temperature. For example, for every 10-degree Celsius increase in the spindle temperature, the current required by the motor may increase by 0.5 amperes. This relationship can be described by a neural network model. The input layer contains two nodes for current and temperature, and multiple neurons are set in the hidden layer for feature extraction. The time series analysis and prediction adopt an autoregressive moving average model. By analyzing the periodic characteristics in the historical data, the possible compensation time points in the future can be predicted. For example, during continuous machining, a compensation adjustment may be required every 2 hours of operation, and this rule can be obtained through time series analysis. The error accumulation trend curve reflects the degree of deterioration of the system performance over time and can be described by an exponential function. Assuming the initial error is 0.01 mm, when the accumulated error exceeds 0.05 mm, the compensation mechanism is triggered. The compensation action is achieved by adjusting the proportional gain coefficient of the servo system. For example, the original gain value of 100 is adjusted to 110. The update of the control parameters is reflected in multiple aspects of the servo system, including the parameter adjustment of the speed loop and the position loop. The parameter of the speed loop may need to increase the integral time constant, from 0.1 s to 0.15 s, to adapt to the change of the system characteristics caused by the temperature change. The parameter of the position loop may need to reduce the proportional gain to ensure the system stability. During the optimization process of the error model, the comprehensive influence of the current and temperature changes on the machining accuracy needs to be considered. For example, when the spindle temperature rises, thermal expansion will cause the offset of the tool center position. At this time, fine adjustment of the positions of each axis needs to be carried out through a compensation algorithm. The calculation of the compensation value is based on the mapping relationship between temperature and displacement, and a polynomial fitting model may need to be established. The realization of dynamic balance requires the system to be able to quickly respond to the changes in working conditions. When the accuracy decreases due to the increase in temperature, the machining accuracy is maintained by adjusting the servo parameters. For example, when the spindle drive power is 5 kWh, for every 5-degree Celsius increase in temperature, the position compensation value needs to increase by 0.02 mm, and the machining dimension stability is maintained through real-time adjustment.

[0027] Step S103, based on the error accumulation trend, determine the time range of the tool change window. If the error accumulation reaches the preset threshold, trigger the compensation action; otherwise, continue to monitor the error change.

[0028] Obtain the error value, calculate the cumulative value, and draw the trend line. Analyze the slope of the trend line, determine the window period range, and locate the threshold point. For the monitored point data, compare with the preset value and calculate the change difference. If the change exceeds the threshold point, trigger the compensation action and output the compensation amount parameter. If the change does not reach the threshold point, continue to monitor the error value and update the cumulative value data. Adjust the equipment operating state according to the compensation amount parameter and generate a new judgment line. Continuously track the new judgment line and repeat the above process to form a closed-loop monitoring system.

[0029] Specifically, the acquisition of the error value is completed through a real-time data acquisition system. For example, during the machining process of a machine tool, continuous sampling is carried out on the position deviation between the spindle and the workpiece. The error data collected within each sampling period is cumulatively calculated to form an error cumulative curve, which can intuitively reflect the error change trend during the machining process. For example, on a precision grinding machine, if the error data is sampled once per second and the machine runs continuously for one hour, the gradual change process of the workpiece surface quality can be clearly seen through the cumulative error curve. The slope analysis of the trend line is crucial for judging the change of equipment performance. Taking the temperature control system of an injection molding machine as an example, when the data detected by the mold temperature sensor shows that the slope of the temperature change trend line exceeds 0.5 degrees per minute, it indicates that the temperature control system may be abnormal. The determination of the window period range is usually based on historical data statistics. For example, for the cooling system, the temperature change in the first ten minutes can be used as an analysis window, and the data is updated in real-time and scrolled. In terms of the determination of the threshold point position, it can be set in combination with actual process requirements. Taking the machining of precision parts as an example, when the cumulative error reaches 80% of the workpiece tolerance zone, this point is set as the threshold point. During the comparison of the monitored point data with the preset value, the allowable range of process fluctuations needs to be considered. For example, in the manufacturing process of semiconductor wafers, for film thickness control, if the difference between the measured value and the target value exceeds plus or minus 3%, a compensation mechanism needs to be triggered. The implementation of the compensation action requires accurate compensation amount parameters. Taking a numerically controlled machine tool as an example, when the dimensional error caused by tool wear is detected, the system will automatically calculate the compensation amount according to the error size. If the tool wear causes the workpiece diameter to decrease by 0.02 mm, 0.01 mm will be added to the radial compensation amount. The adjustment of the equipment operating state needs to consider the dynamic response characteristics of the system. For example, in the pressure control of a press, the compensation action needs to be implemented step by step according to the material characteristics and processing requirements to avoid new fluctuations caused by over-compensation. After the new judgment line is generated, its change trend needs to be continuously tracked. Taking a forging equipment as an example, after the compensation action is executed, the system will generate a new pressure control curve, and by comparing the deviation between the real-time pressure curve and the standard curve, the stability of the compensation effect can be ensured. The formation of the closed-loop monitoring system enables the entire process to be continuously optimized. For example, in the production of injection molded parts, through multiple compensations and adjustments, the system can gradually find the optimal combination of process parameters and improve the product quality stability.

