Full-automatic detection platform with parameter intelligent comparison function

Abstract

The application discloses a full-automatic detection platform with parameter intelligent comparison function, which comprises an electric control cabinet, a marble platform, a positioning assembly, a mobile detection assembly, an infrared scanner, a two-dimensional code scanner and a monitor; a controller scans global reference features of a workpiece through a camera, establishes a real-time pose deviation conversion matrix T and a workpiece real-time coordinate system, then collects local images of each feature to be detected and extracts actual parameters under the coordinate system, finally performs coordinate mapping compensation comparison based on inverse transformation of T, generates a final deviation value and compares the final deviation value with a tolerance to determine whether the workpiece is qualified or not; the intelligent comparison logic of first sensing an actual placement pose of the workpiece and then performing inverse transformation compensation eliminates the misjudgment problem caused by workpiece translation or rotation deviation in traditional detection, supports automatic switching of detection models of two-dimensional codes, integrates a safety light curtain, and realizes full-automatic batch detection with high precision, high flexibility and high safety.

Description

Technical Field

[0001] The present invention relates to the technical field of automated workpiece detection, in particular to a full-automatic detection station with an intelligent parameter comparison function. Background Art

[0002] In the manufacturing industry, for workpieces with specific geometric features, batch detection of the dimensional and positional accuracy of their key features is required. Currently common detection methods include manual sampling inspection using calipers or go-no go gauges, and automatic detection using fixed visual inspection equipment. Fixed visual inspection equipment usually places the workpiece on a high-precision platform, realizes rough positioning of the workpiece by means of mechanical limit baffles, then a pre-programmed camera moves along a fixed path to take photos, and simply compares the measured feature dimensions with preset standard values to judge whether the workpiece is qualified.

[0003] First, manual detection is inefficient, and it is difficult to ensure consistency and repeatability for position detection of multiple notches, and visual fatigue and human errors are prone to occur after long-time work. Second, although the existing fixed visual inspection equipment realizes automation, its comparison logic highly depends on the workpiece being accurately placed at the theoretically designed position. Due to the inherent manufacturing tolerances of the workpiece itself (such as slight bending of the side edge and inclination of the end face), and the fact that the placement force and angle of the operator within the limit baffles are difficult to be completely consistent, there will always be slight translation deviation and rotation deviation between the actual pose of the workpiece during each detection and the theoretical reference. The traditional absolute coordinate comparison method cannot distinguish such placement deviation from the workpiece's own machining deviation, and often misjudges the offset caused by placement as the notch position out of tolerance, resulting in a large number of qualified workpieces being falsely detected as unqualified, or requiring repeated re-placement and re-detection, which seriously affects the detection efficiency and the real statistics of the yield. Third, the notch size specifications of different batches of workpieces may be different, and the existing equipment requires manual re-input of parameters or program replacement when changing batches, which is cumbersome to operate and prone to errors.

[0004] Therefore, in view of the above-mentioned problems, the present invention proposes a full-automatic detection station with an intelligent parameter comparison function. Summary of the Invention

[0005] In order to overcome the problem in the existing detection technology that the comparison result is inaccurate and qualified products are misjudged due to the deviation between the actual placement pose of the workpiece and the theoretical reference, the present invention proposes a full-automatic detection station with an intelligent parameter comparison function, which is used for realizing highly robust intelligent comparison detection of feature parameters by sensing the workpiece pose deviation in real time and performing coordinate mapping compensation.

[0006] The technical solution of the present invention is: a full-automatic detection station with an intelligent parameter comparison function, comprising: The electrical control cabinet is equipped with an industrial-grade PLC or embedded controller with a sampling period of no less than 1ms. The bottom of the electrical control cabinet is fixed to the ground by four height-adjustable M20 support legs with an adjustment range of ±30mm for overall machine leveling. The marble platform is fixed to the upper surface of the electrical control cabinet, and its overall flatness is calibrated to 0.01mm, serving as the absolute physical reference surface. The positioning component includes two first limiting baffles and one second limiting baffle fixed to a marble platform, forming a U-shaped positioning area. The length of the first limiting baffle is 200mm, the length of the second limiting baffle is 300mm, and the mutual perpendicularity tolerance is 0.02mm. The motion detection component includes a horizontal moving slide rail, a horizontal moving slider, a vertical moving cylinder, a camera (5-megapixel global shutter CMOS, 3.45μm pixel size), and a fill light (ring-shaped white LED, adjustable illumination intensity of 2000Lux). The camera and fill light are fixed in relative positions and are jointly mounted on the output end of the vertical moving cylinder.

