An automatic control method and system for sheet metal part processing equipment

Abstract

The application relates to the technical field of bending control, and discloses an automatic control method and system of sheet metal part machining equipment. The method comprises the following steps: binding an angle compensation execution amount and a die cumulative bending frequency to construct a compensation amount original sequence, generating a compensation control parameter; filtering the compensation amount original sequence to extract a trend degradation component and a random fluctuation component, generating a servo motor compensation adjustment amount control instruction; establishing a linear regression relationship according to the trend degradation component to extract a degradation rate, calculating a confidence lower limit of a remaining process service life, generating a stroke control instruction and a bending pressure control instruction; and generating a punch scheduling control instruction, a die changing mechanism action time sequence control instruction or a slider locking shutdown control instruction according to the confidence lower limit. The application solves the problems that the existing automatic control method of a bending machine cannot separate the trend degradation component from the compensation amount historical sequence and cannot quantitatively predict the remaining process service life of a die and implement graded preventive control intervention.

Description

Technical Field

[0001] This application relates to the field of bending control technology, and in particular to an automatic control method and system for sheet metal processing equipment. Background Technology

[0002] Sheet metal bending is a widely used metal sheet forming process in modern manufacturing. A bending machine applies bending pressure to the sheet metal using a punch to shape it to a target angle. Existing automatic control methods for bending machines typically employ a closed-loop compensation control strategy. This involves an angle sensor collecting the measured bending angle after each bending stroke, adding the deviation between the measured angle and the target angle command to the next stroke displacement command, and driving the slider servo motor to execute the next bending action according to the corrected stroke displacement. Alternatively, an offline trained neural network model can be used to predict the springback amount and then add it to the control command, thereby achieving angle deviation compensation for a single bending stroke.

[0003] However, the existing solutions have significant shortcomings. The control time dimension of current automatic control methods for bending machines is limited to a single bending stroke. Each compensation action is based solely on the data collected by the sensors at that moment, without systematically accumulating and analyzing the compensation amounts from each subsequent operation. This results in the control system's inability to distinguish between two completely different sources of deviation: one is random fluctuation caused by differences in thickness tolerance, hardness deviation, and rolling direction between individual sheets in the same batch; the other is a monotonically increasing trend of degradation in compensation amount caused by the geometric wear of the punch as the cumulative number of bends increases. Current solutions treat both types of deviations as a single compensation amount, and the interference of random fluctuations masks the trend of degradation, preventing the control system from sensing the cumulative degree of die wear.

[0004] Existing solutions cannot separate the trend degradation component from the historical sequence of compensation amounts, resulting in the control system lacking the ability to quantitatively predict the punch degradation rate and remaining process life. Without prediction of the remaining process life, the control system cannot proactively adjust the work order scheduling strategy or trigger the die changing mechanism before the die wear reaches a critical state. It can only respond passively after the compensation amount exceeds the limit, which leads to the discovery of a batch of bending scraps. Furthermore, since the prediction of the remaining process life itself has statistical uncertainty, if a confidence interval is not constructed for the prediction results and the lower confidence bound is used as the basis for triggering graded intervention, the timing of intervention will be systematically biased when the degradation rate changes abruptly, causing the graded control strategy to lose reliability at the moment when conservative decision-making is most needed. Summary of the Invention

[0005] This application provides an automatic control method and system for sheet metal processing equipment, which solves the problem that existing automatic control methods for bending machines cannot separate trend degradation components from the historical sequence of compensation amounts, and thus cannot quantitatively predict the remaining usable life of the die and implement graded preventive control intervention. It realizes the transformation of bending machines from single-stroke compensation control to proactive predictive control across batches of the die's entire life cycle.

[0006] In a first aspect, this application provides an automatic control method for sheet metal processing equipment, the automatic control method for sheet metal processing equipment comprising: Step S1: After each bending stroke, the deviation between the measured bending angle collected by the angle sensor and the target angle command is used as the angle compensation execution amount. It is bound to the cumulative bending number of the mold and written into the storage unit. The original sequence of compensation amount is constructed by arranging the cumulative bending number of the mold in ascending order. The angle compensation execution amount is then superimposed on the next stroke displacement command of the slider in a closed-loop feedback manner to generate compensation control parameters. Step S2: Perform moving median filtering on the original compensation amount sequence to extract the trend degradation component. Subtract the original compensation amount sequence from the trend degradation component to obtain the random fluctuation component. Generate a servo motor compensation adjustment control command based on the random fluctuation component to drive the slider servo motor to perform position closed-loop control. Step S3: Establish a linear regression relationship between the trend degradation component and the cumulative bending number of the mold, extract the degradation rate and the initial compensation baseline, calculate the confidence lower bound of the remaining process usable life by combining the pre-stored compensable limit threshold in the punch file, and generate the slider servo motor stroke control command and bending pressure control command based on the degradation rate. Step S4: Based on the preset range of the confidence lower bound of the remaining process life, generate hierarchical punch scheduling control instructions, die changing mechanism action timing control instructions, or slider locking stop control instructions to complete the hierarchical automatic control of the bending process of the bending machine.

[0007] Secondly, this application provides an automatic control system for sheet metal processing equipment, the automatic control system of which includes: The generation module is used to take the deviation between the measured bending angle collected by the angle sensor and the target angle command as the angle compensation execution amount after each bending stroke, bind it with the cumulative bending number of the mold and write it into the storage unit, arrange it in ascending order according to the cumulative bending number of the mold, construct the original sequence of compensation amount, and superimpose the angle compensation execution amount to the next stroke displacement command of the slider in a closed loop feedback manner to generate compensation control parameters. The drive module is used to perform moving median filtering on the original compensation quantity sequence, extract the trend degradation component, subtract the original compensation quantity sequence from the trend degradation component to obtain the random fluctuation component, generate a servo motor compensation adjustment control command based on the random fluctuation component, and drive the slider servo motor to perform position closed-loop control. The analysis module is used to establish a linear regression relationship between the trend degradation component and the cumulative number of bends of the die, extract the degradation rate and the initial compensation baseline, calculate the confidence lower bound of the remaining process usable life by combining the pre-stored compensable limit threshold in the punch file, and generate slider servo motor stroke control command and bending pressure control command based on the degradation rate. The control module is used to generate punch scheduling control commands, die changing mechanism action timing control commands, or slider locking stop control commands in a hierarchical manner based on the preset range where the lower limit of the remaining process usable life is located, so as to complete the hierarchical automatic control of the bending process of the bending machine.