[0030] Step S104, within the tool change window, analyze the timing relationship to adjust the execution order of the compensation action and the machining action, ensuring their precise synchronization in time and avoiding introducing new errors.

[0031] Obtain the timing relationship data within the tool change window, extract the time characteristics of the compensation action and the machining action from the timing relationship data; analyze the time characteristics of the compensation action and the machining action, and determine whether there is a time conflict between the compensation action and the machining action; if there is a time conflict between the compensation action and the machining action, then rearrange the execution order of the compensation action and the machining action according to a preset adjustment strategy; calculate the time synchronization error value of the compensation action and the machining action according to the adjusted execution order; if the time synchronization error value exceeds a preset threshold, further adjust the execution order of the compensation action and the machining action through an optimization algorithm; use a machine learning algorithm to train the mapping relationship model between the time synchronization error and the execution order of the compensation action and the machining action; generate the optimal execution timing plan for the compensation action and the machining action according to the trained mapping relationship model.

[0032] Specifically, the timing relationship of the tool change window is reflected in the interaction process between equipment processing and compensation actions. Taking the continuous drilling of a CNC machine tool as an example, the equipment needs to perform tool change compensation after completing a certain number of drillings. Under normal circumstances, after every twenty drilling operations, the system will detect the tool wear status and perform compensation adjustment. The extraction of time characteristics mainly focuses on three dimensions: drilling time, tool change time, and compensation action time. For example, a typical drilling cycle is thirty seconds, the tool change time is sixty seconds, and the compensation adjustment time is twenty seconds. The judgment of time conflict focuses on analyzing the overlapping situation between processing and compensation actions. Taking the production of an injection molding machine as an example, the mold clamping time conflicts with the compensation action, which may lead to product quality problems. The preset adjustment strategy usually adopts the priority sorting method, arranging the compensation action to be executed during the mold clamping gap period. For example, when the mold is in the open state, the system uses this idle time to complete the compensation action. The calculation of the time synchronization error value needs to consider the smoothness of action connection. Taking a welding equipment as an example, there is a coordination relationship between the torch position compensation and the welding action. The system calculates the time difference between the compensation action and the welding action by recording the timestamps of the torch movement trajectory points. When the compensation action is advanced or delayed by more than one hundred milliseconds, an optimization adjustment mechanism will be triggered. Machine learning algorithms optimize production efficiency by establishing an association model between time characteristics and execution order. Taking the palletizing of an industrial robot as an example, the system collects a large amount of timing data of actions such as gripping, transporting, and placing. Through algorithms such as support vector machines, the corresponding relationship between the action execution order and the production beat is trained. The model can predict the time synchronization error under different execution orders, so as to select the optimal solution. The generation of the optimal execution timing plan needs to balance production efficiency and equipment stability. Taking a CNC machining center as an example, when parameters such as the spindle speed and feed rate change, the system will correspondingly adjust the execution timing of the compensation action. Through the analysis of historical data, it is found that arranging the compensation action during the transition period of processing parameter adjustment can reduce the impact on machining accuracy. At the same time, according to the material characteristics of the machined workpiece, the system will dynamically adjust the execution frequency of the compensation action to achieve the optimal balance between machining quality and efficiency.

