A method and system for spinning forming control of a curved sheet metal part

Abstract

The present application relates to the technical field of spinning forming, and discloses a spinning forming control method and system for a curved sheet metal part, which comprises the following steps: obtaining target profile data of a sheet metal part to be processed and generating an initial spinning trajectory; controlling a spinning device to spin, collecting multi-source state data of a spinning deformation zone, and judging the springback tendency of the current spinning deformation zone; when the result is high springback risk, applying a normal micro disturbance, collecting transient unloading state data and steady unloading state data, and obtaining a springback response result; when there is an elastic recovery trend, generating a compensation spinning trajectory; and controlling the spinning device to spin based on the compensation spinning trajectory to obtain a curved sheet metal part finished product. The present application compensates for the profile deviation caused by material springback, and improves the dimensional accuracy of the curved sheet metal part.

Description

Technical Field

[0001] This invention relates to the field of spinning forming technology, and more specifically, to a spinning forming control method and system for curved sheet metal parts. Background Technology

[0002] Curved sheet metal parts are widely used in aerospace engines, aerospace structural components, and high-end equipment manufacturing. Typical components such as non-circular engine lips, irregularly shaped fairings, and complex curved surface connectors often exhibit complex spatial curvature variations, thin wall thicknesses, and high dimensional accuracy requirements. To meet the manufacturing needs of complex curved structures, spin forming technology, with its advantages of high forming flexibility, high material utilization, and applicability to the manufacturing of large, thin-walled curved components, has become one of the important processing methods for curved sheet metal parts.

[0003] Existing patent CN118950811A discloses a method for spinning a non-circular aircraft engine lip. It establishes a corresponding spatial geometric model based on the target surface of the part and uses a surface discretization method to convert the target surface into multiple discrete trajectory points. Then, a CNC spinning trajectory is generated based on these discrete trajectory points. The spinning equipment controls the relative movement between the spinning wheel and the workpiece according to the pre-set trajectory, thereby gradually completing the forming of the curved sheet metal part. However, in the actual spinning process, the material is affected by factors such as spinning pressure, temperature changes, the degree of local plastic deformation, and stress release. Different areas often exhibit different deformation response characteristics. Especially in areas with large curvature changes, thin-walled areas, and complex spatial transition areas, the material is prone to significant elastic recovery, i.e., springback. When the degree of material springback differs from the pre-set forming model, the actual forming contour can easily deviate from the target surface, thus affecting the dimensional accuracy, surface continuity, and assembly consistency of the part.

[0004] Therefore, it is necessary to design a spinning forming control method and system for curved sheet metal parts to solve the problems existing in the current technology. Summary of the Invention

[0005] In view of this, the present invention proposes a spinning forming control method and system for curved sheet metal parts, aiming to solve the problem that curved sheet metal parts are prone to springback during actual spinning, which can easily cause the actual forming contour to deviate from the target surface, thereby affecting the dimensional accuracy, surface continuity and assembly consistency of the parts.

[0006] In one aspect, the present invention proposes a spinning forming control method for curved sheet metal parts, comprising: Obtain the target profile data of the sheet metal part to be processed, and generate an initial spinning trajectory based on the target profile data; Based on the initial spinning trajectory, the spinning equipment is controlled to perform spinning, and multi-source state data of the spinning deformation zone is collected during the spinning process; the springback tendency of the current spinning deformation zone is judged based on the multi-source state data, and a springback tendency judgment result is obtained, which includes a high springback risk result and a low springback risk result; When the springback tendency judgment result is the high springback risk result, the spinning equipment is controlled to apply a normal micro-perturbation, and transient unloading state data and steady-state unloading state data are collected during the application of the normal micro-perturbation; the multi-source state data, transient unloading state data and steady-state unloading state data are correlated and analyzed to obtain the springback response result; When the rebound response result indicates an elastic recovery trend, a normal compensation trajectory segment is generated based on the current spinning position, and the normal compensation trajectory segment is merged into the initial spinning trajectory to obtain the compensated spinning trajectory. Based on the compensated spinning trajectory, the spinning equipment is controlled to perform spinning to obtain the finished curved sheet metal part.

[0007] Furthermore, the multi-source state data includes spinning pressure data, temperature data, and spinning wheel displacement data.

[0008] Furthermore, when determining the springback tendency of the current spinning deformation zone based on the multi-source state data and obtaining the springback tendency determination result, the following steps are included: Extract the normal rotational force component from the rotational force data, extract the temperature state of the deformation zone from the temperature data, and extract the normal displacement tracking state from the wheel displacement data; When the normal rotational pressure component meets the high springback risk condition, the temperature state of the deformation zone meets the low plastic deformation condition, and the normal displacement tracking state meets the low following condition, it is determined to be the high springback risk result; otherwise, it is determined to be the low springback risk result.

[0009] Furthermore, when applying normal micro-perturbations to the spinning equipment, the following are included: The control wheel performs a short-term retraction action along the normal direction away from the mandrel. The retraction displacement of the short-term retraction action is less than the normal downward pressure of the current spinning pass, and the duration of the short-term retraction action is less than the preset disturbance duration.

[0010] Furthermore, when acquiring transient unloading state data and steady-state unloading state data during the application of the normal micro-perturbation, the process includes: Within the time window during which the rotary wheel performs the short-term retraction action, the rotary wheel force change signal and normal displacement change signal are collected as the transient unloading state data; after the rotary wheel completes the short-term retraction action and remains in a stable state for a preset static time, the rotary wheel residual force signal and normal stable displacement signal are collected as the steady-state unloading state data.