[0008] Thirdly, an automatic control device for sheet metal processing equipment is provided, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor invokes the instructions in the memory to cause the automatic control device for sheet metal processing equipment to execute the aforementioned automatic control method for sheet metal processing equipment.

[0009] Fourthly, a computer-readable storage medium is provided, wherein instructions are stored therein, which, when executed on a computer, cause the computer to perform the above-described automatic control method for sheet metal processing equipment.

[0010] The technical solution provided in this application constructs an original sequence of compensation amounts by binding the deviation between the measured bending angle collected by the angle sensor and the target angle command after each bending stroke of the bending machine with the cumulative bending count of the die. This original sequence of compensation amounts is then superimposed on the displacement command of the next stroke of the slider using a closed-loop feedback method to generate compensation control parameters. This allows the bending machine to correct the displacement command of the next stroke in real time based on the measured deviation after each bending stroke, solving the problem of traditional bending machines relying on manual experience to adjust the compensation amount and failing to continuously accumulate deviation data during production. Furthermore, the original compensation amount sequence is subjected to moving median filtering to extract the trend degradation component. The difference between the original compensation amount sequence and the trend degradation component yields a random fluctuation component. Based on this random fluctuation component, a servo motor compensation adjustment control command is generated to drive the slider servo motor to perform position control. This enables the control system to automatically distinguish and handle two different types of deviation sources for the first time. The extraction of the random fluctuation component allows the servo motor's compensation adjustment range to be dynamically adjusted according to the batch characteristics of the sheet material, avoiding interference and masking of the trend degradation signal by random fluctuations.

[0011] This application establishes a linear regression relationship between the trend degradation component and the cumulative number of bends in the die to extract the degradation rate and the initial compensation baseline. It then calculates the lower bound of the remaining usable process life based on the pre-stored compensable limit threshold in the punch file. Based on the degradation rate, it generates stroke control commands for the slider servo motor and bending pressure control commands, extending the automatic control time dimension of the bending machine from a single bending stroke to the entire lifecycle of the die across batches. The introduction of the degradation rate allows the control system to proactively correct the stroke control parameters and bending pressure parameters before the punch wear reaches a critical state. The construction of the lower bound provides statistically conservative assurance for tiered intervention decisions, ensuring that intervention actions can still be completed before the die life is exhausted in the worst-case scenario, even when the degradation rate undergoes a step change. Finally, based on the preset interval where the lower bound of the remaining usable process life falls, it generates tiered punch scheduling control commands, die changing mechanism action sequence control commands, or slider locking shutdown control commands, directly translating the prediction results into specific control actions for the bending machine's actuators. This achieves a shift from a passive response to scrap to a proactive preventative intervention control mode. Attached Figure Description

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

[0013] Figure 1 This is a schematic diagram of an embodiment of the automatic control method for sheet metal processing equipment in this application. Figure 2 This is a schematic diagram comparing the original sequence of compensation amount and the trend degradation component in an embodiment of this application. Detailed Implementation

[0014] This application provides an automatic control method and system for sheet metal processing equipment. The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0015] For ease of understanding, the specific process of the embodiments of this application is described below. Please refer to [link / reference]. Figure 1 One embodiment of the automatic control method for sheet metal processing equipment in this application includes: Step S1: After each bending stroke, the deviation between the measured bending angle collected by the angle sensor and the target angle command is used as the angle compensation execution amount. It is bound to the cumulative bending number of the mold and written into the storage unit. The original sequence of compensation amount is constructed by sorting the cumulative bending number of the mold in ascending order. The angle compensation execution amount is then superimposed on the next stroke displacement command of the slider in a closed-loop feedback manner to generate compensation control parameters. Specifically, the angle compensation execution amount is the difference between the measured bending angle collected by the angle sensor after each bending stroke and complete unloading of the punch, and the target angle command stored in the CNC unit for that stroke. This reflects the forming angle deviation caused by the elastic rebound of the sheet metal. This deviation value is written into the storage unit using the cumulative bending count of the die output by the bending machine counter at the end of each bending stroke as the index key. It is continuously added as production progresses, forming an original sequence of compensation amounts with the cumulative bending count of the die as the horizontal axis and the angle compensation execution amount of each stroke as the vertical axis. This difference value is superimposed on the next stroke displacement command of the slider in the opposite direction, becoming the compensation control parameter. After being written into the servo driver position control register, it is converted into a drive pulse signal and output to the slider servo motor, driving the slider to perform the bending action according to the corrected stroke displacement in the next bending stroke.

[0016] Step S2: Perform moving median filtering on the original compensation quantity sequence to extract the trend degradation component. Subtract the original compensation quantity sequence from the trend degradation component to obtain the random fluctuation component. Generate servo motor compensation adjustment control command based on the random fluctuation component to drive the slider servo motor to perform position closed-loop control. Specifically, the original compensation sequence contains two types of skewers of different natures. One type originates from the geometric wear of the die as the cumulative number of bends increases, manifesting as a monotonically increasing trend in the compensation execution amount, known as the trend degradation component. The other type originates from the differences in thickness tolerance, hardness deviation, and rolling direction angle among individual plates in the same batch, manifesting as a component that randomly fluctuates around the trend degradation component, known as the random fluctuation component. The moving median filter uses the median of a consecutive number of data points as the filter output for that window position. A window length of 50 bends is recommended. The reason for using the median instead of the mean is that the median is robust to single extreme outliers and will not skew the trend estimate due to sudden angle changes caused by defects in individual plates. Subtracting the trend degradation component value from each point in the original compensation sequence yields the random fluctuation component. The standard deviation is calculated for all data points of the random fluctuation component. When the standard deviation exceeds the preset fluctuation threshold, a compensation adjustment upper and lower limit correction command is written to the servo driver.