[0007] Infrared scanners are mounted on the support columns on both sides to form a safety light curtain, with a response time of ≤20ms. A QR code scanner is also fixed to the support columns, with a reading distance of 30-80mm and a resolution of 0.2mm.

[0008] The monitor, electrically connected to the controller, is used to display images and detection data.

[0009] The controller has the following built-in detection logic, and the execution steps are as follows: S1, the workpiece is placed within the U-shaped area with the slotted end facing outwards and the unslotted end abutting the second limiting baffle. The camera moves to the global reference shooting position and captures images including the contact line between the unslotted end of the workpiece and the second limiting baffle, and the gaps between the side edge lines and the first limiting baffle. The edge is then calculated using a sub-pixel edge extraction algorithm (0.1 pixel precision): Translational deviation along the X-axis Δx (unit: mm); Translational deviation along the Y-axis Δy (unit: mm); Rotational deviation about the Z-axis Δθ (in degrees); Establish the real-time pose deviation transformation matrix T: ; At the same time, a real-time coordinate system for the workpiece is established, with the theoretical reference point of the workpiece (center of the grooveless end face) as the origin, the X-axis along the length direction, and the Y-axis along the width direction.

[0010] S2, the controller controls the horizontal slide rail to move at a speed of 20mm / s according to the pre-stored detection path (generated based on theoretical design coordinates), and the vertical cylinder drives the camera to automatically focus at each positioning point (focusing time ≤0.3s).

[0011] Take partial images of each rectangular groove on the workpiece, and collect parameters including the center coordinates of the groove ( , ), slot length Groove width ; All the above parameters are recorded in the workpiece's real-time coordinate system as follows: .

[0012] S3 retrieves the standard parameter model corresponding to the current batch from the QR code scanning results, including the theoretical coordinates of each slot. , Theoretical length Theoretical breadth .

[0013] Apply inverse transformation compensation to each measured coordinate value: ; Calculate the final alignment deviation value: ; ; ; ; S4, preset tolerance thresholds include position deviation ≤ ±0.10mm, length deviation ≤ ±0.15mm, and width deviation ≤ ±0.10mm. If all deviation values ​​of all rectangular slots are within the tolerance range, the workpiece is judged as "qualified"; otherwise, it is judged as "unqualified", and the specific slot position and deviation value that exceeds the tolerance are highlighted on the monitor.

[0014] The beneficial effects of this invention are: 1. This invention first scans the global reference features of the workpiece to establish a real-time pose deviation transformation matrix T and the real-time coordinate system of the workpiece. Then, it performs local acquisition of each feature to be inspected and performs compensation calculation based on the inverse transformation of T in the coordinate mapping comparison. This eliminates the problem of qualified products being misjudged as unqualified due to translational and rotational deviations between the actual placement posture of the workpiece and the theoretical reference in traditional fixed vision inspection equipment. It accurately removes the systematic error introduced by placement from the measurement results, so that even if the workpiece is placed in a rough state with ±2.0mm translation and ±3.0° rotation in the concave positioning area, it can still obtain the detection accuracy comparable to that of precision fixture positioning, thereby greatly reducing the requirements for operator skills and front-end process accuracy.

[0015] 2. This invention sets up a concave positioning area on a marble platform, consisting of two first limiting baffles and one second limiting baffle, and uses a QR code scanner to automatically identify workpiece batches and call up the corresponding standard parameter model. This allows for automatic updates of the detection path and comparison benchmark when switching production for workpieces of different specifications, without the need for manual re-entry of parameters or replacement of the program. Simply scanning the QR code automatically updates the detection path and comparison benchmark. This solves the problems of cumbersome production changeover operations and easy detection failures due to incorrect parameter settings in existing equipment under multi-variety, small-batch production modes, and significantly improves the flexibility and production efficiency of the equipment. Attached Figure Description

[0016] Figure 1 The diagram shown is a schematic representation of the overall three-dimensional structure of the present invention; Figure 2 The diagram shown is a schematic representation of the internal structure of the fully automated testing station of the present invention. Figure 3 The diagram shown is a schematic representation of the structure of the motion detection component of the present invention. Figure 4 The diagram shown is a schematic representation of the positioning component structure of the present invention. Figure 5 The diagram shown is a schematic representation of the structure of the safety component of this invention. Figure 6 The diagram shown illustrates the workflow of this invention.