[0011] Furthermore, when performing correlation analysis on the multi-source state data, transient unloading state data, and steady-state unloading state data to obtain the rebound response result, the following steps are included: The transient unloading state data includes abrupt changes in the wheel force and normal displacement, which are correlated and compared with the multi-source state data including the wheel pressure data and wheel displacement data, to obtain transient response information. The steady-state unloading state data includes residual wheel force and normal stable displacement, which are correlated and compared with the multi-source state data including the wheel pressure data and wheel displacement data, to obtain steady-state response information. When the transient response information and the steady-state response information both satisfy the rebound identification condition, the elastic recovery trend is determined to exist.

[0012] Furthermore, when generating the normal compensation trajectory segment based on the current spinning position, it includes: The normal overpressure compensation amount is determined based on the rebound response result and the spinneret displacement data; the normal compensation trajectory segment is generated by extending the preset compensation length along the spinneret feed direction, starting from the current spinning position.

[0013] Furthermore, after controlling the spinning equipment to perform spinning based on the compensated spinning trajectory, the method further includes: After the current spinning pass is completed, the actual forming profile of the current pass is obtained; the actual forming profile is compared with the target profile data to obtain the global profile deviation; the initial spinning trajectory of subsequent spinning passes is adjusted by overall normal offset based on the global profile deviation.

[0014] Furthermore, it also includes: performing the forming process based on a multi-pass spinning method, and after each pass, updating the normal compensation amount, springback judgment condition, and spinneret displacement judgment condition corresponding to the subsequent spinning passes according to the global contour deviation.

[0015] Compared with existing technologies, the advantages of this invention are as follows: By acquiring target surface data to generate an initial spinning trajectory and collecting multi-source state data of the spinning deformation zone in real time during the spinning process, the springback tendency of the current deformation zone is judged. This allows for early identification of areas where springback may occur during the forming process, improving the ability to perceive the material deformation state. When a high risk of springback is detected, normal micro-perturbation is applied to the spinning equipment, and correlation analysis is performed using transient and steady-state unloading data to obtain springback response results. This allows for the acquisition of elastic recovery characteristics during the actual unloading of the material, thereby achieving proactive identification of springback behavior. This avoids the lag problem caused by relying solely on post-forming detection results for compensation, improving the accuracy and real-time performance of springback identification. A normal compensation trajectory segment is generated based on the springback response results and integrated into the initial spinning trajectory. This allows the spinning trajectory to be dynamically adjusted according to the actual material deformation state, thereby compensating for surface deviations caused by material springback, reducing the error between the actual formed contour and the target surface, and improving the dimensional accuracy and surface consistency of curved sheet metal parts.

[0016] On the other hand, this application also provides a spin forming control system for curved sheet metal parts, for applying the above-mentioned spin forming control method for curved sheet metal parts, including: The preforming unit is configured to acquire target profile data of the sheet metal part to be processed and generate an initial spinning trajectory based on the target profile data; The judgment unit is configured to control the spinning equipment to perform spinning based on the initial spinning trajectory, and to collect multi-source state data of the spinning deformation zone during the spinning process; to judge the springback tendency of the current spinning deformation zone based on the multi-source state data, and to obtain a springback tendency judgment result, wherein the springback tendency judgment result includes a high springback risk result and a low springback risk result. The analysis unit is configured to control the spinning equipment to apply a normal micro-perturbation when the springback tendency judgment result is the high springback risk result, and to collect transient unloading state data and steady-state unloading state data during the application of the normal micro-perturbation; and to perform correlation analysis on the multi-source state data, transient unloading state data and steady-state unloading state data to obtain the springback response result; The compensation unit is configured to generate a normal compensation trajectory segment based on the current spinning position when the rebound response result indicates an elastic recovery trend, and to merge the normal compensation trajectory segment into the initial spinning trajectory to obtain a compensated spinning trajectory. The output unit is configured to control the spinning equipment to perform spinning based on the compensated spinning trajectory, thereby obtaining a finished curved sheet metal part.

[0017] It is understandable that the above-mentioned spinning forming control methods and systems for curved sheet metal parts have the same beneficial effects, and will not be elaborated further here. Attached Figure Description

[0018] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 A flowchart of a spin forming control method for curved sheet metal parts provided in an embodiment of the present invention; Figure 2 A logic flowchart of the spinning forming control process for curved sheet metal parts provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of the multi-station spinning forming device for curved sheet metal parts provided in an embodiment of the present invention; Figure 4 This is a functional block diagram of a spin forming control system for curved sheet metal parts provided in an embodiment of the present invention.

