Motion control method for servo press

Abstract

The invention discloses a motion control method for a servo press, which relates to the technical field of industrial automation and numerical control forming equipment control, and comprises the following steps: S1, acquiring real-time data through a peripheral equipment conveyor belt and a state sensor of a manipulator, and acquiring a synchronous signal and a response delay value; determining a current production line fluctuation condition to obtain an initial adjustment parameter of the intermittent period; s2, processing signal fluctuation by adopting a fuzzy control algorithm according to the obtained initial adjustment parameters, prolonging the dwell time if response delay exceeds a preset threshold value, otherwise, keeping original division, and obtaining an accurate action time and dwell time division scheme; according to the motion control method for the servo press, the production takt and the product yield are improved while the equipment safety is guaranteed, and the application advantages of the servo press in a high-takt intermittent automatic production line are fully exerted.

Description

Technical Field

[0001] This invention relates to the field of industrial automation and CNC forming equipment control technology, specifically to a motion control method for a servo press. Background Technology

[0002] Servo presses are widely used in automated manufacturing fields such as multi-station progressive die stamping due to their programmable slider motion, adjustable motion curves, and fast response speed. In these applications, servo presses typically operate intermittently, requiring coordination with peripheral equipment such as conveyor belts, feeding mechanisms, and loading / unloading robots in each stamping cycle to achieve precise material feeding according to step distances, stamping formation, and workpiece removal. Therefore, the motion control accuracy of the servo press within the intermittent cycle directly affects the operating cycle time of the entire production line, equipment safety, and product forming quality.

[0003] Existing motion control methods for servo presses mostly employ control strategies based on fixed cycles or single synchronous signal triggers. This involves driving the slider to complete the downward stamping and return movements according to a predetermined motion curve under preset time parameters. However, in actual automated production processes, the operating status of peripheral equipment often fluctuates, such as delays in the feeding mechanism's response or changes in the timing of robotic arm movements. When the production line cycle time changes, fixed cycles or single trigger methods struggle to reflect the real-time status of peripheral equipment in a timely manner, easily leading to asynchrony between the servo press's stamping movements and the feeding step distance or material handling actions. Especially in multi-station progressive die stamping scenarios, the servo press needs to complete a full stamping movement for each material feeding step. If the start time or duration of the stamping cycle is not precisely controlled, stamping timing deviations can easily occur, leading to material accumulation, abnormal die stress, part dimensional deviations, and even the risk of interference between the robotic arm and the slider. To avoid these problems, existing production lines often reduce overall operating speed to improve system stability, thus limiting the high cycle time and high efficiency advantages of servo presses. Furthermore, existing motion control methods face the technical challenge of balancing motion accuracy and stability during intermittent operation. On the one hand, the predetermined slide motion curve needs to be precisely executed during the stamping phase to meet the speed, position, and pressure requirements of the forming process; on the other hand, the slide position needs to be reliably maintained during the rest phase, or the slide needs to complete its return stroke and remain stably in the standby position within a very short time. Simultaneously, the start time, action time, and rest time of the entire intermittent cycle need to be dynamically adjusted based on the real-time operating status of peripheral equipment, rather than relying solely on fixed parameter settings. However, existing technologies generally lack a motion control method that can comprehensively utilize peripheral equipment status information to dynamically divide the stamping cycle and, based on this, perform closed-loop regulation of the servo press slide motion. This makes it difficult for servo presses to simultaneously achieve high-precision motion control, high synchronous reliability, and stable intermittent operation in complex automated production environments, leaving room for further improvement. Summary of the Invention

[0004] The purpose of this invention is to provide a motion control method for a servo press, thereby solving the problems existing in the prior art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a motion control method for a servo press, comprising: S1, acquiring real-time data through the status sensors of peripheral equipment conveyor belts and robotic arms, obtaining synchronization signals and response delay values, and determining the current production line fluctuation to obtain initial adjustment parameters for the intermittent cycle; S2, processing signal fluctuations using a fuzzy control algorithm based on the obtained initial adjustment parameters, determining if the response delay exceeds a preset threshold and extending the pause time, otherwise maintaining the original division, to obtain a precise action time and pause time division scheme; S3, extracting the cycle start time from the obtained division scheme, comparing it with the servo press slider position sensor data, determining the starting point of the motion curve to obtain a high-precision cycle control sequence; S4, performing closed-loop adjustment on the obtained control sequence using a PID control algorithm, comparing the actual slider position with the predetermined curve deviation, determining if the deviation is greater than a threshold and adjusting the servo motor speed, otherwise maintaining the current speed, to obtain an optimized stamping motion curve execution path; S5, monitoring real-time status changes of peripheral equipment based on the optimized execution path, obtaining updated synchronization signals, determining the slider's position holding requirements during the pause time to obtain a rapid instruction sequence for returning to the standby position.

[0006] Preferably, step S1 includes acquiring the pulse frequency of the peripheral equipment conveyor belt and the load displacement of the robot arm, converting the pulse frequency and load displacement into multi-dimensional real-time data; performing clock alignment processing on the multi-dimensional real-time data to extract a synchronization signal containing a trigger timestamp; calculating the response delay of the actuator based on the synchronization signal; performing a difference calculation between the response delay and a preset reference time to determine the production line fluctuation reflecting operational stability; and obtaining the initial adjustment parameters of the intermittent period by matching the production line fluctuation with a pre-stored logic mapping table.