[0017] Explanation of reference numerals in the attached diagram: 1. Electrical control cabinet; 2. Marble platform; 3. First limit baffle; 4. Support plate; 5. Detection cabinet; 6. Horizontal moving slide rail; 7. Horizontal moving slider; 8. Vertical moving cylinder; 9. Camera; 10. Fill light; 11. Second limit baffle; 12. Support column; 13. Infrared scanner; 14. Heat pipe; 15. QR code scanner. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Please see Figures 1-5 This invention provides an embodiment of a fully automated testing station with intelligent parameter comparison function: This invention includes an electrical control cabinet 1, a marble platform 2, a positioning component, a motion detection component, a safety component, a data input component, and a monitor. The electrical control cabinet 1 houses an industrial-grade controller, which employs a programmable logic controller or an embedded industrial computer. Its sampling period is set to no more than 1 millisecond to ensure real-time processing of signals from the camera 9, infrared scanner 13, QR code scanner 15, and various motion actuators. Four height-adjustable support legs are evenly distributed at the bottom of the electrical control cabinet 1. By rotating the adjusting nuts on the support legs, the height of each support leg can be independently adjusted within a range of ±30 mm, allowing operators to level the entire electrical control cabinet 1 on uneven workshop floors. A marble platform 2 is fixedly connected to the upper surface of the electrical control cabinet 1 by multiple sets of high-strength bolts. This marble platform 2 undergoes precision grinding and aging treatment, ensuring its overall flatness is no less than 0.01 mm within any 1000 mm × 1000 mm area. The marble platform 2 not only serves as the bearing surface for the workpiece to be inspected, but also provides an absolutely reliable physical reference surface for the moving module of camera 9. Since the moving slide rail of camera 9 is indirectly fixed to the edge of marble platform 2 through support plate 4 and inspection cabinet 5, the flatness of marble platform 2 directly determines the straightness of the moving path of camera 9 and the reference consistency of workpiece surface feature measurement.

[0020] Please see Figure 2 and Figure 4A positioning assembly is fixedly installed on the upper surface of the marble platform 2. This positioning assembly specifically includes two first limiting baffles 3 and one second limiting baffle 11. The two first limiting baffles 3 are parallel to each other and spaced a predetermined distance apart. The second limiting baffle 11 is perpendicular to the first limiting baffles 3. The same end of each of the two first limiting baffles 3 is perpendicularly attached to both end faces of the second limiting baffle 11 and fixed with screws. The three baffles together form a U-shaped positioning area. To ensure positioning accuracy, the perpendicularity tolerance between the first limiting baffles 3 and the second limiting baffle 11 is controlled within 0.02 mm. The aluminum heat pipe 14 workpiece to be inspected has a specific shape: its top view is rectangular, its side view is I-shaped, and multiple equidistant or unequally spaced rectangular grooves are cut along one side. Before inspection, the operator needs to place the workpiece in the U-shaped positioning area with the end with the rectangular groove facing the operating side (i.e., the side facing the person standing), and the flat end face without the groove tightly against the inner side of the second limiting baffle 11. At this time, there will be a small gap between the two sides of the workpiece and the first limiting baffle 3. This gap is caused by the manufacturing tolerance of the workpiece itself and slight differences in the operator's placement method. The positioning component only plays a coarse positioning role. It does not need to force the workpiece to be fixed in an absolutely ideal position, but only to restrict the workpiece within a known and limited deviation range (within ±2 mm along the X and Y axes, and within ±3 degrees around the Z axis). This greatly reduces the requirements for the accuracy of the workpiece's previous processing and the operator's placement skills.

