Intelligent self-adaptive variable cross-section panel robot roll forming equipment and roll forming method

Abstract

The application discloses a kind of intelligent self-adapting variable cross-section panel robot roll forming equipment and roll forming method, including robot module, flexible tooling module, sensing module and control module. Robot module end installs roll forming tool head;Flexible tooling module includes pneumatic clamping unit and flexible support unit;Sensing module includes pressure sensor and position sensor;Control module runs fuzzy PID adaptive control algorithm to adjust the clamping pressure of pneumatic clamping unit, and control flexible support unit keeps synchronous follow-up support with robot module end in roll forming process. The application solves the forming precision and quality defect problem caused by nonlinear hysteresis of pneumatic system and sheet metal suspension in variable cross-section forming, improves the clamping force response speed and support adaptability, realizes high flexibility, high-precision variable cross-section panel roll forming.

Description

Technical Field

[0001] This invention relates to the field of metal sheet plastic processing technology, and in particular to a robotic roll forming system and method for variable cross-section sheets, namely, an intelligent adaptive robotic roll forming equipment and roll forming method for variable cross-section sheets. Background Technology

[0002] With the increasing demands for lightweight, integrated, and high-performance products in modern high-end manufacturing, the application of variable cross-section sheet metal parts in engineering structures such as automobiles and aerospace has significantly increased. These parts have gradually varying height, width, or curvature along their length, achieving better mechanical property distribution and higher material utilization compared to traditional constant cross-section parts. However, the complexity of their geometric features leads to a geometric increase in manufacturing difficulty, placing stringent requirements on the flexibility and precision of the forming process.

[0003] Industrial robot roll forming technology, as an emerging progressive sheet metal forming process, utilizes a high-precision industrial robot to hold a roll forming tool head and, in conjunction with flexible tooling, performs sequential localized plastic forming on metal sheets. Compared to traditional die stamping and continuous roll forming, this technology offers extremely high process flexibility and design freedom. It can adapt to the needs of multi-variety, small-batch, and customized production without the need to develop dedicated molds, and is particularly suitable for the rapid manufacturing of high-strength steel thin-walled structural parts.

[0004] However, existing robotic roll forming processes are primarily designed for machining parts with uniform cross-sections. Their pneumatic clamping systems generally employ traditional PID control algorithms with fixed parameters, which struggle to effectively address the inherent characteristics of pneumatic systems, such as high gas compressibility and severe nonlinearity. This results in lag in clamping force response, significant overshoot, and insufficient disturbance resistance. Furthermore, existing tooling often uses fixed-width molds or passive support mechanisms, failing to provide adaptive support as the sheet metal cross-section changes. This causes the edges and web areas of the sheet metal to be suspended during variable cross-section forming, leading to forming defects such as web warping and edge wrinkling. These problems severely restrict the accuracy and quality stability of robotic roll forming of high-strength steel parts with variable cross-sections. Summary of the Invention

[0005] In view of the above problems, an intelligent adaptive variable cross-section sheet metal robot roll forming equipment and roll forming method are proposed to overcome or at least partially solve the above problems. Specifically: A smart adaptive variable cross-section sheet metal robotic roll forming equipment (also known as a robotic roll forming system for variable cross-section sheets) includes: The robot module has a rolling tool head installed at its end; The flexible tooling module includes a pneumatic clamping unit and a flexible support unit. The sensing module includes a pressure sensor and a position sensor. The control module runs a fuzzy PID adaptive control algorithm to adjust the clamping pressure of the pneumatic clamping unit to a constant value and to control the flexible support unit to maintain synchronous follow-up support with the end of the robot module during the roll forming process.

[0006] Optionally, the flexible tooling module also includes a base with a linear guide rail. The pneumatic clamping unit and the flexible support unit are mounted on the linear guide rail and their relative positions can be adjusted along the linear guide rail. The pneumatic clamping unit is used to fix the metal sheet, and the flexible support unit is used to provide support for the metal sheet during the roll forming process.

[0007] Optionally, the flexible support unit includes at least one set of support sliders that can translate along a direction perpendicular to the metal sheet feed direction, and a servo drive mechanism that drives the support sliders to move. The control module controls the servo drive mechanism to adjust the position of the support sliders by outputting drive control signals, so as to achieve synchronous follow-up support between the flexible support unit and the end of the robot module.

[0008] Optionally, the flexible support unit and the end of the robot module maintain synchronous follow-up support, including: during the roll forming process, the control module calculates the target support position of the flexible support unit according to the real-time position coordinates of the end of the robot module, and drives the support slider to move to the target support position through the drive control signal, so as to maintain continuous contact support to the inner wall of the metal sheet.

[0009] Optionally, the metal sheet can be any one of martensitic steel, DP800 duplex steel or QP1180 steel, the forming force applied by the rolling tool head during the rolling process ranges from 100N to 4000N, and the clamping pressure applied by the pneumatic clamping unit is constant at 600N.

[0010] Optionally, the sensing module communicates bidirectionally with the pneumatic clamping unit and the flexible support unit; the sensing module also includes a vision sensor installed at the end of the robot module, which is used to collect the springback data of the metal sheet after roll forming, and the control module is also configured to correct the roll forming trajectory of the robot module based on the springback data.

[0011] Optionally, when the control module runs the fuzzy PID adaptive control algorithm, it takes the pressure deviation and the rate of change of the pressure deviation between the clamping pressure data collected by the pressure sensor and the target clamping pressure value as input, calculates the PID parameter correction online based on the preset fuzzy control rule table, and outputs the adjustment control signal to the electro-proportional valve of the pneumatic clamping unit to stabilize the clamping pressure.

[0012] Optionally, the control module is also configured to: during the roll forming process, when the clamping pressure data collected by the pressure sensor fluctuates beyond a preset threshold, trigger the flexible support unit to perform support position compensation movement.

[0013] Optionally, the flexible support unit determines the initial posture parameters based on the pre-formed bending creases and bending angle data, and adjusts the support posture in real time based on the position data collected by the position sensor during the roll forming process.

[0014] A robotic roll forming method for variable cross-section plates includes: S1. Based on the thickness parameters and bending angle requirements of the metal sheet to be processed, generate the rolling trajectory of the robot module and the initial posture of the flexible support unit. S2. Clamp the metal sheet in the pneumatic clamping unit and adjust the clamping pressure to a preset constant value through the control module; S3. The control robot module rolls the metal sheet according to the rolling trajectory. During the rolling process, the control module dynamically adjusts the clamping pressure of the pneumatic clamping unit based on the pressure and position data collected by the sensor module, and drives the flexible support unit to keep synchronous and follow the support of the robot module end.

[0015] This invention achieves rapid, overshoot-free closed-loop adjustment of pneumatic clamping pressure by integrating a fuzzy PID adaptive control algorithm with flexible support follow-up control, effectively suppressing the impact of inherent nonlinear hysteresis of the pneumatic system on forming stability. Simultaneously, the flexible support unit moves synchronously with the robot end effector, eliminating the suspended area of ​​the sheet metal during the variable cross-section forming process, thus fundamentally suppressing web warping and edge wrinkling defects. The roll forming equipment only requires adjustment of control parameters to adapt to the production of variable cross-section parts of different specifications, significantly shortening the changeover cycle and providing a highly flexible and high-precision equipment solution for the precision roll forming of high-strength steel variable cross-section components. Attached Figure Description

[0016] To more clearly illustrate the technical solution of the present invention, the accompanying drawings used in the description of the present invention will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a schematic diagram of an intelligent adaptive variable cross-section plate robot roll forming equipment provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the variable cross-section roll forming process provided in an embodiment of the present invention; Figure 3This is a schematic diagram of the flexible tooling adjustment angle provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of the pneumatic servo circuit provided in an embodiment of the present invention; Figure 5 This is a schematic diagram of an example processed sheet metal provided in an embodiment of the present invention; Figure 6 This is a Simulink simulation comparison diagram of fuzzy PID control and traditional PID control provided in the embodiments of the present invention; Figure 7 This is a flowchart of the fuzzy PID closed-loop control provided in an embodiment of the present invention. Detailed Implementation

[0018] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0019] This invention provides an intelligent adaptive variable cross-section plate robotic roll forming equipment, which may specifically include: Robot module 1, with a rolling tool head 2 installed at the end of the robot module; The flexible tooling module includes a pneumatic clamping unit 4 and a flexible support unit 3. The sensing module includes a pressure sensor and a position sensor. The control module runs a fuzzy PID adaptive control algorithm to adjust the clamping pressure of the pneumatic clamping unit 4 to a constant value and control the flexible support unit 3 to maintain synchronous follow-up support with the end of the robot module during the roll forming process.

