A steel structure metal sheet cutting and stacking device and method
By using real-time detection and a flipping mechanism to adjust the orientation of parts, the problem of out-of-plane warping of steel structure parts caused by cutting heat input was solved, and the stable stacking and hoisting of heavy steel structure parts were achieved.
Patent Information
- Application Number
- CN202611473361.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-21
- Publication Date
- 2026-11-17
AI Technical Summary
Steel structure parts cut by CNC flame, plasma, or laser cutting will warp out of plane due to uneven distribution of heat input along the thickness of the plate. This will cause the parts to be unable to be pushed and reworked after being placed in place, and the accumulation of interlayer false height will affect the stability of binding and hoisting.
By detecting the out-of-plane shape of the parts in real time before blanking, the orientation of the parts is adjusted by the flipping mechanism to make the effective orientation of adjacent parts the same, thereby reducing the accumulation of interlayer false height. The out-of-plane contour detection unit, shape calculation unit, flipping mechanism and orientation decision unit work together to ensure that the difference in deflection amplitude between adjacent parts is minimized.
It effectively suppresses the accumulation of false height between layers, improves the stability of stacking and the reliability of hoisting, avoids rework, and is suitable for the stacking process of heavy steel structure parts.
Smart Images

Figure CN123105306A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of steel structure component processing equipment and automatic palletizing control technology, specifically to a palletizing device and method for parts after hot cutting of steel structure metal plates. Background Technology
[0002] When CNC flame, plasma, or laser cutting is used, the heat input is unevenly distributed along the thickness of the plate, resulting in out-of-plane warping of the cut steel structural parts. This manifests as angular deformation, plate bending, and end curling, with individual piece deflection ranging from 2mm to 8mm. The parts must be stacked layer by layer for bundling and hoisting, with each piece weighing tens to hundreds of kilograms. Once in place, they cannot be pushed or reworked, and the quality control of interlayer bonding is irreversible.
[0003] In existing technologies, mechanical guidance and alignment mechanisms constrain the horizontal orientation of parts when they enter the palletizing station, addressing in-plane deviations caused by loading offsets and conveyor misalignments, but do not involve the out-of-plane shape of the parts themselves; the controlled variable of vision or laser positioning combined with horizontal coordinate control is still the position of the parts in the horizontal plane, which does not affect the interlayer height caused by undulations in the thickness direction; the whole-stack pressing and weighting only passively compress the gaps with external force without reducing the gaps themselves, and the pressing force is limited when the stack reaches tens of tons, and the rebound after unloading still causes the binding to loosen and the transfer to be misaligned.
[0004] Furthermore, the stacking of lightweight board materials such as cardboard, corrugated board, and wood involves alternating flipping of adjacent layers. This method is suitable for lightweight components that can be pushed or compressed, requiring adjacent layers to counteract the bending of the entire stack. The interlayer gaps of heavy, rigid components are determined by geometric fit; two components with the same curvature only fit together when their orientations are aligned, while those with opposite orientations leave the middle section unsupported. This teaching is exactly the opposite of the approach in this case and lacks item-by-item testing and layer-by-layer pairing decisions.
[0005] In summary, existing methods are limited by in-plane control, passive compaction, and the assumption of lightweight components. The technical problem to be solved by this invention is to provide a stacking device and method that, based on the out-of-plane morphology measured piece by piece, selects the relative orientation of two adjacent pieces before unloading to suppress the accumulation of interlayer false height, under the condition that the cutting of heavy steel structure parts cannot be pushed and reworked after placement. Summary of the Invention
[0006] The technical problem this invention aims to solve is that steel structural parts cut by CNC flame, plasma, or laser cutting commonly exhibit out-of-plane warping due to uneven heat input along the thickness direction. The upward orientation of the parts as they fall from the cutting table grid onto the conveyor is random, and the calibrated orientation... In a batch of parts, they are randomly distributed (approximately half with convex surfaces facing upwards and half with concave surfaces facing upwards). If no orientation control is applied during stacking, the effective orientations of adjacent parts will be nearly 50% opposite, introducing approximately twice the deflection amplitude per layer, accumulating linearly with the number of layers. This leads to overall stack height discrepancies, loosening of bundles due to springback, and misalignment during lifting. Furthermore, the large mass of individual parts and the inability to be pushed or reworked after placement make the feedback loop of placing, inspecting, and then correcting the position technically impossible under these conditions. Therefore, this invention uses the gap between adjacent parts as the quantity to be processed and the placement orientation of the next part as the control variable. Before the parts are placed, the orientation that minimizes the gap is calculated based on the measured out-of-plane morphology of each part. By selectively flipping the parts, the effective orientations of all parts are unified to be the same, thereby suppressing the accumulation of interlayer height discrepancies without making any positional corrections to the already placed layers.
[0007] The technical solution of this invention: A steel structure metal sheet cutting and palletizing device of this invention includes a palletizing frame, a gripping head, a lifting drive mechanism, a transfer mechanism, a palletizing platform, and a transition conveyor platform. It also includes an out-of-plane contour detection unit, a morphology calculation unit, a flipping mechanism, a layer gap register, an orientation decision unit, and a control cabinet. Preferably, the palletizing platform is placed below the palletizing frame, receiving parts layer by layer to form a stack; the transition conveyor platform receives parts output from the previous hot cutting process; the gripping head, driven by the lifting drive mechanism and the transfer mechanism, transfers the parts from the transition conveyor platform to the top of the palletizing platform for release. The out-of-plane contour detection unit is mounted above the transition conveyor platform. When a part is gripped but not yet dropped, it measures the height sequence of the upper surface of the part along several scanning lines. The morphology calculation unit calculates the out-of-plane morphology of the part based on the height sequence, obtaining its calibrated morphology and deflection amplitude. With calibration orientation Among them, the orientation is determined A value of +1 or -1 indicates whether the part is convex or concave when placed in its natural delivery position, respectively. The stack gap register stores the cumulative stack gap of the already stacked portion of the stack. And the effective orientation of the previously placed part. Its deflection amplitude Effective orientation Placement orientation factor With calibration orientation The product. The orientation decision unit reads the effective orientation of the previous part from the layer gap register. With flexural amplitude The placement orientation factor of the current part is calculated using the following formula. This reduces the gap between the current part and the previous part. Take the minimum value:
[0008] The placement orientation factor calculated by the flipping mechanism based on the orientation decision unit The parts are rotated 180 degrees or kept in their natural delivery posture, and then released by the gripper head to the top of the stacking table, so that the effective orientation of adjacent parts is the same and the layer gap is minimized. This is the difference in deflection amplitude between two adjacent parts. The control cabinet is electrically connected to the out-of-plane contour detection unit, shape calculation unit, flipping mechanism, layer gap register, orientation decision unit, lifting drive mechanism, and transfer mechanism, and records the layer gap of the current layer after each part is released. Accumulated to cumulative gap .
[0009] Preferably, the out-of-plane contour detection unit is a line laser contour sensor group arranged laterally along the transition conveyor table. The projection plane of the line laser contour sensor group is perpendicular to the upper surface of the part, and each line laser contour sensor forms several parallel scanning lines along the feed direction of the part.
[0010] Preferably, the morphological calculation unit performs least-squares quadratic surface fitting on the height sequence, and uses the sign of the quadratic term obtained from the fitting to determine the orientation. The maximum height of each measuring point relative to the fitted surface is used as the deflection amplitude. .
[0011] Preferably, the morphology calculation unit further determines whether the calibrated morphology is a surface bend or an end curl based on the height sequence, and when the calibrated morphology is an end curl, the difference in the height of the measuring points on both ends from the surface is used as the deflection amplitude. .
[0012] Preferably, the transition conveyor is also equipped with a buffer station, where several parts to be encoded are temporarily stored; the orientation decision unit selects the deflection amplitude value from the several parts to be encoded. Deflection amplitude of the previous part The part with the smallest difference is selected as the current part.
[0013] Preferably, the orientation decision unit calculates the cumulative gap using the following formula. and in the accumulated gaps When the value exceeds the set threshold, the placement orientation factor of the current part is used instead. The opposite value is simultaneously indicated by a warning signal from the control cabinet:
[0014] Preferably, the orientation decision unit also reads the nesting data from the previous thermal cutting process, predicts the order in which each part arrives at the transition conveyor based on the nesting data, and pre-programs the placement orientation factors for each part accordingly. sequence.
[0015] Preferably, the flipping mechanism includes a rectangular plate fixed to the bottom telescopic end of the lifting drive mechanism, two hanging plates extending downward from both ends of the rectangular plate, a rotary frame rotatably mounted between the two hanging plates via a connecting shaft, a rotary motor mounted on the side wall of the hanging plates with its output shaft coaxially connected to the connecting shaft on one side of the rotary frame, an electric push rod mounted on the side wall of the rotary frame, a U-shaped frame fixed to the telescopic end of the electric push rod, a rubber block sealed within the rectangular holes on the upper and lower inner walls of the U-shaped frame, and an injection pump mounted on the side wall of the U-shaped frame and communicating with the inner cavity of the U-shaped frame; after the injection pump injects hydraulic oil into the inner cavity of the U-shaped frame, the rubber block expands and contracts along the rectangular holes under the action of oil pressure and adaptively conforms to the upper and lower surfaces of the part, clamping the part in the U-shaped frame; the rotary motor drives the rotary frame together with the clamped part to rotate 180 degrees around a horizontal axis parallel to the feed direction of the part. The gripping head includes a hydraulic cylinder and an electric suction cup. The cylinder body of the hydraulic cylinder is mounted on the top of the rectangular plate, and the electric suction cup is mounted on the bottom telescopic end of the hydraulic cylinder. Therefore, the gripping head moves up and down with the lifting drive mechanism but does not rotate with the slewing frame.
[0016] Preferably, when the out-of-plane contour detection unit fails, the orientation decision unit will place an orientation factor. If the value is fixed at +1, meaning that none of the parts are flipped over, the control cabinet will issue an alarm signal and record the number of layers that have been placed.
[0017] The parts are hot-cut steel plates with a thickness of 4mm to 40mm, with a deflection amplitude of... The thickness is 2mm to 8mm, and the number of stack layers is set from 10 to 40.
[0018] A method for cutting and stacking sheet metal for steel structures, using the aforementioned sheet metal cutting and stacking device, includes the following steps: S1, accumulating the layer gap in the layer gap register... S1. Zero the reference transfer position and set it above the palletizing table; S2. The part output from the previous thermal cutting process arrives at the transition conveyor table, and the out-of-plane contour detection unit measures the height sequence of the upper surface of the part along several scanning lines; S3. The shape calculation unit calculates the calibration shape and deflection amplitude of the part based on the height sequence. With calibration orientation S4. The orientation decision unit reads the effective orientation of the previously placed part from the gap register. With flexural amplitude Press to make the interlayer gap The minimum value is used to calculate the placement orientation factor of the current part. S5. The gripper head grips the current part, and the flipping mechanism follows the placement orientation factor. S6. The part is rotated 180 degrees or kept in its natural delivery posture; S7. The lifting drive mechanism and the transfer mechanism drive the gripper head to move the part to the top of the palletizing table and release it, and the part falls onto the palletizing table or the upper surface of the formed stack; S8. The orientation decision unit calculates the layer gap of the current layer based on the effective orientation and deflection amplitude of the current part and the previous part. And add it to the accumulated gap in the gap register. S8. Repeat steps S2 to S7 until the stack reaches the set number of layers, wherein adjacent parts are stacked in the same direction (both with convex surfaces facing up or both with concave surfaces facing up), and the gap between each layer is... The cumulative lamination is the difference in deflection amplitude between two adjacent parts. It does not accumulate linearly with the set number of layers.