[0017] Step S3: Establish a linear regression relationship based on the trend degradation component and the cumulative bending number of the die, extract the degradation rate and the initial compensation baseline, and calculate the confidence lower bound of the remaining process usable life by combining the pre-stored compensable limit threshold in the punch file. Generate the slider servo motor stroke control command and bending pressure control command based on the degradation rate. Specifically, the linear regression relationship between the trend degradation component and the cumulative number of bends in the die is solved using the least squares method. With the values ​​of each data point in the trend degradation component as the dependent variable and the corresponding cumulative number of bends in the die as the independent variable, the slope and intercept of the regression line are obtained by minimizing the sum of squared residuals. The slope represents the degradation rate, measured in degrees per 10,000 bends, and the intercept represents the initial compensation baseline. When the goodness of fit is lower than a preset fitting threshold, a quadratic polynomial regression is used for refitting. The remaining usable process life is calculated by dividing the difference between the compensable limit threshold and the initial compensation baseline by the degradation rate, and then subtracting the current cumulative number of bends in the die. The compensable limit threshold is pre-stored by the punch manufacturer in the punch file and represents the upper limit of springback that the bending machine can correct through stroke compensation. Exceeding this value, the bending machine cannot maintain bending accuracy. The standard deviation of the regression residuals is divided by the degradation rate to obtain the predicted standard deviation of the remaining usable process life. Multiplying this by a preset confidence coefficient and subtracting it from the remaining usable process life yields the lower confidence bound. A confidence coefficient of 1.96 corresponds to a 95% confidence level.

[0018] Step S4: Based on the preset range where the lower limit of the remaining usable process life is located, generate punch scheduling control instructions, die changing mechanism action timing control instructions, or slider locking stop control instructions in stages to complete the staged automatic control of the bending process of the bending machine.

[0019] Specifically, the lower bound of the remaining usable life of the process is used as the basis for determining graded intervention, and three preset intervals are set. When the lower bound is greater than the first preset intervention threshold, a punch scheduling control command is generated and written to the work order scheduling module, restricting the punch to only accept bending work orders with lower precision levels. When the lower bound is between the first and second preset intervention thresholds, the expected remaining usable time is obtained by dividing the lower bound by the bending frequency of the current shift. Half of this time is taken as the latest die change start time, and a die change mechanism action sequence control command is generated and written to the die change mechanism. When the lower bound is not greater than the second preset intervention threshold, a stop control command is output to the slider locking mechanism to lock the slider to a safe height. At the same time, the initial compensation baseline and geometric parameters of the spare punch are read from the punch file, the geometric parameters are written to the punch parameter register of the CNC unit, and the initial compensation baseline is written to the servo driver compensation reference register, completing the reset of control parameters after punch switching. The specific values ​​of the three thresholds are determined by the production process requirements and the equipment die change cycle, and are preset by the process engineer based on the actual equipment situation before being written to the punch file.

[0020] In one specific embodiment, step S1 includes: The angle compensation execution amount is obtained by subtracting the measured bending angle collected by the angle sensor after each bending stroke from the target angle command stored in the CNC unit. The measured bending angle is the stable angle value when the plate elastically recovers after the punch is unloaded, and the target angle command is the design angle corresponding to the current bending process. The angle compensation execution amount is bound to the cumulative number of bends of the mold output by the bending machine counter at the end of the current bending stroke in the form of key-value pairs. The bound key-value pairs are written into the storage unit and sorted in ascending order with the cumulative number of bends of the mold as the sorting index to obtain the original sequence of compensation amount. The angle compensation execution amount is superimposed on the next stroke displacement command of the slider, wherein the superposition direction is opposite to the sign of the angle compensation execution amount, and the superimposed stroke displacement command is used as the compensation control parameter and written into the position control register of the servo driver. The compensation control parameters are converted into drive pulse signals by the servo driver and output to the slider servo motor. The slider is driven to perform the next bending action according to the stroke displacement corresponding to the compensation control parameters. After the next bending stroke is completed, the angle sensor is triggered again to collect the measured bending angle and update the original sequence of compensation amount.

[0021] Specifically, the angle compensation execution amount is the difference between the stable measured angle collected by the angle sensor after the bending stroke of the sheet metal ends and the punch is completely unloaded, and the target angle command in the CNC unit for that stroke. Its physical meaning is the deviation of the sheet metal from the design angle due to elastic rebound, measured in degrees. The opposite sign of the superposition direction to the angle compensation execution amount means that when the measured angle is greater than the target angle, the deviation value is positive, and the next stroke displacement command needs to increase the stroke on the original basis to deepen the punch's downward pressure depth; therefore, the superposition amount takes a negative direction. Conversely, when the deviation value is negative, the superposition amount takes a positive direction to reduce the downward pressure depth, thereby causing the measured bending angle to converge towards the target angle. The angle compensation execution amount is bound to the cumulative bending number of the die in key-value pairs. In the key-value pair, the cumulative bending number of the die is used as the search key, and the angle compensation execution amount is used as the corresponding value. After both are written into the storage unit, they are arranged in ascending order by key value, so that the original sequence of compensation amounts truly reflects the change process of die wear with the accumulation of bending times over time.

[0022] After the compensation control parameters are written into the servo driver's position control register, the servo driver converts the displacement in the register into drive pulse signals of corresponding frequency and quantity, and outputs them to the encoder feedback loop of the slider servo motor. This drives the servo motor to perform precise positioning for the next bending stroke according to the corrected target displacement. After the slider reaches the bottom position of the correction stroke and completes the bending of the sheet metal, the punch returns to a safe height, and the sheet metal elastically rebounds. The angle sensor collects the measured bending angle again at the moment the punch is unloaded. The difference between this angle and the current target angle command is used to obtain a new angle compensation execution amount. This amount is then bound to the latest cumulative bending count of the die output by the counter and added to the storage unit, completing one cycle update of the original sequence of compensation amounts. This causes the sequence length to increase by one data point with each completion of the bending stroke.

[0023] Figure 2 This is a schematic diagram comparing the original sequence of compensation amount and the trend degradation component in an embodiment of this application. Figure 2 The horizontal axis represents the cumulative number of bends of the die, and the vertical axis represents the angle compensation execution amount. The thin gray line represents the original sequence of compensation amount, reflecting the change in the deviation between the measured bending angle collected by the angle sensor and the target angle command after each bending stroke, which includes random fluctuations caused by individual differences in the same batch of sheet metal. The thick black line represents the trend degradation component output after applying a moving median filter to the original compensation amount sequence, reflecting the trend of monotonically increasing angle compensation execution amount due to the increase in the cumulative number of bends caused by the geometric wear of the punch. The difference between the two curves is the random fluctuation component. It can be seen that the trend degradation component increases from about 0.8° initially to about 2.5° with the cumulative number of bends of the die, showing a clear monotonically increasing trend, which verifies the cumulative effect of die wear on springback.

