Intelligent servo control system of multi-shaft synchronous driving metal bending forming machine

Abstract

The invention relates to the technical field of metal processing control, in particular to an intelligent servo control system of a multi-shaft synchronous driving metal bending forming machine. The data acquisition module is used for acquiring a forming angle error of a historical metal plate during each bending and processing parameters of the current metal plate, and providing a data basis for subsequent springback analysis; the springback characteristic analysis module is used for mining springback resistance characteristics of metal under a complex structure based on historical data, determining the overall springback resistance degree during each bending, and mapping and predicting a current forming angle error under the current bending times; the error correction module is combined with the current process parameters to calculate the predicted springback value, the corrected springback value is determined through comparative analysis of the current forming angle error and the predicted springback value, and the overcompensation risk is avoided; and finally, the machining pressure control module dynamically adjusts the bending shaft pressure according to the corrected springback value, and the machining precision is remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of metal processing control, and in particular to an intelligent servo control system for a multi-axis synchronously driven metal bending and forming machine. Background Art

[0002] With technological advancements, mechanical design has become increasingly complex. Metal bending machines are devices that bend metal sheets or tubes into specific shapes using dies, pressure, or rolling. They can produce metal parts that meet specific requirements based on complex design drawings. Multi-axis synchronously driven bending machines achieve precise bending of metal sheets through the coordinated operation of multiple motion axes. During the bending process, precise control of the punch's trajectory and applied pressure is required to ensure the quality of the final product.

[0003] There is stress inside the metal sheet, and the metal itself has a certain toughness, so after it is bent, it will produce a certain amount of rebound under the action of the internal stress and toughness of the metal. The existing technology only calculates the applied pressure based on a preset rebound angle when controlling the pressure of the bending axis (the axis that provides pressure) of the metal bending machine; however, as the shape of the metal becomes more complex, its internal metal structure will provide a certain degree of stability for the metal, making the degree of rebound smaller. Therefore, if the compensation pressure is still calculated based on a fixed rebound angle, over-compensation will occur, resulting in quality problems such as over-bending in the final product. Summary of the Invention

[0004] In order to solve the technical problem that as the shape of metal becomes more complex, its internal metal structure will provide a certain degree of stability for the metal, making the rebound degree smaller, and therefore if the compensation pressure is still calculated based on a fixed rebound angle, over-compensation will occur, resulting in the final product having a quality problem of overbending (the specific problem needs to be written), the purpose of the present invention is to provide a multi-axis synchronous drive metal bending forming machine intelligent servo control system, the technical solution adopted is as follows:

[0005] The present invention proposes an intelligent servo control system for a multi-axis synchronously driven metal bending machine, the system comprising:

[0006] A data acquisition module is used to obtain the preset bending angle of each historical metal sheet and the actual bending angle after each bending and calculate the historical forming angle error based on the error; obtain the processing parameters of the current metal sheet during the current bending;

[0007] The springback characteristic analysis module is used to calculate the overall springback resistance of each historical metal sheet during the same bending cycle based on the differences and fluctuation characteristics between the historical forming angle errors. The target value is determined among all the overall springback resistances based on the number of bends of the current metal sheet, thereby obtaining the current forming angle error.

[0008] An error correction module is configured to obtain a predicted springback amount of the current bend according to the current processing parameters of the metal sheet; and determine a corrected springback amount based on the difference between the current forming angle error and the predicted springback amount;

[0009] The processing pressure control module is used to control the processing pressure of the bending axis of the bending forming machine based on the corrected springback amount under the current number of bending times of the current metal sheet.

[0010] Furthermore, the method for obtaining the historical forming angle error includes:

[0011] The difference between the preset bending angle of each historical metal sheet at each bending and the corresponding actual bending angle is used as the historical forming angle error.

[0012] Furthermore, the method for obtaining the overall rebound resistance includes:

[0013] Select any bending process as the target bending process;

[0014] Under the target bending process, the historical forming angle error of each historical metal sheet is numerically adjusted based on the hyperbolic tangent function to obtain the springback resistance index;

[0015] Analyze the differences and fluctuation characteristics between the springback resistance indicators of all historical metal sheets under the target bending process, and determine the springback consistency of each historical metal sheet under the target bending process;

[0016] Determine a reference weight for each historical metal sheet based on the overall characteristics of the springback consistency of each historical metal sheet under all bending processes;

[0017] The reference weights of the historical metal sheets are used to perform weighted averaging on the springback resistance indices of the historical metal sheets under the target bending process, and the weighted results are used as the overall springback resistance of the historical metal sheets under the target bending process.