[0007] Preferably, step S2 includes acquiring initial adjustment parameters and real-time acquired signal fluctuations, extracting the amplitude change rate of the signal fluctuations; inputting the amplitude change rate into a fuzzy control algorithm model to calculate a membership function to obtain a fuzzy control output value; comparing the fuzzy control output value with a preset time series to determine a response delay value; if the response delay value exceeds a preset threshold, extending the pause time on the original time axis to obtain a corrected pause time; and using the corrected pause time to synchronize and align the action time on the original time axis to obtain a precise action time and pause time division scheme.

[0008] Preferably, step S3 includes acquiring a high-frequency sampling data stream containing a division scheme of the estimated time interval and the slider position sensor; performing time-domain segmentation on the high-frequency sampling data stream according to the division scheme to obtain a set of position data segments to be analyzed covering the complete stamping cycle; generating a smooth gradient feature sequence for the set of position data segments to be analyzed; if the smooth gradient feature sequence exceeds a preset static state fluctuation threshold, determining the coarse motion start time and extracting the initial segment micro dataset; performing polynomial fitting on the initial segment micro dataset to construct a local analytical model; determining the corrected high-precision motion curve start point by solving the curvature abrupt change point of the local analytical model; realigning the original position data segments based on the high-precision motion curve start point to generate standardized single-cycle motion trajectory data; and constructing a standard reference model based on the standardized single-cycle motion trajectory data to obtain a high-precision cycle control sequence.

[0009] Preferably, step S4 includes acquiring real-time displacement data collected by a displacement sensor and converting the real-time displacement data into a pulse feedback signal; calculating the difference between the pulse feedback signal and the theoretical value of a predetermined curve to obtain a deviation value; performing closed-loop adjustment calculation on the deviation value to determine a control voltage for correcting the error; converting the control voltage into a drive frequency, and using the drive frequency to adjust the servo motor to approximate the predetermined curve to obtain an optimized stamping motion curve execution path.

[0010] Preferably, step S5 includes collecting optimized execution path data and extracting state change features reflecting fluctuations in the operation of peripheral devices; using the state change features to obtain an updated synchronization signal, and defining a slider resting window based on the synchronization signal; calculating a holding torque value for the position drift risk within the slider resting window; constructing a return trajectory planning model using the holding torque value as the initial constraint, and solving the return trajectory planning model to generate a fast instruction sequence to return to the standby position.

[0011] Preferably, the method further includes S6: selecting key time nodes from the obtained instruction sequence, verifying the step distance correspondence of the multi-station mold, determining if the step distance and stroke frequency do not match, recalculating the start time, otherwise confirming synchronization, and obtaining a stable control scheme for the entire intermittent cycle. Specifically, this includes acquiring the real-time instruction sequence of the multi-station mold control system, parsing the pulse frequency and displacement encoding data contained in the instruction sequence, extracting the key time nodes characterizing the switching of mold station actions, mapping the key time nodes to a preset periodic time coordinate system, and calculating the synchronization deviation value of the mold station during the movement process by combining the stamping frequency and feeding step distance parameters.

[0012] Preferably, step S6 further includes, if the synchronization deviation value exceeds a preset range, deriving a correction trigger time based on the synchronization deviation value, and generating a correction instruction containing new timing parameters based on the correction trigger time; executing the correction instruction to reconstruct the control logic of the intermittent cycle, locking the step size and stroke synchronization state, and obtaining a stable control scheme for the entire intermittent cycle.

[0013] Preferably, the process further includes S7: driving the servo press to execute using the obtained stable control scheme, collecting equipment feedback data during the execution process, and determining the switching accuracy between actions and pauses to obtain the final production line synchronization optimization result. Specifically, this includes obtaining a preset stable control command to drive the servo press to run and obtain the execution trajectory; determining equipment feedback data based on the real-time pulse response generated by the execution trajectory; extracting the critical time points between actions and pauses from the equipment feedback data, and judging the timing deviation based on the critical time points.

[0014] Preferably, step S7 further includes determining the switching accuracy between action and pause based on load fluctuations if the timing deviation exceeds a preset threshold; mapping the switching accuracy to a collaborative instruction to obtain an operating cycle; and obtaining the final production line synchronization optimization result through multi-camera matching of the operating cycle.

[0015] As can be seen from the above technical solution, the present invention has the following beneficial effects: This motion control method for servo presses acquires real-time status data from peripheral devices such as conveyor belts and robotic arms, obtaining synchronization signals and response delay information. Based on this, it dynamically determines and adjusts the start time, action time, and pause time of intermittent stamping cycles, achieving high-precision synchronous control between the servo press's stamping action and peripheral devices. By introducing a fuzzy control algorithm to process signal fluctuations and rationally dividing the action and pause times, it effectively improves the adaptability and stability of cycle control. Furthermore, by combining slider position sensor data to determine the starting point of the motion curve and using a PID control algorithm to perform closed-loop adjustment of the slider's motion, the actual motion trajectory can accurately follow the predetermined stamping curve, thereby ensuring the forming accuracy of the stamping process. Simultaneously, by verifying the relationship between the multi-station die step distance and the number of strokes and continuously correcting the synchronization state, it avoids problems such as asynchronous feeding, mechanical interference, and abnormal die stress. While ensuring equipment safety, it improves production cycle time and product yield, fully leveraging the advantages of servo presses in high-cycle, intermittent automated production lines. Attached Figure Description