[0019] The beneficial effects of this invention are: 1. This invention uses the gap between two adjacent pieces, rather than the coordinate deviation of a single piece, as the quantity to be processed. It rewrites the interlayer false height, which was originally determined by the outward warping of a single piece, as the difference in the deflection amplitude of two adjacent pieces. When the effective orientations of two adjacent pieces are the same, the gap is the difference in their deflection amplitudes. When the effective orientations are opposite, the gap is the sum of their deflection amplitudes. This invention unifies the effective orientations of two adjacent pieces before blanking, so that the controllable quantity of a single layer changes from the sum of the deflection amplitudes of the two pieces to the difference in their deflection amplitudes. Therefore, the accumulated gap no longer increases linearly with the number of layers.
[0020] 2. This invention uses the placement orientation of the next part, rather than the position of the already placed layers, as the control variable, so that all control actions are applied to the part that has not yet been placed. The orientation change is completed by the flipping mechanism before release. Heavy rigid parts cannot be pushed or reworked after being placed. The feedback route with position correction as the core is not feasible under this working condition. Therefore, the pre-decision of this invention can take effect under irreversible conditions. It does not pursue improving the repeatability of the placement of a single part, nor does it make any position correction to the already placed layers. The quantities and control variables processed are different from the existing in-plane control routes. 3. This invention achieves co-directional stacking by unifying the effective orientation of two adjacent parts, which is opposite to the teaching direction of alternating flipping of lightweight sheet materials. It also utilizes the low dispersion of the deflection amplitude of the same batch to compress the residual lamination to the order of magnitude of the amplitude difference. Parts from the same batch and with the same cutting process have similar deflection amplitudes but a dispersion much smaller than the average value. After alignment, the residual lamination is the difference between the deflection amplitudes of two adjacent parts, and this difference is much smaller than the average deflection amplitude. The stronger the batch-specific system warping, the smaller the proportion of the difference relative to the average value, and the more prominent the suppression effect. This is contrary to the conventional approach of treating all system deviations as unfavorable factors and eliminating them. Attached Figure Description
[0021] Figure 1 This is a schematic diagram of the overall layout of the present invention; Figure 2This is a schematic diagram of the out-of-plane contour detection station structure of the present invention; Figure 3 This is a schematic diagram illustrating the interlayer gap cancellation principle of adjacent layers with consistent orientation in this invention; Figure 4 This is a comparison curve of the cumulative gap as a function of the number of layers in this invention; Figure 5 This is a control logic block diagram of the present invention; Figure 6 This is a flowchart of the palletizing method of the present invention; Figure 7 This is a schematic diagram of the flipping mechanism of the present invention.
[0022] In the diagram: 1. Palletizing frame; 2. Gripping head; 3. Lifting drive mechanism; 4. Transfer mechanism; 5. Palletizing table; 6. Out-of-plane contour detection unit; 7. Shape calculation unit; 8. Flipping mechanism; 9. Layer gap register; 10. Orientation decision unit; 11. Control cabinet; 12. Part; 13. Stack body; 14. Transition conveyor table; 81. Rectangular plate; 82. Hanging plate; 83. Rotary frame; 84. Rotary motor; 85. Electric push rod; 86. U-shaped frame; 87. Rubber block; 88. Injection pump; 21. Hydraulic cylinder; 22. Electric suction cup. Detailed Implementation
[0023] The following combination Figures 1 to 7 The specific embodiments of the present invention will be further described in detail below. It should be noted that the composition, connection relationship and movement mode of each mechanism described below are given item by item on the premise of sufficient disclosure, and the reference numerals appearing therein are consistent with the reference numerals in the case concept.
[0024] A steel structure metal sheet cutting and palletizing device includes a palletizing frame 1, a gripping head 2, a lifting drive mechanism 3, a transfer mechanism 4, a palletizing table 5, a transition conveyor table 14, an out-of-plane contour detection unit 6, a shape calculation unit 7, a flipping mechanism 8, a layer gap register 9, an orientation decision unit 10, and a control cabinet 11. Figure 1As shown. The installation positions of each component from top to bottom and from the center outwards are as follows: the transfer mechanism 4 is a gantry-type translation mechanism or a suspended translation trolley that spans the transition conveyor platform 14 and the palletizing platform 5. Its translation guide rail is mounted on the top of the palletizing frame 1 and is fixedly connected to the palletizing frame 1. The slide block reciprocates horizontally along the guide rail; the lifting drive mechanism 3 is assembled on the slide block of the transfer mechanism 4 with a linear guide rail pair and a ball screw pair or a gear rack pair. The body is rigidly fixed to the slide block; the gripping head 2 is a vacuum adsorption gripping head. Its specific structure is shown in the corresponding section below. It is suspended at the bottom; the flipping mechanism 8 passes through its... The rectangular plate 81 receives the lifting output end of the lifting drive mechanism 3. The upper surface of the rectangular plate 81 is bolted to the flange of the lifting output end. Two hanging plates 82 extend downward from both ends of the rectangular plate 81. A rotating frame 83 is rotatably installed between the two hanging plates 82 via a connecting shaft. The rotating motor 84 is mounted on the side wall of the hanging plate 82. The cylinder body of the hydraulic cylinder 21 of the gripping head 2 is mounted on the top of the rectangular plate 81. Its telescopic end extends downward through the rectangular plate 81 and is connected to the electric suction cup 22. Therefore, the gripping head 2 moves up and down with the lifting drive mechanism 3, but does not rotate with the rotating frame 83 of the flipping mechanism 8. On the ground side: the palletizing frame 1 is a portal frame steel structure, with the lower ends of the two columns fixed to the workshop floor via a base plate and pre-embedded bolts, and the top beam spans directly above the palletizing platform 5; the palletizing platform 5 is a load-bearing platform, fixed to the center below the door frame of the palletizing frame 1 via anchor bolts, and its upper surface forms the stacking reference surface of the stack body 13; the transition conveyor 14 is a roller conveyor or chain conveyor line, arranged along the feeding direction of the parts 12 on the feeding side of the palletizing frame 1, with its end near the palletizing frame 1 extending to the material picking position; the out-of-plane contour detection unit 6 is fixed on the detection bracket beam spanning the transition conveyor 14, with the probe facing downwards; the morphology calculation unit 7, the layer gap register 9, and the orientation decision unit 10 are all installed in the same control cabinet 11; the control cabinet 11 is arranged on the ground next to the palletizing frame 1.
[0025] The palletizing table 5 is fixed to the workshop floor by anchor bolts, and its center is located in the lower projection area of the inner frame of the palletizing machine 1. The upstream end of the transition conveyor table 14, which is the end away from the palletizing machine 1, is connected to the discharge port of the previous cutting machine, and its downstream end, which is the end close to the palletizing machine 1, extends to the material picking position and is adjacent to the feeding side of the palletizing machine 1. The two are at the same height or connected by a downward slope, so that the parts 12 can smoothly enter the transition conveyor table 14 after being output from the cutting machine. The body of the lifting drive mechanism 3 and the slide of the transfer mechanism 4 are rigidly fixed by bolt pairs. The output end is fixed to the middle of the upper surface of the rectangular plate 81 of the flipping mechanism 8 via a flange. The electric suction cup 22 of the gripping head 2 is hung below the rectangular plate 81 via a hydraulic cylinder 21, thus forming a pick-and-place chain of "slide seat - lifting drive mechanism 3 - rectangular plate 81 - hydraulic cylinder 21 - electric suction cup 22 - part 12". The flipping mechanism 8 moves up and down with the lifting drive mechanism 3. Its rotating frame 83 is located in the middle of the transfer path of the gripping head 2, forming a passage station between the gripping head 2's pick-up position and release position, and completes the flipping after the part 12 is picked up and before it is released. The gripping head 2 thus has two movements: it moves horizontally along the translation guide rail with the transfer mechanism 4 and moves vertically up and down with the lifting drive mechanism 3. It can also be lifted and lowered twice relative to the rectangular plate 81 by the hydraulic cylinder 21, but it does not rotate with the rotating frame 83. The detection bracket of the out-of-plane contour detection unit 6 is fixedly connected to the frame column of the transition conveyor table 14, or is independently placed on the ground to isolate the vibration of the conveyor line. Its probe faces downwards and is directly opposite the upper surface of the part 12. The rotary motor 84 is mounted on the side wall of the hanging plate 82. Its output shaft is coaxially connected to the connecting shaft on one side of the rotary frame 83. The other end of the connecting shaft is supported on the hanging plate 82 on the other side via a bearing seat, so that the rotary frame 83 can rotate around the axis of the connecting shaft. This axis is arranged horizontally, parallel to the feed direction of the part 12, and located on the symmetrical mid-plane of the two hanging plates 82. The control cabinet 11 is electrically connected to the out-of-plane contour detection unit 6, the shape calculation unit 7, the flipping mechanism 8, the layer gap register 9, the orientation decision unit 10, the lifting drive mechanism 3, and the transfer mechanism 4 via cables. The control cabinet 11 integrates the calculation functions of the shape calculation unit 7, the layer gap register 9, and the orientation decision unit 10. The morphology calculation unit 7, the gap register 9, and the orientation decision unit 10 can be physically processor modules or independent controllers within the control cabinet 11, and the data interaction between the three is completed through the internal bus of the control cabinet 11.
[0026] During operation, the preceding cutting machine feeds the cut part 12 along the feed direction to the transition conveyor table 14; the transfer mechanism 4 drives the lifting drive mechanism 3 and the gripper head 2 to move horizontally above the transition conveyor table 14, the lifting drive mechanism 3 drives the gripper head 2 to descend, and the electric suction cup 22 picks up the part 12 and rises back; if the orientation decision unit 10 determines that the current part 12 needs to be flipped, the transfer mechanism 4 first sends the part 12 to the clamping position of the flipping mechanism 8, the hydraulic cylinder 21 lowers the electric suction cup 22 along with the part 12 into the frame-shaped inner cavity of the rotary frame 83, and the electric push rod 85 pushes the U-shaped frame 86 to the edge of the part 12; the injection pump 88 injects hydraulic oil into the inner cavity of the U-shaped frame 86, and the rubber... Block 87 extends along the rectangular hole under oil pressure and adaptively fits the upper and lower surfaces of part 12, clamping part 12 within the U-shaped frame 86. Subsequently, the electric suction cup 22 releases its adsorption and is lifted out of the rotation envelope by the hydraulic cylinder 21. The rotary motor 84 drives the rotary frame 83 to rotate 180 degrees along with part 12. After rotation, the hydraulic cylinder 21 lowers the electric suction cup 22 to re-adsorb the flipped part 12. The injection pump 88 draws oil to retract the rubber block 87, and the electric push rod 85 retracts to make room for the U-shaped frame 86. The hydraulic cylinder 21 delivers part 12 to the underside of the rotary frame 83, and finally, it is transferred by the gripping head 2. If flipping is not required, part 12 is transferred directly by the gripping head 2 without rotation. Afterward, the gripping head 2 moves the oriented part 12 horizontally above the stacking table 5 and releases it. Parts 12 are stacked layer by layer to form a stack 13. The entire transfer stroke is determined by the translation stroke of the transfer mechanism 4, typically the sum of the projected side length of part 12 and the distance between the two units; the lifting stroke is determined by the grab height, release height and the stack height of the stack 13, increasing with each layer already placed.