[0019] The equipment includes: 1. spinning equipment; 2. robotic arms; 3. vacuum sealing welding equipment; 4. vacuum heat treatment equipment; 5. quality inspection equipment; and 6. control equipment. Detailed Implementation

[0020] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the disclosure to those skilled in the art. It should be noted that, unless otherwise specified, embodiments and features in the embodiments of the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0021] In some embodiments of this application, see Figure 1-2 As shown, a spin forming control method for curved sheet metal parts is proposed, including: S100: Acquire the target profile data of the sheet metal part to be processed, and generate an initial spinning trajectory based on the target profile data; S200: Based on the initial spinning trajectory, the spinning equipment is controlled to perform spinning, and multi-source state data of the spinning deformation zone is collected during the spinning process; the springback tendency of the current spinning deformation zone is judged based on the multi-source state data, and the springback tendency judgment result is obtained. The springback tendency judgment result includes high springback risk result and low springback risk result. S300: When the springback tendency judgment result is a high springback risk result, control the spinning equipment to apply normal micro-perturbation, and collect transient unloading state data and steady-state unloading state data during the application of normal micro-perturbation; perform correlation analysis on multi-source state data, transient unloading state data and steady-state unloading state data to obtain springback response results; S400: When the rebound response result indicates an elastic recovery trend, a normal compensation trajectory segment is generated based on the current spinning position, and the normal compensation trajectory segment is merged into the initial spinning trajectory to obtain the compensated spinning trajectory. S500: Based on the compensation spinning trajectory control, the spinning equipment performs spinning to obtain finished curved sheet metal parts.

[0022] In this embodiment, the curved sheet metal part is preferably an aero-engine lip, an air intake guide, a combustion chamber connector, or other thin-walled curved surface components with spatial curvature variation characteristics. Since such parts are prone to springback due to the elastic-plastic recovery of the material during spin forming, this embodiment improves the final forming accuracy by real-time monitoring of the forming state during the spin forming process and adjusting the subsequent spin forming trajectory according to the actual deformation state of the material.

[0023] In step S100, before starting the spinning process, the target surface data corresponding to the sheet metal part to be processed is first acquired. The target surface data is used to describe the final design shape of the part, including three-dimensional surface model data, surface contour data, reference coordinate data, and allowable tolerance data. Among them, the three-dimensional surface model data is used to describe the overall spatial shape of the part; the surface contour data is used to describe the curvature changes at different locations; the reference coordinate data is used to establish the positional correspondence between the part model and the coordinate system of the spinning equipment; and the allowable tolerance data is used to limit the allowable deviation range between the final formed contour and the design contour.

[0024] Specifically, the target surface data originates from the 3D model established during the product design phase. The control equipment first reads the 3D model file and converts the model to the machine tool coordinate system used by the spinning equipment. Then, using the central axis of the spinning mandrel as the reference axis, the target surface is processed in layers.

[0025] Specifically, the control equipment slices the target surface layer by layer along the axial direction of the part, forming multiple cross-sectional profiles. Each cross-sectional profile corresponds to the actual shape of the target surface at a specific height. After the cross-sections are extracted, the control equipment extracts profile points for each profile, obtaining a set of profile points describing the target surface. Subsequently, the control equipment establishes a continuous trajectory based on the spatial relationship between adjacent profile points. For areas with large curvature changes, a smaller trajectory discretization interval is used; for areas with gentler curvature changes, a larger trajectory discretization interval is used. This reduces the amount of trajectory data while ensuring trajectory accuracy.

[0026] The trajectory discrete interval is preferably determined through a trial production process. Specifically, a sample with the same material, plate thickness, and structural characteristics as the target product can be selected for experimental spinning, and the surface error formed under different discrete interval conditions can be compared. When the discrete interval continues to decrease, if the improvement in surface accuracy is less than the preset accuracy improvement requirement, then the corresponding discrete interval is determined as the process discrete interval. For example, for nickel-based high-temperature alloy curved sheet metal parts with a thickness of 1.2mm to 2.0mm, an axial discrete interval of 1mm to 3mm can be preferably used for trajectory generation; for areas with drastic curvature changes, the discrete interval is further reduced to improve local trajectory accuracy.

[0027] After discretizing the trajectory, the control equipment further combines the motion axis parameters of the spinning equipment to convert each discrete trajectory point into control commands that the spinning equipment can recognize, thereby generating the initial spinning trajectory. The initial spinning trajectory includes not only the spatial motion position of the spinning wheel, but also the feed direction, spindle rotation state, and pass execution sequence at the corresponding position.

[0028] In step S200, the control device sends the initial spinning trajectory to the spinning equipment and controls the spinning wheel to gradually plastically form the workpiece according to the predetermined trajectory. As the spinning wheel continues to feed, the workpiece material undergoes localized plastic deformation under the combined action of the spinning wheel and the mandrel. During this process, the stress state, thermal state, and deformation state accumulated in different areas are not entirely the same, therefore the control device needs to acquire the current state information of the deformation zone in real time.

[0029] In this embodiment, the multi-source state data includes spinning pressure data, temperature data, and spinning wheel displacement data. The spinning pressure data reflects the force generated during the contact between the spinning wheel and the workpiece. Preferably, a pressure sensor is installed on the spinning wheel support mechanism or the spinning wheel mounting base to collect real-time data on changes in the force on the spinning wheel. The temperature data reflects the thermal state of the current deformation zone. Preferably, an infrared temperature measuring device or thermal imaging device is arranged near the spinning deformation area to perform non-contact temperature acquisition of the spinning deformation zone. The spinning wheel displacement data reflects the actual tracking of the spinning wheel to the target trajectory. Preferably, the real-time position of the spinning wheel is obtained by an encoder or displacement sensor installed on each motion axis and compared with the corresponding trajectory position.