[0033] Step S105, aiming at the complexity of the motion trajectory, a non-linear optimization algorithm is adopted to optimize the execution path of the compensation action and reduce the conflict with the processing action.

[0034] Obtain target motion trajectory data, and use a trajectory complexity analysis module to analyze the motion trajectory data to obtain trajectory complexity characteristic information; establish a non-linear trajectory optimization model according to the trajectory complexity characteristic information; obtain the execution path data of the compensation action, and input the execution path data into the non-linear trajectory optimization model; dynamically adjust the execution path of the compensation action according to the calculation result of the non-linear trajectory optimization model to obtain an optimized compensation execution path; determine whether there is a path conflict between the optimized compensation execution path and the target machining path; if there is a path conflict, return to the non-linear trajectory optimization model to recalculate the compensation execution path; until there is no path conflict, output the optimized compensation execution path as the final compensation execution path; generate a trajectory optimization compensation control instruction according to the final compensation execution path to complete the optimization compensation of the motion trajectory.

[0035] Specifically, the complexity characteristics of motion trajectory data include key indicators such as the curvature change of the motion path, the fluctuations of speed and acceleration. For example, during the machining process of a machine tool cutter, the complexity of the approaching trajectory of the cutter from the workpiece surface to the cutting point is reflected in the continuous change of the path curvature, and the feed speed and cutting depth need to be precisely controlled. When dealing with such trajectory optimization, non-linear optimization algorithms usually adopt methods such as gradient descent or genetic algorithms. Taking the machining of an arc contour as an example, the execution path of the compensation action needs to consider the dimensional deviation caused by tool wear. By establishing an optimization model that includes tool radius, feed speed, and cutting force, the optimal compensation path is calculated. When inputting the data of the execution path of the compensation action, the key lies in accurately obtaining the tool wear amount and machining error data. For example, when a lathe is turning the inner ring of a bearing, the tool wear amount obtained through an on-line measurement system is 0.03 mm, and based on this, the offset and adjustment angle of the compensation path are calculated. The optimization model will balance machining efficiency and compensation accuracy to generate the best path plan. The adjustment of the compensation path needs to consider the workpiece material characteristics and cutting parameters. Taking the machining of steel parts with carbide tools as an example, when the cutting speed increases, the tool wear intensifies, resulting in an increase in the compensation amount. At this time, it is necessary to recalculate the optimization path considering the influence of thermal deformation. Through the iterative optimization process, the compensation path is gradually adjusted until the accuracy requirements are met. In terms of path conflict detection, the interference problems between the tool and the workpiece, fixture, etc. are mainly considered. For example, in five-axis simultaneous machining, when the tool switches from one surface to another, there may be an interference risk, and it is necessary to detect in real time and re-plan the trajectory. Spatial geometric algorithms are used for collision detection to ensure the safety of the compensation path. The generation of the final execution path needs to comprehensively consider machining accuracy and efficiency. When machining turbine blades, the optimized path can improve the surface roughness from the original 1.6 microns to 0.8 microns, and the machining time only increases by 5%. Trajectory optimization not only improves machining quality, but also extends tool life and reduces production costs. The complete motion trajectory optimization and compensation process embodies the idea of adaptive control. By obtaining the machining state data in real time and dynamically adjusting the compensation strategy, the tool always maintains the optimal motion state. This method can effectively cope with various interference factors during the machining process and ensure the stability and consistency of machining quality.

[0036] Step S106: According to the optimized compensation path, adjust the control parameters of the servo motor in real time to ensure the smooth execution of the compensation action during the machining process.

[0037] Obtain the optimized compensation path, and extract the coordinate information of each path point from the compensation path; according to the coordinate information, determine the execution trajectory of the compensation action to obtain the action flow curve; use the least squares method to fit the action flow curve to obtain a smooth execution trajectory function; according to the smooth execution trajectory function, calculate the adjustment amount values at each moment; according to the adjustment amount values, combined with the dynamic response characteristics of the servo motor, determine the change range of the control parameters; during the machining process, collect the operation status data of the servo motor in real time; according to the operation status data, judge whether the actual value of the control parameter exceeds the change range; if the actual value exceeds the change range, recalculate the adjustment amount value and update the control parameter.