[0021] Please see Figure 2 and Figure 3A vertically extending support plate 4 is fixedly connected to the upper edge of the marble platform 2. A testing cabinet 5, roughly inverted U-shaped, is fixedly connected to the upper end of the support plate 4, spanning across the marble platform 2. A horizontal moving slide rail 6 is fixedly installed on the inner wall of the testing cabinet 5. The length direction of the horizontal moving slide rail 6 is parallel to the length direction of the marble platform 2. The slide rail uses a high-precision ball linear guide pair and is equipped with a closed-loop stepper motor drive system. An encoder provides real-time position feedback, enabling the slide rail to achieve a repeatability accuracy of ±0.01mm. A horizontal moving slider 7 is slidably connected to the horizontal moving slide rail 6. This slider is driven by a stepper motor via a synchronous belt or lead screw, allowing it to reciprocate linearly along the slide rail in the horizontal direction. A vertical moving cylinder 8 is fixedly installed on the horizontal moving slider 7. The cylinder body of the vertical moving cylinder 8 is vertically downward or vertically upward, and its output end is connected to a piston rod. A camera 9 and a supplementary light 10 are simultaneously fixedly connected to the end of the piston rod. Camera 9 uses a 5-megapixel global shutter CMOS image sensor with a pixel size of 3.45μm. Paired with a fixed-focus industrial lens, it can achieve a physical resolution of 0.0145mm per pixel within a 40mm × 30mm field of view. The fill light 10 is a ring-shaped white LED light source with stepless adjustment of its intensity from 0-500Lux. The relative positions of camera 9 and fill light 10 are completely fixed, both being fastened to the same transition plate. This transition plate is then connected to the piston rod end of vertical movement cylinder 8. Therefore, when vertical movement cylinder 8 is activated, camera 9 and fill light 10 move up and down synchronously as a whole, always maintaining a fixed angle between the fill light direction and the lens optical axis. This integrated lifting design ensures consistent illumination conditions on the inspected feature surface at different working heights, thereby improving the repeatability of image acquisition.

[0022] Please see Figure 5Two support columns 12 are vertically fixed on both sides of the front end of the testing platform (i.e., the operating side). The bottom of each support column 12 is connected to the side wall of the electrical control cabinet 1 or the side reinforcement plate of the marble platform 2 by bolts. An infrared scanner 13 is fixedly installed at the middle height of each support column 12. The two infrared scanners 13 are arranged opposite each other, one as the transmitter and the other as the receiver, forming a horizontal safety light curtain between them. This safety light curtain is located at a certain height above the marble platform 2, and the plane of the light curtain intersects the marble platform 2 vertically. That is, if any part of the human body crosses the operating front edge of the testing platform and enters the testing area above the marble platform, the safety light curtain will be blocked. The controller is electrically connected to the infrared scanner 13. When the safety light curtain is detected to be blocked, the testing platform is immediately prohibited from starting a new testing action, or if the testing is in progress, the current testing process is immediately interrupted and all moving parts (horizontal moving slide rail 6, vertical moving cylinder 8, etc.) are controlled to return to the safe position.

[0023] A QR code scanner 15 is fixedly installed at the upper end of each support column 12. This QR code scanner 15 can quickly read QR code labels printed or affixed to the surface of the aluminum heat pipe 14 workpiece or on the turnover pallet. The QR code scanner 15 is electrically connected to the controller, and its scanning results are transmitted to the controller via a standard serial communication protocol. During operation, before the formal start of the inspection program, the QR code scanner 15 is first controlled to read the batch code corresponding to the workpiece currently placed on the inspection table. Then, based on the batch code, the standard parameter model matching the batch is automatically retrieved from the storage unit inside the controller. The standard parameter model includes at least the theoretical number of rectangular slots on the workpiece in the batch, the theoretical position coordinates (X and Y coordinates) of each rectangular slot relative to the theoretical design datum of the workpiece, and the theoretical length and width of each rectangular slot. Different batches of workpieces may have different numbers of slots, slot spacing, or slot size specifications. By automatically identifying and retrieving the corresponding model through QR codes, the potential for errors from manual parameter input is completely avoided, and rapid switching between different batches is also achieved without stopping the machine for recalibration or program modification.