[0030] After obtaining the aforementioned multi-source state data, the control equipment begins springback tendency analysis. First, the normal rotational pressure component is extracted from the rotational pressure data. The normal rotational pressure component refers to the pressure component perpendicular to the workpiece surface, which reflects the degree of stress accumulation within the material. Subsequently, the temperature state of the deformation zone is extracted from the temperature data. The temperature state of the deformation zone reflects the plastic flow capability of the material in the current region. Simultaneously, the normal displacement tracking state is extracted from the wheel displacement data. Specifically, the control equipment compares the actual normal position of the wheel with the theoretical trajectory normal position and obtains the corresponding displacement tracking deviation.

[0031] Furthermore, the control equipment determines whether the current region meets the high springback risk condition based on the normal spinning pressure component. The high springback risk condition is preferably obtained through process calibration. Specifically, the correspondence between spinning pressure and final springback amount can be established through multiple sets of spinning tests. When the statistical results show that the springback amount corresponding to a certain pressure range continuously exceeds the allowable deviation of the target profile, this pressure range is defined as a high springback risk range. Simultaneously, the control equipment determines whether the low plastic deformation condition is met based on the temperature state of the deformation zone. The low plastic deformation condition is preferably established based on a material process database. For example, for GH4169 nickel-based superalloy, its corresponding hot spinning process temperature range can be pre-established. When the current deformation zone temperature is lower than the lower limit of the process temperature range, it is determined that the material's plastic flow capacity is insufficient, thus meeting the low plastic deformation condition. The control equipment also determines whether the low following condition is met based on the normal displacement tracking status. Specifically, the control equipment statistically analyzes the trajectory tracking deviation change trend over multiple consecutive sampling periods. When the trajectory deviation continuously increases and the deviation change trend exceeds the preset allowable range, the current region is determined to meet the low following condition. The preset allowable range is preferably obtained through historical processing data. Specifically, the distribution range of trajectory deviation during the processing of qualified products can be statistically analyzed, and the corresponding range can be used as the normal following interval; if it exceeds this interval, it is judged as a low following state.

[0032] When the normal spinning force component meets the high springback risk condition, the temperature state of the deformation zone meets the low plastic deformation condition, and the normal displacement tracking state meets the low following condition, the control device determines that there is a significant springback risk in the current spinning region and outputs a high springback risk result. Otherwise, it determines that the springback risk in the current region is low and outputs a low springback risk result.

[0033] In step S300, when the control device determines that there is a low risk of springback in the current spinning deformation zone, it continues to spin according to the initial spinning trajectory.

[0034] When the control equipment determines that there is a high risk of springback in the current spinning deformation zone, it does not immediately make large-scale modifications to the entire pass trajectory. Instead, it first applies a small, short-duration, and controllable normal micro-perturbation to the current spinning position to actively stimulate the unloading response in the current deformation zone. Based on this unloading response, it identifies whether there is a significant elastic recovery trend in the region. The purpose of this is to ensure that the control equipment does not only judge the springback risk based on static state data, but also further verifies the springback sensitivity through the material response after the actual perturbation, thereby improving the targeting and reliability of subsequent compensation control.

[0035] Specifically, the so-called normal micro-disturbance refers to a small, short-duration retraction movement of the spinning wheel along the normal direction of the workpiece surface during normal feed. This movement does not alter the overall forming path of the pass, but rather observes the timely response and recovery trend of the workpiece material after localized unloading through a small unloading process. To avoid this disturbance affecting the already formed basic shape, the retraction displacement should be significantly smaller than the normal downward pressure of the current spinning pass. In other words, the retraction only partially releases the current pressure state, rather than completely disengaging from workpiece contact.

[0036] The retraction displacement and duration of the short-term retraction action are not fixed but are pre-calibrated based on equipment capabilities, material thickness, material type, and part curvature characteristics. Specifically, multiple disturbance tests can be conducted using samples made of the same or similar material as the sheet metal part to be processed. Under different retraction displacements and durations, it can be observed whether the material exhibits excessive unloading, surface vibration, or local instability. The disturbance parameters suitable for the current product can then be determined by combining the final springback results. Preferably, the retraction displacement can be a small portion of the downward pressure of the current method, typically just enough to trigger an unloading response without disrupting the forming continuity. The duration should be based on the ability to acquire stable response changes and should not be too long to avoid interfering with the normal forming rhythm.

[0037] In a preferred embodiment, for thin-walled curved sheet metal parts, the short-time retraction action can be set as a very short normal retraction process, for example, completing the three stages of retraction, holding, and recovery within a short time window. This allows for localized release of some load on the material without significantly disturbing the overall forming path. The preferred value of this parameter is usually determined experimentally, and in principle, it should "elicit a material response without causing new forming errors."

[0038] Furthermore, within a preset time window during the short-term retraction action of the spinning wheel, the control device continuously collects signals of sudden changes in the force and normal displacement of the spinning wheel, and uses these as transient unloading state data. This preset time window is preferably set based on the response speed and sampling frequency of the spinning equipment, and should generally cover the brief change interval before and after the retraction action to capture the dynamic process of sudden pressure drop and displacement change. To ensure the comparability of the collected signals, the length of this window can be determined experimentally, and in principle, it should be able to fully reflect the rapid changes in the initial stage of unloading.