[0020] Specifically, robot module 1 can be a KUKA robot, with a rolling head 2 fixedly mounted on its end flange. The rolling head 2 is used to apply rolling forming force to the surface of the metal sheet during the movement of robot module 1. Robot module 1 can drive the rolling head 2 to move sequentially along the surface of the sheet 6 according to a preset motion trajectory. Local plastic deformation occurs in the contact area between the rolling head 2 and the sheet 6, thereby achieving progressive forming of the sheet. The motion trajectory of robot module 1 can be pre-planned and generated by the control module based on the geometric parameters of the part to be processed, and converted into a sequence of instructions that robot module 1 can execute.

[0021] The flexible tooling module includes a pneumatic clamping unit 4 and a flexible support unit 3.

[0022] The pneumatic clamping unit 4 is used to clamp and fix the metal sheet to be processed before the roll forming begins, ensuring that there is no relative displacement between the metal sheet and the robot module 1 during the entire roll forming process. The pneumatic clamping unit 4 can be composed of a cylinder, grippers, and a matching pneumatic circuit. The pneumatic circuit is equipped with an electro-proportional valve. By adjusting the control voltage of the electro-proportional valve, the gas pressure entering the cylinder can be changed, thereby adjusting the clamping pressure applied by the grippers to the edge of the sheet.

[0023] The flexible support unit 3 can be arranged below or to the side of the metal sheet to provide adjustable support for the sheet during roll forming. The flexible support unit 3 includes a support member movable in at least one direction and a servo drive mechanism that drives the support member. The support member contacts the inner wall or bottom surface of the sheet, and its position can be changed to adapt to changes in the sheet cross-section. The support position of the flexible support unit 3 is dynamically calculated and driven by the control module based on the real-time position of the robot module's end effector.

[0024] In some embodiments of the present invention, the flexible support unit 3 may also be composed of several independent support rods, without a displacement sensor. The width adjustment of the support rods is displayed by mechanical scale readings. Those skilled in the art can choose this option themselves.

[0025] The sensing module includes a pressure sensor and a position sensor.

[0026] A pressure sensor is installed in the pneumatic circuit of the pneumatic clamping unit 4 to collect the gas pressure in the working chamber of the cylinder in real time and convert the pressure signal into an electrical signal for transmission to the control module. The control module determines the actual state of the current clamping pressure based on the deviation between the pressure value fed back by the pressure sensor and the target pressure value.

[0027] The position sensor is installed on the moving part of the flexible support unit 3 to detect the current position of the support component in real time and transmit the position signal to the control module, providing feedback for the closed-loop position control of the flexible support unit 3.

[0028] The control module can run a fuzzy PID adaptive control algorithm. During the roll forming process, the control module receives pressure data collected by the pressure sensor, calculates the pressure deviation between the current clamping pressure and the preset constant pressure value, and the rate of change of this pressure deviation. These two variables are used as inputs for fuzzy inference. Based on pre-established fuzzy control rules, fuzzy inference is performed online to calculate the proportional coefficient correction, integral coefficient correction, and derivative coefficient correction. The corrected PID parameters are then used to calculate the control signal output to the electro-proportional valve. This control signal can be an analog voltage signal or a pulse width modulation signal, controlling the valve opening of the electro-proportional valve, thereby adjusting the gas pressure entering the cylinder and stabilizing the clamping pressure of the pneumatic clamping unit 4 at a preset constant value.

[0029] Meanwhile, the control module can also calculate the target support position of the flexible support unit 3 based on the real-time position coordinates of the robot module's end effector at each sampling moment, combined with a pre-set support position mapping relationship. The control module compares the target support position with the current actual position fed back by the position sensor, generates a drive control signal through a position closed-loop control algorithm, and outputs it to the servo drive mechanism of the flexible support unit 3 to drive the support component to move to the target support position. During the roll forming process, this position calculation and drive process is executed cyclically with a fixed control cycle, so that the support component of the flexible support unit 3 always follows the movement of the robot module's end effector and synchronously adjusts its support posture, forming continuous contact support for the inner wall or bottom surface of the sheet metal.

[0030] In one or more embodiments of the present invention, the flexible tooling module further includes a base 5, on which a linear guide rail is provided. The pneumatic clamping unit 4 and the flexible support unit 3 are mounted on the linear guide rail and their relative positions can be adjusted along the linear guide rail. The pneumatic clamping unit 4 is used to fix the metal sheet, and the flexible support unit 3 is used to provide support for the metal sheet during the roll forming process.

[0031] Base 5 serves as the fundamental load-bearing component of the flexible tooling module. It can be made of metal and fixedly installed on the working surface or platform. The upper surface of base 5 is machined with a mounting plane, on which linear guide rails are fixedly laid. The extension direction of the linear guide rails is parallel to the feed direction of the metal sheet. There can be two linear guide rails, arranged parallel to each other. The cross-sectional shape of the linear guide rails is rectangular or trapezoidal, and the upper surface and both sides of the linear guide rails are precision ground to form sliding mating surfaces.

[0032] The pneumatic clamping unit 4 may include a clamping unit base, a cylinder, grippers, and a clamping unit slider.

[0033] The clamping unit slider can be fixedly installed on the bottom surface of the clamping unit base, forming a sliding fit with the sliding mating surface of the linear guide rail. The clamping unit slider can reciprocate along the extension direction of the linear guide rail. The clamping unit base can also be equipped with a clamping unit locking mechanism, which is used to lock the clamping unit base onto the linear guide rail after the pneumatic clamping unit 4 is adjusted to the target position, preventing positional displacement due to vibration or force during roll forming. A cylinder is fixedly installed on the upper surface of the clamping unit base, with its piston rod end connected to a gripper. The front end of the gripper has a clamping surface for contacting the edge of the metal sheet. The number of pneumatic clamping units 4 is at least two sets, with the two sets of pneumatic clamping units 4 respectively arranged at the two side edges of the metal sheet to symmetrically clamp and fix the metal sheet from both sides.

[0034] The flexible support unit 3 includes a support unit base, a support component, a servo drive mechanism, and a support unit slider.

[0035] The support unit slider is fixedly mounted on the bottom surface of the support unit base, forming a sliding fit with the sliding mating surface of the linear guide rail. The support unit slider can reciprocate along the extension direction of the linear guide rail. A support unit locking mechanism is also provided on the support unit base, which is used to lock and fix the support unit base to the linear guide rail after the flexible support unit 3 is adjusted to the target position. A servo drive mechanism is fixedly mounted on the upper surface of the support unit base, including a servo motor and a transmission assembly. The transmission assembly converts the rotational motion of the servo motor into the linear motion of the support member. The support member is mounted on the output end of the transmission assembly. Driven by the servo drive mechanism, the support member can translate in a direction perpendicular to the metal sheet feed direction, thereby changing the contact position between the support member and the inner wall or bottom surface of the metal sheet.

[0036] Before the roll forming process begins, the operator can loosen the clamping unit locking mechanism and the support unit locking mechanism according to the cross-sectional shape and size parameters of the metal sheet to be processed, and manually push the pneumatic clamping unit 4 and the flexible support unit 3 to the preset initial position along the linear guide rail, so that the cylinder jaws of the pneumatic clamping unit 4 are aligned with the edge clamping position of the metal sheet, and the support member of the flexible support unit 3 is aligned with the initial support position of the metal sheet.

[0037] After the position adjustment is completed, the operator locks the clamping unit locking mechanism and the support unit locking mechanism to complete the initial position setting of the flexible tooling module.

[0038] During the roll forming process, the pneumatic clamping unit 4 applies constant clamping pressure to the edge of the metal sheet through the cylinder jaws to fix the metal sheet, and the flexible support unit 3 drives the support component to translate through the servo drive mechanism to provide continuous contact support in the area where the cross-section of the metal sheet changes.

[0039] The relative positions of the pneumatic clamping unit 4 and the flexible support unit 3 can be flexibly adjusted along the linear guide rail, enabling the flexible tooling module to adapt to the processing requirements of variable cross-section plates of different specifications and sizes, and to achieve rapid reconfiguration of the tooling layout without replacing hardware components.

[0040] In one or more embodiments of the present invention, the flexible support unit 3 includes at least one set of support sliders that can be translated along a direction perpendicular to the metal sheet feeding direction, and a servo drive mechanism that drives the support sliders to move. The control module controls the servo drive mechanism to adjust the position of the support sliders by outputting a drive control signal, so as to achieve synchronous follow-up support between the flexible support unit 3 and the end of the robot module.