[0018] Furthermore, the method for obtaining the rebound consistency includes:

[0019] Under the target bending process, among the springback resistance indices of all historical metal sheets, the standard score of the springback resistance index of each historical metal sheet is calculated, and the absolute value of the standard score is negatively correlated mapped and normalized as the springback consistency of each historical metal sheet under the target bending process.

[0020] Furthermore, the method for obtaining the reference weight includes:

[0021] The average of the springback consistency of each historical metal sheet under all bending processes is used as the reference weight of each historical metal sheet.

[0022] Furthermore, the method for obtaining the current forming angle error includes:

[0023] Among all the overall springback resistances, the overall springback resistance with the same number of bends as the current metal sheet is taken as the target value of the current metal sheet;

[0024] The target value is calculated using the inverse function of the hyperbolic tangent function to obtain the current forming angle error of the current metal sheet under the current bending condition.

[0025] Furthermore, the method for obtaining the predicted rebound amount includes:

[0026] Under simulation conditions, multiple sets of simulation processing parameters and corresponding simulation springback amounts are obtained;

[0027] The simulated machining parameters are used as input and the simulated springback amount is used as output to train the neural network, thereby obtaining a springback prediction model.

[0028] The processing parameters of the current metal sheet under the current number of bending times are used as the input of the springback prediction model, and the predicted springback amount is output.

[0029] Furthermore, the neural network adopts an SSA-LSTM network.

[0030] Furthermore, the method for obtaining the corrected rebound amount includes:

[0031] The ratio of the difference between the predicted springback amount and the current forming angle error to the predicted springback amount is used as the adjustment ratio;

[0032] The product of the adjustment ratio and the predicted springback amount is used as the corrected springback amount.

[0033] Furthermore, the controlling of the processing pressure of the bending axis of the bending forming machine based on the corrected springback amount under the current number of bending times of the current metal sheet includes:

[0034] The correction pressure of the bending axis is determined according to the preset pressure of the bending axis, the correction rebound amount, and the dynamic rebound coefficient. The formula model of the correction pressure includes:

[0035]

[0036] Among them, h b Indicates the correction pressure of the bending axis; h s Indicates the preset pressure of the bending axis; K indicates the dynamic rebound coefficient; Δθ indicates the corrected rebound amount; θ s Indicates the predicted springback amount of the current bend;

[0037] Under the current bending times of the current metal sheet, the processing pressure of the bending axis is set to the corresponding correction pressure so as to perform bending processing on the current metal sheet.

[0038] The present invention has the following beneficial effects:

[0039] In the data acquisition module, the historical forming angle errors in the historical metal sheet processing process are obtained to provide a data basis for the subsequent analysis process, and the processing parameters of the current metal sheet during the current bending are also obtained. Due to its own characteristics, metal parts will rebound when subjected to pressure bending. However, as the shape of the metal becomes more complex, its internal metal structure will provide a certain degree of stability for the metal, that is, it will have a certain degree of resistance to rebound. In order to avoid the occurrence of over-bending caused by over-compensation of pressure, the rebound feature analysis module analyzes the differences and fluctuation characteristics of the historical forming angles of all historical metal sheets, determines the overall rebound resistance of the historical metal sheets at each bending, and maps to determine the current forming angle error of the current metal sheet. The current forming angle error is used to characterize the possible bending angle of the current metal sheet under the current bending. Furthermore, in the error correction module, the present invention obtains the predicted springback amount of the current bend based on the current metal sheet processing parameters, analyzes the difference between this predicted springback amount and the current forming angle error, and determines the corrected springback amount. This corrected springback amount takes into account the metal's own springback compensation during the springback process, thus more accurately reflecting the potential springback conditions of the current metal sheet. Therefore, in the final processing pressure control module, the processing pressure of the bending machine's bending axis is controlled based on the corrected springback amount, effectively improving the accuracy of the processing pressure and greatly reducing the occurrence of over-bending in the finished product. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0041] Figure 1 This is a system block diagram of an intelligent servo control system for a multi-axis synchronously driven metal bending and forming machine provided by one embodiment of the present invention;

[0042] Figure 2 This is a flow chart of a method for obtaining overall rebound resistance provided by one embodiment of the present invention;

[0043] Figure 3 This is a schematic diagram of the angular relationship between the predicted springback amount and the current forming angle error provided by one embodiment of the present invention;

[0044] Figure 4 The present invention is a schematic diagram of the system structure of an intelligent servo control system for a multi-axis synchronously driven metal bending and forming machine provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0045] To further illustrate the technical means and effectiveness of the present invention in achieving its intended objectives, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail the specific implementation, structure, features, and effectiveness of an intelligent servo control system for a multi-axis synchronously driven metal bending machine according to the present invention. In the following description, references to "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0046] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0047] The specific scheme of the intelligent servo control system for a multi-axis synchronously driven metal bending and forming machine provided by the present invention is described in detail below with reference to the accompanying drawings.