[0024] In one specific embodiment, step S2 includes: Using the original sequence of compensation amount as input, the window length of the moving median filter is set, and the median of the angle compensation execution amount of the continuous window length starting from the current data point in the original sequence of compensation amount is taken. The median output value of each window position is arranged in ascending order according to the cumulative bending number of the mold to obtain the trend degradation component. The angle compensation execution amount of each data point in the original compensation amount sequence is subtracted from the value of the corresponding position in the trend degradation component to obtain the random fluctuation component. A normality test is performed on all data points of the random fluctuation component, and the test statistic is compared with the preset normality threshold. When the test statistic is lower than the preset normality threshold, an alarm signal for abnormal performance dispersion of the sheet material batch is output to the CNC unit. The standard deviation is calculated based on all data points of the random fluctuation component. The standard deviation is compared with the preset fluctuation threshold. When the standard deviation exceeds the preset fluctuation threshold, the difference between the standard deviation and the preset fluctuation threshold is used as the compensation adjustment amount to generate the servo motor compensation adjustment amount control command. The servo motor compensation adjustment control command is written into the servo driver, which drives the slider servo motor to perform position control according to the upper and lower limits of the adjustment amount corresponding to the servo motor compensation adjustment control command, thereby completing the adjustment of the angle deviation caused by random fluctuation components in the current bending process.

[0025] Specifically, a window length of 50 bends is recommended for the moving median filter. This value is based on the fact that the trend change in mold wear is a slow-time-scale process, and the time span corresponding to 50 bends is sufficient to smooth out random fluctuations caused by individual differences in a single batch of sheet metal. At the same time, the timeliness of the trend degradation component will not decrease due to an excessively long window. The trend degradation component is the smoothed sequence output after the original compensation quantity sequence is filtered by the moving median filter. Its physical meaning is the component that reflects the systematic monotonically increasing trend of the compensation quantity caused by mold geometric wear after removing random fluctuations. The median is used instead of the mean because the median is robust to extreme outliers within the window, and sudden angle changes caused by single sheet metal defects or equipment vibration will not skew the trend estimation results. The random fluctuation component is the difference between the data points of each data point in the original sequence of compensation quantity and the corresponding position value of the trend degradation component. It reflects the random disturbance of the compensation quantity caused by thickness tolerance, hardness deviation and rolling direction deviation between individual plates in the same batch. Its statistical distribution follows a normal distribution with a mean of zero under normal working conditions. The normality test is carried out by the Shapiro-Wilk test, and the preset normality threshold is 0.95. When the test statistic is lower than 0.95, it indicates that the distribution of the random fluctuation component has non-normal characteristics and the performance dispersion of the plate batch is abnormal. At this time, an alarm signal is output to the CNC unit to prompt manual inspection of the mechanical properties of the plate.

[0026] The standard deviation of the random fluctuation component reflects the dispersion of the compensation amount caused by individual differences in the same batch of sheet metal. It is recommended to set the preset fluctuation threshold to six times the accuracy class of the angle sensor. Taking a bending machine angle sensor accuracy class of ±0.05 degrees as an example, the preset fluctuation threshold should be 0.30 degrees. When the standard deviation exceeds 0.30 degrees, it indicates that the random fluctuation has exceeded the reasonable range of the equipment's compensation capability, and the upper and lower limits of the servo motor's compensation adjustment amount need to be narrowed and corrected. The compensation adjustment amount correction is the difference between the standard deviation and the preset fluctuation threshold. This difference is written into the servo driver as the constraint boundary of the compensation adjustment amount control command. When the servo motor executes position closed-loop control, it limits the single compensation adjustment amount within the corrected upper and lower limits to prevent abnormal jumps in the compensation amount caused by excessive random fluctuations from being transmitted to the slider stroke output, thereby keeping the bending angle deviation within a controllable range.

[0027] In one specific embodiment, step S3 establishes a linear regression relationship based on the trend degradation component and the cumulative number of bends in the die, and extracts the degradation rate and the initial compensation baseline, including: Using the values ​​of each data point of the trend degradation component as the dependent variable and the cumulative number of bends of the mold corresponding to each data point as the independent variable, least squares linear regression is performed on all data points of the trend degradation component. The slope of the regression line is determined as the degradation rate, and the intercept of the regression line is determined as the initial compensation baseline. The difference between each data point of the trend degradation component and the corresponding value of the regression line is calculated to obtain the regression residual. The goodness of fit is calculated based on the regression residual. The goodness of fit is compared with the preset fitting threshold. When the goodness of fit is lower than the preset fitting threshold, all data points of the trend degradation component are switched to quadratic polynomial fitting. The coefficient of the first term obtained by refitting is determined as the degradation rate, and the constant term obtained by refitting is determined as the initial compensation baseline. Write the degradation rate, initial compensation baseline, and regression residual into the convex mold file.

[0028] Specifically, the degradation rate is the unit change of the trend degradation component with the increase of the cumulative bending number of the die, measured in degrees per 10,000 bends. Physically, it represents the increase in the angle compensation execution amount corresponding to each additional 10,000 bending strokes of the die, reflecting the rate of geometric wear of the punch. The initial compensation baseline is the intercept of the regression line when the cumulative bending number of the die is zero, reflecting the basic springback compensation amount of the punch in the initial stage of use, determined by the material's elastic properties and the initial geometric parameters of the die. Least squares linear regression determines the slope and intercept of the regression line by minimizing the sum of squared residuals between each data point of the trend degradation component and the corresponding estimated value of the regression line. This method possesses the optimal unbiased estimation property under the premise that there is a linear relationship between the trend degradation component and the cumulative bending number of the die. Goodness of fit reflects the degree to which the regression line fits the data points of the trend degradation component. The value ranges from zero to one. It is recommended to set the preset goodness of fit threshold to 0.85. If it is lower than 0.85, it means that the linear model cannot fully describe the changing pattern of the trend degradation component. It is necessary to switch to quadratic polynomial fitting to capture the nonlinear trend of the accelerated degradation stage. After switching, the coefficient of the first term of the quadratic polynomial corresponds to the degradation rate, and the constant term corresponds to the initial compensation baseline. The naming should be consistent with the linear regression results.