[0039] After the spinning wheel completes a short-term retraction action and remains in a stable state for a preset resting time, the control equipment continues to collect the residual force signal and normal stable displacement signal of the spinning wheel, and uses them as steady-state unloading data. The preset resting time is preferably determined through process testing. The setting principle is to ensure that the material has sufficient time to complete preliminary elastic recovery after partial unloading, but not so long as to affect the next forming cycle. Usually, the resting time can be determined based on the material thickness, thermal state, and equipment response capability. For hot spinning, the resting time should also be set in conjunction with the material temperature decay rate to avoid temperature changes interfering with the springback response judgment.

[0040] Specifically, transient unloading data refers to data collected when the rotary wheel just begins to retract, while the material is still in the rapid stress release phase; steady-state unloading data refers to data collected after the retraction is complete and the material has entered a relatively stable recovery phase. The former mainly reflects the stress release intensity of the material at the moment of rapid unloading, while the latter mainly reflects the residual recovery trend retained by the material after stable unloading. Using both types of data together can more comprehensively identify springback behavior and avoid misjudgments based solely on signal fluctuations at a single moment.

[0041] Specifically, the time window and preset resting time are also determined experimentally. First, based on the equipment's sampling frequency, the rotary wheel's response speed, and the material's recovery speed, a sampling window can be set to fully cover the retraction process and the brief response changes before and after it. Then, a short resting phase is set after this window to observe the further recovery of the material after partial unloading. If the resting time is too short, the material has not yet fully exhibited stable recovery characteristics, which is not conducive to judgment; if the resting time is too long, it will affect the processing cycle and may introduce interference factors such as temperature changes. Therefore, the preset resting time is preferably a length that reflects the stable recovery trend without significantly affecting the production cycle.

[0042] In a preferred embodiment, when the control device performs a short-term retraction action, it simultaneously acquires the sudden change signals of the rotating wheel's force and normal displacement, and uses these as transient unloading state data. After the short-term retraction action ends, it continues to acquire the residual force signal and normal stable displacement signal of the rotating wheel over a period of time, and uses these as steady-state unloading state data. Through this staged acquisition method, the response characteristics of the material in two different stages, "immediately after unloading" and "after unloading stabilization," can be captured separately, thereby more accurately distinguishing between ordinary fluctuations and true springback.

[0043] Furthermore, when performing correlation analysis on multi-source state data, transient unloading state data, and steady-state unloading state data to obtain the rebound response results, the control device does not simply judge a single signal, but analyzes the corresponding relationship between the forming state before the disturbance and the unloading response after the disturbance to determine whether the material has obvious elastic recovery characteristics.

[0044] Specifically, the control equipment first correlates and compares the sudden change signals of the rotating wheel force and normal displacement in the transient unloading state data with the rotating pressure data and rotating wheel displacement data in the multi-source state data to form transient response information. The purpose of this step is to determine whether a significant rapid unloading response occurs in the current deformation zone when the rotating wheel retracts briefly. If the rotating pressure drops rapidly in a short time, and the displacement signal shows a significant sudden change, it indicates that the material in this region has released a large amount of instantaneous stress after being disturbed.

[0045] Subsequently, the control equipment correlates and compares the residual force signal and normal stable displacement signal of the spinning wheel in the steady-state unloading data with the spinning force data and spinning wheel displacement data in the multi-source state data to form steady-state response information. This step is used to further determine whether the material still maintains a significant recovery trend after unloading. If there is still significant residual force or displacement recovery in the steady state, it usually indicates that the elastic recovery in that region has not been completely completed, and there is a strong possibility of rebound.

[0046] Specifically, the springback identification condition is preferably determined through experimental calibration. Multiple sets of disturbance tests can be conducted on materials, thicknesses, and curvature characteristics similar to or the same as the sheet metal part to be processed, and the correspondence between the transient and steady-state responses under different disturbance conditions and the final springback results is recorded. Based on the statistical results, signal combinations that stably correspond to a clear springback area are selected as the springback identification condition. In other words, this condition is not directly specified by a single fixed value, but is determined comprehensively through sample testing, historical processing data, and final surface error analysis.

[0047] When both transient and steady-state response information meet the springback identification criteria, the control device can determine that there is an elastic recovery trend in the current region. Preferably, "meeting the criteria" means that the rapid stress release characteristics in the transient stage and the continuous recovery characteristics in the steady-state stage appear simultaneously, mutually confirming each other and reducing misjudgments. Through this two-stage response identification method, this embodiment can more accurately identify potential springback-sensitive areas during spinning, providing a reliable basis for the subsequent generation of normal compensation trajectory segments.

[0048] In a preferred embodiment, for a nickel-based superalloy curved sheet metal part, when the spinning wheel performs a short-term normal retraction in a high-risk area, the pressure signal shows a significant decrease, while the normal displacement continues to change in the recovery direction after retraction, and the residual force signal has not completely subsided during the static phase. This indicates a significant elastic recovery trend in the area. In this case, the control device uses the response result of this area as the basis for generating the subsequent compensation trajectory, so as to make targeted corrections to the spinning trajectory in subsequent steps.

[0049] In step S400, when the control device determines, based on the aforementioned steps, that the current spinning region exhibits a significant elastic recovery trend, it indicates that the region has a high probability of springback after unloading. If processing continues along the original initial spinning trajectory, the final formed profile is prone to deviating from the target surface. Therefore, this embodiment does not directly abandon the current trajectory, but rather generates a normal compensation trajectory segment with a compensating effect based on the current spinning position. By appropriately modifying the local trajectory, the deviation caused by material springback is offset.