[0038] Specifically, in a multi-axis linkage machining system, the optimization of the compensation path plays a key role in machining accuracy. Taking a five-axis linkage machining center as an example, when obtaining the path point coordinates, the position data of five axes need to be recorded simultaneously. Suppose when machining a complex surface, a series of discrete points are collected on the tool center point trajectory, and each point contains the coordinates of the workbench rotation angle, swing angle and three linear axes. The determination of the action flow curve involves spatial kinematic analysis. Taking the example of a tool machining a blade, during the process of the tool moving from the blade root to the blade tip, the workbench needs to continuously adjust the angle to maintain the best cutting posture between the tool and the workpiece surface. This complex spatial motion can be described by establishing the mapping relationship between the workpiece coordinate system and the machine tool coordinate system. During the least squares fitting process, the continuity and smoothness of the curve need to be considered. For example, when machining an arc, the discrete points collected may fluctuate due to factors such as vibration. Through the least squares fitting, a smooth arc trajectory can be obtained, eliminating the influence of sampling noise. Specifically, the sampling points can be fitted into a cubic spline curve, which not only ensures the continuity of the position but also ensures the smooth transition of speed and acceleration. The calculation of the adjustment amount needs to consider the dynamic characteristics of the machine tool. Taking the response characteristics of the servo motor as an example, when the feed speed suddenly increases from 100 millimeters per minute to 500 millimeters per minute, the actual rotation speed of the motor will have a certain lag. By analyzing the acceleration characteristics of the motor, the compensation amount during the speed change process can be calculated in advance to avoid trajectory deviation caused by insufficient dynamic response. The change range of the control parameters is closely related to the machine tool performance. Taking the current limit of the servo motor as an example, during high-speed cutting, if the actual current exceeds 80% of the rated value, even if the planned trajectory is reasonable, the feed speed needs to be appropriately reduced to prevent the motor from overloading. At the same time, the acceleration of each axis also needs to be monitored to ensure it is within a safe range. Real-time data collection and processing are the keys to ensuring the compensation effect. For example, during the machining process, the speed, current and other status information of the servo driver are obtained through a high-speed data acquisition card, and the sampling frequency can reach 1000 times per second. When a sudden change in the current or a fluctuation in the speed of a certain axis is detected, the system can quickly calculate a new compensation value to ensure machining accuracy. This closed-loop control method can effectively cope with external disturbances such as cutting force changes.

[0039] In step S107, the error model is dynamically updated based on the data from the spindle temperature sensor and the tool wear monitoring device, and the time points and execution paths of the compensation actions are continuously optimized.

[0040] Obtain the real-time temperature value collected by the spindle temperature sensor and the wear value data collected by the tool wear monitoring device; input the temperature value and the wear value into a preset error model library; determine whether the temperature value and the wear value exceed a preset threshold; if so, trigger the calculation of the error value, determine the compensation amount and the execution path according to the calculated error value; use an optimizer to optimize the compensation amount and the execution path to obtain an optimized compensation action; dynamically adjust the time points and execution paths in the optimized compensation action to obtain an adjusted compensation action; and feedback the adjusted compensation action to the error model library for real-time update.