[0048] See also Figure 1 , which shows a system block diagram of a multi-axis synchronously driven metal bending forming machine intelligent servo control system provided by an embodiment of the present invention. The system includes: a data acquisition module 101, a rebound feature analysis module 102, an error correction module 103, and a processing pressure control module 104.

[0049] The data acquisition module 101 is used to obtain the preset bending angle of each historical metal sheet and the actual bending angle after each bending and obtain the historical forming angle error based on the error; and obtain the processing parameters of the current metal sheet during the current bending.

[0050] A metal bending machine is a device that bends metal sheets or tubes into specific shapes using dies, pressure, or rolling. Multi-axis synchronously driven bending machines achieve precise bending of metal sheets through the coordinated operation of multiple motion axes. To achieve the complex shapes corresponding to the machine's design drawings, precise control of the punch's trajectory and applied pressure is required, necessitating a more accurate intelligent servo control system.

[0051] The intelligent servo control system is a fully digital drive that integrates servo drive technology, PLC technology, and motion control technology. It can achieve high-precision, high-dynamic response motion control through algorithm optimization and adaptive adjustment. In this process, a multi-level feedback network is formed through technologies such as grating scales, magnetic scales, and laser displacement sensors to describe the synergistic effect of axes such as the feed axis (X), bending axis (Y), and rotation axis (C). Then, based on the dynamic relationship between the master and slave axes and generalized predictive control technology, the multi-axis synchronous drive of the metal bending machine is realized. Among them, the feed axis is responsible for controlling the horizontal movement of the metal sheet, the bending axis is the core axis for performing the bending action, and is used to drive the bending die to apply pressure to the metal sheet to cause it to deform, thereby achieving the desired bending angle. The rotation axis is used to adjust the angle of the bending die.

[0052] After the metal parts are designed, the metal bending machine will formulate the corresponding processing parameters according to the design requirements, and then the metal bending machine will process the metal sheet according to the preset processing parameters; therefore, in the data acquisition module, the processing parameters of the current metal sheet during the current bending can be obtained first. The processing parameters can specifically include the processing pressure of each axis (feeding axis (X), bending axis (Y), rotation axis (C)), preset bending angle, advancement speed, etc., which can all be obtained through the built-in FPGA chip of the servo driver, the sensor on the bending machine robot arm, and the laser rangefinder.

[0053] However, after the metal sheet is processed, due to the elasticity and other properties of the metal, the metal parts will rebound at a certain angle after leaving the metal bending machine. Therefore, in the actual processing process, in order to compensate for the rebound characteristics, greater pressure is often applied to make the metal sheet bend at a larger angle, and then through the rebound of the metal, the bending angle of the metal is finally returned to the preset bending angle. However, as the structure of the metal parts becomes increasingly complex, the various parts of the metal sheet are pulled by the existing structure, causing the rebound angle to gradually become smaller. If a fixed compensation pressure is continued to be used at this time, the metal will bend excessively due to insufficient rebound.

[0054] Therefore, in this module, the actual bending angles (after the springback process) and preset bending angles of multiple historical metal sheets after each bending during the historical processing process are also obtained (the preset bending angles are obtained according to the requirements of the metal parts, and the actual bending angles can be obtained through later measurements). Based on the error between the two, the historical forming angle error of each historical metal sheet after each bending is determined. The historical forming angle error can be used to reflect the angle deviation caused by the pressure applied during each historical bending under the characteristics of metal springback. Based on the data of these historical metal sheets, the rebound characteristics of the metal can be analyzed in the subsequent process, so as to adjust the construction pressure of the current metal sheet under the current number of bending times.

[0055] Preferably, in one embodiment of the present invention, the method for obtaining the historical forming angle error includes:

[0056] The difference between the preset bending angle and the corresponding actual bending angle of each historical metal sheet at each bending is taken as the historical forming angle error. When the historical forming angle error is positive, it means that the pressure is too large, resulting in the final historical metal sheet product being over-bending; conversely, when the historical forming angle error is negative, it means that the rebound is too large and the pressure is too small, resulting in the final historical metal sheet product being too small in curvature; and when the historical forming angle is 0, it means that the pressure is just enough to make the final historical metal sheet product meet the desired requirements.