[0041] The support slider can be a block-shaped component made of metal, with its upper surface having a support surface for contacting the inner wall or bottom surface of the metal sheet. This support surface is hardened to enhance wear resistance. The bottom of the support slider may have a sliding fit structure that mates with a linear guide pair fixedly mounted on the support unit base. The extension direction of the linear guide pair is perpendicular to the feed direction of the metal sheet. Under the constraint of the linear guide pair, the support slider can only perform linear reciprocating motion in a single direction perpendicular to the feed direction of the metal sheet.

[0042] The servo drive mechanism may include a servo motor, a coupling, a ball screw pair, and a screw nut seat.

[0043] The servo motor can be fixedly mounted on the support unit base, and its output shaft is connected to the lead screw shaft of the ball screw assembly via a coupling. The nut of the ball screw assembly is fixedly connected to the lead screw nut seat, and the lead screw nut seat is rigidly connected to the bottom of the support slider via a connecting plate.

[0044] When the servo motor receives the drive control signal output by the control module, the servo motor drives the ball screw pair to rotate. The ball screw pair converts the rotational motion of the screw shaft into the linear motion of the nut along the screw axis. The nut drives the support slider to translate in a direction perpendicular to the feeding direction of the metal sheet through the screw nut seat and connecting plate.

[0045] The servo motor integrates an encoder, which detects the rotor angular displacement of the servo motor in real time and feeds the angular displacement signal back to the servo driver. The servo driver performs closed-loop position adjustment based on the deviation between the angular displacement signal and the target position command, thereby controlling the support slider to move precisely to the target position.

[0046] During the roll forming process, the control module can perform support position calculation and control command output at a fixed control cycle.

[0047] Within each control cycle, the control module first acquires the real-time position coordinates of the robot module's end effector. These coordinates are transmitted from the robot module 1's control system to the control module via a communication interface. The control module calculates the target support position of the flexible support unit 3 corresponding to the current position of the robot module's end effector based on a pre-established mapping relationship. This mapping relationship is pre-calculated and stored in the control module's storage unit based on the variable cross-sectional geometry of the metal sheet and pre-formed bending creases and angle data. The control module compares the target support position with the actual position of the support slider fed back by the position sensor, calculates the position deviation, and generates a speed command value using a position proportional-integral-derivative (PID) control algorithm. This speed command value is converted from digital to analog signal to form an analog voltage-based drive control signal. The drive control signal is transmitted via a signal cable to the servo driver of the servo drive mechanism. The servo driver drives the servo motor to rotate according to the drive control signal, which in turn drives the support slider to the target support position via a ball screw pair.

[0048] As the end effector of the robot module moves continuously along the rolling trajectory, its real-time position coordinates are constantly updated. Within each control cycle, the control module repeatedly executes the aforementioned position calculation, deviation calculation, and drive control signal output process, continuously adjusting the support slider's position to ensure that the support surface of the slider remains in contact with the inner wall or bottom surface of the metal sheet, following the movement of the robot module's end effector. When the robot module's end effector moves in a region where the metal sheet's cross-section widens, the control module controls the support slider to translate outwards along a direction perpendicular to the feed direction to expand the support range; conversely, when the robot module's end effector moves in a region where the metal sheet's cross-section narrows, the control module controls the support slider to translate inwards along a direction perpendicular to the feed direction to shrink the support range.

[0049] The position adjustment of the support slider is synchronized with the movement of the end effector of the robot module in time. The support surface of the support slider always applies continuous support force to the inner wall or bottom surface of the metal sheet to prevent the web of the metal sheet from warping due to lack of bottom support under the action of roll forming force, and at the same time prevent the edge area of ​​the metal sheet from wrinkling due to lack of lateral constraint.

[0050] In one or more embodiments of the present invention, the flexible support unit 3 and the end of the robot module maintain synchronous follow-up support, including: during the roll forming process, the control module calculates the target support position of the flexible support unit 3 according to the real-time position coordinates of the end of the robot module, and drives the support slider to move to the target support position through the drive control signal, so as to maintain continuous contact support to the inner wall of the metal sheet.

[0051] The control module can pre-store a mapping table between the end position coordinates of the robot module and the target support position of the flexible support unit 3.

[0052] The process of establishing the mapping table may include: first, extracting the cross-sectional width and height values ​​at each discrete cross-sectional position along the length of the metal sheet based on the cross-sectional geometric design data of the metal sheet to be processed; then, calculating the target position value of the support slider corresponding to each discrete cross-sectional position based on the installation position of the flexible support unit 3 in the tooling coordinate system and the movable range of the support slider; and finally, storing the discrete position coordinates of the end of the robot module and the corresponding target position value of the support slider in the storage unit of the control module in the form of a data table.

[0053] During the roll forming process, for the robot module end position coordinates that are not directly listed in the mapping table, the control module uses a linear interpolation algorithm to calculate the corresponding support slider target position value. The linear interpolation algorithm calculates the interpolated position value between two adjacent discrete position points based on the distance ratio between the current coordinates of the robot module end and the coordinates of the two discrete points.

[0054] At the start of each control cycle, the control module obtains the real-time position coordinates of the robot module's end effector in the workpiece coordinate system from the robot controller of robot module 1. The workpiece coordinate system is a Cartesian coordinate system fixed to the metal sheet, established before the roll forming process begins through a workpiece calibration program. Robot module 1 controller sends data packets containing the robot module's end effector position coordinates to the control module at fixed communication intervals. The data packet format includes a timestamp, X-axis coordinate value, Y-axis coordinate value, and Z-axis coordinate value fields. Upon receiving the data packet, the control module parses it, extracts the real-time position coordinate values ​​of the robot module's end effector, and transforms the real-time position coordinates from the workpiece coordinate system to the tooling coordinate system based on a pre-stored coordinate transformation matrix between the workpiece coordinate system and the tooling coordinate system. The tooling coordinate system is a Cartesian coordinate system fixed to the base 5 of the flexible tooling module. The coordinate transformation matrix is ​​pre-determined and stored in the control module through a hand-eye calibration program.

[0055] The control module inputs the converted coordinates of the robot module's end effector coordinate system to the mapping relationship calculation module. The mapping relationship calculation module calculates the corresponding target support position value for the flexible support unit 3 based on the stored mapping relationship table and a linear interpolation algorithm. The target support position value is represented by the displacement of the support slider on the linear guide pair, with the zero point of the displacement being the displacement count value when the support slider is at its innermost extreme position. The control module compares the calculated target support position value with the actual position value of the support slider as fed back in real time by the position sensor. The position sensor outputs the displacement of the support slider in the form of orthogonal pulse signals. The position sensor can be a grating ruler displacement sensor or a magnetic grating ruler displacement sensor mounted on the side of the linear guide pair. The position comparator in the control module calculates the position deviation value between the target support position value and the actual position value, which is expressed as the displacement difference in mm.

[0056] The position deviation value is input to the position proportional-integral-derivative (PID) controller. The proportional, integral, and derivative coefficients of the PID controller are pre-tuned and stored in the parameter storage area of ​​the control module based on the dynamic characteristics of the servo drive mechanism and transmission components. The PID controller calculates the speed command value based on the position deviation value, which represents the rotational angular velocity that the servo motor should output. The speed command value is converted into an analog voltage signal by a digital-to-analog converter (DAC), which serves as the drive control signal. The amplitude of the analog voltage signal is linearly proportional to the speed command value, and the polarity of the analog voltage signal indicates the rotation direction of the servo motor.

[0057] The drive control signal can be transmitted to the analog input port of the servo driver of the servo drive mechanism via a shielded signal cable. The servo driver internally contains a speed closed-loop control loop and a current closed-loop control loop. After receiving the drive control signal, the servo driver first converts the corresponding analog voltage value into a speed setpoint, and then drives the servo motor to rotate at the speed corresponding to the speed setpoint through the speed closed-loop control loop. The rotational motion of the servo motor is transmitted to the lead screw shaft of the ball screw pair through a coupling. The rotation of the lead screw shaft drives the nut of the ball screw pair to move along the lead screw axis. The nut, through the lead screw nut seat and connecting plate, drives the support slider to translate in a direction perpendicular to the metal sheet feed direction. The actual position of the support slider is detected in real time by a position sensor and fed back to the control module, forming a position closed-loop control.

[0058] During the roll forming process, the above-mentioned cycle of position calculation, deviation calculation, drive control signal generation and position closed-loop control is repeatedly executed with a fixed control cycle. The time length of the control cycle can be freely set according to actual needs.