[0024] The monitor is an industrial touchscreen, installed on the front side of the inspection cabinet 5 or fixed near the inspection table via a cantilever bracket for easy operation. Electrically connected to the controller, the monitor displays not only real-time images and inspection data transmitted from the camera 9, but also serves as a human-machine interface for receiving operator commands (starting inspection, viewing historical records, modifying tolerance parameters, etc.). The monitor's display interface is divided into three main areas: an image display area, a deviation data area, and a status indicator area. The image display area shows a real-time partial image of the workpiece currently captured by the camera 9, and overlays the theoretical contour read from the standard parameter model with a highlighted color, allowing the operator to intuitively compare the differences between the actual features and the theoretical positions. The deviation data area displays the measured dimensions, theoretical dimensions, and final comparison deviation values ​​of each rectangular groove in a list format, row by row. For deviations exceeding the preset tolerance range, the system automatically marks them with a prominent background or font color. The status indicator area displays the current inspection progress, equipment operating status, safety light curtain status, and the final pass / fail judgment result.

[0025] Please see Figure 6 Furthermore, the detection logic of the intelligent parameter comparison function inside the controller is explained below: S1, after the operator roughly places the workpiece in the U-shaped positioning area on the marble platform 2 (with the grooveless end face abutting against the second limiting baffle 11 and the two side edges located between the two first limiting baffles 3), and presses the "start" button on the monitor, the controller first executes the pose perception program. The controller controls the horizontal moving slide rail 6 to move the camera 9 to the pre-calibrated global reference shooting position, which is usually located about 40mm above the grooveless end face of the workpiece. Then, the controller lowers the camera 9 to the preset imaging height through the vertical moving cylinder 8 and automatically triggers the supplementary light 10 to illuminate with an illuminance of 2,000 lux, and then takes an image containing the key reference features of the workpiece. These reference features include the mating line between the grooveless end face of the workpiece and the second limiting baffle 11 (actually a distinct edge formed due to material or color differences) and the gaps between the two side edge lines of the workpiece and the inner surfaces of the two first limiting baffles 3 (manifested as two dark or bright bands). The image processing module inside the controller uses a sub-pixel edge extraction algorithm, which can locate the precise position of the above features in the image, with a positioning accuracy of 0.1 pixels. By calculating the position of the bonding line and the width of the gap on both sides, the controller can calculate the minute deviations of the workpiece in three degrees of freedom relative to the theoretical design datum when it is actually placed: The translational deviation Δx (in millimeters) along the X-axis, the translational deviation Δy (in millimeters) along the Y-axis, and the rotational deviation Δθ (in degrees) about the Z-axis perpendicular to the marble platform 2 are calculated. If the entire bonding line is offset to the left by 0.1 millimeters relative to its theoretical position, then Δx is -0.1 millimeters; if the gap on the left is significantly larger than the gap on the right, it indicates that the workpiece has undergone clockwise or counterclockwise deflection, thus the value of Δθ is calculated. The controller then establishes a three-dimensional homogeneous coordinate transformation matrix T based on these three deviations, the expression of which is: .

[0026] Meanwhile, the controller also establishes a real-time coordinate system for the workpiece that dynamically changes with the actual position and posture of the workpiece. This coordinate system takes the theoretical design reference point of the workpiece as the origin, with the X-axis along the length of the workpiece (from the ungrooved end to the grooved end) and the Y-axis along the width of the workpiece (from one edge to the other edge).

[0027] S2, after completing pose perception and establishing the transformation matrix T and the workpiece's real-time coordinate system, the controller enters the batch inspection stage. The controller starts controlling the horizontal moving slide rail 6 based on the detection path information pre-stored in the internal storage unit. This detection path is generated based on the workpiece's theoretical design coordinates and includes a series of image capture points, each corresponding to the theoretical center position of a rectangular groove. The horizontal moving slide rail 6 drives the camera 9 to move sequentially to each theoretical image capture point at a constant speed of 20 millimeters per second. After reaching each point, the vertical moving cylinder 8 causes the camera 9 to move slightly up and down to achieve autofocus, then triggers the camera 9 to capture a magnified local image containing the complete outline of a single rectangular groove. Since the camera 9's field of view is only 40mm × 30mm, while the entire workpiece may be 3-5m long, segmented shooting is necessary. In each local image, the controller again calls the sub-pixel edge extraction algorithm to accurately calculate multiple geometric parameters of the rectangular groove, the pixel coordinates of the groove center in the image, and then transforms them to the actual coordinate values ​​in the workpiece's real-time coordinate system (denoted as...). , The pixel distance between the two endpoints of the slot along its length is converted into the actual length. The pixel distance between the two edges of the notch in the width direction, converted into the actual width. All the above parameters are recorded as a vector in the real-time workpiece coordinate system established in step S1. It is important to note that because the workpiece's real-time coordinate system is established based on the workpiece's actual pose, even if the workpiece as a whole undergoes translation or rotation, the actual coordinate values ​​of each slot will remain unchanged. , It is also measured in a coordinate system that "moves with the workpiece", which eliminates the influence of placement deviation on the original measurement value, making subsequent comparisons more direct.