[0029] The regression residual is a sequence of differences between the values ​​of each data point of the trend degradation component and the corresponding estimated value of the regression line. Its standard deviation reflects the dispersion of the trend degradation component near the predicted value of the regression model. The smaller the dispersion, the more stable the degradation trend and the higher the prediction accuracy. The reason for writing the regression residual, degradation rate, and initial compensation baseline together into the punch file is that the standard deviation of the regression residual is a necessary input for calculating the standard deviation of the remaining usable life prediction. Storing all three together in the punch file ensures data consistency for the same punch across different production batches and avoids calculation errors caused by scattered data storage. The punch file is a persistent storage record indexed by the punch number, containing fields such as punch geometric parameters, cumulative bending count, degradation rate, initial compensation baseline, regression residual, and compensable limit threshold. The write operation is performed after each regression calculation, overwriting the historical records of the same field.

[0030] In one specific embodiment, step S3, which calculates the lower bound of the remaining usable process life by combining the pre-stored compensable limit threshold in the punch file, includes: Read the compensable limit threshold, degradation rate, initial compensation baseline, and current cumulative number of bends of the die from the punch file. Divide the difference between the compensable limit threshold and the initial compensation baseline by the degradation rate, and then subtract the current cumulative number of bends of the die to obtain the remaining usable process life. Regression residuals are read from the punch file. The standard deviation of regression residuals is calculated based on the regression residuals. The standard deviation of regression residuals is divided by the degradation rate to obtain the predicted standard deviation of remaining process usable life. The predicted standard deviation of remaining process usable life is multiplied by a preset confidence coefficient to obtain the prediction deviation. The remaining usable process life is subtracted from the predicted deviation to obtain the lower confidence bound of the remaining usable process life. The lower confidence bound of the remaining usable process life and the width of the confidence interval are written into the punch file.

[0031] Specifically, the compensable limit threshold is the upper limit of springback that the bending machine can correct through stroke compensation, pre-stored in the punch file by the punch manufacturer. The unit is degrees. Physically, it represents the maximum angular deviation that the bending machine can compensate for by increasing the punch's depth of cut without changing process parameters. Beyond this value, the bending machine's stroke compensation mechanism can no longer maintain bending accuracy. A typical value is 3.0 degrees, determined by the punch manufacturer based on the punch material, geometry, and the bending machine's mechanical limit parameters, and then written into the punch file. The calculation logic for the remaining usable process life is as follows: divide the difference between the compensable limit threshold and the initial compensation baseline by the degradation rate to obtain the total number of bends required for the trend degradation component to grow from the initial state to the compensable limit threshold. Then subtract the current cumulative number of bends of the die to obtain the remaining number of bends the punch can maintain bending accuracy at the current degradation rate, in times. The standard deviation of the regression residual reflects the degree of dispersion of the trend degradation component near the estimated value of the regression model. Dividing it by the degradation rate yields the standard deviation of the remaining usable life prediction. This operation transforms the prediction uncertainty of the compensation dimension into the prediction uncertainty of the number of bends dimension through the error propagation principle.

[0032] The preset confidence coefficient is 1.96, corresponding to a 95% one-sided confidence level under a normal distribution. This value is chosen because, in industrial control scenarios, a 95% confidence level is a conventional engineering approach that balances predictive conservatism with mold utilization. A value that is too small will lead to an overly conservative lower bound, resulting in premature mold retirement and waste; a value that is too large will lead to an overly optimistic lower bound, creating a risk of scrap at the end of the mold's lifespan. Multiplying the predicted standard deviation of the remaining usable process life by 1.96 yields the prediction bias. Subtracting the remaining usable process life from the prediction bias yields the lower bound of the remaining usable process life. This lower bound represents a pessimistic estimate of the remaining usable bending times of the punch at a 95% confidence level. Using this pessimistic estimate as the criterion for tiered intervention ensures that intervention actions can be completed before the punch's lifespan is exhausted, even in the worst-case scenario. The confidence interval width is the difference between the upper and lower confidence bounds of the remaining process life. The upper confidence bound is determined by the sum of the remaining process life and the prediction deviation. The confidence interval width is compared with the preset validity threshold to determine whether the current amount of prediction data is sufficient. When the confidence interval width exceeds the preset validity threshold, the tiered intervention decision is suspended, and intervention is carried out after the data accumulates until the confidence interval narrows.

[0033] In one specific embodiment, step S3, which generates slider servo motor stroke control commands and bending pressure control commands based on the degradation rate, includes: The remaining process life confidence lower bound is read from the punch file. The remaining process life confidence lower bound is compared with the preset life threshold. When the remaining process life confidence lower bound is lower than the preset life threshold, the trend degradation compensation amount is calculated based on the product of the degradation rate and the current cumulative number of bends of the die. The trend degradation compensation amount is superimposed on the stroke displacement reference value to obtain the corrected stroke displacement amount. The corrected stroke displacement amount is written into the position control register of the servo driver to generate the slider servo motor stroke control instruction. The bending pressure correction amount is calculated based on the degradation rate and the current cumulative number of bends of the mold. The bending pressure correction amount is then added to the bending pressure reference value to obtain the corrected bending pressure value. The corrected bending pressure value is then written into the hydraulic system pressure control valve to generate a bending pressure control command.