[0057] It should be noted that, in this embodiment of the present invention, it is believed that the internal structure of the metal sheet will provide a certain stabilizing effect on the metal, so that the degree of rebound becomes smaller. Therefore, over-compensation will occur when calculating the pressure based on a fixed rebound angle, resulting in quality problems of over-bending in the final metal product. Therefore, the default values ​​of the historical forming angles are all non-negative numbers, so the historical forming angle error specifically characterizes the bending angle of each historical metal sheet after each bending.

[0058] In this embodiment of the present invention, the type of metal sheet should remain consistent.

[0059] The springback characteristic analysis module 102 is used to calculate the overall springback resistance of all historical metal sheets during each bending based on the differences and fluctuation characteristics between the historical forming angle errors for all historical metal sheets under the same bending; and to determine the target value among all the overall springback resistances according to the number of bending times of the current metal sheet, thereby obtaining the current forming angle error.

[0060] After the mechanical axis exits the bending process, the metal sheet will rebound to a certain extent, resulting in the bending angle not reaching the required design angle. Therefore, according to these rebound conditions, the bending pressure (compensation pressure) is further formulated, and then under the action of the preset pressure and the compensation pressure, the metal sheet is bent to the required design angle. However, in the past processing process, the compensation of the processing pressure of the metal sheet was directly obtained by relying on the rebound angle predicted by simulation. However, the rebound of the metal will be affected by the structure of the metal sheet. Therefore, when the structure of the metal parts becomes more and more complex, the various parts of the metal sheet are pulled by the existing structure, resulting in its rebound angle gradually becoming smaller, so over-bending often occurs. This is because in the simulation prediction, most of the rebound conditions caused by the material itself are considered, and the rebound is not considered. The metal structure causes the metal's springback to fail to meet expectations. In this process, the smaller the over-bending angle is, the more similar the springback after this bending is to the predetermined springback, and the less the metal's springback characteristics are affected by the structure of the metal component. Therefore, in this module, for all historical metal sheets, under the same bending, the differences and fluctuation characteristics between the historical forming angle errors are analyzed, and the overall springback resistance of the historical metal sheets at each bending is calculated to reflect the influence of metal structural components on the springback characteristics; then, according to the number of bending times of the metal component, the target value is determined in all the overall springback resistances, so as to calculate the current forming angle error of the current metal sheet under the current number of bendings, and the current forming angle error reflects the possible over-bending angle under the current number of bendings.

[0061] First, the differences and fluctuation characteristics of the historical forming angle errors of the historical metal sheets under the same bending are analyzed to calculate the overall springback resistance of the historical metal sheets during each bending.

[0062] Preferably, in one embodiment of the present invention, the method for obtaining the overall rebound resistance includes:

[0063] See also Figure 2 , which shows a flow chart of a method for obtaining overall rebound resistance in one embodiment of the present invention, the method comprising the following steps:

[0064] Step S201: a bending process is selected as a target bending process. Under the target bending process, a historical forming angle error of each historical metal sheet is numerically adjusted to obtain a springback resistance index.

[0065] In order to facilitate subsequent explanation and illustration, one bending process is selected from all the bending processes of all historical metal sheets as the target bending process.

[0066] Based on the analysis in step S1, it can be seen that when the historical forming angle error is positive and the larger it is, it means that the pressure is too high, resulting in the final historical metal sheet product being over-bent, that is, the internal structure of the metal is more resistant to the rebound characteristics. Therefore, in the target bending process, the historical forming angle error of each historical metal sheet is numerically adjusted based on the hyperbolic tangent function to achieve positive proportional normalization and obtain the resistance to rebound index. At this time, the larger the resistance to rebound index is, the greater the resistance of the internal structure of the metal to the rebound characteristics.

[0067] Step S202: Analyze the differences and fluctuation characteristics between the springback resistance indices of all historical metal sheets under the target bending process, and determine the springback consistency of each historical metal sheet under the target bending process.

[0068] Although over-bending is a common phenomenon, there are always some cases due to processing errors. For example, when a metal sheet has some internal defects, the rebound of its bending from the initial bending will be different from that of other metal sheets. In this case, the springback resistance of the metal sheet is inaccurate, and its reference value needs to be appropriately reduced.