[0059] At each moment the robot module's end effector moves continuously along the rolling trajectory, the control module updates the target support position value based on the latest real-time position coordinates of the robot module's end effector and drives the support slider to move to the updated target support position. The support surface of the support slider always adjusts its support position following the movement of the robot module's end effector, thereby maintaining continuous contact support to the inner wall of the metal sheet. Continuous contact support means that the support surface of the support slider and the inner wall or bottom surface of the metal sheet are always in physical contact, with no gap between the support surface and the sheet surface. The support force applied by the support surface to the sheet is perpendicular to the sheet surface and its magnitude meets the process requirements, thus counteracting the vertical component of the rolling forming force applied to the sheet surface by the rolling tool head 2, preventing the sheet from deforming and detaching from the support surface under the action of the rolling forming force.

[0060] When the end effector of the robot module moves along the rolling track to the area where the cross section of the metal sheet widens, the target support position value calculated by the control module gradually increases. The drive control signal controls the servo motor to rotate forward, and the support slider moves outward along a direction perpendicular to the feed direction. The support surface moves outward synchronously as the cross section of the sheet widens. When the end effector of the robot module moves along the rolling track to the area where the cross section of the metal sheet narrows, the target support position value calculated by the control module gradually decreases. The drive control signal controls the servo motor to rotate in reverse, and the support slider moves inward along a direction perpendicular to the feed direction. The support surface moves inward synchronously as the cross section of the sheet narrows.

[0061] The movement speed of the support slider is matched with the component of the movement speed of the robot module end along the rolling trajectory perpendicular to the feed direction. The matching relationship is determined by the mapping table and the derivative relationship in the linear interpolation algorithm, so that the position change of the support slider and the position change of the robot module end maintain a synchronous correspondence in space.

[0062] In a preferred embodiment of the present invention, the metal sheet can be any one of martensitic steel, DP800 duplex steel or QP1180 steel, the forming force applied by the rolling tool head 2 during the rolling process ranges from 100N to 4000N, and the clamping pressure applied by the pneumatic clamping unit 4 is constant at 600N.

[0063] Specifically, the metal sheet can be any one of martensitic steel, duplex steel DP800, or QP1180 steel, with a thickness ranging from 1.0 mm to 4.0 mm. The forming force applied to the surface of the metal sheet by the rolling head 2 during the rolling process ranges from 100 N to 4000 N, while the clamping pressure applied to the edge of the metal sheet by the pneumatic clamping unit 4 is constant at 600 N. A force sensor can be installed at the end of the robot module, located between the end flange of the robot module and the rolling head 2. The force sensor is used to detect the actual rolling pressure value applied by the rolling head 2 to the surface of the metal sheet in real time, and converts the actual rolling pressure value into an electrical signal that is transmitted to the control module. The control module performs constant force rolling control based on the actual rolling pressure value fed back by the force sensor.

[0064] In one or more embodiments of the present invention, the force sensor can be a six-dimensional force sensor or a single-axis force sensor. The six-dimensional force sensor can simultaneously detect force components and torque components in three orthogonal directions, while the single-axis force sensor only detects unidirectional force components along the axis of the rolling tool head 2 or along the normal direction of the metal sheet. Those skilled in the art can freely choose according to actual needs.

[0065] The control module stores the rolling pressure setting value. The rolling pressure setting value is input by the operator through the human-machine interface before the rolling forming starts, according to the material grade, thickness and bending angle requirements of the metal sheet. The rolling pressure setting value can be selected in the range of 100N to 4000N.

[0066] During the roll forming process, the control module compares the actual roll pressure value obtained in each sampling period with the set roll pressure value, and calculates the roll pressure deviation between the actual roll pressure value and the set roll pressure value. The control module determines whether the absolute value of the roll pressure deviation exceeds a preset roll pressure allowable error threshold. When the actual roll pressure value exceeds the sum of the set roll pressure value and the roll pressure allowable error threshold, the control module determines that the current actual roll pressure value is out of the set range and needs to perform a lifting compensation action to reduce the roll pressure.

[0067] The lifting compensation action can be implemented in the following way: The control module calculates the lifting compensation displacement of the robot module's end effector along the normal direction of the metal sheet, moving away from the metal sheet surface, based on the rolling pressure deviation value. The calculation of the lifting compensation displacement uses a proportional control algorithm; the lifting compensation displacement equals the product of the rolling pressure deviation value and a preset force-position conversion coefficient. The value of the force-position conversion coefficient can be pre-calibrated based on the equivalent stiffness of the metal sheet during the rolling forming process and stored in the parameter storage area of ​​the control module. The control module superimposes the calculated lifting compensation displacement onto the Z-axis coordinate component of the preset rolling trajectory position command of the robot module's end effector within the current control cycle. The Z-axis direction is perpendicular to the normal direction of the metal sheet surface, and the superimposed Z-axis coordinate value is the corrected Z-axis coordinate value. The control module sends the target position command of the robot module's end effector, containing the corrected Z-axis coordinate value, to the robot controller of robot module 1 via the communication interface. The robot controller drives the joints of robot module 1 to move according to the received target position command, causing the robot module's end effector to perform a slight lifting action, the direction of which is away from the metal sheet surface.

[0068] After the robot module's end effector lifts, the contact deformation between the rolling head 2 and the metal sheet surface decreases, and the actual rolling pressure applied by the rolling head 2 to the metal sheet surface decreases accordingly. The control module continuously monitors the actual rolling pressure value fed back by the force sensor. When the actual rolling pressure value falls back to the range limited by the rolling pressure set value and the rolling pressure allowable error threshold, the control module stops calculating and outputs the lifting compensation displacement. The robot module's end effector maintains the corrected position and continues to move along the rolling trajectory. When the actual rolling pressure value is lower than the difference between the rolling pressure set value and the rolling pressure allowable error threshold, the control module determines that the current actual rolling pressure value is below the set range and needs to perform a downward compensation action to increase the rolling pressure.

[0069] The implementation of the downward compensation action is similar to that of the lifting compensation action. The control module calculates the downward compensation displacement, which is directed toward the surface of the metal sheet. The control module adds the downward compensation displacement to the Z-axis coordinate component and sends it to the robot controller. The end of the robot module performs a micro-downward action to increase the contact deformation between the rolling tool head 2 and the surface of the metal sheet, thereby restoring the actual rolling force value to the set range.

[0070] The aforementioned constant-force rolling control process is continuously executed with a fixed control cycle during the rolling forming process. The control cycle is synchronized with the sampling cycle of the force sensor, ensuring that the control module can respond promptly and perform compensation actions when the actual rolling force value deviates from the set range, so that the forming force during the rolling forming process is always maintained within the set range. The clamping pressure applied by the pneumatic clamping unit 4 is constant at 600N. The clamping pressure is adjusted in a closed loop through a fuzzy PID adaptive control algorithm to ensure constancy. The closed-loop control of the clamping pressure and the constant-force control of the rolling force operate independently and in parallel without interfering with each other.

[0071] In a preferred embodiment of the present invention, the sensing module communicates bidirectionally with the pneumatic clamping unit 4 and the flexible support unit 3; the sensing module also includes a vision sensor installed at the end of the robot module, the vision sensor is used to collect the springback data of the metal sheet after roll forming, and the control module is also configured to correct the roll forming trajectory of the robot module 1 based on the springback data.

[0072] Specifically, the sensing module can send pressure data collected by the pressure sensor and position data collected by the position sensor to the control module. Simultaneously, it receives parameter configuration commands from the control module and forwards them to the electro-proportional valve of the pneumatic clamping unit 4 and the servo driver of the flexible support unit 3. The bidirectional communication protocol can use CANopen or EtherCAT.

[0073] The sensing module also includes a vision sensor installed at the end of the robot module. The vision sensor can be an industrial camera or a laser profile sensor, and is fixedly installed on the mounting bracket on the side of the rolling tool head 2, with the optical axis direction perpendicular to the surface of the metal sheet.

[0074] After each rolling forming pass, robot module 1 moves the vision sensor along the surface of the formed area, scanning it. The vision sensor collects three-dimensional point cloud data or two-dimensional contour curve data of the metal sheet surface at a fixed sampling interval and transmits the collected data to the control module. The control module runs a springback calculation program, comparing the actual contour data collected by the vision sensor with the theoretical contour data to calculate the springback value of the metal sheet in the cross-sectional height direction or bending angle direction.