[0041] Specifically, the spindle temperature sensor is an important device for measuring and monitoring temperature changes during the machining process. Usually, a thermocouple sensor is adopted, which can accurately detect the temperature changes of key parts such as spindle bearings and motors. For example, in the spindle of a high-speed machining center, when the rotational speed reaches 12,000 revolutions per minute, the temperature of the bearing part is generally controlled within 60 degrees Celsius. The temperature sensor acquires real-time data at a sampling frequency of 0.5 seconds and transmits the data to the system controller. The tool wear monitoring device usually consists of two parts: an optical sensor and a force sensor. The optical sensor measures the change in the edge profile of the tool by laser projection to monitor the wear amount. The force sensor measures the change in the cutting force. When the tool wear intensifies, the cutting force will increase significantly. Taking a milling cutter as an example, when the flank wear value reaches 0.3 millimeters, a warning signal will be triggered. The error model library is a database containing various machining error models, and corresponding error compensation models are established for different working conditions. The temperature error model determines the key measurement points through thermal deformation analysis and establishes the corresponding relationship between temperature and thermal deformation amount. For example, for every one-degree rise in the spindle temperature rise, it may cause an axial thermal expansion of 0.01 millimeters. The wear error model is established based on the correlation analysis between tool wear and machining accuracy. Threshold setting needs to consider the machining process requirements and equipment characteristics. The temperature threshold is usually set at 90% of the upper limit of the normal operating temperature of the equipment. If the normal operating temperature is 70 degrees, the threshold is set at 63 degrees. The tool wear threshold is determined according to the machining accuracy requirements. During precision machining, the tool wear amount may be required not to exceed 0.1 millimeters. The error value calculation is based on the temperature and wear overlimit, and operations are carried out in combination with the error model. For example, after a certain machining center has been running continuously for two hours, the spindle temperature reaches 65 degrees, exceeding the threshold. At the same time, the detected tool wear value is 0.15 millimeters. The system automatically calculates the axial error value to be 0.03 millimeters and the radial error value to be 0.02 millimeters. The determination of the compensation amount and the execution path needs to consider the dynamic characteristics of the machine tool. For axial errors, a pre-compensation method can be adopted, and the compensation amount is added in advance in the machining program. Radial errors require real-time compensation, which is achieved by adjusting the position of the feed axis. The execution path of the compensation action needs to meet the requirement of smooth acceleration and deceleration transition. The optimizer adopts an adaptive algorithm and dynamically adjusts the compensation strategy according to the historical compensation effect. If it is found that a certain compensation action causes the surface roughness of the machining to deteriorate, the system will automatically reduce the compensation rate and increase the transition time. The optimized compensation scheme will be stored in the model library for reference in subsequent similar working conditions.

[0042] Step S108, during the machining process, the error accumulation situation is monitored in real time. If the error exceeds the preset range, the compensation action is re-triggered to ensure the machining accuracy.

[0043] Obtain real-time monitoring data during the machining process, and extract the error values from the real-time monitoring data; determine whether the error values exceed a preset threshold. If they exceed the preset threshold, accumulate the error values to obtain an accumulated error amount; calculate a compensation amount based on the accumulated error amount, and determine a compensation action point; at the compensation action point, adjust the position data of the machining point according to the compensation amount to obtain adjusted position data; obtain the machining data corresponding to the adjusted position data, and calculate the precision value of the machining data; determine whether the precision value is within a preset range. If it is not within the preset range, return to execute the compensation action; record the final machining data, and close the acquisition of the real-time monitoring data and the triggering of the compensation action.

[0044] Specifically, real-time monitoring data is crucial for improving machining accuracy during precision machining. The spindle vibration sensor can collect axial and radial vibration values in real time, the temperature sensor monitors the temperature rise change of the spindle, and the force sensor detects the change of machining force. For example, in milling machining, when obtaining the spindle temperature value and rising from the initial temperature of 25°C to 45°C, thermal deformation may cause the tool center to shift by 0.02 mm. For the judgment of error values, a reasonable preset threshold needs to be set. For example, when machining precision shaft parts, the radial runout tolerance is usually 0.01 mm. When the monitored error exceeds this value, the system starts to record the accumulated amount. This accumulated amount includes the superposition effects of various factors such as thermal deformation and tool wear. The calculation of the compensation amount needs to consider the trend of error accumulation. Assuming that in the continuous machining process, the error increases by 0.005 mm for each product machined. The system will predict that the error will exceed the threshold when machining the fifth product and initiate compensation in advance when machining the fourth product. The selection of the compensation action point is usually carried out when the feed direction changes or during the tool idle stroke to reduce the impact on the machining surface quality. The adjustment of the position data adopts a multi-level compensation strategy. When compensating for the first time, it is corrected by 80% of the accumulated error to avoid over-compensation. For example, when the axial runout reaches 0.03 mm, the first compensation is 0.024 mm, and then decide whether to perform secondary compensation after observing the effect. After compensation, the precision value is re-detected through an on-line measurement system. The determination of the precision value needs to comprehensively consider multiple indicators such as dimensional accuracy and geometric tolerance. Taking the machining of a go-no-go gauge as an example, its flatness requirement is 0.005 mm, and the roundness requirement is 0.008 mm. These indicators need to be re-verified after compensation. If a certain indicator does not meet the requirements, the system will fine-tune the compensation strategy, such as changing the compensation direction or adjusting the compensation amount. The recording of the final machining data includes comparison data before and after compensation, the compensation path, the precision improvement effect, etc. These data can be used for optimizing the machining parameters of subsequent similar workpieces to form a closed-loop feedback. The closing of the monitoring point and the triggering point needs to consider the production rhythm and is executed after completing the last qualified product and confirming the process stability. By establishing a complete data record, the machining error laws of different batches and different time periods can be analyzed, providing a basis for process improvement.