[0016] Figure 1 This is a flowchart of the motion control method for a servo press according to the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments 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, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] like Figure 1 As shown, the present invention provides a technical solution: a motion control method for a servo press, comprising: S1. Real-time data is collected through the status sensors of the peripheral equipment conveyor belt and robot arm to obtain synchronization signals and response delay values, and the current fluctuation of the production line is determined to obtain the initial adjustment parameters of the intermittent cycle. S2. Based on the obtained initial adjustment parameters, a fuzzy control algorithm is used to process signal fluctuations. If the response delay exceeds the preset threshold, the pause time is extended; otherwise, the original division is maintained, thus obtaining an accurate action time and pause time division scheme. S3. Extract the cycle start time from the obtained division scheme, compare it with the data of the servo press slider position sensor, determine the starting point of the motion curve to obtain a high-precision cycle control sequence. S4. The PID control algorithm is used to perform closed-loop regulation on the obtained control sequence. By comparing the deviation between the actual slider position and the predetermined curve, if the deviation is greater than the threshold, the servo motor speed is adjusted; otherwise, the current speed is maintained to obtain the optimized stamping motion curve execution path. S5. Based on the optimized execution path, monitor the real-time status changes of peripheral devices, obtain the updated synchronization signal, and determine the position holding requirements of the slider during the pause time to obtain a fast instruction sequence to return to the standby position. S6. Select key time nodes from the obtained instruction sequence, verify the step distance correspondence of the multi-station mold, and determine if the step distance and the number of strokes do not match. If so, recalculate the start time; otherwise, confirm synchronization and obtain a stable control scheme for the entire intermittent cycle. S7. Drive the servo press to execute using the obtained stable control scheme, collect equipment feedback data during the execution process, and determine the switching accuracy between action and pause to obtain the final production line synchronization optimization result.

[0019] This motion control method is based on the coupling principle of multi-equipment collaboration on the production line and high-precision cycle control of the servo press. By setting status sensors at peripheral equipment on the production line, the system collects real-time data on the conveyor belt's operating status, the robot's pick-and-place actions, and corresponding synchronization signals, calculating the response delay value relative to the servo press's actions. This quantifies the fluctuations of the production line under different load and cycle time conditions. Initial adjustment parameters generated based on these fluctuations provide fundamental data support for subsequent intermittent cycle division. Building upon this, a fuzzy control algorithm is introduced to process the response delay and its changing trend. Fuzzy rules map the relationship between "delay level" and "pause time adjustment amount," enabling the system to adaptively adjust the ratio of action time to pause time when facing nonlinear fluctuations and uncertain disturbances, thus obtaining a more accurate cycle division scheme. Subsequently, by extracting the cycle start time from the cycle division scheme and aligning it with the real-time feedback data from the servo press's slider position sensor, the starting point of the motion curve is determined. This ensures that each control cycle establishes a high-precision cycle control sequence based on the slider's actual physical position, avoiding accumulated errors. During the execution phase, a PID control algorithm is used to perform closed-loop adjustment of the control sequence. By continuously comparing the deviation between the actual position of the slider and the predetermined motion curve, the servo motor speed is dynamically adjusted to ensure that the slider's motion trajectory stably conforms to the target curve, thereby forming an optimized stamping motion curve execution path. Furthermore, when the stamping action is completed and enters the pause phase, the real-time status changes of peripheral equipment are continuously monitored, and synchronization signals are updated. Based on the slider's holding requirements during the pause time, a rapid command sequence for returning to the standby position is generated, ensuring that the slider quickly returns to the standby position while meeting mold safety and process requirements. Finally, by verifying the correspondence between key time nodes in the rapid command sequence and the step distance of the multi-station mold, it is ensured that the number of strokes and the step distance are strictly matched. If a mismatch is detected, the cycle start time is readjusted, thus forming a stable and reliable intermittent cycle control scheme. The accuracy of action and pause switching is evaluated through execution feedback data, achieving optimization of the overall production line synchronization performance.

[0020] By introducing peripheral equipment status sensors and a synchronization signal analysis mechanism, real-time perception and quantification of production line fluctuations are achieved, freeing the motion control of the servo press from dependence on fixed cycle times and significantly improving the system's adaptability to changes in operating conditions. Fuzzy control algorithms are used to adaptively adjust the intermittent cycle, effectively reducing the risk of cycle mismatch caused by response delay fluctuations. Precise alignment of slider position sensor data with the cycle start time improves cycle control accuracy and reduces cumulative errors. Combined with PID closed-loop regulation, dynamic optimization of the stamping motion curve is achieved, improving the smoothness and repeatability of slider motion. A multi-station die step distance verification mechanism ensures consistency between the number of stampings and the die step distance, reducing die damage and product defect rates. Overall, this improves the synchronization stability, processing accuracy, and operational reliability of the production line.

[0021] S1 includes acquiring the pulse frequency of the peripheral equipment conveyor belt and the load displacement of the robot arm, converting the pulse frequency and load displacement into multi-dimensional real-time data; performing clock alignment processing on the multi-dimensional real-time data to extract a synchronization signal containing a trigger timestamp; calculating the response delay of the actuator based on the synchronization signal; performing a difference calculation between the response delay and a preset reference time to determine the production line fluctuation reflecting operational stability; and matching the production line fluctuation with a pre-stored logic mapping table to obtain the initial adjustment parameters of the intermittent cycle.