[0028] S3, the controller first retrieves the standard parameter model corresponding to the batch from the storage unit based on the QR code batch information retrieved in step S1. This model contains the theoretical design values ​​for each rectangular slot: theoretical center coordinates ( , (Under theoretical design standards), theoretical length Theoretical width Then, the controller takes the actual coordinate values ​​of each rectangular slot measured in step S2. Inverse transform compensation is applied for calculation. Specifically, the following formula is used: ,in It is the inverse matrix of the real-time pose deviation transformation matrix T established in step S1.

[0029] It should be noted that, These are coordinate values ​​measured in the real-time coordinate system of a skewed workpiece, while The function is to rotate and translate this coordinate value in the reverse direction, mapping it back to the coordinate value under the theoretical design reference. After the inverse transformation, the compensated coordinate value is obtained. For length and width These two dimensional parameters, being direction-dependent scalars, are affected by rotational deviations. The measured length and width are actually projections of the workpiece's actual dimensions onto the imaging plane of camera 9. Therefore, the inverse transformation also requires corresponding trigonometric function corrections based on the rotation angle Δθ. Finally, the controller calculates the compensated deviation values ​​for each rectangular groove: Position X-direction deviation ; Position Y-direction deviation ; Length deviation ; Width deviation .

[0030] These deviation values ​​are the true reflections of the workpiece's machining accuracy after eliminating workpiece placement errors.

[0031] S4, the controller judges the pass / fail status of the above deviation values ​​according to the preset tolerance range. In a typical inspection application, the preset tolerance threshold is set as: positional deviation (including...) and The length deviation shall not exceed ±0.01mm. Width deviation not exceeding ±0.15mm The tolerance should not exceed ±0.10mm. These thresholds can be modified via the monitor according to the quality requirements of different batches of workpieces. If every type of deviation value of all rectangular grooves on the current workpiece is within the corresponding tolerance range, the controller determines the workpiece as "qualified"; if any deviation of any rectangular groove exceeds the tolerance range, it is determined as "unqualified". The determination result, along with detailed deviation data, is sent to the monitor. The monitor displays the measured size, theoretical size, and final comparison deviation value of each groove in a list format in the deviation data area, and displays the final conclusion in large font in green "qualified" or red "unqualified" in the status indicator area. For unqualified workpieces, the system will also highlight the corresponding out-of-tolerance location in the image display area with a flashing red box, so that the operator can quickly locate the problem characteristics, and the entire inspection process for a single workpiece is completed.

[0032] To further verify the detection accuracy and reliability of this invention, numerous repeatable and comparative tests were conducted in a real manufacturing environment. The test subjects were a batch of aluminum heat pipes 14, each with six equidistant rectangular slots machined on it. The nominal length of each slot was 12.00 mm, the nominal width was 3.50 mm, and the center distance between adjacent slots was 25.00 mm. Thirty workpieces were selected, each placed by the same operator with different offsets and rotations within a U-shaped positioning area. The detection table of this invention was then used for inspection, with a high-precision coordinate measuring machine serving as a standard reference. The test results showed that the standard deviation of the slot position deviation (after compensation) measured by the detection table of this invention for the same workpiece under different placement postures was less than 0.02 mm, the standard deviation of the length deviation was less than 0.03 mm, and the standard deviation of the width deviation was less than 0.02 mm, demonstrating excellent consistency.