[0050] Specifically, the generation of the normal compensation trajectory segment is not simply a matter of shifting the entire spinning wheel position inward. Instead, it involves comprehensively determining the normal overpressure compensation amount by combining the springback response results of the current region with the spinning wheel displacement data. The so-called normal overpressure compensation amount refers to the additional normal correction amount pre-applied at the current spinning position to compensate for any subsequent elastic recovery in that region. Essentially, this compensation amount provides a pre-existing "overforming" correction space, allowing the material to spring back to a state close to the target profile after unloading.

[0051] In practical settings, the normal overpressure compensation amount is preferably determined jointly based on the springback response results, the wheel displacement data, and the historical process database. Specifically, the control equipment can first statistically analyze the experimental data of similar materials, thicknesses, and curvature regions, and analyze the deviation changes between the final formed profile and the target surface under different compensation conditions. Based on this, a correspondence between the compensation amount and the actual springback amount can be established. In other words, the normal overpressure compensation amount is not a fixed value, but a dynamically determined correction amount based on the springback characteristics of the current region. For regions with large springback, the compensation amount can be appropriately increased; for regions with small springback, the compensation amount can be appropriately decreased to avoid introducing new overforming errors.

[0052] The principle for setting the normal overpressure compensation amount is: it should be sufficient to offset the expected springback in the current area, but not so large as to cause localized damage, surface wrinkling, or mandrel interference. An initial compensation range is usually obtained through multiple sets of spinning tests, and then refined based on the current springback response during formal processing. For example, for a thin-walled curved sheet metal part, tests revealed that when a local area shows a significant elastic recovery trend, if the corresponding normal compensation amount is too small, dimensional deviations will still exist after springback; if the compensation amount is too large, slight overpressure marks will appear locally. Therefore, the final compensation amount is usually determined after achieving a balance between springback offsetting and forming stability.

[0053] In a preferred embodiment, the control device extends a preset compensation length along the spinning feed direction from the current spinning position to generate a normal compensation trajectory segment. This preset compensation length defines the effective range of the compensation trajectory. Its function is not to completely rewrite the entire trajectory, but rather to locally correct only the current rebound-sensitive area and its preceding and following transition areas to maintain the continuity and smoothness of the trajectory. In other words, the compensation length determines how long the compensation effect lasts along the trajectory, while the compensation amount determines how much the trajectory deviates in the normal direction; both together determine the local compensation effect.

[0054] The preset compensation length is also determined through process experiments and part structural characteristics. Specifically, the spatial length covered by the compensation trajectory segment can be determined based on the local curvature variation range of the target part, the length of the springback-sensitive area, and the equipment response capability. For areas with large local curvature variations, the compensation length can be appropriately extended to ensure a smooth transition; for areas with small local variations, the compensation length can be appropriately shortened to reduce unnecessary intervention. Preferably, the compensation length is designed to completely cover the current springback-affected area, avoiding both insufficient compensation and excessively wide compensation range that could affect overall efficiency.

[0055] After generating the normal compensation trajectory segment, the control device merges it into the initial spinning trajectory to form a compensated spinning trajectory. This merging is not simply inserting the compensation segment into the original trajectory; more importantly, it ensures spatial and motion continuity between the compensation segment and the preceding and following trajectories, preventing abrupt changes or pauses during spinning. Preferably, the merging process can be achieved by smoothing the transition at the connection point, ensuring the compensation trajectory segment maintains the same motion trend as the original trajectory during entry and exit, thus avoiding new processing fluctuations. For example, in a local forming area of ​​a nickel-based high-temperature alloy curved sheet metal part, the control device determines, based on the springback response, that the area exhibits a significant elastic recovery trend and, combined with the current spinning wheel displacement, determines the need for a certain amount of normal overpressure compensation. Subsequently, the control device sets a local compensation trajectory along the feed direction, centered on the current spinning position, causing the spinning wheel to move slightly closer to the mandrel in this area compared to the original trajectory. This allows the area to essentially return to the target surface after subsequent unloading, thereby reducing the final contour error.

[0056] Specifically, when the control device determines, based on the aforementioned steps, that there is no obvious elastic recovery trend in the current spinning area, it continues spinning according to the initial spinning trajectory.

[0057] In step S500, after the compensated spinning trajectory is generated, the control device sends the trajectory to the spinning equipment and continues to control the spinning equipment to perform the subsequent forming process according to the compensated spinning trajectory. Since the trajectory has taken into account the aforementioned springback response results, the spinning equipment can make targeted corrections to the springback-sensitive area when executing the compensated trajectory, thereby improving the fit between the final formed contour and the target surface.

[0058] It should be noted that spinning based on a compensated spinning trajectory does not mean that subsequent processing remains completely unchanged. On the contrary, in multi-pass spinning scenarios, new contour deviations may exist after each pass. Therefore, this embodiment preferably detects and corrects the forming result after each pass. Specifically, after the current spinning pass is completed, the control device acquires the actual forming profile of the current pass and compares the actual forming profile with the target surface data to obtain the global contour deviation.

[0059] Global contour deviation refers to the spatial deviation between the overall contour of the part after the current spinning pass and the target surface. This deviation can be used to determine whether the current compensation is sufficient, and it can also guide the trajectory correction of subsequent passes. Based on the global contour deviation, the control equipment adjusts the overall normal offset of the initial spinning trajectory of subsequent spinning passes, making the subsequent trajectory more closely match the actual forming state. This avoids processing along the exact same trajectory in every pass, thus forming a closed-loop control for continuous correction.