[0027] The geometric center of the palletizing table 5 is taken as the installation reference origin O. The feed direction of part 12 is taken as the longitudinal direction. The direction from the transition conveyor 14 to the palletizing frame 1 is taken as the positive direction. The horizontal direction perpendicular to the feed direction is taken as the transverse direction. The vertical direction passing through point O is taken as the height direction. The height at the origin O is the installation reference zero position. Under this reference, the positional relationship and mating dimensions of each structure are as follows.
[0028] The height of the palletizing platform 5 is set to zero. The conveying surface of the transition conveyor 14 is higher than the platform of the palletizing platform 5, and the height difference between the two is 800mm to 1500mm. This height difference is not less than the sum of the maximum thickness of a single part (40mm), the maximum deflection (8mm), and the obstacle crossing safety distance. The obstacle crossing safety distance is 50mm to 150mm to ensure that when the gripper head 2 picks up the part 12 and lifts it to the transfer height, the lowest point of the part 12 does not interfere with the already stacked body 13 on the palletizing platform 5. The height of the lower edge of the top crossbeam of the palletizing frame 1 is not less than the sum of the full stack height, the maximum thickness of a single part, and the obstacle crossing safety distance. The full stack height is determined by multiplying the set number of layers (10 to 40 layers) by the sum of the thickness of a single part and the gap between single layers. The clearance between the lower edge of the crossbeam and the translation guide above it must also accommodate the external height of the lifting drive mechanism 3 when it is at the upper limit of its stroke. Since the rectangular plate 81, hanging plate 82, rotating frame 83 and hydraulic cylinder 21 of the flipping mechanism 8 are all raised and lowered together with the lifting drive mechanism 3, the height of the lower edge of the top beam must also include the total height occupied by the above components in the vertical direction when the lifting drive mechanism 3 is at the upper limit of its stroke.
[0029] The transition conveyor 14 extends to the picking position at one end near the palletizer frame 1. The distance between the longitudinal centerline of the picking position and the centerline of the palletizer 5, plus the maximum projected side length of part 12 along the feed direction, is the translational stroke required by the transfer mechanism 4. When the slide moves from the picking position to the release position, the center of the gripper head 2 moves exactly from the geometric center of part 12 to directly above the center of the palletizer 5. The end of the transition conveyor 14 away from the palletizer frame 1 is longitudinally connected to the discharge port of the preceding cutting machine. The centerlines of the two are collinear, and the conveying surfaces are at the same height or connected by a slope section of no more than 5%. The palletizer 5 is located in the center between the two columns of the palletizer frame 1, and its longitudinal centerline is in the same vertical plane as the centerline of the translational trajectory of the slide of the transfer mechanism 4.
[0030] Lateral positional relationship: The lateral centerline of the conveying surface of the transition conveyor 14 is parallel to the lateral centerline of the palletizing platform 5, and their projections in the horizontal plane coincide. Therefore, the lateral orientation of part 12 is determined by the coincidence of the two centerlines throughout the entire process from the material pick-up position to the release position, without the need for additional lateral adjustment. The translation trajectory of the gripper head 2 in the horizontal plane coincides with this common centerline, and the translation deviation is limited to within ±3mm by the straightness of the guide rail and the lateral guide wheel. The two columns of the palletizing frame 1 are symmetrically arranged on both sides of this common centerline, and their net distance is not less than the sum of the maximum lateral projection side length of part 12 and the width of the working channel on both sides. The width of the working channel on one side is 500mm to 1200mm. The two columns of the detection bracket of the out-of-plane contour detection unit 6 are also straddling both sides of this common centerline, and its crossbeam is located 600mm to 1200mm above the conveying surface of the transition conveyor 14.
[0031] The flatness of the mounting surface of the linear guide pair of the lifting drive mechanism 3 is no greater than 0.1 mm / m, and the parallelism between the guide rail axis and the normal of the palletizing table 5 is no greater than 0.2 mm / m, to ensure that the horizontal posture deviation of the gripper head 2 at the release point of part 12 is no greater than 1 mm / m; the perpendicularity of the translation guide rail of the transfer mechanism 4 to the above-mentioned normal is no greater than 0.2 mm / m; the parallelism between the rotation axis of the rotating frame 83 connecting shaft and the feed direction of part 12 is no greater than 0.3 mm / m; the flatness of the palletizing table 5 is no greater than 1 mm / m, to ensure the fitting reference of the first layer of the stack 13. The control cabinet 11 is placed on the side of the palletizing frame 1, and the clear distance on the cabinet door side is no less than 800 mm; the power cables and signal cables between the cabinet and each actuator are laid along the cable trays on the column of the palletizing frame 1, the signal lines and power lines are run in separate trays, and the sensor signal lines are shielded twisted pairs and grounded at one end.
[0032] The assembly sequence of each structure from bottom to top is as follows: First, lay out the lines on the workshop floor and fix the anchor bolts of the palletizing platform 5 using the grouting method. Position the palletizing platform 5 and level it to the aforementioned flatness requirements. Then, hoist the two columns of the palletizing frame 1 and initially tighten the base plate bolts. After correcting the verticality of the columns, tighten them completely. Next, hoist the top crossbeam so that the two columns and the crossbeam form a portal frame. Then, install the translation guide rail of the transfer mechanism 4 and adjust its straightness with shims. Install the slide and drive unit. Then, connect the body of the lifting drive mechanism 3 to the slide and use its guide rail pair to correct the parallelism between the lifting axis and the normal line of the palletizing platform 5. Finally, assemble the flipping mechanism 8, that is, first fix the middle of the upper surface of the rectangular plate 81 to the lifting output flange of the lifting drive mechanism 3, and then install the two lifting... Plate 82 extends downwards from both ends of rectangular plate 81 and is fixedly connected. Then, a rotating frame 83 and a connecting shaft are installed between the two hanging plates 82. The rotating motor 84 is installed on the outer wall of the hanging plate 82, and the parallelism between the axis of the connecting shaft and the feed direction of part 12 is corrected. Finally, the cylinder bodies of two hydraulic cylinders 21 are symmetrically installed on both sides of the fixed connection point at the middle of the top of the rectangular plate 81, so that their telescopic ends pass through the rectangular plate 81 and are connected to the hanging seat of the electric suction cup 22. The parallelism between the suction surface and the table surface of the stacking table 5 is checked. After that, the transition conveyor table 14 is installed, and its center line is corrected with the discharge port of the previous cutting machine as a reference, so that the elevation of its conveying surface meets the above-mentioned height difference. Finally, the detection bracket and the line laser profile sensor group are hoisted, and the coplanarity of each projection plane is corrected with calibration blocks. After all the mechanical parts are installed, the control cabinet 11 is positioned, the modules inside the cabinet are inserted and wired, and the single-machine jogging, no-load cycle and load calibration are completed.
[0033] The palletizing frame 1, palletizing platform 5, and transition conveyor 14 are three components that are directly connected to the ground and bear the weight of part 12 and stack 13. Their structural forms and mutual cooperation are as follows.
[0034] The palletizing frame 1 consists of two box-shaped welded columns and one box-shaped welded crossbeam connected by end-face connecting plates and high-strength bolt pairs to form a portal frame. Base plates are welded to the lower ends of the columns, and each base plate is connected to the ground with no less than four chemical anchors or pre-embedded bolts. A leveling bolt group is installed between the base plate and the ground, with a 30mm to 50mm grout layer reserved. After the column verticality is corrected to no more than 1mm / m, the anchors are tightened and non-shrink grout is injected. The upper surface of the lower flange of the crossbeam is the mounting surface of the translation guide rail of the transfer mechanism 4. This surface is machined to ensure a flatness of no more than 0.1mm / m. Stiffening ribs are welded to the back of the crossbeam at two concentrated areas of the sliding wheel pressure. A horizontal tie rod is also installed between the two columns near the top to improve the lateral stiffness of the portal frame against the horizontal inertial force when the transfer mechanism 4 starts and stops. Since the entire weight of the flipping mechanism 8 and the gripping head 2, as well as the reaction torque when the rotating frame 83 rotates, are transmitted to the palletizing frame 1 by the lifting drive mechanism 3 through the transfer mechanism 4, the arrangement of the stiffening ribs of the crossbeam and the position of the column tie rods must be checked according to the combined load of the self-weight of the flipping mechanism 8 and the rotational inertia force under the full-stack picking condition. The number of anchor bolts and the strength of the grouting layer of the column base plate are also determined according to the combined load.
[0035] The palletizing platform 5 is welded from a steel frame and a steel plate platform with a thickness of not less than 20mm. The platform and the frame below are fully welded and reinforced with plug welds. The upper surface of the platform serves as the stacking reference surface, and during installation, it is leveled to a flatness of no more than 1mm / m. The platform has recessed strapping grooves around its perimeter, with a depth of 20mm to 40mm and a width of 60mm to 100mm. After a full stack, the strapping straps can pass through these grooves for securing. A clearance of not less than 500mm is left between the outer columns of the platform to allow for the entry of lifting clamps and forklift forks. The palletizing platform 5 and the palletizing frame 1 are two independent ground-based components, not rigidly connected, to prevent vibrations from the transfer conveyor 14 and the transfer mechanism 4 from being transmitted to the weighing or locking elements via the palletizing frame 1. The palletizing platform 5 can also be equipped with a weighing sensor for verifying the actual number of layers.
[0036] The transition conveyor table 14 consists of several sets of rollers or chain plates and a support-type frame. The generatrix on the rollers forms the conveying surface, the elevation of which is determined according to the height difference described in item (d) and finely adjusted by the leveling shims at the bottom of the support legs. An adjustable guide rod is provided on each side of the conveying direction. The distance between the two rods is adjusted according to the maximum lateral projection side length of part 12 plus a margin of not less than 50mm. The top of the guide surface of the guide rod is more than 5mm lower than the upper surface of part 12 to ensure that there is no interference when the gripper head 2 picks up the material vertically. A buffer block and a proximity switch are provided at the end of the conveyor table near the previous cutting machine. After part 12 is in place, the proximity switch gives a signal to allow picking up the material.
[0037] The buffer station is an independently driven roller conveyor section on the transition conveyor 14 located upstream of the detection station, or a parallel buffer platform connected to this section. Driven by a separate geared motor, it can start and stop independently of the main conveyor section and keep part 12 stationary for selection. Several cylinder-driven, liftable stops are installed along the conveying direction on this section to separate and release parts 12 from the buffer. The length of the buffer station is determined by the number of parts it can accommodate, taking the sum of the projected lengths of 3 to 8 parts, which is used by the orientation decision unit 10 to select the deflection amplitude value from the parts to be selected. The one most similar to the previous one.