[0034] Specifically, the preset lifespan threshold is the boundary for triggering corrections to the stroke control parameters and bending pressure control parameters. It is recommended to set this value as the product of the average bending frequency of the current batch and the planned maintenance cycle of the die. Physically, this represents the lower limit of the number of bends the punch can still complete within the planned maintenance window. A value below this indicates that the degree of punch degradation has affected bending accuracy, requiring active correction of the stroke control parameters and bending pressure control parameters. The trend degradation compensation amount is calculated by multiplying the degradation rate by the current cumulative number of bends of the die. Physically, this represents the increase in the trend of springback caused by die wear from the time the punch was put into use until the current cumulative number of bends. This value reflects the cumulative effect of punch geometric wear on the bending angle. The trend degradation compensation amount is superimposed on the stroke displacement reference value in the same direction as the angle compensation execution amount, i.e., superimposed in the direction of increasing the punch's downward pressure depth. The stroke displacement reference value is the standard stroke displacement amount corresponding to the current bending process when the punch has not worn. This value is pre-calculated and stored by the CNC unit based on the target angle of the current bending process and the initial geometric parameters of the punch.

[0035] The bending pressure correction reflects the additional hydraulic pressure compensation required to maintain the same bending angle after the punch wears, leading to an increase in the contact area between the punch and the sheet metal and a decrease in contact stiffness. It is calculated by multiplying the degradation rate by the cumulative number of bends in the current die, then multiplying by the hydraulic pressure-angle conversion factor. This factor is calibrated by the bending machine manufacturer based on the cross-sectional area of ​​the hydraulic cylinder and the geometric parameters of the punch, and stored in the CNC unit parameter library. The bending pressure reference value is the standard hydraulic pressure value corresponding to the current bending process when the punch is not worn. This value is pre-calculated and stored by the CNC unit based on the sheet metal thickness, material, and target bending angle. The bending pressure correction is added to the bending pressure reference value to obtain the corrected bending pressure value. This value is written into the pressure setting register of the hydraulic system pressure control valve. The pressure control valve adjusts the throttling opening of the hydraulic oil circuit according to the setting value in the register, driving the hydraulic cylinder to apply bending pressure to the punch according to the corrected bending pressure value. This ensures that the punch can still apply sufficient forming force to the sheet metal to achieve the target bending angle even when worn.

[0036] In one specific embodiment, step S4 includes: The remaining process life confidence lower bound is read from the punch file. The remaining process life confidence lower bound is compared with the first preset intervention threshold. When the remaining process life confidence lower bound is greater than the first preset intervention threshold, a punch scheduling control instruction is generated based on the current punch accuracy level limit range. The punch scheduling control instruction is written into the work order scheduling module to restrict the punch to only accept bending work orders corresponding to the accuracy level limit range. The lower bound of the remaining available process life is compared with the second preset intervention threshold. When the lower bound of the remaining available process life is greater than the second preset intervention threshold but not greater than the first preset intervention threshold, the estimated remaining available time is calculated based on the ratio of the lower bound of the remaining available process life to the bending frequency of the current shift. Half of the estimated remaining available time is determined as the latest mold change start time. Based on the latest mold change start time and the spare punch number, the mold change mechanism action timing control instruction is generated and written into the mold change mechanism. The remaining process life is compared with the confidence lower bound and the second preset intervention threshold. When the confidence lower bound of the remaining process life is not greater than the second preset intervention threshold, a slider locking stop control command is output to the slider locking mechanism to drive the slider locking mechanism to lock the slider to a safe height position. The initial compensation baseline and geometric parameters of the spare punch are read from the punch file. The geometric parameters are written into the punch parameter register of the CNC unit, and the initial compensation baseline is written into the compensation reference register of the servo driver. After the punch is switched, the graded automatic control parameters of the bending process of the bending machine are reset.

[0037] Specifically, both the first and second preset intervention thresholds are threshold values ​​in the dimension of bending times. The first preset intervention threshold is greater than the second preset intervention threshold. They divide the remaining usable process life confidence lower bound into three intervals to correspond to three levels of intervention intensity. It is recommended that the first preset intervention threshold be three times the average bending output of the current shift. The physical meaning is that the punch will reach its life limit within three shifts. At this time, the punch still has the ability to accept low-precision work orders. The precision level limit range is determined by the process engineer based on the actual compensation precision margin calculated according to the current degradation rate and written into the punch file. After the punch scheduling control instruction is written into the work order scheduling module, the work order scheduling module filters the work order queue according to the precision level limit range in the instruction and prohibits high-precision work orders with target angle tolerance bands smaller than the lower limit of the precision level limit range from being assigned to the punch. The second preset intervention threshold is recommended to be the average bending output of the current shift. The physical meaning is that the punch will reach its life limit within a shift. At this time, the die change preparation should be started immediately. The bending frequency of the current shift is calculated by dividing the cumulative number of bends from the start of the current shift to the current time by the corresponding duration. The remaining usable process life confidence lower bound is divided by the bending frequency of the current shift to obtain the expected remaining usable time. Taking half of this time as the basis for the latest die change start time is because the die change operation itself requires a certain time window to complete. Taking half can ensure that the die change operation has enough buffer time before the punch life is exhausted.

[0038] The spare punch number is automatically matched by the tool database when receiving the timing control command for the die-changing mechanism. The matching rule is to retrieve the punch number from the tool database that has the same specifications and model as the current punch and the lowest cumulative bending count. The matching result is written as the spare punch number into the timing control command for the die-changing mechanism. The die-changing mechanism starts the die-changing action at the corresponding time according to the latest die-changing start time in the command and the spare punch number. When the lower bound of the remaining usable process life is not greater than the second preset intervention threshold, a slider locking stop control command is output to the slider locking mechanism. After the current bending stroke ends, the slider locking mechanism locks the slider to a safe height position. The safe height position is the position where the net distance between the bottom of the punch and the top surface of the die is not less than the maximum thickness of the sheet metal. This position is calibrated and stored in the CNC unit by the bending machine manufacturer at the factory. After the punch switching is completed, the geometric parameters of the spare punch are read from the punch file and written into the punch parameter register of the CNC unit. The CNC unit recalculates the stroke displacement reference value and bending pressure reference value of each bending process according to the new geometric parameters. The initial compensation baseline of the spare punch is written into the compensation reference register of the servo drive. The servo drive re-executes the closed-loop feedback superposition of the angle compensation execution amount with the new initial compensation baseline as the starting point. The original sequence of compensation amount is re-initialized from the zero value of the cumulative bending number of the die with the spare punch number as the new index.