[0060] In multi-pass spinning, the preferred method is to update the subsequent control parameters after each pass, including the normal compensation, springback judgment condition, and spinneret displacement judgment condition. This is because as the forming process progresses, the stress state, temperature state, and deformation state of the material change, and the judgment conditions of the previous pass may not be entirely applicable to the next pass. Therefore, the control equipment needs to readjust the control parameters of subsequent passes based on the global contour deviation of the current pass to ensure that the judgment and compensation logic remains consistent with the actual forming state.

[0061] The update of the normal compensation amount is mainly used to correct the degree of offsetting local springback in subsequent passes; the update of the springback judgment condition is mainly used to adapt to the changes in the material's springback sensitivity under different passes; and the update of the spinneret displacement judgment condition is used to correct the spinneret's tracking requirements for the trajectory, making subsequent judgments more accurate. In other words, these three conditions are not set once and then fixed, but are gradually modified as the forming process progresses, so that the entire spinning control process is always in a dynamic closed loop.

[0062] In a preferred embodiment, if a slight outward deviation is found in a certain local area after the previous processing, the control equipment can appropriately increase the normal overpressure compensation in subsequent processing steps, and simultaneously relax or tighten the springback judgment conditions for the corresponding area, so that the next processing step can continue forming on a more suitable compensation basis. After multiple iterations, the final curved sheet metal part can be closer to the target surface requirements.

[0063] Specifically, see Figure 3 As shown, the multi-station spinning forming apparatus for curved sheet metal parts includes a spinning device 1, a robotic arm 2, a vacuum sealing welding device 3, a vacuum heat treatment device 4, a quality inspection device 5, and a control device 6. The spinning device 1 is used to complete the spinning forming of the curved sheet metal parts; the robotic arm 2 is used for transferring the workpiece between stations; the vacuum sealing welding device 3 is used for welding and sealing the formed workpiece; the vacuum heat treatment device 4 is used for heat treatment of the welded workpiece; the quality inspection device 5 is used for contour inspection and quality inspection of the final product; and the control device 6 is used for executing and controlling the data signals in the multi-station spinning forming apparatus. The curved sheet metal part spinning forming control method of this application is applied to the spinning forming stage and is realized through the execution control of the spinning device 1 by the control device 6.

[0064] Based on another preferred embodiment described above, see [link to preferred embodiment]. Figure 4 As shown, this embodiment provides a spin forming control system for curved sheet metal parts, used to apply the above-described spin forming control method for curved sheet metal parts, including: The preforming unit is configured to acquire the target profile data of the sheet metal part to be processed and generate an initial spinning trajectory based on the target profile data; The judgment unit is configured to control the spinning equipment to spin based on the initial spinning trajectory, and to collect multi-source state data of the spinning deformation zone during the spinning process; to judge the springback tendency of the current spinning deformation zone based on the multi-source state data, and to obtain the springback tendency judgment result, which includes a high springback risk result and a low springback risk result; The analysis unit is configured to control the spinning equipment to apply normal micro-perturbation when the springback tendency judgment result is a high springback risk result, and to collect transient unloading state data and steady-state unloading state data during the application of normal micro-perturbation; and to perform correlation analysis on multi-source state data, transient unloading state data and steady-state unloading state data to obtain springback response results; The compensation unit is configured to generate a normal compensation trajectory segment based on the current spinning position when the rebound response result indicates an elastic recovery trend, and to merge the normal compensation trajectory segment into the initial spinning trajectory to obtain the compensated spinning trajectory. The output unit is configured to control the spinning equipment based on the compensated spinning trajectory to spin the finished curved sheet metal part.

[0065] In summary, by acquiring target surface data to generate an initial spinning trajectory and collecting multi-source state data of the spinning deformation zone in real time during the spinning process, the springback tendency of the current deformation zone can be judged. This allows for early identification of areas where springback may occur during the forming process, improving the ability to perceive the material deformation state. When a high risk of springback is detected, normal micro-perturbation is applied to the spinning equipment, and correlation analysis is performed using transient and steady-state unloading data to obtain springback response results. This allows for the acquisition of elastic recovery characteristics during the actual unloading of the material, thereby achieving proactive identification of springback behavior. This avoids the lag problem caused by relying solely on post-forming detection results for compensation, improving the accuracy and real-time performance of springback identification. A normal compensation trajectory segment is generated based on the springback response results and integrated into the initial spinning trajectory. This allows the spinning trajectory to be dynamically adjusted according to the actual material deformation state, compensating for surface deviations caused by material springback, reducing the error between the actual formed contour and the target surface, and improving the dimensional accuracy and surface consistency of curved sheet metal parts.

[0066] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the present invention.