[0038] The gripping head 2 is a vacuum adsorption type gripping head, consisting of a hydraulic cylinder 21, an electric suction cup 22, a hanging base, a suction cup beam, several vacuum suction cups, partition plates, vacuum pipelines and solenoid valve groups, an elastic buffer mechanism, a vacuum pressure sensor, mechanical grippers, a fall arrestor, and a position detection element. Multiple hydraulic cylinders 21 are mounted symmetrically on both sides of the central fixing point at the top of the rectangular plate 81 of the flipping mechanism 8 via flange seats, with their telescopic ends extending downwards through the rectangular plate 81. The electric suction cup 22 is the adsorption execution part of the gripping head 2, consisting of a hanging base, a suction cup beam, several vacuum suction cups, partition plates, vacuum pipelines and solenoid valve groups, an elastic buffer mechanism, and a vacuum pressure sensor, and is mounted on the bottom telescopic end of the hydraulic cylinder 21 via the hanging base. The suction cup beam is a rectangular box-shaped welded beam, the length of which is determined by the maximum projected side length of part 12 in the transverse direction. Several vacuum suction cups are arranged in a matrix on the lower surface of the suction cup beam to form an adsorption array. The array is divided into several independent adsorption zones by partition plates along the longitudinal and transverse directions. Each adsorption zone is controlled by a vacuum pipeline and a solenoid valve group. Each vacuum pipeline is independently led to the manifold box at the top of the suction cup beam. An elastic buffer mechanism is located between the hanging base and the suction cup beam. Mechanical grippers are arranged in pairs at both ends of the suction cup beam. Anti-fall baffles are located around the lower surface of the suction cup beam and extend downward. The vacuum pressure sensor is installed in the manifold box, and the position detection element is installed on the lower surface of the suction cup beam.
[0039] The upper end of the hanging base is bolted to the bottom extension flange of the hydraulic cylinder 21. The cylinder body of the hydraulic cylinder 21 is mounted on the top of the rectangular plate 81 of the flipping mechanism 8. Therefore, the gripping head 2 is not mounted on the rotating frame 83 and does not rotate with the rotating frame 83. The hanging point of the hanging base coincides with the projection of the geometric center of the part 12 in the horizontal plane, so that the plate surface is subjected to symmetrical force when the gripping head 2 picks up and puts down the part 12. The elastic buffer mechanism consists of several compression springs and guide columns. The springs are fitted on the guide columns. The upper end of the guide column passes through the hanging base and the lower end is fixed to the upper surface of the suction cup beam, so that the suction cup beam can float elastically in the vertical direction relative to the hanging base. The floating stroke is 10mm to 30mm. Several vacuum suction cups are mounted on the lower surface of the suction cup beam via ball joints. The ball joints allow each suction cup to rotate flexibly around its own axis, ensuring that they can still adhere to each other even when there is a local angle between the suction cup opening plane and the upper surface of part 12. The lower end face of the suction cup opening forms the adsorption surface, and the position of the adsorption surface in the plane is arranged according to the matrix pitch of the adsorption array, with a pitch ranging from 200mm to 500mm. The partition plates are thin plates vertically welded to the lower surface of the suction cup beam, with their lower edges lower than the plane where the suction cup opening is located. Each partition plate forms several non-interconnected adsorption cavities, allowing vacuum adsorption to be established independently in each zone. The vacuum pipeline and solenoid valve assembly converge into a main pipe via a manifold box. The main pipe runs along the upper surface of the suction cup beam to one side of the hanging seat, then ascends along the cylinder side wall of the hydraulic cylinder 21 and the cable tray of the rectangular plate 81, passes through the slide and slide seat of the lifting drive mechanism 3, and finally exits from the cable tray of the transfer mechanism 4. Since neither the gripping head 2 nor the rectangular plate 81 rotates, the main pipe does not need to pass through the center hole of any rotating component, nor does it need to be equipped with a rotary joint. The signal line of the vacuum pressure sensor and the control line of the solenoid valve assembly are laid together with the main pipe and introduced into the control cabinet 11. Pairs of mechanical grippers are placed at both ends of the suction cup beam along the transverse direction and extend downwards. Their gripping surfaces are perpendicular to the transverse direction, and the gripping center height is flush with the adsorption surface. Under normal working conditions, the mechanical grippers are retracted at both ends of the suction cup beam and do not participate in load bearing. They only open and support the edge of the part 12 when vacuum adsorption fails. The anti-fall baffle extends downward around the lower surface of the suction cup beam, with its lower edge below the adsorption surface. The extension is 10mm to 20mm. In the event of an accidental loss of pressure during vacuum adsorption, it blocks the edge of part 12 from all sides, restricting its horizontal sliding.
[0040] During material handling, the gripping head 2 descends with the lifting drive mechanism 3 to contact the upper surface of the suction cup opening and the upper surface of the part 12. At this time, the elastic buffer mechanism is compressed, and the suction cup beam floats upward relative to the hanging seat, so that each suction cup adheres to the upper surface of the part 12 with a constant clamping force, avoiding additional deformation of the plate surface due to rigid pressure. Subsequently, the vacuum pump is started, and the solenoid valve group is opened one by one according to the zone. Each vacuum suction cup forms a negative pressure in the adsorption zone and sucks up the part 12. Because part 12 has out-of-plane warping, when their effective orientations are opposite, the contact surfaces of adjacent parts form reverse arc surfaces. If a single vacuum chamber is used for the entire surface, air leakage will occur between the suction cup opening on the concave side and the plate surface, making it impossible to establish a sufficient negative pressure to bear the load. This gripping head 2 uses partition plates to divide the adsorption surface into several non-interconnected adsorption chambers. The planar dimensions of the adsorption chambers range from 200mm to 600mm. As long as one suction cup is in contact and sealed in any adsorption area, a negative pressure can be established in that area. The remaining poorly contacted adsorption areas are individually cut off by the solenoid valve assembly. Therefore, the negative pressure is not affected by air leakage in the uncontacted areas, and its adaptability to flexed plate surfaces is better than that of a single vacuum chamber adsorption for the entire surface. In addition, the ball joint allows each suction cup to be aligned one by one as the plate surface tilts, further increasing the effective contact area. The vacuum pressure sensor monitors the vacuum level in the manifold box. When the vacuum level is lower than the set threshold or the vacuum level in a certain area is continuously lower than the average value of the other areas by a set difference, it is determined that the area is leaking and only that area is cut off, while the other areas maintain adsorption. If the vacuum level of the whole machine is lower than the set threshold, it is determined that the vacuum adsorption has failed. At this time, the mechanical gripper opens and supports the edge of the part 12, and the anti-fall baffle limits it from all sides. The lifting drive mechanism 3 puts the part 12 back to the transition conveyor table 14 or the nearest safe position according to the degradation rule to prevent the part 12 from falling. Typical parameters: vacuum level ranges from -60 kPa to -85 kPa; number of adsorption zones ranges from 2 to 6; number of suction cups is determined by the area of the adsorption surface and the holding force of a single suction cup, ranging from 4 to 24. The mechanical grippers and anti-fall baffles are pressure-loss protection components integrated into the gripping head 2, belonging to two different parts from the U-shaped frame 86 and rubber block 87 of the flipping mechanism 8, and their activation times are also different: the mechanical grippers only open to support the edge of the plate when the gripping head 2 holds the part 12 and vacuum adsorption fails; during the rotation of the flipping mechanism 8, the electric suction cup 22 of the gripping head 2 has released its adsorption and risen outside the rotation envelope, at which point the U-shaped frame 86 and rubber block 87 bear the weight of the part 12 and the rotational inertial load. The gripping force of the mechanical grippers is not less than twice the weight of the part 12, and the extension of the anti-fall baffle is not less than half the thickness of the part 12. The hydraulic cylinder 21 provides a secondary lifting stroke for the gripping head 2 relative to the rectangular plate 81, with a stroke ranging from 150mm to 400mm. This stroke is used to send the electric suction cup 22 along with the part 12 into the frame-shaped inner cavity of the rotary frame 83, and to release the part 12 below the rotary frame 83 after rotation. When the part is suspended or descending under load, the rodless chamber of the hydraulic cylinder 21 is locked by a hydraulically controlled check valve, so the part 12 will not slide down on its own when the power is cut off or the oil pipe is broken.
[0041] The out-of-plane contour detection unit 6 consists of several line laser contour sensors, sensor mounting bases, detection brackets, sensor power supply and signal cables, and analog-to-digital conversion and synchronous triggering modules. The line laser contour sensors preferably employ triangulation laser contour sensors, with each sensor outputting a sequence of elevation heights above the surface along a single scanning line.
[0042] The line laser profile sensor assembly is arranged laterally along the transition conveyor table 14, which is perpendicular to the feed direction of the part 12. It is fixed to the detection bracket via sensor mounting bases. The two ends of the detection bracket are bolted to the frame columns of the transition conveyor table 14, or installed on an independent foundation to isolate the transmission of vibration from the conveyor line to the sensors, so that the sensor assembly is suspended above the transition conveyor table 14. The sensor mounting base is an L-shaped corner mount. Its vertical surface is screwed to the waist-shaped hole on the lower wing of the crossbeam of the detection bracket, and its horizontal surface supports the sensor body. The mounting spacing of each sensor can be adjusted laterally along the crossbeam through the waist-shaped hole, and the working distance can be finely adjusted along the height direction. A heat insulation plate is placed between the mounting base and the sensor, and a protective cover is installed to block the residual heat and arc light from cutting. The power supply and signal cables of the sensors are gathered by a drag chain to the junction box on one side of the detection bracket, and then introduced into the shape calculation unit 7 in the control cabinet 11 by shielded cables. The laser projection plane of each line laser profile sensor is perpendicular to the upper surface of part 12. That is, the plane containing the laser fan is coplanar with the normal to the upper surface of part 12 and forms a 90-degree angle with the upper surface of part 12. Thus, the height value measured by the sensor directly corresponds to the distance of the upper surface of part 12 from the reference plane. The line laser profile sensors are arranged at intervals along the feed direction of part 12, forming several parallel scanning lines. The spacing between adjacent scanning lines is the scanning line spacing.
[0043] Measurement is performed while part 12 is being picked up by gripper head 2 but before it is dropped onto palletizing station 5. Specifically, it is triggered when gripper head 2 has lifted part 12 from transition conveyor 14 and is hovering it above the inspection station. This condition is chosen for two reasons: first, after part 12 is lifted, there is no pallet 13 obstructing its view below, allowing the out-of-plane contour detection unit 6 to obtain the full surface view of part 12 without obstruction; second, the measurement action is parallel to the gripping and conveying actions, without separately occupying the palletizing cycle, meaning the inspection is performed during the conveying stroke and does not affect the dropping cycle of each part 12. Typical parameters: 3 to 5 scan lines, 200mm to 400mm scan line spacing, 200Hz to 2000Hz sampling frequency per sensor, measurement resolution not less than 0.2mm (meaning the minimum resolvable increment of the height of a single point above the surface is not greater than 0.2mm), and the measurement range covers the surface undulations of part 12 from 0mm to 50mm in the thickness direction. When the projected side length of part 12 exceeds the coverage of the scanning lines, the transfer mechanism 4 makes one or more steps at the detection station so that each scanning line traverses the entire length of part 12.
[0044] The morphology calculation unit 7 consists of a processor, a fitting calculation program, a morphology discrimination module, and a data interface. Physically, it can be an industrial control computer or a software module in an embedded controller within the control cabinet 11. It receives the height sequence of the out-of-plane contour detection unit 6 through the data interface and sends the calculation results to the layer gap register 9 and the orientation decision unit 10 through the data interface.