[0039] The automatic control method for sheet metal processing equipment in the embodiments of this application has been described above. The automatic control system for sheet metal processing equipment in the embodiments of this application is described below. One embodiment of the automatic control system for sheet metal processing equipment in the embodiments of this application includes: The generation module is used to take the deviation between the measured bending angle collected by the angle sensor and the target angle command as the angle compensation execution amount after each bending stroke, bind it with the cumulative bending number of the mold and write it into the storage unit, arrange it in ascending order according to the cumulative bending number of the mold, construct the original sequence of compensation amount, and superimpose the angle compensation execution amount to the next stroke displacement command of the slider in a closed loop feedback manner to generate compensation control parameters. The drive module is used to perform moving median filtering on the original compensation quantity sequence, extract the trend degradation component, subtract the original compensation quantity sequence from the trend degradation component to obtain the random fluctuation component, generate a servo motor compensation adjustment control command based on the random fluctuation component, and drive the slider servo motor to perform position closed-loop control. The analysis module is used to establish a linear regression relationship between the trend degradation component and the cumulative number of bends of the die, extract the degradation rate and the initial compensation baseline, calculate the confidence lower bound of the remaining process usable life by combining the pre-stored compensable limit threshold in the punch file, and generate slider servo motor stroke control command and bending pressure control command based on the degradation rate. The control module is used to generate punch scheduling control commands, die changing mechanism action timing control commands, or slider locking stop control commands in a hierarchical manner based on the preset range where the lower limit of the remaining process usable life is located, so as to complete the hierarchical automatic control of the bending process of the bending machine.

[0040] This invention also provides an automatic control device for sheet metal processing equipment, which can be a server. The automatic control device includes a processor, memory, display screen, input device, network interface, and database connected via a system bus. The processor, designed as a computer, provides computing and control capabilities. The memory of the automatic control device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database of the automatic control device stores the data corresponding to this embodiment. The network interface of the automatic control device is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements the above-described method.

[0041] The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when the instructions are executed on a computer, cause the computer to perform the steps of the automatic control method of the sheet metal processing equipment.

[0042] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0043] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the existing solution, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause an automatic control device (which may be a personal computer, server, or network device, etc.) of a sheet metal processing equipment to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0044] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An automatic control method for sheet metal processing equipment, characterized in that, The method includes: Step S1: After each bending stroke, the deviation between the measured bending angle collected by the angle sensor and the target angle command is used as the angle compensation execution amount. It is bound to the cumulative bending number of the mold and written into the storage unit. The original sequence of compensation amount is constructed by arranging the cumulative bending number of the mold in ascending order. The angle compensation execution amount is then superimposed on the next stroke displacement command of the slider in a closed-loop feedback manner to generate compensation control parameters. Step S2: Perform moving median filtering on the original compensation amount sequence to extract the trend degradation component. Subtract the original compensation amount sequence from the trend degradation component to obtain the random fluctuation component. Generate a servo motor compensation adjustment control command based on the random fluctuation component to drive the slider servo motor to perform position closed-loop control. Step S3: Establish a linear regression relationship between the trend degradation component and the cumulative bending number of the mold, extract the degradation rate and the initial compensation baseline, calculate the confidence lower bound of the remaining process usable life by combining the pre-stored compensable limit threshold in the punch file, and generate the slider servo motor stroke control command and bending pressure control command based on the degradation rate. Step S4: Based on the preset range of the confidence lower bound of the remaining process life, generate hierarchical punch scheduling control instructions, die changing mechanism action timing control instructions, or slider locking stop control instructions to complete the hierarchical automatic control of the bending process of the bending machine.

2. The automatic control method for sheet metal processing equipment according to claim 1, characterized in that, Step S1 includes: The angle compensation execution amount is obtained by subtracting the measured bending angle collected by the angle sensor after each bending stroke from the target angle command stored in the CNC unit. The measured bending angle is the stable angle value when the plate elastically recovers after the punch is unloaded, and the target angle command is the design angle corresponding to the current bending process. The angle compensation execution amount is bound to the cumulative number of times the mold is bent at the end of the current bending stroke by the bending machine counter in the form of key-value pairs. The bound key-value pairs are written into the storage unit and sorted in ascending order with the cumulative number of times the mold is bent as the sorting index to obtain the original sequence of compensation amount. The angle compensation execution amount is superimposed on the next stroke displacement command of the slider, wherein the superposition direction is opposite to the sign of the angle compensation execution amount, and the superimposed stroke displacement command is used as the compensation control parameter and written into the position control register of the servo driver. The compensation control parameters are converted into drive pulse signals by the servo driver and output to the slider servo motor. The slider is driven to perform the next bending action according to the stroke displacement corresponding to the compensation control parameters. After the next bending stroke is completed, the angle sensor is triggered again to collect the measured bending angle and update the original sequence of compensation amount.

3. The automatic control method for sheet metal processing equipment according to claim 1, characterized in that, Step S2 includes: Using the original sequence of compensation amounts as input, the window length of the moving median filter is set, and the median of the angle compensation of the continuous window length starting from the current data point in the original sequence of compensation amounts is taken. The median output values ​​of each window position are arranged in ascending order according to the cumulative bending number of the mold to obtain the trend degradation component. The angle compensation execution amount of each data point in the original compensation amount sequence is subtracted from the value of the corresponding position in the trend degradation component to obtain the random fluctuation component. A normality test is performed on all data points of the random fluctuation component, and the test statistic is compared with a preset normality threshold. When the test statistic is lower than the preset normality threshold, an alarm signal for abnormal performance dispersion of the sheet material batch is output to the CNC unit. The standard deviation is calculated based on all data points of the random fluctuation component. The standard deviation is compared with a preset fluctuation threshold. When the standard deviation exceeds the preset fluctuation threshold, the difference between the standard deviation and the preset fluctuation threshold is used as the compensation adjustment amount correction amount to generate a servo motor compensation adjustment amount control command. The servo motor compensation adjustment control command is written into the servo driver, which drives the slider servo motor to perform position control according to the upper and lower limits of the adjustment amount corresponding to the servo motor compensation adjustment control command, thereby completing the adjustment of the angle deviation caused by random fluctuation components in the current bending process.