[0075] The control module corrects the rolling trajectory of robot module 1 in the next rolling forming pass based on the calculated springback value. The correction method can be as follows: multiply the springback value by a preset bending compensation coefficient to obtain the trajectory correction amount. The bending compensation coefficient is pre-calibrated according to the metal sheet material grade and thickness, with a value range of 1.0 to 2.0. The trajectory correction amount is then superimposed onto the theoretical rolling trajectory of the current pass, causing the robot module's end effector to generate a bending displacement opposite to the springback direction during the next rolling pass, thereby compensating for the shape deviation caused by springback. The corrected rolling trajectory is then used to generate joint motion commands executable by robot module 1 through motion planning and sent to the robot controller for execution.

[0076] In one or more embodiments of the present invention, when the control module runs the fuzzy PID adaptive control algorithm, the pressure deviation and the rate of change of the pressure deviation between the clamping pressure data collected by the pressure sensor and the target clamping pressure value are used as inputs. Based on the preset fuzzy control rule table, the PID parameter correction is calculated online, and the adjustment control signal is output to the electro-proportional valve of the pneumatic clamping unit 4 to stabilize the clamping pressure.

[0077] Specifically, the pressure sensor can be installed in the air inlet or cylinder chamber of the pneumatic clamping unit 4. The output signal of the pressure sensor is a 4mA to 20mA current signal or a 0V to 10V voltage signal, and the signal amplitude is linearly related to the measured gas pressure. The analog-to-digital conversion unit in the control module samples and quantizes the pressure sensor signal at a sampling frequency of 100Hz to 1000Hz to obtain the digital quantity of clamping pressure data. The control module compares the clamping pressure data acquired in the current sampling period with the pressure value corresponding to the preset target clamping pressure value of 600N, and calculates the pressure deviation E. The calculation formula is E = target pressure value - actual pressure value. The control module further calculates the rate of change of pressure deviation EC. The calculation formula is EC = E(k) - E(k-1), where E(k) is the pressure deviation value of the current sampling period, E(k-1) is the pressure deviation value of the previous sampling period, and the sampling period interval is used to normalize the dimension of the rate of change.

[0078] The control module performs fuzzification processing on the calculated pressure deviation E and pressure deviation change rate EC. The fuzzification process includes two steps: quantization mapping and membership degree calculation.

[0079] Quantization mapping maps the actual range of pressure deviation E to a preset fuzzy universe of discourse, which is a set of discrete integers. The actual range of pressure deviation change rate EC is also mapped to the same fuzzy universe of discourse. After quantization mapping, the control module calculates the membership degree of the mapped values ​​to each preset fuzzy subset. Fuzzy subsets can be represented by seven linguistic variable values: negative large, negative medium, negative small, zero, positive small, positive medium, and positive large. The membership function can be a triangular function or a Gaussian function. The parameters of the triangular membership function include the center value and the base width, while the parameters of the Gaussian membership function include the mean and standard deviation. These parameters are loaded during control module program initialization.

[0080] The fuzzy inference engine performs fuzzy inference operations using the Mamdani inference method, based on a pre-established fuzzy control rule table and the membership degrees of E and EC to each fuzzy subset calculated within the current sampling period. The fuzzy control rule table can contain forty-nine control rules, each rule taking the form of a conditional statement: "If E is A and EC is B, then ΔKp is C, ΔKi is D, and ΔKd is E".

[0081] The design of fuzzy control rules follows the dynamic response law of aerodynamic systems. Specific principles may include: firstly, normalizing the input pressure deviation to a range of "-5, 5". When the absolute value of the pressure deviation E is in the initial response phase (2, 6), the rule outputs a positive proportional coefficient correction ΔKp (e.g., increasing ΔKp from its initial value to 0.4) to accelerate the system response, while setting the integral coefficient correction ΔKi to zero to avoid overshoot caused by integral saturation, and taking a differential coefficient correction ΔKd in the range of 0 to 0.02 to prevent the differential term from being overly sensitive to noise; when the absolute value of the pressure deviation E is in the range of [0.5, 2], the rule outputs a proportional coefficient correction ΔKp of -0.3 to reduce overshoot tendency, while increasing the integral coefficient correction ΔKi to gradually eliminate steady-state error, and setting the micro... The proportional coefficient correction ΔKd is set to 0.04 to balance response speed and stability. When the absolute value of the pressure deviation E is in the range of [0, 0.5], which is close to steady state, the rule outputs a proportional coefficient correction ΔKp of -0.5 to avoid oscillation. The integral coefficient correction ΔKi is adjusted to its preset upper limit to quickly eliminate residual steady error. The differential coefficient correction ΔKd is adjusted according to the sign of the pressure deviation change rate EC. When EC is positive, that is, the deviation is increasing, ΔKd is set to 0.06 to suppress the deviation from continuing to increase. When EC is negative, that is, the deviation is decreasing, ΔKd is set to 0.02 to avoid overcompensation causing oscillation.

[0082] The fuzzy inference operation calculates the applicability of each activated rule, which is the smaller of the membership degrees E and EC in the rule's antecedents. The inference engine truncates or multiplies the membership functions of the output quantities ΔKp, ΔKi, and ΔKd based on the applicability of each rule, and performs a union operation on the inference results of all rules for the same output quantity to obtain the total fuzzy output quantities of ΔKp, ΔKi, and ΔKd. Defuzzification is performed using the centroid method, which calculates the abscissa value of the centroid of the area enclosed by the membership function curve of the total fuzzy output quantity. This abscissa value is the precise correction value for ΔKp, ΔKi, and ΔKd.

[0083] The control module adds the correction values ​​ΔKp, ΔKi, and ΔKd obtained from the defuzzification to the current PID parameter reference values. The reference values ​​are the initial values ​​of the proportional coefficient, integral coefficient, and derivative coefficient, which are pre-tuned according to the characteristics of the aerodynamic system. This results in the corrected real-time proportional coefficient Kp, integral coefficient Ki, and derivative coefficient Kd.

[0084] The control module calculates the output control quantity based on the real-time PID parameters and the current pressure deviation E. The calculation formula for the output control quantity is the incremental PID algorithm formula: Δu(k) = Kp·[E(k) - E(k-1)] + Ki·E(k) + Kd·[E(k) - 2E(k-1) + E(k-2)], where Δu(k) is the output increment for the current control cycle, and E(k), E(k-1), and E(k-2) are the pressure deviation values ​​for the current sampling cycle, the previous sampling cycle, and the sampling cycle before that, respectively. The control module adds the calculated output increment Δu(k) to the control signal value output in the previous control cycle to obtain the control signal value for the current control cycle.

[0085] The control signal value can be a digital quantity ranging from 0 to 65535, corresponding to an analog voltage output range of 0 to 10V. The digital-to-analog converter within the control module converts the digital control signal value into an analog voltage signal of 0 to 10V, which serves as the adjustment control signal. This adjustment control signal is transmitted via a shielded cable to the control port of the electro-proportional valve in the pneumatic circuit of the pneumatic clamping unit 4. The electro-proportional valve linearly adjusts its opening according to the amplitude of the input analog voltage. The valve is completely closed when the analog voltage is zero volts and fully open when the analog voltage is ten volts; the intermediate voltage value is proportional to the valve opening. Changes in the valve opening alter the gas flow rate and pressure entering the cylinder, thereby adjusting the clamping pressure applied by the grippers of the pneumatic clamping unit 4 to the edge of the metal sheet, causing the actual clamping pressure to converge towards the target clamping pressure value of 600N.

[0086] The aforementioned fuzzy PID adaptive control process is repeated within each control cycle, which can be set from 10ms to 50ms, synchronized with the pressure sensor sampling cycle. When the absolute value of the pressure deviation E is greater than the preset threshold, the fuzzy inference engine outputs a correction amount of approximately 0.4 based on the rule base to quickly reduce the deviation. As the pressure deviation E gradually decreases and enters the steady-state region, the fuzzy inference engine gradually enhances the integral action and suppresses the derivative action, ensuring that the actual clamping pressure smoothly approaches the target value without overshoot. When the system is subjected to external disturbances causing instantaneous fluctuations in the clamping pressure, the abrupt changes in the pressure deviation E and the rate of change of deviation EC trigger the fuzzy rules to output the corresponding parameter correction amount, allowing the clamping pressure to quickly recover to the target value.

[0087] In one or more embodiments of the present invention, the control module is further configured to: trigger the flexible support unit 3 to perform support position compensation movement when the clamping pressure data collected by the pressure sensor fluctuates beyond a preset threshold during the roll forming process.