[0045] Only some preferred embodiments of the present invention are listed above, but the present invention is not limited thereto, and many improvements and transformations can be made. As long as the improvements and transformations are made on the basis of the basic principles of the present invention, they shall be regarded as falling within the protection scope of the present invention.

Claims

1. A method for error compensation in multi-axis linkage machining of a mold base hole system, characterized in that: The method comprises the following steps: S101, obtaining the motion trajectory data of each axis during multi-axis linkage machining, recording the current fluctuation information of the servo motor in real time, and establishing a dynamically changing error model by combining the data of the spindle temperature sensor and the tool wear monitoring device; S102, analyzing the nonlinear characteristics of servo current fluctuation and spindle temperature change according to the error model, calculating the predicted value of the time point of the compensation action, and determining the trend of error accumulation; S103, based on the error accumulation trend, determine the time range of the tool change window, if the error accumulation reaches a preset threshold, trigger a compensation action, otherwise continue to monitor the error change; S104, in the tool change window, adjusting the execution order of the compensation action and the processing action through timing relationship analysis to ensure that the two are precisely synchronized in time to avoid introducing new errors; S105. In view of the complexity of the motion trajectory, a nonlinear optimization algorithm is used to optimize the execution path of the compensation action and reduce the conflict with the processing action; S106, adjusting the control parameters of the servo motor in real time according to the optimized compensation path to ensure that the compensation action is smoothly executed during the processing; S107, dynamically updating the error model through data from the spindle temperature sensor and the tool wear monitoring device, and continuously optimizing the timing and execution path of the compensation action; S108. During the processing, the error accumulation is monitored in real time. If the error exceeds the preset range, the compensation action is re-triggered to ensure the processing accuracy.

2. According to claim 1, the error compensation method for multi-axis linkage machining of mold base hole system is characterized in that: The S101 includes: Acquire the motion trajectory data of each axis during multi-axis linkage machining, and use a high-speed data acquisition module to record the real-time position information of the motion axis; Extracting current fluctuation data from the control system of the servo motor and combining it with the motion trajectory data to perform synchronous time stamp alignment; The temperature data is collected by the spindle temperature sensor and integrated with the data of the tool wear monitoring device to generate temperature wear related information; Establishing an initial parameter set of a dynamic error model according to the motion trajectory data, the current fluctuation data and the temperature wear association information; If the deviation between the motion trajectory data and the current fluctuation data exceeds a preset threshold, adjusting the weight coefficient of the dynamic error model; The dynamic error model is optimized by using a regression algorithm to obtain a correction parameter set of the error model; According to the corrected dynamic error model, the real-time error compensation value of each axis during the machining process is generated.

3. According to claim 1, the error compensation method for multi-axis linkage machining of mold base hole system is characterized in that: S102 includes: Obtain real-time current data and spindle temperature data of the servo system; Filtering the real-time current data and the spindle temperature data to obtain smoothed current values ​​and temperature values; Inputting the smoothed current value and the temperature value into a pre-established error model to calculate a characteristic parameter of a nonlinear relationship between the smoothed current value and the temperature value; According to the nonlinear relationship characteristic parameters, a time series analysis method is used to predict the time point of the compensation action to obtain a time point prediction value sequence; For the time point prediction value sequence, determining an error cumulative trend curve by a cumulative error calculation method; If the error accumulation trend curve exceeds a preset threshold, a compensation mechanism is triggered to generate a compensation action instruction; According to the compensation action instruction, the servo system control parameters are adjusted to achieve a dynamic balance between the real-time current data and the spindle temperature data; The adjusted control parameters and system operating status are recorded, and the relationship characteristic parameters in the error model are updated.