4. The automatic control method for sheet metal processing equipment according to claim 1, characterized in that, In step S3, a linear regression relationship is established based on the trend degradation component and the cumulative bending number of the mold to extract the degradation rate and the initial compensation baseline, including: Using the values ​​of each data point of the trend degradation component as the dependent variable and the cumulative number of bends of the mold corresponding to each data point as the independent variable, least squares linear regression is performed on all data points of the trend degradation component. The slope of the regression line is determined as the degradation rate, and the intercept of the regression line is determined as the initial compensation baseline. The difference between each data point of the trend degradation component and the corresponding value of the regression line is calculated to obtain the regression residual. The goodness of fit is calculated based on the regression residual. The goodness of fit is compared with a preset fitting threshold. When the goodness of fit is lower than the preset fitting threshold, all data points of the trend degradation component are switched to quadratic polynomial fitting. The coefficient of the first term obtained by refitting is determined as the degradation rate. The constant term obtained by refitting is determined as the initial compensation baseline. Write the degradation rate, the initial compensation baseline, and the regression residual into the convex mold file.

5. The automatic control method for sheet metal processing equipment according to claim 4, characterized in that, In step S3, the confidence lower bound of the remaining usable process life is calculated by combining the pre-stored compensable limit threshold in the punch file, including: Read the compensable limit threshold, the degradation rate, the initial compensation baseline, and the current cumulative number of bends of the die from the punch file. Divide the difference between the compensable limit threshold and the initial compensation baseline by the degradation rate, and then subtract the current cumulative number of bends of the die to obtain the remaining usable process life. The regression residual is read from the punch file, the standard deviation of the regression residual is calculated based on the regression residual, the standard deviation of the regression residual is divided by the degradation rate to obtain the predicted standard deviation of the remaining process usable life, and the predicted standard deviation of the remaining process usable life is multiplied by a preset confidence coefficient to obtain the prediction deviation. The remaining usable process life is subtracted from the predicted deviation to obtain the lower confidence bound of the remaining usable process life. The lower confidence bound of the remaining usable process life and the width of the confidence interval are written into the punch file.

6. The automatic control method for sheet metal processing equipment according to claim 5, characterized in that, Step S3, which generates slider servo motor stroke control commands and bending pressure control commands based on the degradation rate, includes: The remaining usable process life confidence lower bound is read from the punch file. The remaining usable process life confidence lower bound is compared with the preset life threshold. When the remaining usable process life confidence lower bound is lower than the preset life threshold, the trend degradation compensation amount is calculated based on the product of the degradation rate and the current cumulative number of bends of the die. The trend degradation compensation amount is superimposed on the stroke displacement reference value to obtain the corrected stroke displacement amount. The corrected stroke displacement amount is written into the position control register of the servo driver to generate the slider servo motor stroke control command. The bending pressure correction amount is calculated based on the degradation rate and the current cumulative number of bends of the mold. The bending pressure correction amount is then added to the bending pressure reference value to obtain the corrected bending pressure value. The corrected bending pressure value is then written into the hydraulic system pressure control valve to generate a bending pressure control command.

7. The automatic control method for sheet metal processing equipment according to claim 1, characterized in that, Step S4 includes: The remaining process life confidence lower bound is read from the punch file, and the remaining process life confidence lower bound is compared with the first preset intervention threshold. When the remaining process life confidence lower bound is greater than the first preset intervention threshold, a punch scheduling control instruction is generated based on the current punch accuracy level limit range. The punch scheduling control instruction is written into the work order scheduling module to restrict the punch to only accept bending work orders corresponding to the accuracy level limit range. The remaining usable life of the process is compared with the second preset intervention threshold. When the remaining usable life of the process is greater than the second preset intervention threshold and not greater than the first preset intervention threshold, the estimated remaining usable time is calculated based on the ratio of the remaining usable life of the process to the bending frequency of the current shift. Half of the estimated remaining usable time is determined as the latest mold change start time. Based on the latest mold change start time and the spare punch number, a mold change mechanism action timing control instruction is generated and written into the mold change mechanism. The remaining process lifespan confidence lower bound is compared with the second preset intervention threshold. When the remaining process lifespan confidence lower bound is not greater than the second preset intervention threshold, a slider locking stop control command is output to the slider locking mechanism to drive the slider locking mechanism to lock the slider to a safe height position. The initial compensation baseline and geometric parameters of the spare punch are read from the punch file. The geometric parameters are written into the punch parameter register of the CNC unit, and the initial compensation baseline is written into the compensation reference register of the servo driver. The automatic control parameters of the bending process of the bending machine are reset after the punch is switched.

8. An automatic control system for sheet metal processing equipment, characterized in that, An automatic control method for implementing sheet metal processing equipment as described in any one of claims 1-7, wherein the automatic control system of the sheet metal processing equipment comprises: The generation module is used to take the deviation between the measured bending angle collected by the angle sensor and the target angle command as the angle compensation execution amount after each bending stroke, bind it with the cumulative bending number of the mold and write it into the storage unit, arrange it in ascending order according to the cumulative bending number of the mold, construct the original sequence of compensation amount, and superimpose the angle compensation execution amount to the next stroke displacement command of the slider in a closed loop feedback manner to generate compensation control parameters. The drive module is used to perform moving median filtering on the original compensation quantity sequence, extract the trend degradation component, subtract the original compensation quantity sequence from the trend degradation component to obtain the random fluctuation component, generate a servo motor compensation adjustment control command based on the random fluctuation component, and drive the slider servo motor to perform position closed-loop control. The analysis module is used to establish a linear regression relationship between the trend degradation component and the cumulative number of bends of the die, extract the degradation rate and the initial compensation baseline, calculate the confidence lower bound of the remaining process usable life by combining the pre-stored compensable limit threshold in the punch file, and generate slider servo motor stroke control command and bending pressure control command based on the degradation rate. The control module is used to generate punch scheduling control commands, die changing mechanism action timing control commands, or slider locking stop control commands in a hierarchical manner based on the preset range where the lower limit of the remaining process usable life is located, so as to complete the hierarchical automatic control of the bending process of the bending machine.

9. An automatic control device for sheet metal processing equipment, characterized in that, The device includes a memory and a processor, the memory storing a computer program that can run on the processor, and the processor executing the computer program to implement the automatic control method for the sheet metal processing equipment according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is run by the processor, it causes the processor to execute the automatic control method for the sheet metal processing equipment as described in any one of claims 1 to 7.

Images