[0069] Therefore, under the target bending process, the differences and fluctuation characteristics between the springback resistance indicators of all historical metal sheets are analyzed, and the springback consistency of each historical metal sheet under the target bending process is determined, which facilitates the determination of the reference weight of each historical metal sheet in the subsequent process.

[0070] Under the target bending process, among all the historical springback resistance indices of the metal sheets, the standard score of the springback resistance index of each historical metal sheet is calculated. In a set of data, the standard score is calculated as the quotient obtained by dividing the difference between the original data and the overall mean by the standard deviation. The standard deviation is used as the unit to measure how many standard deviations the original data is above its mean, or how many standard deviations it is below the mean. The larger the value is, the higher the position is relative to the mean. Conversely, the smaller the value is, the lower the position is relative to the mean. In this embodiment of the present invention, the main purpose is to reflect the fluctuation characteristics, so The absolute value of the standard score of the springback resistance index of each historical metal sheet under the target bending process is obtained. At this time, the larger the absolute value, the greater the degree of deviation from the average level, the stronger the extremeness, and the lower the consistency. Therefore, the absolute value is negatively correlated and normalized to achieve logical relationship correction, and the springback consistency of each historical metal sheet under the target bending process is obtained. At this time, the greater the springback consistency, the higher the consistency of the springback resistance index of a historical metal sheet with other historical metal sheets under the same bending process, and thus its reference value in subsequent calculations will be higher. The negative correlation mapping and normalization processing here can be used with the formula exp(-x), where exp() represents an exponential function with the natural constant e as the base, and x represents the independent variable.

[0071] It should be noted that, in this embodiment of the present invention, it is considered that the value of the standard deviation cannot be 0.

[0072] At this point, through the above process, the springback consistency of each historical metal sheet in each bending process can be obtained.

[0073] Step S203: determining a reference weight of each historical metal sheet based on the overall characteristics of the springback consistency of each historical metal sheet in all bending processes.

[0074] Based on the analysis in step S202, it can be seen that when the rebound consistency of a historical metal sheet in a certain bending process is greater, it means that its similarity with the resistance to rebound of other historical metal sheets in the same bending process is higher, then the possibility of abnormality will be reduced, and the reference value will be improved. Therefore, the average of the rebound consistency of each historical metal sheet in all bending processes is used as the reference weight of each historical metal sheet. The larger the reference weight, the more its resistance to rebound index can represent the resistance to rebound of the metal in the historical processing process, and the higher the credibility.

[0075] Step S204: integrating the reference weights of the historical metal sheets and the springback resistance indices of all the historical metal sheets under the target bending process, thereby determining the overall springback resistance of the historical metal sheets under the target bending process.

[0076] After the above steps, each historical metal sheet now has a reference weight, so the reference weight can be used to integrate the springback resistance indicators of all historical metal sheets under the target bending process, thereby obtaining the overall springback resistance of the historical metal sheets under the target process.

[0077] The reference weights of historical metal sheets are used to weightedly average the springback resistance indices of historical metal sheets under the target bending process. Under the target bending process, the reference weight of each historical metal sheet is multiplied by the springback resistance index. The larger the product, the higher the reference value of the historical metal sheet and the stronger the springback resistance. The sum of the products corresponding to all historical metal sheets, that is, the final weighted result, is used as the overall springback resistance of the historical metal sheets under the target bending process. The greater the overall springback resistance, the greater the resistance of the internal structural characteristics of the metal parts to springback under the target bending process of the historical processing process.

[0078] At this point, by analyzing the fluctuations and difference characteristics between the historical forming angle errors of the historical metal sheets under the same bending, the overall springback resistance of the historical metal sheets during each bending can be obtained. This index uses the reference weight as a confidence index to adjust the springback resistance index of the historical metal sheets, so it can more accurately reflect the degree of resistance of the internal structural characteristics of the metal parts to springback. At the same time, given that there is a certain functional mapping relationship between the springback resistance index and the historical forming angle error in this embodiment of the present invention, the target value can be determined in all overall springback resistances according to the number of bending times of the current metal sheet, so as to be used to predict and calculate the current forming angle error of the current metal sheet under the current bending.

[0079] Preferably, in one embodiment of the present invention, the method for obtaining the current forming angle error includes:

[0080] Based on the above steps, in the historical processing process, each bend has an overall springback resistance, so among all the overall springback resistances, the overall springback resistance with the same number of bends as the current metal sheet is used as the target value of the current metal sheet.