[0045] The data input terminal of the morphology calculation unit 7 is electrically connected to the signal output terminal of the out-of-plane contour detection unit 6, and the data output terminal is electrically connected to the layer gap register 9 and the orientation decision unit 10, respectively. Its calculation process is completed within the control cabinet 11 without a separate independent mechanical structure. Physically, the morphology calculation unit 7 is installed in the control cabinet 11 in the form of a data acquisition board and a computing host: the acquisition board is inserted into the back panel slot of the middle layer chassis of the control cabinet 11, and the computing host is fixed to the DI unit in the same compartment by guide rail clips. On the guide rail, the two are interconnected via a backplane bus; the shielded cable from the out-of-plane contour detection unit 6 is introduced from the aviation socket on the side wall of the cabinet, connected to the terminal block of the board, and then sent to the computing host. The flat signal line runs along the cable tray inside the cabinet and is laid separately from the power line at the bottom of the cabinet. The morphology calculation unit 7, the layer gap register 9, and the orientation decision unit 10 are located in the same cabinet. All data interaction between the three is completed through the backplane bus inside the cabinet, without any wiring brought out outside the cabinet.
[0046] The morphology calculation unit 7 performs the following mathematical process on the height sequence from the out-of-plane contour detection unit 6.
[0047] First, record each measuring point as... Where x_j and y_j are the coordinates of the measuring point in the plane of part 12, and z_j is the height of the corresponding measuring point relative to the reference plane. A least-squares quadratic surface is fitted to all measuring points, and the fitted surface is defined as: ; Solve for the coefficients a, b, c, d, e, and f according to the criterion of minimizing the sum of squared residuals, that is, let ; The minimum value is taken to obtain the coefficient vector. This solution is a linear least squares problem, which is calculated in real time by the morphological solution unit 7 according to the normal equations.
[0048] Secondly, the orientation is determined by the signs of the quadratic terms a, b, and c. Specifically, examine the sign of curvature of the fitted surface at the center of part 12: if the surface is convex upwards near its stationary point, meaning the center of the fitted surface is higher than the projection of its surrounding boundaries, then part 12 will be convex upwards when placed in its natural delivery posture, denoted as... If the surface is concave, meaning the middle part is lower than the boundary projection, then it is concave upwards, denoted as... Equivalently, the trace of the Hessian matrix of the fitted surface can be calculated. The symbol: trace corresponds to an upward convex shape. The trace is negative, corresponding to the concave shape. .
[0049] Next, the maximum height of each measuring point relative to the fitted surface is taken as the deflection amplitude. ,Right now ; Should Always take a positive value, which represents the upper limit of the amplitude of the undulations outside the 12 surfaces of the part.
[0050] Finally, the calibration morphology is determined as either surface bending or end warping. The criterion is the spatial distribution of the residual, which is the height of the measuring point relative to the fitted surface: if the residual exhibits a monotonic or approximately monotonic curvature change along the feed direction of part 12 and is continuously distributed across the entire surface, it is determined to be surface bending; if the residual is approximately zero in the middle but shows a concentration of opposite signs at both ends of part 12 along the feed direction (i.e., one end is positive and the other is negative, or both ends are in the same direction but the middle is flat), it is determined to be end warping. When end warping is determined, the deflection amplitude is... Instead of taking the maximum residual over the entire area, the difference in the height above the surface between the measuring points at both ends is taken, i.e. Where z_{end 1} and z_{end 2} are the heights above the surface of representative measuring points at both ends of part 12 along the feed direction. The reason for this treatment is that the middle part of the end warping edge is flat, and the maximum residual over the entire area may fall at the local slag or edge burrs, while the difference in heights above the surface at both ends can better represent the true warping amplitude of the end warping edge.
[0051] The flipping mechanism 8 consists of a rectangular plate 81, a hanging plate 82, a rotating frame 83, a rotating motor 84, an electric push rod 85, a U-shaped frame 86, a rubber block 87, and an injection pump 88; it is connected to the lifting output end of the lifting drive mechanism 3 via the rectangular plate 81, and to the electric suction cup 22 of the gripping head 2 via the hydraulic cylinder 21. The rectangular plate 81 is a horizontally placed load-bearing plate. Its upper surface is bolted to the lifting output flange of the lifting drive mechanism 3 at the middle, and its two ends are fixed to the upper ends of the two hanging plates 82. The two hanging plates 82 are parallel plates with their surfaces perpendicular to the feeding direction of the part 12 and extending downward from both ends of the rectangular plate 81. The rotary frame 83 is a rectangular frame component with connecting shafts extending from both sides along the feeding direction. The common axis of the two connecting shafts is the rotation axis. The rotary frame 83 is rotatably installed between the two hanging plates 82 via the two connecting shafts. The rotary motor 84 is fixed to the outer wall of one hanging plate 82, and its output shaft is coaxially connected to the connecting shaft on that side. The connecting shaft on the other side is supported on the other hanging plate 82 via a bearing seat. The side wall of the rotating frame 83 is equipped with several electric push rods 85. In this embodiment, two are used and placed on the two opposite side walls of the rotating frame 83. Each of their telescopic ends is fixed with a U-shaped frame 86. The two U-shaped frames 86 are placed on the two opposite edges of the part 12. The upper and lower inner walls of the U-shaped frame 86 each have a rectangular hole. A rubber block 87 is sealed in the hole. The inner cavity of the U-shaped frame 86 is filled with hydraulic oil. The side wall of the U-shaped frame 86 is equipped with an injection pump 88. The oil inlet and outlet of the injection pump 88 are connected to the inner cavity of the U-shaped frame 86 through an oil passage. The two injection pumps 88 are synchronously controlled by the control cabinet 11 so that the clamping force on both sides of the part 12 is established synchronously.
[0052] The rectangular plate 81 serves as the assembly interface between the flipping mechanism 8 and the lifting drive mechanism 3, and also as the load-bearing component for all parts of the flipping mechanism 8 and the gripping head 2. Its upper surface is bolted to the lifting output flange of the lifting drive mechanism 3 at its center, and its two ends are fixed to the upper ends of the two hanging plates 82. The top of the plate also houses the cylinder bodies of two hydraulic cylinders 21. The net distance between the two hanging plates 82 is greater than the width of the rotary frame 83 along the feed direction, and also greater than the projected side length of the part 12 along the feed direction, ensuring that neither end of the part 12 collides with the hanging plates 82 when the rotary frame 83 rotates. The rectangular frame plane of the rotary frame 83 is initially horizontal, with connecting shafts extending from both sides along the feed direction. These shafts are horizontal, parallel to the feed direction of the part 12, and located on the symmetrical mid-plane of the two hanging plates 82. The rotary motor 84 is fixed to the outer wall of one side of the hanging plate 82. Its output shaft is directly connected to the connecting shaft on the same side via a coupling, and there is no reducer in the transmission chain. Since the rotary motor 84 is mounted on the non-rotating hanging plate 82, its power cable and control line are laid along the hanging plate 82 and the rectangular plate 81 and then merge into the ram cable groove of the lifting drive mechanism 3. It does not rotate with the rotating frame 83 throughout the entire process, so there is no need to set a rotary joint or a center cable hole. The cables and oil pipes of the electric push rod 85 and the injection pump 88 on the rotating frame 83 are first laid along the frame of the rotating frame 83 to the connecting shaft, and then led out through the axial cable hole in the center of the connecting shaft to the outside of the hanging plate 82, and rotate together with the rotating frame 83. At the same time, mechanical limit blocks are set at the relative positions of the hanging plate 82 and the rotating frame 83 to limit the over-rotation of the rotating frame 83. The cylinder of the electric push rod 85 is mounted on the side wall of the rotary frame 83. Its extension and retraction direction is horizontal and perpendicular to the edge of the plate of part 12. The extension and retraction end is fixed to the bottom wall of the U-shaped frame 86. The opening of the U-shaped frame 86 faces the edge of the plate of part 12. The rectangular holes on its upper and lower inner walls are vertically connected. The holes are sealed with rubber blocks 87. The contact surfaces of the rubber blocks 87 face the upper and lower surfaces of part 12. The cylinder of the injection pump 88 is mounted on the side wall of the U-shaped frame 86. Its oil inlet and outlet are connected to the inner cavity of the U-shaped frame 86. The inner cavity is filled with hydraulic oil. The U-shaped frame 86 and the rubber block 87 together constitute a hydraulic adaptive clamping pair: the rubber block 87 has a travel of not less than 8mm along the rectangular hole, and its material is oil-resistant rubber or polyurethane, so it can adaptively fit the out-of-plane undulations of the upper and lower surfaces of the part 12 one by one, thus accommodating both the curved surface and the warped edge of the plate on the same U-shaped frame 86; the contact length of its contact surface on the upper and lower surfaces of the part 12 is not less than one-third of the edge length of the part 12. A lip seal and a return spring are provided between the rectangular hole and the rubber block 87: the lip seal prevents hydraulic oil from leaking out of the rectangular hole, and the return spring pushes the rubber block 87 back to its initial position after the injection pump 88 is depressurized, so that the part 12 is released from the constraint.The hydraulic cylinder 21 of the gripping head 2 is mounted on the top of the rectangular plate 81. Therefore, neither the gripping head 2 nor the electric suction cup 22 is mounted on the rotating frame 83 and does not rotate with the rotating frame 83. The frame-shaped inner cavity of the rotating frame 83 provides space for the flipping of the part 12. The electric suction cup 22 can descend into or rise out of the opening above the inner cavity without interference. The clamping center of the part 12 coincides with the projection of the rotation axis, so that when the part 12 is flipped, it flips around the longitudinal horizontal axis passing through its own geometric mid-surface. The driving torque is minimized, and the projection position of the part 12 in the horizontal plane remains unchanged before and after flipping. The contact surface of the rubber block 87 is provided with anti-slip texture to prevent slippage when clamping the part 12.