Claims

1. A method for controlling the spinning process of curved sheet metal parts, characterized in that, include: Obtain the target profile data of the sheet metal part to be processed, and generate an initial spinning trajectory based on the target profile data; The spinning equipment is controlled to perform spinning based on the initial spinning trajectory, and multi-source state data of the spinning deformation zone are collected during the spinning process; The springback tendency of the current spinning deformation zone is judged based on the multi-source state data to obtain the springback tendency judgment result, which includes a high springback risk result and a low springback risk result. When the springback tendency judgment result is the high springback risk result, the spinning equipment is controlled to apply a normal micro-perturbation, and transient unloading state data and steady-state unloading state data are collected during the application of the normal micro-perturbation. The rebound response results are obtained by performing correlation analysis on the multi-source state data, transient unloading state data and steady-state unloading state data. When the rebound response result indicates an elastic recovery trend, a normal compensation trajectory segment is generated based on the current spinning position, and the normal compensation trajectory segment is merged into the initial spinning trajectory to obtain the compensated spinning trajectory. Based on the compensated spinning trajectory, the spinning equipment is controlled to perform spinning to obtain the finished curved sheet metal part.

2. The spin forming control method for curved sheet metal parts according to claim 1, characterized in that, The multi-source state data includes spinning pressure data, temperature data, and spinning wheel displacement data.

3. The spinning forming control method for curved sheet metal parts according to claim 2, characterized in that, The springback tendency of the current spinning deformation zone is determined based on the multi-source state data. The springback tendency determination result includes: Extract the normal rotational force component from the rotational force data, extract the temperature state of the deformation zone from the temperature data, and extract the normal displacement tracking state from the wheel displacement data; When the normal rotational pressure component meets the high springback risk condition, the temperature state of the deformation zone meets the low plastic deformation condition, and the normal displacement tracking state meets the low following condition, it is determined to be the high springback risk result; otherwise, it is determined to be the low springback risk result.

4. The spinning forming control method for curved sheet metal parts according to claim 1, characterized in that, When applying normal micro-perturbations to the spinning equipment, the following are included: The control wheel performs a short-term retraction action along the normal direction away from the mandrel. The retraction displacement of the short-term retraction action is less than the normal downward pressure of the current spinning pass, and the duration of the short-term retraction action is less than the preset disturbance duration.

5. The spin forming control method for curved sheet metal parts according to claim 4, characterized in that, When acquiring transient unloading state data and steady-state unloading state data during the application of the normal micro-perturbation, the following are included: Within the time window during which the rotary wheel performs the short-term retraction action, the rotary wheel force change signal and normal displacement change signal are collected as the transient unloading state data; after the rotary wheel completes the short-term retraction action and remains in a stable state for a preset static time, the rotary wheel residual force signal and normal stable displacement signal are collected as the steady-state unloading state data.

6. The spin forming control method for curved sheet metal parts according to claim 5, characterized in that, When performing correlation analysis on the multi-source state data, transient unloading state data, and steady-state unloading state data to obtain the rebound response results, the following are included: The transient unloading state data includes abrupt changes in the wheel force and normal displacement, which are correlated and compared with the multi-source state data including the wheel pressure data and wheel displacement data, to obtain transient response information. The steady-state unloading state data includes residual wheel force and normal stable displacement, which are correlated and compared with the multi-source state data including the wheel pressure data and wheel displacement data, to obtain steady-state response information. When the transient response information and the steady-state response information both satisfy the rebound identification condition, the elastic recovery trend is determined to exist.

7. The spin forming control method for curved sheet metal parts according to claim 6, characterized in that, When generating the normal compensation trajectory segment based on the current spinning position, it includes: The normal overpressure compensation amount is determined based on the rebound response result and the spinneret displacement data; the normal compensation trajectory segment is generated by extending the preset compensation length along the spinneret feed direction, starting from the current spinning position.

8. The spin forming control method for curved sheet metal parts according to claim 1, characterized in that, After spinning is performed using a spinning device controlled by the compensated spinning trajectory, the process further includes: After the current spinning pass is completed, the actual forming profile of the current pass is obtained; the actual forming profile is compared with the target profile data to obtain the global profile deviation; the initial spinning trajectory of subsequent spinning passes is adjusted by overall normal offset based on the global profile deviation.

9. The spin forming control method for curved sheet metal parts according to claim 8, characterized in that, Also includes: The forming process is performed based on a multi-pass spinning method, and after each pass, the normal compensation amount, springback judgment condition, and spinneret displacement judgment condition corresponding to the subsequent spinning passes are updated according to the global contour deviation.

10. A spin forming control system for curved sheet metal parts, used for applying the spin forming control method for curved sheet metal parts as described in any one of claims 1-9, characterized in that, include: The preforming unit is configured to acquire target profile data of the sheet metal part to be processed and generate an initial spinning trajectory based on the target profile data; The judgment unit is configured to control the spinning equipment to perform spinning based on the initial spinning trajectory, and to collect multi-source state data of the spinning deformation zone during the spinning process; The springback tendency of the current spinning deformation zone is judged based on the multi-source state data to obtain the springback tendency judgment result, which includes a high springback risk result and a low springback risk result. The analysis unit is configured to control the spinning equipment to apply a normal micro-perturbation when the springback tendency judgment result is the high springback risk result, and to collect transient unloading state data and steady-state unloading state data during the application of the normal micro-perturbation. The rebound response results are obtained by performing correlation analysis on the multi-source state data, transient unloading state data and steady-state unloading state data. The compensation unit is configured to generate a normal compensation trajectory segment based on the current spinning position when the rebound response result indicates an elastic recovery trend, and to merge the normal compensation trajectory segment into the initial spinning trajectory to obtain a compensated spinning trajectory. The output unit is configured to control the spinning equipment to perform spinning based on the compensated spinning trajectory, thereby obtaining a finished curved sheet metal part.

Images