[0081] In view of the fact that in this embodiment of the present invention, the mapping relationship between the historical metal sheet's resistance to rebound index and the historical forming angle error is provided by the hyperbolic tangent function, the inverse function of the hyperbolic tangent function is used to calculate the target value to obtain the current forming angle error of the current metal sheet under the current bending. The current forming angle error is used to characterize the possible bending angle of the current metal sheet under the current bending.

[0082] The error correction module 103 is used to obtain the predicted springback amount of the current bend according to the current processing parameters of the metal sheet; and determine the corrected springback amount based on the difference between the current forming angle error and the predicted springback amount.

[0083] The processing pressure of metal parts is determined based on the pre-calculated springback angle. In the previous steps, the possible bending angle of the current metal sheet under the current bending is calculated and predicted, that is, the current forming angle error. This value is the deviation of the metal part structure that causes the springback of the metal sheet to fail to reach the original springback angle, that is, the reduced springback amount. Therefore, the corrected springback amount can be determined based on the difference between the current forming angle error and the pre-calculated springback angle.

[0084] Prior to this, it is necessary to first determine the pre-calculated rebound angle. Given that after the metal parts are designed, the metal bending forming machine will specify the corresponding processing parameters according to the design requirements, and after the processing is completed, the metal sheet will have a certain rebound after leaving the metal bending forming machine, so in this embodiment of the present invention, it is believed that there is a certain correspondence between the processing parameters of the metal bending forming machine and the rebound angle, so a rebound prediction model can be constructed based on this, so as to obtain the predicted rebound amount of the current bend according to the processing parameters of the current metal sheet under the current number of bends.

[0085] Preferably, in one embodiment of the present invention, the method for obtaining the predicted rebound amount includes:

[0086] Under simulation conditions, combined with the design drawings of the metal parts (bending angle, inner arc radius, and plate thickness, etc.), the motion trajectory of each axis of the metal bending machine is generated through B-spline curves. At the same time, the bending process is simulated using the finite element method. In this process, the possible metal springback under various processing parameters is simulated, thereby obtaining multiple sets of simulated processing parameters and the corresponding simulated springback amounts.

[0087] Then the simulated processing parameters are used as input and the simulated springback amount is used as output to train the SSA-LSTM neural network, and the trained model is used as the springback prediction model.

[0088] Finally, the processing parameters of the current metal sheet under the current number of bending times are used as the input of the springback prediction model to output the predicted springback amount.

[0089] It should be noted that the training process of a neural network is a well-known technology and the specific process will not be described here in detail.

[0090] At this point, we can get the predicted springback of the current metal sheet under the current number of bends. Since the metal structure reduces the springback of the metal sheet after bending, that is, the current forming angle error, please refer to Figure 3, which shows a schematic diagram of the angular relationship between the predicted springback amount and the current forming angle error in one embodiment of the present invention; due to the toughness of the metal and the structural characteristics of the metal, the degree of weakening of this springback is a relatively fixed ratio, so the corrected springback amount can be calculated based on the difference between the current forming angle error and the predicted springback amount, thereby avoiding the subsequent calculation of the compensation pressure. The pressure value is too large, which causes the final metal part to bend.

[0091] Preferably, in one embodiment of the present invention, the method for obtaining the corrected rebound amount includes:

[0092] In this embodiment of the present invention, the current forming angle error is considered to be the weakened springback amount, so the current forming angle error should always be smaller than the predicted springback amount. Therefore, the ratio of the difference between the predicted springback amount and the current forming angle error to the predicted springback amount is used as the adjustment ratio, and the adjustment ratio is also the ratio by which the predicted springback amount should be reduced.

[0093] Finally, the product of the adjustment ratio and the predicted springback amount is used as the corrected springback amount. The corrected springback amount at this time is a more accurate actual springback amount after comprehensive analysis based on the springback situation after metal bending and the metal's own springback resistance.

[0094] The processing pressure control module 104 is used to control the processing pressure of the bending axis of the bending machine based on the corrected springback amount at the current bending number of the current metal sheet.

[0095] Based on the above module, the corrected springback amount of the current metal sheet under the current number of bending times can be obtained. Then, pressure compensation based on this indicator can complete the springback compensation under complex structures, thereby controlling the processing pressure of each axis of the bending machine during the processing of the current metal sheet.