[0053] The rotation axis of the flipping mechanism 8 is parallel to the feeding direction of the part 12, that is, horizontal and along the conveying direction. Therefore, when the rotating frame 83 rotates 180 degrees around the horizontal axis, the part 12 flips in the vertical plane, and its upper surface changes from facing up to facing down or from facing down to facing up, while the projection orientation of the part 12 in the horizontal plane remains unchanged. The clamping, rotating and handover actions are performed in the following order, and the load path during the flipping is separate from the load path during the picking and placing: (a) Positioning and clamping: After the gripping head 2 sucks up the part 12 and rises back to a safe height, the transfer mechanism 4 sends the part 12 to the clamping position of the flipping mechanism 8; the hydraulic cylinder 21 lowers the electric suction cup 22 along with the part 12 into the frame-shaped inner cavity of the rotating frame 83, and the two electric push rods 85 extend synchronously, pushing the two U-shaped frames 86 to the two opposite plate edges of the part 12 respectively; then the two injection pumps 88 inject hydraulic oil into the inner cavity of the U-shaped frame 86 synchronously, and the rubber block 87 extends along the rectangular hole under oil pressure and adaptively fits the upper and lower surfaces of the part 12, elastically clamping the part 12 in the U-shaped frame 86. (ii) Load bearing: During rotation, the entire weight of part 12 and the rotational inertia load are borne by the friction of the U-shaped frame 86 and the rubber block 87 through the oil pressure self-locking of the contact surface. The clamping force of the rubber block 87 on part 12 is not less than twice the weight of part 12, and the effective clamping length of the contact surface is not less than one-third of the length of the edge of part 12, so as to ensure that part 12 does not slip or retract when it is vertically suspended at a 90-degree position. During rotation, the electric suction cup 22 of the gripping head 2 has been released from adsorption and risen outside the rotation envelope. There is no constraint between it and part 12, and it does not participate in load bearing. The rotational reaction torque of the rotary frame 83 is jointly borne by the lead screw self-locking of the electric push rod 85, the holding torque of the rotary motor 84, and the bearing seat at the connecting shaft. The rotary motor 84 is a servo motor with a power failure brake. When the rotary frame 83 is in position, or when the motor loses power or is accidentally disconnected, the brake locks the rotary frame 83 in that angle position to prevent the part 12 from rotating on its own and changing its orientation. (III) Rotation: The rotary motor 84 drives the rotary frame 83, together with the electric push rod 85, U-shaped frame 86, rubber block 87, and part 12, to rotate 180 degrees via the coupling, or to remain at the zero position without flipping. (iv) Handling and release: After the flipping is in place, the electric push rod 85 remains extended, the injection pump 88 maintains oil pressure, and the hydraulic cylinder 21 lowers the electric suction cup 22 into the rotation envelope range and re-adsorbs the flipped part 12; after adsorption is established and the vacuum reaches the set value, the injection pump 88 pumps oil to make the rubber block 87 retract under the action of the return spring, and the electric push rod 85 retracts to make the U-shaped frame 86 leave the edge of the plate, and the part 12 is handed over to the gripper head 2 to hold again; then the hydraulic cylinder 21 lowers the part 12 to below the rotating frame 83, and then the gripper head 2 moves it to the top of the stacking table 5 for release.The above sequence ensures that during the entire process of rotating part 12 around the horizontal axis, there is only one mechanism bearing the force. That is, the picking and placing section is supported by the electric suction cup 22 of the gripping head 2, and the rotating section is supported by the U-shaped frame 86 and the rubber block 87. The two switch the supporting body at the junction point. Thus, the gripping head 2 has risen outside the rotating envelope during the rotation. After the rotation, the frame of the rotating frame 83 is always located below part 12. The gripping head 2 and the electric suction cup 22 will not be sandwiched between part 12 and the stacking table 5, and the material drop path is unobstructed. The positioning accuracy after 180-degree rotation is determined by the encoder resolution of the rotary motor 84 and its static stiffness under holding torque. Since there is no reducer in the transmission chain, the rotary motor 84 should preferably be a low-speed, high-torque direct-drive servo motor, or a servo motor plus an indexing positioning plate can be used. The typical positioning repeatability is no more than ±0.5 degrees, which corresponds to a horizontal position offset of no more than 2 mm after the edge of part 12 is flipped with a projected side length of 3000 mm. The flipping action must be completed before the gripper head 2 releases part 12: that is, the orientation decision unit 10 provides the placement orientation factor. Then, the flipping mechanism 8 first performs flipping or holding, and then the gripping head 2 moves and releases the oriented part 12 above the stacking table 5 to ensure that the effective orientation of the part 12 when it falls into the stack 13 is consistent with the control command.
[0054] Both the layer gap register 9 and the orientation decision unit 10 are physically memory and processor modules within the control cabinet 11. The layer gap register 9 is a read / write data memory, and the orientation decision unit 10 is a processor that executes the decision algorithm. They are electrically connected via the internal bus of the control cabinet 11 and are also electrically connected to the shape calculation unit 7 and the flipping mechanism 8. Their positions within the cabinet are adjacent to the shape calculation unit 7: the orientation decision unit 10 is installed as a processor board in an adjacent slot within the same chassis as the shape calculation unit 7 in the control cabinet 11; the layer gap register 9 is a power-off retained data memory, fixed to the DI below this compartment. The guide rails are powered by the DC power supply inside the cabinet via an uninterruptible power supply module to ensure the number of stacked layers and the accumulated layer gaps. The status is not lost after an unexpected power outage; the three are interconnected via a backplane bus, and are connected to the flipping mechanism 8, lifting drive mechanism 3, and transfer mechanism 4 outside the cabinet via terminal blocks at the bottom of the cabinet and output relays and drive interfaces.
[0055] Segmentation register 9 stores the following field: cumulative segmentation. , is a floating-point number in mm; the effective orientation of the previously placed part 12. The value is +1 or -1; the deflection amplitude of the previously placed part 12. , is a positive value, and the unit is mm; number of layers placed. The values are integers. The orientation decision unit 10 reads these fields after each part 12 is released and updates them after the solution is complete. , and Write back the layer gap register 9; control cabinet 11 will write the layer gap of the current layer after each piece is released. Accumulated to cumulative gap And write it back.
[0056] The decision-making of orientation decision unit 10 is based on the following hierarchical relationship. Let the first... The effective orientation of part 12 that has been placed is Flexural amplitude value , No. +1 current part 12 has a valid orientation. Flexural amplitude value The gap between the two is ; because , Each takes only +1 or -1, the effective orientation of the current item. By placement orientation factor With calibration orientation Decision, that is The orientation decision unit 10 can only control... Its value can be +1 or -1, while Part 12 is naturally delivered in a fixed and randomly distributed manner. That can make That is, two adjacent pieces have the same effective orientation and are stacked in the same direction. At this time, formula (1) becomes ; That is, the gap is reduced to the difference in deflection amplitude between two adjacent pieces. In contrast, if no treatment is done, the effective orientations of two adjacent pieces are opposite ( ),but ; That is, the gap is the sum of the deflection amplitudes of the two pieces, approximately twice the average deflection amplitude.
[0057] The above derivation holds true on the premise that "the calibration morphology functions of two adjacent parts are approximately consistent." This premise holds true for parts from the same batch of thermally cut parts because: parts from the same batch are taken from the same base plate, have the same carbon equivalent, use the same cutting parameters (cutting speed, power, and gas flow rate), and have the same cooling conditions. Therefore, the out-of-plane morphology function, i.e., the curvature distribution of the bend or warp, exhibits batch-specific systematic correlation. The differences between individual parts mainly lie in the deflection amplitude. In terms of size, and shape function Approximately consistent. Therefore, in formula (1)... , The amplitude of the linear combination approximation gap in the projected overlapping area is reasonable. The gap can be compressed to the difference in the deflection amplitude between the two adjacent pieces if the effective orientations are the same.
[0058] When the cache station is enabled, the orientation decision unit 10 selects the deflection amplitude value from a plurality of parts 12 to be coded. Deflection amplitude of the previous part 12 The one with the smallest difference is selected as the current part 12, that is, the nearest neighbor is selected in the cache pool, such that | - | Further reduce, thereby accumulating gaps Press down to an even lower level.
[0059] When the accumulated gaps When the value exceeds a set threshold, such as 100mm or a threshold calculated based on the number of layers, the orientation decision unit 10 changes to placing the orientation factor for the current part 12. The opposite value, that is, when the effective orientation of two adjacent pieces is the same, the orientation is temporarily switched to opposite due to exceeding the limit. The single-layer gap changes from the difference type to the sum type. At the same time, the control cabinet 11 issues an audible and visual alarm or a prompt signal in the form of a message from the host computer so that the operator can check.
[0060] Control cabinet 11 integrates morphology calculation unit 7, gap register 9, and orientation decision unit 10. Their signal links are closed in the following sequence: Out-of-plane contour detection unit 6 measures the height sequence when part 12 is gripped but not yet dropped, and sends it via electrical signal to morphology calculation unit 7 within control cabinet 11; morphology calculation unit 7 calculates the calibrated morphology and deflection amplitude. With calibration orientation Write the result to the gap register 9 and set the deflection amplitude value. With calibration orientation Orientation decision unit 10 is sent to the orientation decision unit 10; the orientation decision unit 10 reads the valid orientation of the previous item from the gap register 9. With flexural amplitude The placement orientation factor of the current part is calculated according to formula (1). ,Will Sending to the execution controller of the flipping mechanism 8 and the transfer mechanism 4; the flipping mechanism 8 presses... After the flipping or holding is completed, the lifting drive mechanism 3 and the transfer mechanism 4 drive the gripper head 2 to transfer and release part 12 above the palletizing table 5; after part 12 is released, the orientation decision unit 10 will determine the layer gap. The accumulated gaps in the gap register 9 and the effective orientation of this document With flexural amplitude Write back the layer gap register 9 for use by the next item. The above link is closed once in each stacking cycle of part 12.
[0061] The following describes a complete cycle from zero to full stack, explaining the actions that occur for each item upon arrival. For ease of explanation, let the target stack number of 13 be . Layers, the initial number of layers placed =0, cumulative gap =0.
[0062] Step 1, Initialization: Control cabinet 11 will accumulate the layer gap in layer gap register 9. Reset to zero, number of layers already placed Set the position to 0 and set the reference transfer position above the palletizing platform 5. The electric push rod 85 of the flipping mechanism 8 is in the retracted position, the two U-shaped frames 86 are in the yielding position, the injection pump 88 is in the depressurized state, and the rotary frame 83 is in the zero position parallel to the feed direction of the part 12.
[0063] Step 2, First part arrives: The first part 12 output from the previous cutting machine arrives at the transition conveyor table 14. The transfer mechanism 4 drives the gripper head 2 to move above the part, and the lifting drive mechanism 3 lowers to pick up the part 12 and lifts it to the inspection station. The out-of-plane contour detection unit 6 measures the height sequence of the upper surface of the part along several scanning lines and sends it to the shape calculation unit 7. The shape calculation unit 7 fits and obtains its calibration shape, such as the surface curvature and deflection amplitude. With calibration orientation In this embodiment, =+1. Since there is no previous item in the interlayer gap register 9, at this time... =0, orientation decision unit 10 takes the first item as the reference and takes the placement orientation factor. =+1, meaning maintaining a natural delivery posture and effective orientation. = · =+1; The flipping mechanism 8 does not flip, and the gripping head 2 releases the first piece to the palletizing table 5, where it lands on the table surface. After release, the layer gap register 9 records... , Number of layers already placed Updated to 1, cumulative gap It remains at 0 because there is no lower layer to form a gap with the first layer.
[0064] Step 3, arrival of the second part: The second part, part 12, arrives at the transition conveyor 14, and is detected and calculated. and In this embodiment, =+1. Orientation decision unit 10 reads the previous layer. =+1、 , for order = =+1, thus making two adjacent pieces have the same effective orientation and are stacked in the same direction, take =sign( · ) = +1, because =+1, therefore, it maintains a natural delivery posture without flipping, effectively orienting. = · =+1. The gripper head 2 releases the second piece above the first piece; both are stacked in the same direction, with their convex surfaces facing upwards. Layer gap. =| - |, Orientation decision unit 10 accumulates it to the cumulative gap. Layer gap register 9 updated , , Updated to 2.