[0088] Specifically, the criterion for judging clamping pressure data fluctuations can be the absolute value of the deviation between the clamping pressure values ​​collected by the pressure sensor over several consecutive sampling periods and the target clamping pressure value. In each control period, the control module compares the current actual clamping pressure value with the target clamping pressure value of 600N and calculates the absolute value of the pressure fluctuation deviation. A preset threshold is stored in the parameter storage area of ​​the control module. For example, when the actual clamping pressure value is lower than 540N or higher than 660N, the control module determines that the clamping pressure data fluctuation exceeds the preset threshold. To avoid false triggering due to instantaneous noise, the control module can use a sliding window filtering algorithm, confirming the trigger condition is met only when the absolute value of the pressure fluctuation deviation exceeds the preset threshold for five consecutive sampling periods.

[0089] Once the triggering condition is met, the control module executes the support position compensation motion triggering program. The control module determines the support position compensation direction and compensation displacement of the flexible support unit 3 based on the sign and amplitude of the pressure fluctuation deviation. When the actual clamping pressure is below 540N, it indicates that the clamping and fixing effect of the pneumatic clamping unit 4 on the metal sheet is weakened, and the sheet may experience slight slippage or loosening. At this time, the control module calculates the positive compensation displacement of the support position, with the compensation direction being to move the support slider closer to the inner wall of the metal sheet to increase the support force on the sheet and compensate for the lack of support caused by insufficient clamping force. When the actual clamping pressure is above 660N, it indicates that excessive clamping pressure may cause indentations or deformation at the edge of the sheet. At this time, the control module calculates the negative compensation displacement of the support position, with the compensation direction being to move the support slider slightly away from the inner wall of the metal sheet to release some support constraints and avoid stress concentration in the sheet due to over-constraint.

[0090] The control module adds the calculated compensation displacement to the target support position value of the flexible support unit 3 within the current control cycle, forming a corrected target support position value. The control module then converts the corrected target support position value into a drive control signal using a position closed-loop control algorithm, outputting it to the servo drive mechanism of the flexible support unit 3. This drives the support slider to move to the corrected position, completing the support position compensation movement. During the support position compensation movement, the control module continuously monitors the clamping pressure data fed back by the pressure sensor. When the clamping pressure data recovers to within a preset threshold range, the control module stops outputting the compensation displacement increment, and the flexible support unit 3 maintains its compensated position and continues to perform the follow-up support task.

[0091] In one or more embodiments of the present invention, the flexible support unit 3 determines the initial posture parameters based on the pre-formed bending creases and bending angle data, and adjusts the support posture in real time based on the position data collected by the position sensor during the roll forming process.

[0092] Pre-formed bending crease and bending angle data can be generated during the roll forming process planning stage. During process planning, based on the target variable cross-sectional geometry of the metal sheet, the sheet is divided into several forming zones along its length. Each zone corresponds to a specific bending crease position line and bending angle value. The bending crease position line defines the trajectory line along which the end-effector roll forming tool 2 of the robot module applies bending deformation to the surface of the metal sheet, and the bending angle value defines the bending angle that the metal sheet should achieve at that trajectory line.

[0093] The initial attitude parameters of the flexible support unit 3 are calculated and determined based on the aforementioned bending crease position line and bending angle value. Specifically: based on the projection coordinates of the bending crease position line in the metal sheet plane, combined with the installation position of the flexible support unit 3 in the tooling coordinate system, the initial reference position of the support slider in the metal sheet feed direction is determined; based on the bending angle value and the metal sheet thickness, the spatial coordinates of the inner wall of the metal sheet at the bending position are calculated through geometric relationships, and then the initial support position value of the support slider perpendicular to the feed direction is calculated, so that the support surface of the support slider is aligned with the inner wall area of ​​the sheet corresponding to the bending crease before the rolling begins.

[0094] The initial attitude parameters include the initial longitudinal position of the support slider along the linear guide and the initial transverse position of the support slider perpendicular to the feed direction. The initial longitudinal position is set by manually pushing the flexible support unit 3 along the linear guide and locking the support unit locking mechanism; the initial transverse position is set by the control module writing the calculated initial support position value into the position control command, driving the servo drive mechanism to move the support slider to this initial position. After the initial attitude is set, the support surface of the support slider makes contact with the inner wall of the metal sheet at the starting position of the bending crease.

[0095] During the roll forming process, a position sensor, a grating ruler displacement sensor mounted on the side of the linear guide pair, collects the actual position data of the support slider in real time, outputting the displacement of the support slider in the form of orthogonal pulse signals. The control module reads the position sensor data at a fixed control cycle to obtain the actual position value of the support slider. Simultaneously, the control module calculates the theoretical target support position value based on the real-time position coordinates of the robot module's end effector. The control module compares the theoretical target support position value with the actual position value to calculate the position following deviation. When the position following deviation exceeds a preset allowable range, the control module generates a drive control signal through a position proportional-integral-derivative controller to drive the servo drive mechanism to adjust the position of the support slider, causing the actual position of the support slider to converge towards the theoretical target support position, thereby achieving real-time adjustment of the support posture.

[0096] This invention also provides a robotic roll forming method for variable cross-section plates, comprising: S1. Based on the thickness parameters and bending angle requirements of the metal sheet to be processed, generate the rolling trajectory of robot module 1 and the initial posture of flexible support unit 3. S2. Clamp the metal sheet in the pneumatic clamping unit 4, and adjust the clamping pressure to a preset constant value through the control module; S3. The control robot module 1 rolls the metal sheet according to the rolling trajectory. During the rolling process, the control module dynamically adjusts the clamping pressure of the pneumatic clamping unit 4 based on the pressure and position data collected by the sensor module, and drives the flexible support unit 3 to keep synchronous follow-up support with the end of the robot module.

[0097] This invention achieves rapid, overshoot-free closed-loop adjustment of pneumatic clamping pressure by integrating a fuzzy PID adaptive control algorithm with flexible support follow-up control, effectively suppressing the impact of inherent nonlinear hysteresis of the pneumatic system on forming stability. Simultaneously, the flexible support unit moves synchronously with the robot end effector, eliminating the suspended area of ​​the sheet metal during the variable cross-section forming process, thus fundamentally suppressing web warping and edge wrinkling defects. The system only requires adjustment of control parameters to adapt to the production of variable cross-section parts of different specifications, significantly shortening the changeover cycle and providing a highly flexible and high-precision equipment solution for the precision roll forming of high-strength steel variable cross-section components.

[0098] The above is the overall concept of the present invention. For ease of understanding, the present invention also provides the following embodiments: Example 1

[0099] This embodiment provides a process for robotic roll forming of variable cross-section parts. The blank used is a martensitic steel plate with a tensile strength of 1000 MPa, an elongation of not less than 5% at room temperature, and a thickness of 1.0 mm. During the forming process, the robot module 1, the roll forming tool head 2, and the flexible tooling module work together to complete the positioning, clamping, and progressive roll forming of the part.

[0100] A schematic diagram of the structure of the robot roll forming flexible tooling is shown below. Figure 1 As shown in the diagram, the variable cross-section roll forming process is as follows: Figure 2 As shown, the flexible tooling can be adjusted in the following ways: Figure 3 As shown, the pneumatic circuit schematic is as follows: Figure 4 As shown, the fuzzy PID closed-loop control flowchart is as follows: Figure 7 As shown, the flexible tooling module includes a base, two pneumatic clamping units 4 disposed at the clamping end, and a flexible support unit 3 composed of independent support surfaces. The two pneumatic clamping units 4 are arranged in a single row along the clamping end. The initial support angle of each independent support rod in the flexible support unit 3 is preset according to the pre-formed bending crease and bending angle, and gradually participates in the support as the clamping end moves upward during each forming process, so as to provide continuous support for the inner wall of the metal sheet.

[0101] The specific process of robot roll forming of variable cross-section parts in this embodiment includes the following steps: S10. Determine the robot's roll forming trajectory: S10-1. Use KUKA robot's dedicated programming language to write robot control programs and set the motion logic and instruction sequence of robot module 1; S10-2. Based on the thickness parameters (1.0mm) of the sheet material to be processed and the bending angle requirements for each pass, calculate the forming trajectory of the rolling tool head 2, mark the trajectory on the surface of the metal sheet material, generate the target path, and adjust the initial angle of each support rod in the flexible tooling according to the bending crease position and bending angle value. S10-3. Without loading the sheet metal, drive robot module 1 to perform an unloaded run according to the generated trajectory data to verify the accuracy and feasibility of the forming trajectory.