4. The error compensation method for multi-axis linkage machining of a mold base hole system according to any one of claims 1 to 3, characterized in that: The S104 includes: Acquire the timing relationship data in the tool change window, and extract the time characteristics of the compensation action and the processing action in the timing relationship data; Analyzing the time characteristics of the compensation action and the processing action to determine whether there is a time conflict between the compensation action and the processing action; If there is a time conflict between the compensation action and the processing action, the execution order of the compensation action and the processing action is rearranged according to a preset adjustment strategy; Calculating the time synchronization error value between the compensation action and the processing action according to the adjusted execution sequence; If the time synchronization error value exceeds a preset threshold, the execution order of the compensation action and the processing action is further adjusted through an optimization algorithm; Using a machine learning algorithm to train a mapping relationship model between the time synchronization error and the execution sequence of the compensation action and the processing action; According to the trained mapping relationship model, an optimal execution timing scheme of the compensation action and the processing action is generated.

5. The error compensation method for multi-axis linkage machining of a mold base hole system according to any one of claims 1 to 3, characterized in that: The S105 includes: Acquire target motion trajectory data, and use a trajectory complexity analysis module to analyze the motion trajectory data to obtain trajectory complexity feature information; Establishing a nonlinear trajectory optimization model according to the trajectory complexity characteristic information; Acquiring execution path data of the compensation action, and inputting the execution path data into the nonlinear trajectory optimization model; According to the calculation result of the nonlinear trajectory optimization model, the execution path of the compensation action is dynamically adjusted to obtain an optimized compensation execution path; Determining whether there is a path conflict between the optimized compensation execution path and the target processing path; If there is a path conflict, return to the nonlinear trajectory optimization model and recalculate the compensation execution path; When there is no path conflict, output the optimized compensation execution path as the final compensation execution path; According to the final compensation execution path, a trajectory optimization compensation control instruction is generated to complete the optimization compensation of the motion trajectory.

6. The error compensation method for multi-axis linkage machining of a mold base hole system according to any one of claims 1 to 3, characterized in that: The S106 includes: Obtaining an optimized compensation path, and extracting coordinate information of each path point from the compensation path; According to the coordinate information, determining the execution trajectory of the compensation action and obtaining the action flow curve; The action flow curve is fitted using the least square method to obtain a smooth execution trajectory function; Calculating the adjustment value at each moment according to the smoothed execution trajectory function; According to the adjustment value and in combination with the dynamic response characteristics of the servo motor, the variation range of the control parameter is determined; During the processing, the operating status data of the servo motor is collected in real time; According to the operating status data, determining whether the actual value of the control parameter exceeds the variation range; If the actual value exceeds the variation range, the adjustment value is recalculated and the control parameter is updated.

7. The error compensation method for multi-axis linkage machining of a mold base hole system according to any one of claims 1 to 3, characterized in that: The S107 includes: Acquire the real-time temperature value collected by the spindle temperature sensor and the wear value data collected by the tool wear monitoring device; Inputting the temperature value and the wear value into a preset error model library; Determining whether the temperature value and the wear value exceed a preset threshold; If yes, the error value calculation is triggered, and the compensation amount and execution path are determined according to the calculated error value; Using an optimizer to optimize the compensation amount and the execution path to obtain an optimized compensation action; Dynamically adjusting the time point and execution path in the optimized compensation action to obtain an adjusted compensation action; The adjusted compensation action is fed back to the error model library for real-time updating.

8. The error compensation method for multi-axis linkage machining of a mold base hole system according to any one of claims 1 to 3, characterized in that: The S108 includes: Acquire real-time monitoring data during the processing, and extract error values ​​from the real-time monitoring data; Determine whether the error value exceeds a preset threshold value, and if so, accumulate the error value to obtain a cumulative error amount; Calculate the compensation amount according to the accumulated error amount and determine the compensation action point; At the compensation action point, adjusting the position data of the processing point according to the compensation amount to obtain adjusted position data; Acquire processing data corresponding to the adjusted position data, and calculate the accuracy value of the processing data; Determine whether the accuracy value is within a preset range, and if not, return to perform compensation action; The final processing data is recorded, and the collection of the real-time monitoring data and the triggering of the compensation action are turned off.

Images