[0065] Step four, the +1 item, general cycle: for item 3 to item 4 Each piece repeats the following actions: upon reaching the transition conveyor 14, it is lifted; the out-of-plane contour detection unit 6 measures its height; and the shape calculation unit 7 calculates its shape. , Orientation decision unit 10 reads the previous layer , and according to The orientation factor is calculated, and the flipping mechanism 8 flips or holds the component accordingly. The gripping head 2 releases the component onto the upper surface of the stack 13, and the orientation decision unit 10 calculates... And accumulated to the cumulative gap Layer gap register 9 write-back , , Increment by 1. Thus, the effective orientation of two adjacent pieces is always the same, and they are stacked in the same direction. The gap between each layer is the difference in the deflection amplitude of the two adjacent pieces.
[0066] Step 5, Over-limit handling: During the cycle, if the accumulated gaps after a certain accumulation... If the value exceeds a set threshold, the orientation decision unit 10 will change the placement orientation factor for that component. The opposite value is obtained and a prompt signal is triggered in control cabinet 11, while the gap register 9 is still updated according to the actual valid orientation.
[0067] Step 6, Finishing the Stack: When the stack has been filled to the required number of layers... Reach the set number of floors The transfer mechanism 4 stops picking up materials, and the stack 13 is completed. At this time, two adjacent parts 12 are stacked in the same direction, that is, both with their convex surfaces facing upwards or both with their concave surfaces facing upwards, and the gaps between each layer are... The cumulative gap is the difference in deflection amplitude between two adjacent elements. Not affected by floor number It is linearly cumulative, but depends on the dispersion of the deflection amplitude within the batch.
[0068] The key parameters involved in this invention, their value ranges, and their functions are as follows, and these values all correspond to the presumptions stated in the case concept: Thickness parameters: Part 12 is a hot-cut steel plate component with a thickness ranging from 4mm to 40mm. This range covers common plate thicknesses for steel structure cutting. The greater the thickness, the more pronounced the out-of-plane warping tendency caused by cutting heat input becomes, and the more necessary the alignment suppression in this solution becomes.
[0069] Deflection amplitude parameter: Deflection amplitude of a single piece The amplitude ranges from 2mm to 8mm. This amplitude determines the order of the interlayer gap introduced per piece when unaligned, approximately 2. It also determines the base number of the residual gap after alignment, that is, the difference in deflection amplitude between two adjacent pieces.
[0070] Layer count parameter: The number of layers for stack 13 can be set from 10 to 40. The more layers, the greater the cumulative gap when not aligned. The more severe the linear growth, the more prominent the suppression effect of this scheme.
[0071] Scan line spacing: The scan line spacing of the out-of-plane contour detection unit 6 is between 200mm and 400mm. The smaller the spacing, the more precise the surface fitting of the bend in the board surface; the larger the spacing, the fewer sensors are needed and the lower the cost. This spacing must ensure that at least 3 scan lines are laid out along the entire length of the part 12 to distinguish the overall curvature of the bend.
[0072] Inspection cycle time: The measurement time for a single piece by the out-of-plane contour detection unit 6 is determined by the number of scan lines, sampling frequency, and feed speed of part 12, typically ranging from 0.3 seconds to 1.5 seconds. Since the measurement is performed in parallel during the gripping and hovering process, this time must not exceed the single-piece handling cycle time to avoid slowing down the palletizing rhythm.
[0073] Transport cycle time: The typical time for transferring and releasing a single item from the transition conveyor 14 to the palletizing station 5 is 6 to 15 seconds, determined by the translation speed of the transfer mechanism 4 and the lifting speed of the lifting drive mechanism 3. The detection cycle time must be nested within the transport cycle time.
[0074] The clamping oil pressure inside the U-shaped frame 86 is between 0.2 MPa and 2.0 MPa, automatically adjusted by the injection pump 88 according to the outlet pressure. The extension stroke of the rubber block 87 along the rectangular hole is not less than 8 mm, used to cover the 2 mm to 8 mm out-of-plane undulations on the upper and lower surfaces of the part 12. The hardness of the contact surface of the rubber block 87 is Shore A 60 to 80, which can both slightly deform and fit the curved surface under oil pressure without leaving indentations on the surface of the part 12. The rated torque of the rotary motor 84 is selected as twice the moment of the weight of the rotary frame 83, electric push rod 85, U-shaped frame 86, rubber block 87 and part 12 about the rotation axis. Its encoder resolution is not less than 4096 lines per revolution, and it is equipped with a power failure brake and holding torque.
[0075] When nesting data feedforward is enabled, the orientation decision unit 10 reads the nesting data from the previous cutting process, predicts the order in which each part 12 arrives at the transition conveyor 14, and pre-arranges the placement of orientation factors. Sequence. This prediction reduces the reliance on real-time computation and, in conjunction with nearest neighbor selection when caching is enabled, further compresses gaps.
[0076] The following are two numerical examples, showing the layer gaps of each piece. With accumulated gaps Calculations were performed item by item to demonstrate the sufficiency of the disclosure and the technical effect. In the embodiment, the deflection amplitude of part 12 is assumed. The alignment criteria, determined piece by piece by the morphological calculation unit 7, are that adjacent effective orientations are the same. ), first item taken =+1.
[0077] In one embodiment, the average flexural amplitude is 4 mm, the batch dispersion is 1 mm, and the stack is up to 20 layers: The deflection values of the 20 parts 12 in this embodiment according to their arrival order The values are as follows, all in mm: 4.0, 5.0, 3.0, 4.5, 3.5, 4.8, 4.0, 3.2, 4.9, 3.1, 4.6, 3.4, 4.7, 4.1, 3.3, 4.4, 3.6, 4.2, 3.8, 4.0.
[0078] Its mean is approximately 4.0 mm, and the deviation of a single piece from the mean is no more than 1.0 mm, satisfying the assumption of a dispersion of 1 mm.
[0079] In the case of non-alignment, the effective orientations of two adjacent pieces are opposite: the gaps between adjacent pieces , cumulative gaps .
[0080] all The sum is 80.1 mm. =4.0mm, =4.0mm, therefore .
[0081] That is, the cumulative gap of the 20-layer stack 13 without orientation is 152.2 mm.
[0082] Orientation case, where adjacent effective orientations are the same: gaps between adjacent pieces The calculations are as follows, in mm: Items 1–2: |5.0−4.0|=1.0, =1.0; Items 2–3: |3.0−5.0|=2.0, =3.0; Items 3–4: |4.5−3.0|=1.5, =4.5; Items 4–5: |3.5−4.5|=1.0, =5.5; Items 5–6: |4.8−3.5|=1.3, =6.8; Items 6–7: |4.0−4.8|=0.8 =7.6; Items 7–8: |3.2−4.0|=0.8, =8.4; Items 8–9: |4.9−3.2|=1.7, =10.1; Items 9–10: |3.1−4.9|=1.8, =11.9; Items 10–11: |4.6−3.1|=1.5, =13.4; Items 11–12: |3.4−4.6|=1.2, =14.6; Items 12–13: |4.7−3.4|=1.3, =15.9; Items 13–14: |4.1−4.7|=0.6, =16.5; Items 14–15: |3.3−4.1|=0.8, =17.3; Items 15–16: |4.4−3.3|=1.1, =18.4; Items 16–17: |3.6−4.4|=0.8, =19.2; Items 17–18: |4.2−3.6|=0.6, =19.8; Items 18–19: |3.8−4.2|=0.4, =20.2; Items 19–20: |4.0−3.8|=0.2, =20.4.
[0083] That is, the cumulative gap of the 20-layer stack 13 after alignment is 20.4mm.
[0084] Comparison: The ratio of unaligned 152.2mm to aligned 20.4mm is 7.46, and the cumulative gap is reduced to less than one-eighth of that of the unaligned gap, which is consistent with the estimated 152mm to 19mm in the case concept.
[0085] Another embodiment, with an average deflection amplitude of 6 mm, a dispersion of 1.5 mm, stacked to 30 layers, and the buffer station option enabled: The deflection values of the 30 parts 12 in this embodiment according to their arrival order The values are as follows, all in mm: 6.0, 7.5, 4.5, 6.8, 5.2, 7.0, 4.8, 6.5, 5.5, 7.2, 4.6, 6.3, 5.0, 6.9, 4.9, 7.1, 5.3, 6.6, 5.8, 7.4, 4.7, 6.1, 5.6, 6.7, 5.1, 6.4, 5.4, 7.3, 4.5, 6.2.
[0086] Its average value is approximately 6.0 mm, and the deviation of a single piece from the average value is no more than 1.5 mm, satisfying the assumption of a dispersion of 1.5 mm.
[0087] Non-alignment cases: All The sum is 178.9 mm. =6.0mm, =6.2mm, cumulative gap .
[0088] That is, the cumulative gap of the 30-layer stack 13 without orientation is 345.6 mm.
[0089] In the case of unbuffered alignment, adjacent effective orientations are the same: the gap between adjacent pieces is... The calculations are as follows, in mm: |6.0→7.5|=1.5, |7.5→4.5|=3.0, |4.5→6.8|=2.3, |6.8→5.2|=1.6, |5.2→7.0|=1.8, |7.0→4.8|=2.2, |4.8→6.5|=1.7, |6.5→5.5|=1.0, |5.5→7.2|=1.7, |7.2→4.6|=2.6, |4.6→6.3|=1.7, |6.3→5.0|=1.3, |5.0→6.9|=1.9, |6.9→4.9|=2.0, |4.9→7. 1|=2.2, |7.1→5.3|=1.8, |5.3→6.6|=1.3, |6.6→5.8|=0.8, |5.8→7.4|=1.6, |7.4→4.7|=2.7, |4.7→6.1|=1.4, |6.1→5.6|=0.5, |5.6→6.7|=1.1, |6.7→5.1|=1.6, |5.1→6.4|=1.3, |6.4→5.4|=1.0, |5.4→7.3|=1.9, |7.3→4.5|=2.8, |4.5→6.2|=1.7.
[0090] The cumulative gaps, calculated by summing the values, are: 1.5, 4.5, 6.8, 8.4, 10.2, 12.4, 14.1, 15.1, 16.8, 19.4, 21.1, 22.4, 24.3, 26.3, 28.5, 30.3, 31.6, 32.4, 34.0, 36.7, 38.1, 38.6, 39.7, 41.3, 42.6, 43.6, 45.5, 48.3, and 50.0, in mm. This means the cumulative gap for the 30-layer stack 13 without buffer alignment is 50.0 mm.
[0091] Orientation and Cache Station Option Activation: The cache station holds several parts 12 to be coded, and the orientation decision unit 10 selects the deflection amplitude value from the parts to be coded. The item with the smallest difference from the previous item is selected as the current item, i.e., nearest neighbor selection. In the extreme case where the buffer can hold all items to be coded and they are sorted in ascending or descending order of deflection amplitude, the item with the smallest difference in deflection amplitude is selected after sorting. The sequence is: 4.5, 4.5, 4.6, 4.7, 4.8, 4.9, 5.0, 5.1, 5.2, 5.3, 5.4, 5.5, 5.6, 5.8, 6.0, 6.1, 6.2, 6.3, 6.4, 6.5, 6.6, 6.7, 6.8, 6.9, 7.0, 7.1, 7.2, 7.3, 7.4, 7.5.