[0022] First, during production line operation, pulse frequency data from the peripheral equipment conveyor belt and load displacement data from the robotic arm during workpiece handling are continuously acquired. The conveyor belt pulse frequency is obtained in real-time from the pulse signals on the conveyor belt drive shaft, representing the conveyor belt's operating cycle time. The robotic arm's load displacement is obtained from the displacement changes during pick-up and unload actions, reflecting the robotic arm's motion load state under current operating conditions. Subsequently, the pulse frequency and load displacement data are synchronously collected according to a unified sampling period and combined to form multi-dimensional real-time data containing conveyor cycle time information and load motion information. The sampling period is determined based on the production line control... The fixed control cycle of the control system is determined to ensure that each set of data corresponds to the same actual operating time period. After obtaining multi-dimensional real-time data, clock alignment processing is performed on the data. Specifically, using the main clock of the control system as a reference, a unified timestamp is added to the conveyor belt pulse data and the robot displacement data. By aligning data points with the same timestamp, time deviations caused by different sampling start times between different devices are eliminated. After clock alignment is completed, the time nodes that trigger changes in conveyor belt movement or robot load are extracted from the aligned multi-dimensional real-time data, and these time nodes are used as the trigger timestamps for synchronization signals. Then, the trigger timestamps of the synchronization signals are matched with the actual operation completed by the actuator. The timing of the actions is compared, and the time difference between the two is calculated to obtain the response delay of the actuator under the current production cycle. This response delay characterizes the actuator's actual ability to follow the action commands of peripheral equipment. Subsequently, the response delay is compared with a pre-set reference time, which is obtained by averaging the actuator response times of multiple consecutive production cycles under stable production line operation. This reference time represents the standard response level of the production line under ideal stable conditions. The difference between the response delay and the reference time yields the production line fluctuation value, which reflects the current stability of the production line operation. When the production line fluctuation value increases... This indicates that the current production cycle has deviated from a stable state. When the fluctuation value of the production line decreases, it indicates that the current production cycle is approaching a stable state. Finally, the fluctuation value of the production line is used as an index parameter and matched with a pre-stored logical mapping table. This logical mapping table is formed during the equipment debugging phase by manually calibrating and storing the reasonable intermittent cycle adjustment amounts corresponding to different production line fluctuation states. Each fluctuation range corresponds to a unique initial adjustment parameter for the intermittent cycle. The initial adjustment parameter for the intermittent cycle that matches the current production line fluctuation is directly output by looking up the table, thus providing a clear and executable initial basis for further precise division of action time and rest time in subsequent steps.

[0023] S2 includes acquiring initial adjustment parameters and real-time collected signal fluctuations, extracting the amplitude change rate of the signal fluctuations; inputting the amplitude change rate into a fuzzy control algorithm model to calculate the membership function to obtain a fuzzy control output value; comparing the fuzzy control output value with a preset time series to determine the response delay value; if the response delay value exceeds a preset threshold, extending the pause time on the original time axis to obtain a corrected pause time; using the corrected pause time to synchronize and align the action time on the original time axis to obtain a precise action time and pause time division scheme.

[0024] After completing step S1 and obtaining the initial adjustment parameters for the intermittent cycle, the initial adjustment parameters and real-time signal fluctuation data collected during the current production cycle are first acquired synchronously. This signal fluctuation data originates from the changes in the synchronization signal of peripheral equipment over a continuous time period, reflecting the instantaneous instability of the production cycle during actual operation. Subsequently, continuous-time differential processing is performed on the signal fluctuation data. By comparing the changes in signal fluctuation amplitude at adjacent sampling times, the amplitude change rate of the signal fluctuation within a unit sampling period is extracted. This amplitude change rate is used to quantitatively describe the speed at which the signal changes from a stable state to a fluctuating state, and its magnitude is directly calculated from the real-time acquired data. The amplitude change rate is obtained through calculation. After obtaining the amplitude change rate, it is input as an input to a pre-constructed fuzzy control algorithm model. This fuzzy control algorithm model is established during the system debugging phase based on actual production line operating experience. It contains membership functions set for different amplitude change rate ranges. By calculating the membership of the input amplitude change rate, the corresponding fuzzy control output value is obtained. This fuzzy control output value reflects the system's need for time adjustment under the current signal fluctuation state. Next, the fuzzy control output value is compared with a pre-set time series, which is a standard set during the system initialization phase based on the equipment's rated cycle time and historical operating data. A time reference sequence is used to determine the response delay value corresponding to the current production state by mapping the fuzzy control output value to the corresponding time position in the time sequence. This response delay value represents the time offset of the actual executed action relative to the ideal cycle time under the current fluctuation conditions. Subsequently, the response delay value is compared with a preset threshold. This threshold is determined during equipment debugging and trial operation by statistically analyzing the maximum allowable time offset over multiple stable operating cycles and taking its upper limit. It is used to distinguish between acceptable and unacceptable delays in the system. When the response delay value exceeds the preset threshold, it is determined that the current production cycle time has exceeded the stable control range. At this point, in the original... Based on the time axis, the pause time is extended proportionally according to the magnitude of the response delay value, thus obtaining the corrected pause time. When the response delay value does not exceed a preset threshold, the original pause time remains unchanged. Finally, the corrected pause time is used to synchronize the action time on the original time axis. Specifically, while keeping the total cycle of a single stamping action unchanged, the distribution ratio of action time and pause time is redistributed, so that the start time of the action is re-aligned with the synchronization signal of the peripheral equipment. This results in a precise, stable, and executable action time and pause time division scheme, providing a clear time reference for subsequent slider motion control.