[0096] Preferably, in one embodiment of the present invention, at the current number of bends of the current metal sheet, the processing pressure of the bending axis of the bending forming machine is controlled based on the corrected springback amount, including:

[0097] When processing metal sheets, an initial theoretical processing pressure will be obtained based on the design drawing parameters (such as material thickness, bending degree, etc.), that is, the preset pressure recorded in this embodiment of the present invention (set according to the implementation scenario); and given that the metal sheet has a certain degree of rebound, a rebound compensation item can be determined based on the dynamic rebound coefficient and the corrected rebound amount, and then the compensation pressure can be obtained. The compensation pressure and the preset pressure are added together to obtain the final corrected pressure.

[0098] Therefore, the corrected pressure of the bending axis is calculated based on the preset pressure of the bending axis, the corrected rebound amount, and the dynamic rebound coefficient. The formula model of the corrected pressure includes:

[0099]

[0100] Among them, h b Indicates the correction pressure of the bending axis; h s Indicates the preset pressure of the bending axis; K indicates the dynamic rebound coefficient; Δθ indicates the corrected rebound amount; θ s Indicates the predicted springback of the current bend.

[0101] In the formula model for correcting pressure, is the rebound compensation term, and 1 is the benchmark, which corresponds to the ideal working condition without rebound. It represents the pressure correction factor, which is finally multiplied by the preset pressure to obtain the final corrected pressure of each axis.

[0102] Under the current number of bending times of the current metal sheet, the processing pressure of the bending axis is set to the corresponding correction pressure to bend the current metal sheet. At the same time, the other axes cooperate with each other and the above process is used for compensation and correction during each bending until the processing of the metal sheet is completed.

[0103] It should be noted that the dynamic rebound coefficient needs to be determined according to the material of the metal sheet in the specific implementation scenario.

[0104] In summary, in the data acquisition module, the historical forming angle errors in the historical metal sheet processing process are obtained to provide a data basis for the subsequent analysis process, and at the same time, the processing parameters of the current metal sheet during the current bending are obtained. Due to its own characteristics, metal parts will rebound when subjected to pressure bending. However, as the shape of the metal becomes more complex, its internal metal structure will provide a certain stability for the metal, that is, it will have a certain resistance to rebound. In order to avoid the occurrence of over-bending caused by over-compensation of pressure, the rebound feature analysis module analyzes the differences and fluctuation characteristics of the historical forming angles of all historical metal sheets, determines the overall rebound resistance of the historical metal sheets at each bending, and maps to determine the current forming angle error under the current number of bending of the metal sheet. The current forming angle error is used to characterize the possible bending angle of the current metal sheet under the current bending. Furthermore, in the error correction module, the embodiment of the present invention obtains the predicted springback amount of the current bend based on the current metal sheet processing parameters, analyzes the difference between this predicted springback amount and the current forming angle error, and determines the corrected springback amount. This corrected springback amount takes into account the metal's own springback compensation during the springback process, and thus can more accurately reflect the potential springback of the current metal sheet. Therefore, in the final processing pressure control module, the processing pressure of the bending machine's bending axis is controlled based on the corrected springback amount, which can effectively improve the accuracy of the processing pressure and greatly reduce the occurrence of over-bending in the finished product.

[0105] It should be noted that the system provided in the above embodiment is only illustrated by the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device can be divided into different functional modules to complete all or part of the functions described above.

[0106] See also Figure 4 , which shows a system structure diagram of a multi-axis synchronously driven metal bending and forming machine intelligent servo control system provided by an embodiment of the present invention, including a processor 400, a memory 401, a bus 402 and a communication interface 403, wherein the processor 400, the communication interface 403 and the memory 401 are connected via the bus 402; wherein the memory 401 may include a high-speed random access memory, the bus 402 may be an ISA bus, a PCI bus or an EISA bus, etc., and the processor 400 may be an integrated circuit chip with signal processing capabilities; the memory 401 stores at least one instruction, at least one program, a code set or an instruction set, and when the at least one instruction, at least one program, a code set or an instruction set is loaded and executed by the processor, the steps of each module in the intelligent servo control system of a multi-axis synchronously driven metal bending and forming machine are realized.

[0107] It should be noted that the order in which the embodiments of the present invention are described above is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0108] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

[0109] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. An intelligent servo control system for a multi-axis synchronously driven metal bending machine, characterized in that: The system includes: A data acquisition module is used to obtain the preset bending angle of each historical metal sheet and the actual bending angle after each bending and calculate the historical forming angle error based on the error; obtain the processing parameters of the current metal sheet during the current bending; The springback characteristic analysis module is used to calculate the overall springback resistance of each historical metal sheet during the same bending cycle based on the differences and fluctuation characteristics between the historical forming angle errors. The target value is determined among all the overall springback resistances based on the number of bends of the current metal sheet, thereby obtaining the current forming angle error. An error correction module is configured to obtain a predicted springback amount of the current bend according to the current processing parameters of the metal sheet; and determine a corrected springback amount based on the difference between the current forming angle error and the predicted springback amount; The processing pressure control module is used to control the processing pressure of the bending axis of the bending forming machine based on the corrected springback amount under the current number of bending times of the current metal sheet.