[0092] The adjacent differences are as follows: 0, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.2, 0.2, 0.1 ... The cumulative layer gaps are calculated by summing the values: 0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0, 1.1, 1.3, 1.5, 1.6, 1.7, 1.8, 1.9, 2.0, 2.1, 2.2, 2.3, 2.4, 2.5, 2.6, 2.7, 2.8, 2.9, 3.0, in mm. That is, when buffering is enabled and the nearest neighbor sorting is performed, the cumulative layer gap for stack 13 (30 layers) is 3.0 mm. In a practically limited buffer window, such as a window of 5 pending codes, successive nearest neighbor selection ensures that the difference between adjacent deflection amplitudes is no greater than 1.0 mm, resulting in a cumulative layer gap. The value is no greater than 29 × 1.0 = 29.0 mm, which is between the lower limit of 3.0 mm for full sorting and 50.0 mm for no cache, and is still better than the case without cache.
[0093] Comparison: Unaligned 345.6mm, aligned without buffer 50.0mm, aligned with buffer from the extreme case of 3.0mm to no more than 29.0mm under a 5-piece window. Compared to unaligned, aligned without buffer reduces the cumulative gap by about one-seventh, and aligned with buffer further reduces it to about one-one-hundred-fifteenth in the extreme case or about one-twelfth under a 5-piece window. Example 2 shows that under conditions where the average deflection amplitude is higher at 6mm and the number of layers is more than 30, the gap suppression effect of this solution increases synchronously with the average value and the number of layers, and the buffer option provides additional compression.
[0094] The two embodiments above together demonstrate that: the alignment stacking changes the gap of each layer from the sum of the deflection amplitudes of the two pieces to the difference of the deflection amplitudes of the two pieces. The cumulative gap no longer increases linearly with the number of layers, but is controlled by the dispersion of the deflection amplitude within the batch. The stronger the batch-specific system warping, the more prominent the suppression effect of this case.
[0095] When the outer contour detection unit 6 fails, such as when the sensor is disconnected or excessively dirty, the orientation decision unit 10 cannot obtain the current deflection amplitude value of the part. With calibration orientation At this point, the degradation rule applies: place the orientation factor. If the value is fixed at +1, meaning that each part 12 remains in its natural delivery posture without being flipped, the control cabinet 11 will issue an alarm signal and record the number of layers placed. This degradation mode does not rely on measured values, but it loses the ability to orient each part individually. The effective orientation of two adjacent parts is randomly distributed according to their natural delivery posture, and they are opposite in about half the probability, introducing a layer gap. The layer gap suppression effect decreases, requiring subsequent verification and handling by the operator.
[0096] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in the present invention should still be covered by the claims of the present invention.
Claims
1. A steel structure metal sheet cutting and palletizing device, comprising a palletizing frame (1), a gripping head (2), a lifting drive mechanism (3), a transfer mechanism (4), a palletizing table (5), and a transition conveyor table (14), characterized in that, It also includes an out-of-plane contour detection unit (6), a shape calculation unit (7), a flipping mechanism (8), a layer gap register (9), an orientation decision unit (10), and a control cabinet (11); the palletizing table (5) is placed below the palletizing frame (1), receiving parts (12) layer by layer to form a stack (13); the transition conveyor table (14) receives the parts (12) output from the previous hot cutting process; the gripping head (2) is driven by the lifting drive mechanism (3) and the transfer mechanism (4). The part (12) is moved from the transition conveyor (14) to the palletizing station (5) and released. The out-of-plane contour detection unit (6) is mounted above the transition conveyor (14). When the part (12) is grasped but not yet dropped, it measures the height sequence of the upper surface of the part (12) along several scanning lines. The shape calculation unit (7) calculates the out-of-plane shape of the part (12) based on the height sequence to obtain the calibration shape and deflection amplitude of the part (12). With calibration orientation The calibration orientation The value is +1 or -1, which respectively indicates that the part (12) is placed with its convex surface facing up or its concave surface facing up when it is placed in its natural delivery posture; the layer gap register (9) stores the cumulative layer gap of the stacked portion of the stack (13). and the effective orientation of the previously placed part (12). Its deflection amplitude The effective orientation The placement orientation factor With the calibration orientation The product; the orientation decision unit (10) reads the effective orientation of the previous part (12) from the gap register (9). With the aforementioned deflection amplitude The placement orientation factor of the current part (12) is calculated using the following formula. This makes the effective orientation of the current part (12) the same as the effective orientation of the previous part (12), thereby reducing the gap between the two layers. Minimum value: interlayer gap ,in , The effective orientation and deflection amplitude of the previously placed part (12) are respectively. , The effective orientation and deflection amplitude of the current part (12) are respectively; the flipping mechanism (8) includes a rectangular plate (81) fixed to the bottom extension end of the lifting drive mechanism (3); the gripping head (2) includes a hydraulic cylinder (21) and an electric suction cup (22). The cylinder body of the hydraulic cylinder (21) is mounted on the top of the rectangular plate (81), and the electric suction cup (22) is mounted on the bottom extension end of the hydraulic cylinder (21). Therefore, the gripping head (2) rises and falls with the lifting drive mechanism (3) but does not rotate with the rotating frame (83) of the flipping mechanism (8); the flipping mechanism (8) calculates the placement orientation factor according to the orientation decision unit (10). The part (12) is rotated 180 degrees or kept in its natural delivery posture, and then released by the gripper head (2) above the stacking table (5), so that two adjacent parts (12) are stacked in the same direction and in close contact, and the gap between the layers is... The difference in deflection amplitude between two adjacent parts (12); the control cabinet (11) is electrically connected to the out-of-plane contour detection unit (6), the shape calculation unit (7), the flipping mechanism (8), the layer gap register (9), the orientation decision unit (10), the lifting drive mechanism (3), and the transfer mechanism (4), and after each part (12) is released, the layer gap of this layer is... Accumulated to the accumulated layer .
2. The steel structure metal sheet cutting and stacking device according to claim 1, characterized in that, The out-of-plane contour detection unit (6) is a line laser contour sensor group arranged laterally along the transition conveyor table (14). The projection plane of the line laser contour sensor group is perpendicular to the upper surface of the part (12), and each line laser contour sensor forms a number of parallel scanning lines along the feeding direction of the part (12).
3. The steel structure metal sheet cutting and stacking device according to claim 1, characterized in that, The morphological calculation unit (7) performs least-squares quadratic surface fitting on the height sequence, and determines the calibration orientation based on the sign of the quadratic term obtained from the fitting. The maximum height of each measuring point relative to the fitted surface is taken as the deflection amplitude. .
4. The steel structure metal sheet cutting and stacking device according to claim 1, characterized in that, The morphology calculation unit (7) further determines whether the calibration morphology is a plate surface bend or an end curl based on the height sequence, and when the calibration morphology is an end curl, the difference in the height above the surface of the measuring points on both ends is used as the deflection amplitude. .
5. The steel structure metal sheet cutting and stacking device according to claim 1, characterized in that, The transition conveyor (14) is also equipped with a buffer station, where several parts (12) awaiting coding are temporarily stored; the orientation decision unit (10) selects the deflection amplitude value from the several parts (12) awaiting coding. The deflection amplitude of the previously described part (12) The one with the smallest difference is selected as the currently described part (12).
6. The steel structure metal sheet cutting and stacking device according to claim 1, characterized in that, The orientation decision unit (10) calculates the cumulative gap using the following formula. and in the accumulated gaps When the value is greater than a set threshold, the placement orientation factor of the current part (12) is changed. The opposite value is simultaneously indicated by a warning signal issued by the control cabinet (11): accumulated gap. .
7. The steel structure metal sheet cutting and stacking device according to claim 1, characterized in that, The orientation decision unit (10) also reads the nesting data of the previous hot cutting process, predicts the order in which each part (12) arrives at the transition conveyor (14) based on the nesting data, and pre-arranges the placement orientation factor of each part (12) accordingly. sequence.
8. The steel structure metal sheet cutting and stacking device according to claim 1, characterized in that, The flipping mechanism (8) further includes two hanging plates (82) extending downward from both ends of the rectangular plate (81), a rotating frame (83) rotatably mounted between the two hanging plates (82) via a connecting shaft, a rotary motor (84) mounted on the side wall of the hanging plate (82) with its output shaft coaxially connected to the connecting shaft on one side of the rotating frame (83), an electric push rod (85) mounted on the side wall of the rotating frame (83), a U-shaped frame (86) fixedly connected to the telescopic end of the electric push rod (85), a rubber block (87) sealed within the rectangular holes on the upper and lower inner walls of the U-shaped frame (86), and an injection pump (88) mounted on the side wall of the U-shaped frame (86) and communicating with the inner cavity of the U-shaped frame (86); the rubber block (87) in The oil pressure injected by the injection pump (88) extends and contracts along the rectangular hole, and adaptively fits the upper and lower surfaces of the part (12) to clamp the part (12) in the U-shaped frame (86); the rotating frame (83) rotates 180 degrees around a horizontal axis parallel to the feeding direction of the part (12). During the rotation, the weight of the part (12) and the rotational inertial load are borne by the U-shaped frame (86) and the rubber block (87). The clamping force of the rubber block (87) on the part (12) is not less than twice the weight of the part (12); the electric suction cup (22) of the gripping head (2) has been released from adsorption and withdrawn outside the rotation envelope during the rotation of the rotating frame (83).
9. The steel structure metal sheet cutting and stacking device according to claim 1, characterized in that, When the out-of-plane contour detection unit (6) fails, the orientation decision unit (10) will change the placement orientation factor. If the value is fixed at +1, that is, each of the parts (12) is not flipped over, the control cabinet (11) will issue an alarm signal and record the number of layers that have been placed.
10. A method for cutting and stacking steel structure metal plates, characterized in that, The steel structure metal sheet cutting and stacking device according to any one of claims 1 to 9 includes the following steps: S1, the accumulated gap in the gap register (9) Zero out and set the reference transfer position above the palletizing station (5); S2. The part (12) output from the previous hot cutting process arrives at the transition conveyor (14), and the out-of-plane contour detection unit (6) measures the height sequence of the upper surface of the part (12) along several scanning lines. S3, the morphology calculation unit (7) calculates the calibration morphology and the deflection amplitude of the part (12) based on the height sequence. With the calibration orientation ; S4. The orientation decision unit (10) reads the effective orientation of the previously placed part (12) from the gap register (9). With the aforementioned deflection amplitude Press to make the gap The minimum value is used to calculate the placement orientation factor of the current part (12). This makes the effective orientation of the current part (12) the same as the effective orientation of the previous part (12); S5. The gripping head (2) grips the current part (12), and the flipping mechanism (8) operates according to the placement orientation factor. Rotate the part (12) 180 degrees or keep it in its natural delivery position; S6. The lifting drive mechanism (3) and the transfer mechanism (4) drive the gripping head (2) to transfer the part (12) to the palletizing platform (5) and release it. The part (12) falls onto the palletizing platform (5) or the upper surface of the formed stack (13). S7. The orientation decision unit (10) calculates the layer gap of the current layer based on the effective orientation of the current part (12) and the previous part (12) and the deflection amplitude. And add it to the accumulated gap in the gap register (9). ; S8. Repeat steps S2 to S7 until the stack (13) reaches a set number of layers, wherein the effective orientations of two adjacent parts (12) are the same and they are stacked in the same direction, and the gaps between each layer are... The cumulative layer gap is the difference in the deflection amplitude between two adjacent parts (12). It does not accumulate linearly with the set number of layers.