[0025] S3 includes acquiring a high-frequency sampling data stream containing a division scheme of the estimated time interval and the slider position sensor; performing time-domain segmentation on the high-frequency sampling data stream according to the division scheme to obtain a set of position data segments to be analyzed covering the complete stamping cycle; generating a smooth gradient feature sequence for the set of position data segments to be analyzed; if the smooth gradient feature sequence exceeds a preset static state fluctuation threshold, determining the approximate motion start time and extracting the initial segment micro dataset; performing polynomial fitting on the initial segment micro dataset to construct a local analytical model; determining the corrected high-precision motion curve start point by solving the curvature abrupt change point of the local analytical model; realigning the original position data segments based on the high-precision motion curve start point to generate standardized single-cycle motion trajectory data; and constructing a standard reference model based on the standardized single-cycle motion trajectory data to obtain a high-precision cycle control sequence. After completing step S2 and obtaining a precise action time and pause time division scheme, the process first acquires the division scheme containing the estimated time intervals and the high-frequency sampling data stream output by the servo press slider position sensor. This high-frequency sampling data stream is slider position information continuously acquired at a fixed high sampling frequency within a complete stamping cycle, used to fully reflect the slider's motion state from rest, start-up, acceleration, stamping, to return. Subsequently, based on the action time and pause time intervals given in the division scheme, the high-frequency sampling data stream is time-domain segmented in chronological order, ensuring that each data segment strictly corresponds to a specific continuous time interval within the stamping cycle, thereby obtaining multiple positions to be analyzed covering the complete stamping cycle. A set of data segments is set up. After obtaining the set of data segments for the location to be analyzed, the position change trend between adjacent sampling points is calculated sequentially for each data segment, and a smooth gradient feature sequence reflecting the continuity of position change is generated based on this. The smoothing process is achieved by averaging the change trends of multiple consecutive sampling points to eliminate the influence of high-frequency sampling noise on the judgment result. Subsequently, the smooth gradient feature sequence is compared with a preset static state fluctuation threshold. The static state fluctuation threshold is determined during the equipment debugging phase by statistically analyzing multiple segments of high-frequency sampling data when the slider is in a completely static state, and taking the upper limit of its maximum natural fluctuation range. When the smooth gradient feature sequence is detected to exceed the threshold for the first time, the threshold is set. When the static fluctuation threshold is reached, it is determined that the slider has transitioned from a static state to a true motion state, and the corresponding time point is identified as the approximate start time of motion. Simultaneously, data of predetermined time lengths are extracted forward and backward from this time point to form a micro-dataset containing initial motion details. Next, polynomial fitting is performed on this micro-dataset. By continuously fitting the slider's position change trend over time in this dataset, a local analytical model that accurately reflects the initial motion pattern is constructed. This local analytical model further analyzes the curvature changes in the position change trend, and by identifying locations where the curvature undergoes significant abrupt changes, the true physical starting position of the slider's transition from static to continuous motion is determined, thus obtaining the corrected... The starting point of the high-precision motion curve is determined. After determining the starting point, the time and position corresponding to the starting point are used as a unified reference. The original position data segments are realigned as a whole, so that the motion trajectory in each stamping cycle starts from the same physical starting point, thereby generating standardized single-cycle motion trajectory data. Finally, based on the standardized single-cycle motion trajectory data, the trajectories of multiple cycles are integrated in a consistent manner to construct a standard reference model that reflects the standard motion characteristics of the slider of the servo press under stable operation. This standard reference model is used as the target trajectory output for subsequent control, thereby obtaining a control sequence for high-precision cycle control, providing an accurate, unified and repeatable motion reference for subsequent closed-loop control steps.

[0026] S4 includes acquiring real-time displacement data collected by a displacement sensor and converting the real-time displacement data into a pulse feedback signal; calculating the difference between the pulse feedback signal and the theoretical value of a predetermined curve to obtain a deviation value; performing closed-loop adjustment calculation on the deviation value to determine a control voltage for correcting the error; converting the control voltage into a drive frequency, and using the drive frequency to adjust the servo motor to approximate the predetermined curve to obtain an optimized stamping motion curve execution path.