[0102] S20. The metal sheet to be processed is placed in the clamping area of ​​the flexible tooling module, and the edges of the sheet are clamped by the pneumatic clamping unit 4. The clamping system uses a fuzzy PID adaptive control algorithm to perform closed-loop adjustment of the clamping force. Specifically, the pressure sensor collects the actual pressure in the cylinder of the pneumatic clamping unit 4 in real time and compares it with the target clamping pressure value of 600N to calculate the pressure deviation E and the pressure deviation change rate EC. E and EC are input to the fuzzy inference engine, which calculates the PID parameter corrections ΔKp, ΔKi, and ΔKd online according to the preset expert rule base. When the initial response error |E|>2, ΔKi=0 and ΔKd (centroid value approximately 0.01) is output within the range [0,0.02], while ΔKp increases to approximately 0.4 to accelerate the response and avoid integral saturation and differential impact. When the error is in the medium range of 0.5<|E|<2, ΔKp=-0.3 is output to reduce the proportional action and ΔKi is increased, while differential compensation ΔKd=0.04 is set. When the error |E|<0.5 approaches steady state, ΔKp further decreases and approaches the lower limit of -0.5, ΔKi increases to its preset upper limit, and ΔKd is dynamically allocated between 0.02 and 0.06 based on the sign of the error change rate EC and the threshold (with ±0.5 as the key limit) to suppress overshoot and eliminate steady-state error. The corrected PID parameters are used to calculate the control signal output to the electro-proportional valve, adjust the valve opening, and thus control the cylinder pressure to stabilize the clamping pressure at 600N. The above-mentioned processes of pressure acquisition, deviation calculation, fuzzy inference, parameter correction, and control output are executed cyclically with a fixed control cycle to form a closed-loop pressure control.

[0103] S30. Based on the preset trajectory verified in step S10, control robot module 1 to perform solid roll forming operation on the metal sheet. A force sensor is installed at the end of the robot module. The force sensor detects the actual rolling pressure exerted by the rolling tool head 2 on the sheet surface in real time and transmits the rolling pressure signal to the control module. The control module performs constant force roll forming control based on the rolling pressure feedback: when the actual rolling pressure is detected to exceed the set value (the maximum forming force is set to 1000N in this embodiment), the control module calculates the slight lifting compensation displacement of the robot module end along the normal direction of the sheet away from the sheet surface, and adds this compensation amount to the position command of the robot module end, driving robot module 1 to perform a lifting action to reduce the rolling pressure and correct the actual rolling pressure back to the set range.

[0104] In this embodiment, the steel plate is securely clamped by pneumatic grippers, and the rolling tool head 2 rolls the steel plate step by step according to the shape requirements of the target part, finally producing a variable cross-section part, such as... Figure 5 As shown. After forming, the resulting part was inspected and found to be stable in clamping, with no obvious web warping or edge wrinkling defects observed.

[0105] Example 2

[0106] This embodiment provides a process for robotic roll forming of variable cross-section parts, using a 2.5mm thick DP800 duplex steel sheet as the blank. Compared to Embodiment 1, the sheet thickness is increased in this embodiment, thus increasing the forming load during the roll forming process to ensure that the sheet can stably undergo plastic deformation along a preset trajectory and meet the shape accuracy requirements of the target part.

[0107] The flexible tooling module includes a base, two pneumatic clamping units 4 arranged in a single row, and a flexible support unit 3 composed of independent support surfaces. Each independent support rod gradually participates in the support as the clamping end moves upward during each forming process, providing continuous and stable support for the sheet metal.

[0108] The specific process steps in this embodiment are as follows: S11. Determine the robot's roll forming trajectory: S11-1. Write the control program using KUKA robot-specific programming language; S11-2. Based on the thickness (2.5mm) of the DP800 duplex steel sheet and the bending angle requirements for each pass, calculate and mark the forming trajectory of the tool head, generate the target path, and adjust the relevant angles of the tooling. S11-3. Run the machine under no-load conditions without loading any sheet metal to verify the accuracy of the trajectory and the rationality of the timing of each support rod.

[0109] S21. Place the sheet metal at the pneumatic clamping end and use the same fuzzy PID closed-loop control algorithm as in Example 1 to adjust and keep the clamping pressure constant at 600N to ensure the positioning stability of the sheet metal in the width variation zone and the web depth variation zone.

[0110] S31. Perform solid rolling forming according to the verified trajectory. The robot's end effector force sensor achieves constant force rolling control, and performs slight lifting compensation when the actual rolling force exceeds the set value. In this embodiment, the maximum forming force is controlled at 2500N.

[0111] After forming, the resulting variable cross-section parts were found to be stable in clamping, with no obvious web warping or edge wrinkling. This indicates that even with increased sheet thickness and forming force, this process can still achieve good forming quality.

[0112] Example 3

[0113] This embodiment provides a process for robotic roll forming of variable cross-section parts, using a 4.0mm thick QP1180 steel plate as the blank. QP1180 steel contains martensite and retained austenite, has a high yield strength, and its springback after forming typically increases with increasing strength. For thicker plates and higher strength materials, this embodiment makes adaptive adjustments to the flexible support participation method, trajectory compensation, and forming load control.

[0114] The flexible tooling module structure is the same as in Example 1. The difference from Example 1 is that, under the load condition of a maximum forming force of 3500N, the flexible support unit 3 gradually participates in the support as the clamping end moves down at a preset angle, forming continuous support for the inner wall of the sheet material, so as to reduce the contour deviation caused by elastic recovery after the thick plate is rolled.

[0115] Specific process steps: S12. Determine the robot's roll forming trajectory: S12-1. Write control programs using the KUKA robot programming language; S12-2. Based on the thickness (4.0mm) of QP1180 steel plate, material deformation characteristics, and bending angle requirements for each pass, calculate and mark the tool head forming trajectory, generate the target path, and adjust the tooling angle; for local areas prone to springback, reserve appropriate compensation in the trajectory planning. S12-3. Run the machine under no-load conditions without loading the sheet metal to verify the accuracy of the trajectory and the rationality of the timing of each support rod.

[0116] S22. Clamp the sheet metal into the pneumatic clamping unit 4, and use fuzzy PID closed-loop control to keep the clamping pressure constant at 600N.

[0117] S32. Perform solid roll forming according to the calibration trajectory. During the roll forming process, each independent support rod in the flexible support unit 3 gradually participates in the support in a preset order, forming continuous support for the inner wall of the sheet metal, providing sufficient reaction force support in the turning area and the area of ​​cross-sectional change, and suppressing local instability. The force sensor realizes constant force roll forming control. In this embodiment, the maximum forming force is controlled to be 3500N.

[0118] Post-forming inspection revealed that the resulting parts were stably clamped, with no obvious web warping or edge wrinkling, and contour deviations were effectively controlled. This indicates that the process is well-suited for thicker plates and higher strength materials.

[0119] Example 4

[0120] This embodiment uses a variable cross-section automotive B-pillar as the processing object. Except for the part's geometric parameters, the other process conditions are the same as in Embodiment 1. The cross-sectional dimensions of this variable cross-section automotive B-pillar continuously change along its length, with the width gradually changing from 80mm at one end to 150mm at the other end. The web depth also changes synchronously along the length, forming a complex variable cross-section structure in which both the width and web depth vary.

[0121] The flexible tooling module includes a base, two pneumatic clamping units 4 arranged in a single row, and a flexible support unit 3 composed of independent support surfaces. Based on pre-formed bending creases, bending angles, and width and web depth parameters at each characteristic section, the initial and final postures of the flexible tooling are predetermined, and the initial support angles of each independent support rod are set. Simultaneously, a correspondence is established along the length of the part between the robot's travel position and the order in which each independent support rod participates, enabling the flexible support unit 3 to adjust its effective support height and position in real time according to the robot's current position.

[0122] Specific process steps: S13. Determine the robot's roll forming trajectory: S13-1. Write control programs using the KUKA robot programming language; S13-2. Based on the width variation parameters, web depth variation parameters, sheet thickness, and bending angle requirements of each pass for the variable cross-section section of the automotive B-pillar, calculate and mark the forming trajectory of the tool head, and adjust the relevant angles of the tooling. S13-3. Run the machine under no-load conditions without loading any sheet metal to verify the accuracy of the trajectory and the rationality of the timing of each support rod.

[0123] S23. Clamp the sheet metal and maintain the clamping pressure at 600N using fuzzy PID closed-loop control.