2. The intelligent servo control system for a multi-axis synchronously driven metal bending machine according to claim 1, characterized in that: The method for obtaining the historical forming angle error includes: The difference between the preset bending angle of each historical metal sheet at each bending and the corresponding actual bending angle is used as the historical forming angle error.

3. The intelligent servo control system for a multi-axis synchronously driven metal bending machine according to claim 1, characterized in that: The method for obtaining the overall rebound resistance includes: Select any bending process as the target bending process; Under the target bending process, the historical forming angle error of each historical metal sheet is numerically adjusted based on the hyperbolic tangent function to obtain the springback resistance index; Analyze the differences and fluctuation characteristics between the springback resistance indicators of all historical metal sheets under the target bending process, and determine the springback consistency of each historical metal sheet under the target bending process; Determine a reference weight for each historical metal sheet based on the overall characteristics of the springback consistency of each historical metal sheet under all bending processes; The reference weights of the historical metal sheets are used to perform weighted averaging on the springback resistance indices of the historical metal sheets under the target bending process, and the weighted results are used as the overall springback resistance of the historical metal sheets under the target bending process.

4. The intelligent servo control system for a multi-axis synchronously driven metal bending machine according to claim 3, characterized in that: The method for obtaining the rebound consistency includes: Under the target bending process, among the springback resistance indices of all historical metal sheets, the standard score of the springback resistance index of each historical metal sheet is calculated, and the absolute value of the standard score is negatively correlated mapped and normalized as the springback consistency of each historical metal sheet under the target bending process.

5. The intelligent servo control system for a multi-axis synchronously driven metal bending machine according to claim 3, characterized in that: The method for obtaining the reference weight includes: The average of the springback consistency of each historical metal sheet under all bending processes is used as the reference weight of each historical metal sheet.

6. The intelligent servo control system for a multi-axis synchronously driven metal bending machine according to claim 3, characterized in that: The method for obtaining the current forming angle error includes: Among all the overall springback resistances, the overall springback resistance with the same number of bends as the current metal sheet is taken as the target value of the current metal sheet; The target value is calculated using the inverse function of the hyperbolic tangent function to obtain the current forming angle error of the current metal sheet under the current bending condition.

7. The intelligent servo control system for a multi-axis synchronously driven metal bending machine according to claim 1, characterized in that: The method for obtaining the predicted rebound amount includes: Under simulation conditions, multiple sets of simulation processing parameters and corresponding simulation springback amounts are obtained; The simulated machining parameters are used as input and the simulated springback amount is used as output to train the neural network, thereby obtaining a springback prediction model. The processing parameters of the current metal sheet under the current number of bending times are used as the input of the springback prediction model, and the predicted springback amount is output.

8. The intelligent servo control system for a multi-axis synchronously driven metal bending machine according to claim 7, characterized in that: The neural network adopts SSA-LSTM network.

9. The intelligent servo control system for a multi-axis synchronously driven metal bending machine according to claim 1, characterized in that: The method for obtaining the corrected rebound amount includes: The ratio of the difference between the predicted springback amount and the current forming angle error to the predicted springback amount is used as the adjustment ratio; The product of the adjustment ratio and the predicted springback amount is used as the corrected springback amount.

10. The intelligent servo control system for a multi-axis synchronously driven metal bending machine according to claim 1, characterized in that: The controlling of the processing pressure of the bending axis of the bending forming machine based on the corrected springback amount under the current number of bending times of the current metal sheet comprises: The correction pressure of the bending axis is determined according to the preset pressure of the bending axis, the correction rebound amount, and the dynamic rebound coefficient. The formula model of the correction pressure includes: Among them, h b Indicates the correction pressure of the bending axis; h s Indicates the preset pressure of the bending axis; K indicates the dynamic rebound coefficient; Δθ indicates the corrected rebound amount; θ s Indicates the predicted springback amount of the current bend; Under the current bending times of the current metal sheet, the processing pressure of the bending axis is set to the corresponding correction pressure so as to perform bending processing on the current metal sheet.

Images