[0027] After completing step S3 and obtaining the high-precision periodic control sequence, the actual movement of the slider throughout the entire stamping cycle is continuously sampled by a displacement sensor installed on the servo press slider to obtain real-time displacement data reflecting the real-time position change of the slider. The sampling frequency of the displacement sensor is consistent with the control cycle of the servo system to ensure the timeliness and completeness of the feedback data. Subsequently, the real-time displacement data is converted into a corresponding pulse feedback signal according to the interface requirements of the servo drive system. The pulse feedback signal, in the form of pulse number or pulse density, equivalently represents the displacement change of the slider per unit time and is used for comparison with the target trajectory within the control system at a unified scale. After obtaining the pulse feedback signal, it is compared point-by-point with the theoretical value of a predetermined curve at the corresponding time node. The difference between the two is calculated to obtain the deviation value. The theoretical value of the predetermined curve originates from the high-precision periodic control sequence constructed in step S3 and is used to characterize the target position that the slider should reach at each moment under ideal control conditions. The deviation value is used for quantitative feedback. The deviation of the actual movement trajectory of the slider from the target trajectory is measured. Subsequently, the deviation value is input into the closed-loop adjustment calculation process for continuous correction. The closed-loop adjustment calculation comprehensively calculates a control voltage to offset the current deviation based on the magnitude and trend of the deviation. The magnitude of this control voltage is directly determined by the deviation amplitude, and its direction indicates the adjustment direction for accelerating or decelerating the slider movement, thus ensuring a one-to-one correspondence between the control output and the error state. After obtaining the control voltage, it is converted into a drive frequency signal that matches the input requirements of the servo driver. This drive frequency directly controls the speed change of the servo motor. By increasing or decreasing the drive frequency, the output speed of the servo motor changes accordingly, thereby driving the slider's movement trajectory to approach the predetermined curve. Through the continuous cyclic execution of the above real-time feedback, deviation calculation, and drive adjustment, the slider is dynamically corrected around the predetermined curve throughout the entire stamping process, ultimately forming a stable, smooth, and highly consistent stamping motion curve execution path, providing reliable assurance for stamping accuracy and operational stability.

[0028] S5 includes collecting optimized execution path data and extracting state change features reflecting fluctuations in the operation of peripheral devices; using the state change features to obtain updated synchronization signals and defining a slider rest window based on the synchronization signals; calculating the holding torque value for the position drift risk within the slider rest window; constructing a return trajectory planning model with the holding torque value as the initial constraint, and solving the return trajectory planning model to generate a fast instruction sequence to return to the standby position.

[0029] After completing step S4 and obtaining the optimized stamping motion curve execution path, the optimized execution path data is first collected. This data includes the actual position change trend of the slider after stamping and its corresponding time information. Through continuous analysis of this execution path data, state change features reflecting changes in the operational stability of peripheral equipment are extracted. Specifically, these state change features manifest as the rhythm offset and response delay changes of the peripheral equipment's synchronization signal before and after stamping. Subsequently, the synchronization state of the peripheral equipment is re-evaluated using these state change features to obtain an updated synchronization signal. This updated synchronization signal accurately reflects the coordination state between the peripheral equipment and the servo press under the current production rhythm conditions. After obtaining the updated synchronization signal, based on its distribution on the time axis, the time interval during which the slider can remain stationary after completing the stamping action is clearly defined, thus determining the slider's resting window. This resting window represents the position range during which the slider is allowed to remain stationary without affecting the production rhythm and mold safety. After determining the slider resting window... Subsequently, an assessment is conducted to evaluate the potential positional drift risk of the slider within the pause window due to its own gravity, transmission backlash, or driving inertia. By comprehensively considering the slider's current actual position, pause duration, and velocity changes at the end of the execution path, the holding torque value required to offset this positional drift risk is calculated. This holding torque value is the minimum effective holding output determined through multiple pause tests during the equipment debugging phase, and is adjusted accordingly based on the current operating conditions. After obtaining the holding torque value, it is introduced as an initial constraint into the return trajectory planning model. The return trajectory planning model describes the correspondence between displacement and time during the slider's return from the pause position to the standby position. In this model, by using the holding torque value as an initial constraint, it ensures that the slider overcomes static resistance and smoothly enters the return motion state during the start-up return phase. Subsequently, the return trajectory planning model is solved to generate a set of return control commands arranged in chronological order, thereby forming a rapid command sequence for driving the slider to return to the standby position quickly and stably, providing reliable motion preparation conditions for the synchronous start-up of the next production cycle.

[0030] S6 includes acquiring the real-time command sequence of the multi-station mold control system, parsing the pulse frequency and displacement encoding data contained in the command sequence, and extracting the key time nodes characterizing the switching of mold station actions; mapping the key time nodes to a preset periodic time coordinate system, and calculating the synchronization deviation value of the mold station during the movement process by combining the stamping frequency and feeding step distance parameters; if the synchronization deviation value exceeds a preset range, deriving the correction trigger time based on the synchronization deviation value, and generating a correction command containing new timing parameters based on the correction trigger time; executing the correction command to reconstruct the control logic of the intermittent cycle, locking the step distance and stamping frequency synchronization state, and obtaining a stable control scheme for the entire intermittent cycle.