[0124] S33. Perform solid roll forming according to the verification trajectory. During the roll forming process, when robot module 1 is located in the narrow end region with a cross-sectional width of 80mm, the corresponding independent support rod participates in the support with a support height matching the web depth of that region; when robot module 1 enters the transition region where the cross-sectional width and web depth change synchronously, adjacent independent support rods gradually participate in the support in a preset order, so that the support position transitions continuously along the cross-sectional change direction; when robot module 1 moves to the wide end region with a cross-sectional width of 150mm, the corresponding independent support rod continuously supports the inner wall of the sheet metal with a support height matching the web depth of that region. In this embodiment, the maximum forming force is controlled at 1000N.

[0125] Post-forming inspection revealed that the resulting variable cross-section automotive B-pillar component was stably clamped, exhibiting no significant web warping or edge wrinkling, and demonstrating good contour continuity within the area where width and web depth changed synchronously. This indicates that by matching and adjusting the support height and position according to the robot's travel position using the flexible support unit 3, web warping during the forming process of complex variable cross-section parts can be effectively suppressed, while ensuring the continuity of the roll forming contour.

[0126] Comparative Example 1 The difference between this comparative example and Example 1 lies in the control algorithm of the pneumatic servo circuit. This comparative example uses a traditional PID control algorithm with fixed parameters. The proportional coefficient, integral coefficient, and derivative coefficient remain constant after initial setting and do not have online adaptive adjustment capabilities. The specific process steps include: writing a program using the KUKA robot programming language; marking the forming trajectory based on the sheet thickness and bending angle and verifying it under no-load conditions; clamping the sheet on the pneumatic clamping end of a flexible tooling using traditional PID control; and forming according to the verified trajectory.

[0127] like Figure 6 As shown, the test data indicates that pneumatic systems using traditional PID control struggle to balance dynamic response speed and stability across the entire range due to the compressibility and nonlinear characteristics of gases. Compared to the fuzzy PID adaptive control in Example 1, the step response rise time of this comparative system is significantly longer, while the response time of Example 1 is shortened by approximately 37%. Furthermore, this comparative system exhibits overshoot oscillations when the pressure approaches the target value, while Example 1 achieves overshoot-free control across the entire range from 0 to 0.7 MPa.

[0128] Comparative Example 2 The difference between this comparative example and Example 1 lies in the tooling structure. This comparative example uses a traditional rigid tooling (fixed-width mold), whose support surface width is constant and cannot be adjusted according to changes in the cross-section of the sheet metal. Specific process steps: A program is written using the KUKA robot programming language and the trajectory is verified under no-load conditions; the sheet metal is fixed on the traditional rigid mold; forming is performed according to the verified trajectory.

[0129] In the wide section area of ​​the sheet metal, the rigid tooling can provide some support, but in the area where the section transitions from narrow to wide or in the area of ​​section twisting, the edges and web of the sheet metal are suspended and unsupported. Due to the lack of effective bottom support, the rolling pressure causes the web flatness to deviate significantly, resulting in obvious web warping. During the flange edge forming process, the lack of lateral constraints and follow-up supports causes the internal compressive stress of the sheet metal to cause wrinkling at the edges. Compared with Example 1, the flatness error of the part produced in this comparative example is ±0.5mm, and the wrinkling wave height of the flange edge is 0.2mm; while the flatness error of the part using the flexible tooling of this invention is reduced to ±0.05mm, and the wrinkling phenomenon is completely eliminated or the wave height is reduced to 0.1mm.

[0130] Comparative Example 3 The difference between this comparative example and Example 1 lies in the forming process. This comparative example uses a traditional roll forming process, the specific steps of which include: feeding the pre-cut sheet material into the roll forming machine and positioning it; applying continuous pressure to the sheet material through a series of rollers to gradually deform it into the desired cross-sectional shape; and performing leveling, trimming and deburring after forming.

[0131] For variable cross-section features, traditional roll forming processes require the design and manufacture of multiple sets of dedicated roll dies, along with special variable cross-section mechanisms. Taking a variable cross-section automotive B-pillar component as an example, traditional roll forming requires the development of five sets of rolls, with a mold opening cost of approximately 350,000 yuan and a cycle of at least six months. When product design changes, the original dies are essentially scrapped or require complex readjustment. In contrast, robot-assisted flexible roll forming eliminates the need for dedicated dies. It can adapt to the production of different specifications of products simply by modifying the robot's trajectory program and adjusting the tooling pneumatic parameters. The development cycle is shortened to a few weeks, significantly reducing manufacturing costs and changeover time, and exhibiting extremely high reconfigurability.

[0132] The above provides a detailed description of the intelligent adaptive variable cross-section plate robot roll forming equipment and roll forming method. Specific examples have been used to illustrate the principle and implementation of the invention. The description of the above embodiments is only for the purpose of helping to understand the method and core idea of ​​the invention. At the same time, for those skilled in the art, there will be changes in the specific implementation and application scope based on the idea of ​​the invention. Therefore, the content of this specification should not be construed as a limitation of the invention.

Claims

1. An intelligent adaptive variable cross-section sheet metal robotic roll forming equipment, characterized in that, The equipment includes: Robot module, wherein a rolling tool head is mounted at the end of the robot module; A flexible tooling module, comprising a pneumatic clamping unit and a flexible support unit; The sensing module includes a pressure sensor and a position sensor; The control module operates a fuzzy PID adaptive control algorithm to adjust the clamping pressure of the pneumatic clamping unit to a constant value and to control the flexible support unit to maintain synchronous follow-up support with the end effector of the robot module during the roll forming process. The flexible support unit provides support for the metal sheet during the roll forming process. The flexible support unit includes at least one set of support sliders that can translate along a direction perpendicular to the feed direction of the metal sheet, and a servo drive mechanism that drives the support sliders. The control module controls the servo drive mechanism to adjust the position of the support sliders by outputting a drive control signal, so as to achieve synchronous follow-up support between the flexible support unit and the end effector of the robot module. The flexible support unit and the end of the robot module maintain synchronous follow-up support, which includes: during the roll forming process, the control module calculates the target support position of the flexible support unit according to the real-time position coordinates of the end of the robot module, and drives the support slider to move to the target support position through the drive control signal, so as to maintain continuous contact support to the inner wall of the metal sheet.

2. The equipment according to claim 1, characterized in that, The flexible tooling module also includes a base, on which a linear guide rail is provided. The pneumatic clamping unit and the flexible support unit are mounted on the linear guide rail and their relative positions can be adjusted along the linear guide rail. The pneumatic clamping unit is used to fix the metal sheet.

3. The equipment according to claim 2, characterized in that, The metal sheet is any one of martensitic steel, DP800 duplex steel or QP1180 steel. The forming force applied by the rolling tool head during the rolling process ranges from 100N to 4000N, and the clamping pressure applied by the pneumatic clamping unit is constant at 600N.

4. The equipment according to claim 3, characterized in that, The sensing module communicates bidirectionally with the pneumatic clamping unit and the flexible support unit; the sensing module also includes a vision sensor installed at the end of the robot module, the vision sensor is used to collect the springback data of the metal sheet after roll forming, and the control module is also configured to correct the roll forming trajectory of the robot module based on the springback data.

5. The equipment according to claim 4, characterized in that, When the control module runs the fuzzy PID adaptive control algorithm, it takes the pressure deviation between the clamping pressure data collected by the pressure sensor and the target clamping pressure value, and the rate of change of the pressure deviation, as input. Based on the preset fuzzy control rule table, it calculates the PID parameter correction online and outputs the adjustment control signal to the electro-proportional valve of the pneumatic clamping unit to stabilize the clamping pressure.

6. The equipment according to claim 5, characterized in that, The control module is also configured to: during the roll forming process, when the clamping pressure data collected by the pressure sensor fluctuates beyond a preset threshold, trigger the flexible support unit to perform support position compensation movement.

7. The equipment according to claim 6, characterized in that, The flexible support unit determines its initial posture parameters based on the pre-formed bending creases and bending angle data, and adjusts its support posture in real time based on the position data collected by the position sensor during the roll forming process.

8. A roll forming method using the intelligent adaptive variable cross-section plate robot roll forming equipment as described in claim 1, characterized in that, The method includes: S1. Based on the thickness parameters and bending angle requirements of the metal sheet to be processed, generate the rolling trajectory of the robot module and the initial posture of the flexible support unit. S2. Clamp the metal sheet in the pneumatic clamping unit, and adjust the clamping pressure to a preset constant value through the control module; S3. Control the robot module to roll the metal sheet according to the rolling trajectory. During the rolling process, the control module dynamically adjusts the clamping pressure of the pneumatic clamping unit according to the pressure and position data collected by the sensor module, and drives the flexible support unit to keep synchronous follow-up support with the end of the robot module.

Images