[0031] After completing step S5 and generating a rapid instruction sequence to return to the standby position, the currently executing instruction sequence is first obtained in real time from the multi-station mold control system. This instruction sequence is a set of control instructions output sequentially by the control system as the mold completes feeding, positioning, and forming actions at each station. Subsequently, the instruction sequence is parsed line by line to extract the pulse frequency data and displacement code data. The pulse frequency data reflects the cycle speed of mold feeding and station switching, while the displacement code data characterizes the actual displacement state of the mold between different stations. After parsing, based on the pulse frequency change points and the locations where the displacement code switches, the sequence accurately characterizes the mold station. The key time node for switching from one action state to the next is defined as follows: the key time node corresponds to the actual trigger moment when the mold completes one feeding cycle or one station switch. Next, the key time node is mapped to a pre-set periodic time coordinate system. This system is established based on the single stamping cycle of the servo press and is used to uniformly describe the relative positional relationship between the stamping action and the mold feeding action on the same time axis. After mapping, the time position of the mold station is calculated throughout the entire movement process, combining the stamping frequency parameter and the feeding step distance parameter under the current production state. The stamping frequency parameter is obtained by statistically analyzing the number of stampings completed per unit time. The material step distance parameter is preset based on the mold structure and process requirements to limit the feeding displacement corresponding to each stamping. By comparing the actual time position of the mold station in the cycle time coordinate system with the theoretically expected position, the synchronization deviation value of the mold station action relative to the stamping action is calculated. This synchronization deviation value quantitatively reflects the degree of matching between the step distance and the number of stamping strokes. Subsequently, the synchronization deviation value is compared with a preset range, which is an allowable deviation range determined during the equipment debugging phase through statistical analysis of multi-cycle operation data under a state of complete synchronization between the step distance and the number of stamping strokes. When the synchronization deviation value is determined to exceed this preset range, the current mold... There is a risk of asynchrony between the workstation actions and the stamping cycle. In this case, based on the magnitude and direction of the synchronization deviation, the correction trigger time that needs to be advanced or delayed is derived. Based on this correction trigger time, a correction instruction containing new timing parameters is generated. The correction instruction is used to clearly adjust the trigger time of the mold workstation actions in the next or subsequent cycle. Finally, the correction instruction is executed to reconstruct the original intermittent cycle control logic, so that the mold feeding step distance and the stroke of the servo press are re-established in a one-to-one correspondence, thereby locking the synchronization state of the step distance and the stroke, and finally obtaining a stable control scheme covering the entire intermittent cycle, providing time consistency guarantee for the continuous and stable operation of the production line.

[0032] S7 includes acquiring a preset stable control command to drive the servo press to run and obtain an execution trajectory; determining equipment feedback data based on the real-time pulse response generated by the execution trajectory; extracting the critical time points between actions and pauses from the equipment feedback data and judging the timing deviation based on the critical time points; if the timing deviation exceeds a preset threshold, determining the switching accuracy between actions and pauses by associating with load fluctuations; mapping the switching accuracy to a collaborative command to obtain an operating cycle; and obtaining the final production line synchronization optimization result through multi-machine matching of the operating cycle.

[0033] After completing step S6 and obtaining the stable control scheme for the entire intermittent cycle, the servo press is first driven by the preset stable control instructions in the stable control scheme. This causes the servo press to run continuously according to the determined action time and rest time logic, thereby obtaining an execution trajectory reflecting the actual running state of the slider. The execution trajectory is used to completely record the position change process of the slider within one or more stamping cycles. Subsequently, the running state of the servo press is fed back based on the real-time pulse response signal generated by the execution trajectory during operation. The real-time pulse response is directly generated by displacement feedback or drive feedback and is used to characterize the actual motion behavior of the slider at different time points, thus forming equipment feedback data. After obtaining the equipment feedback data, time series analysis is performed on the data to extract the critical time points corresponding to the slider switching from the motion state to the rest state and from the rest state to the motion state. The critical time points are the true boundary positions of action and rest on the time axis. Then, the critical time points are compared with the theoretical switching time set in the stable control scheme to determine whether there is a time advance or time lag in the actual execution of action and rest, and the timing is determined accordingly. The deviation, wherein the timing deviation is used to quantitatively reflect the time consistency between control commands and actual execution; subsequently, the timing deviation is compared with a preset threshold, which is the maximum allowable deviation range determined by statistical analysis of the action and pause switching time of multiple cycles in a stable operating state of the production line during the equipment debugging phase. When it is determined that the timing deviation exceeds the preset threshold, the switching accuracy between the action phase and the pause phase is further confirmed by combining the load fluctuation data collected during the execution process. The switching accuracy is used to characterize whether the action and pause switching remains within a controllable range under load change conditions; after determining the switching accuracy, the switching accuracy is mapped with a pre-established collaborative command relationship to obtain the corresponding operating cycle parameter. The operating cycle is used to describe the time rhythm of the servo press cooperating with peripheral equipment under the current control accuracy level; finally, by matching and analyzing the operating cycle with the actual operating rhythm of multiple positions in the production line, the time consistency of each position in the same cycle is confirmed, thereby obtaining the final synchronization optimization result reflecting the action synchronization and rhythm stability of the entire production line, providing a quantitative evaluation basis for the overall collaborative operation of the production line.

[0034] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A motion control method for a servo press, characterized in that, include: S1. Real-time data is collected through the status sensors of the peripheral equipment conveyor belt and robot arm to obtain synchronization signals and response delay values, and the current fluctuation of the production line is determined to obtain the initial adjustment parameters of the intermittent cycle. S2. Based on the obtained initial adjustment parameters, a fuzzy control algorithm is used to process signal fluctuations. If the response delay exceeds the preset threshold, the pause time is extended; otherwise, the original division is maintained, thus obtaining an accurate action time and pause time division scheme. S3. Extract the cycle start time from the obtained division scheme, compare it with the data of the servo press slider position sensor, determine the starting point of the motion curve to obtain a high-precision cycle control sequence. S4. The PID control algorithm is used to perform closed-loop regulation on the obtained control sequence. By comparing the deviation between the actual slider position and the predetermined curve, if the deviation is greater than the threshold, the servo motor speed is adjusted; otherwise, the current speed is maintained to obtain the optimized stamping motion curve execution path. S5. Based on the optimized execution path, monitor the real-time status changes of peripheral devices, obtain the updated synchronization signal, and determine the position holding requirements of the slider during the pause time to obtain a fast instruction sequence to return to the standby position.

2. The motion control method for a servo press according to claim 1, characterized in that: S1 includes: The pulse frequency of the peripheral equipment conveyor belt and the load displacement of the robot are obtained, and the pulse frequency and load displacement are converted into multi-dimensional real-time data. Clock alignment processing is performed on the multidimensional real-time data to extract the synchronization signal containing the trigger timestamp; The response delay of the actuator is calculated based on the synchronization signal; The difference between the response delay and the preset reference time is used to determine the production line fluctuations that reflect operational stability. The initial adjustment parameters for the intermittent cycle are obtained by matching the pre-stored logical mapping table of the production line fluctuations.

3. The motion control method for a servo press according to claim 1, characterized in that: S2 includes: Obtain initial adjustment parameters and real-time acquired signal fluctuations, and extract the amplitude change rate of the signal fluctuations; The amplitude change rate is input into the fuzzy control algorithm model to calculate the membership function and obtain the fuzzy control output value. The fuzzy control output value is compared with a preset time series to determine the response delay value; If the response delay value exceeds a preset threshold, the pause time on the original timeline is extended to obtain a corrected pause time; The modified pause time is used to synchronize and align the action time on the original timeline to obtain an accurate action time and pause time division scheme.

4. The motion control method for a servo press according to claim 1, characterized in that: S3 includes: Acquire high-frequency sampling data streams containing the estimated time interval division scheme and the slider position sensor; The high-frequency sampling data stream is divided into time domains according to the partitioning scheme to obtain a set of data segments covering the complete stamping cycle at the position to be analyzed. A smooth gradient feature sequence is generated for the set of data segments to be analyzed. If the smooth gradient feature sequence exceeds a preset static state fluctuation threshold, the approximate motion start time is determined and the micro dataset of the start segment is extracted. A polynomial fit is performed on the initial segment micro dataset to construct a local analytical model. The starting point of the corrected high-precision motion curve is determined by solving the curvature abrupt change point of the local analytical model. The original position data segments are realigned using the starting point of the high-precision motion curve as a reference to generate standardized single-cycle motion trajectory data. A standard reference model is constructed based on the standardized single-cycle motion trajectory data to obtain a high-precision periodic control sequence.

5. The motion control method for a servo press according to claim 1, characterized in that: S4 includes: Acquire real-time displacement data collected by the displacement sensor and convert the real-time displacement data into a pulse feedback signal; The difference between the pulse feedback signal and the theoretical value of the predetermined curve is calculated to obtain the deviation value; Perform closed-loop adjustment calculations on the deviation value to determine the control voltage used to correct the error; The control voltage is converted into a drive frequency, and the servo motor is adjusted using the drive frequency to approximate the predetermined curve, thereby obtaining an optimized stamping motion curve execution path.

6. The motion control method for a servo press according to claim 1, characterized in that: S5 includes: Collect optimized execution path data and extract state change characteristics that reflect fluctuations in the operation of peripheral devices; The updated synchronization signal is obtained using the state change characteristics, and the slider rest window is defined based on the synchronization signal; Calculate the holding torque value to account for the position drift risk within the slider rest window; A return trajectory planning model is constructed using the holding torque value as the initial constraint, and the return trajectory planning model is solved to generate a fast instruction sequence to return to the standby position.

7. The motion control method for a servo press according to claim 1, characterized in that, This also includes S6, selecting key time nodes from the obtained instruction sequence, verifying the step distance correspondence for multi-station dies, determining if the step distance and stroke count do not match, recalculating the start time, otherwise confirming synchronization, and obtaining a stable control scheme for the entire intermittent cycle, specifically including: The real-time command sequence of the multi-station mold control system is obtained, the pulse frequency and displacement code data contained in the command sequence are analyzed, and the key time nodes characterizing the switching of mold station actions are extracted. The key time nodes are mapped to a preset periodic time coordinate system, and the synchronization deviation of the mold station during the movement is calculated by combining the stamping frequency and feeding step distance parameters.

8. The motion control method for a servo press according to claim 7, characterized in that: S6 further includes: If the synchronization deviation value exceeds the preset range, the correction trigger time is deduced based on the synchronization deviation value, and a correction instruction containing new timing parameters is generated based on the correction trigger time. The correction instruction is executed to reconstruct the control logic of the intermittent cycle, lock the step size and stroke synchronization state, and obtain a stable control scheme for the entire intermittent cycle.

9. A motion control method for a servo press according to claim 7, characterized in that, It also includes S7, which drives the servo press to execute through the obtained stable control scheme, collects equipment feedback data during the execution process, and determines the switching accuracy between actions and pauses to obtain the final production line synchronization optimization result, specifically including: The preset stable control commands are obtained to drive the servo press to run, and the execution trajectory is obtained; The device feedback data is determined based on the real-time pulse response generated by the execution trajectory. The critical time points of action and pause are extracted from the feedback data of the device, and the timing deviation is determined based on the critical time points.

10. A motion control method for a servo press according to claim 9, characterized in that: The S7 also includes: If the timing deviation exceeds a preset threshold, the switching accuracy between action and pause is determined by the associated load fluctuation. The switching precision is mapped to cooperative instructions to obtain the operating cycle; The final production line synchronization optimization result is obtained through multi-machine position matching of the aforementioned operating cycle.

Images