[0033] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A fully automatic testing station with intelligent parameter comparison function, characterized in that, include: An electrical control cabinet (1) is provided with a controller inside the electrical control cabinet (1); A marble platform (2), which is fixed to the upper end of the electrical control cabinet (1), has a flatness of 0.01mm and is used to support the workpiece to be inspected; The positioning component is fixed on the marble platform (2) and is used to limit the workpiece so that the workpiece has a unique and definite coarse positioning placement state. The moving detection component includes a horizontal moving slide rail (6), a vertical moving cylinder (8), and a camera (9) installed at the output end of the vertical moving cylinder (8), which is used to acquire images of the workpiece during the moving process; The monitor, electrically connected to the controller, is used to display images and detection data; The controller is used to perform the following intelligent comparison steps: S1, scan the pre-set global reference features of the workpiece through the camera (9), calculate the real-time pose deviation of the workpiece relative to the theoretical design reference in the actual placement state according to the scanning results, establish the real-time pose deviation transformation matrix, and establish the real-time coordinate system of the workpiece that dynamically changes with the actual pose of the workpiece. S2, control the camera (9) to move along the preset path, collect local images of all features to be inspected on the workpiece, and extract the actual geometric parameters and position parameters of each feature to be inspected in the real-time coordinate system of the workpiece; S3, compare the extracted actual parameters with the standard parameter model pre-stored in the controller by coordinate mapping. This coordinate mapping comparison is based on the real-time pose deviation to compensate the actual parameters and generate the final comparison deviation value after eliminating the placement error. S4 outputs the final comparison deviation value to the monitor and compares it with the preset tolerance range to generate a pass / fail judgment signal.

2. The fully automatic detection station with intelligent parameter comparison function according to claim 1, characterized in that: The positioning component includes two first limiting baffles (3) and a second limiting baffle (11) fixed to the upper surface of the marble platform (2); one end of the two first limiting baffles (3) is perpendicularly attached to both ends of the second limiting baffle (11) to form a concave positioning area; the workpiece to be tested is placed in the concave positioning area with its grooveless end face attached to the second limiting baffle (11) and its end with a rectangular groove facing the operating side.

3. The fully automatic detection station with intelligent parameter comparison function according to claim 1, characterized in that: The moving detection component also includes a supplementary light (10), which is fixed in relative position to the camera (9), and both are fixed to the output end of the vertical moving cylinder (8); the horizontal moving slide rail (6) is fixed to the inner wall of the detection cabinet (5), the detection cabinet (5) is fixed to the upper edge of the marble platform (2) through the support plate (4), and the vertical moving cylinder (8) is slidably connected to the horizontal moving slide rail (6) through the horizontal moving slider (7).

4. The fully automatic detection station with intelligent parameter comparison function according to claim 1, characterized in that: The testing station also includes support columns (12) set on both sides of the front end of the testing station. Each support column (12) is fixed with an infrared scanner (13). The two infrared scanners (13) shoot at each other to form a safety light curtain located on the operating side of the testing station. The controller is electrically connected to the infrared scanner (13). When any part of the human body is detected to cross the safety light curtain and enter the area that intersects with the marble platform (2) in the vertical direction, the detection station is prohibited from starting or the detection action being performed is stopped immediately.

5. The fully automatic detection station with intelligent parameter comparison function according to claim 1, characterized in that: The testing station also includes a QR code scanner fixed to the support column (12), which is electrically connected to the controller; Before starting the inspection, the batch code on the workpiece is scanned with a QR code scanner, and the standard parameter model corresponding to the batch is retrieved from the internal storage unit based on the scanning result. The standard parameter model includes at least the theoretical number of rectangular grooves, the theoretical position coordinates of each rectangular groove, the theoretical length and the theoretical width.

6. The fully automatic detection station with intelligent parameter comparison function according to claim 1, characterized in that, The "workpiece global reference features" mentioned in step S1 include: the fitting line features between the grooveless end face of the workpiece and the second limiting baffle (11), and the gap features between the edge lines on both sides of the workpiece and the two first limiting baffles (3); the real-time pose deviation includes translation deviation along the X-axis, translation deviation along the Y-axis, and angular deviation around the Z-axis.

7. The fully automatic detection station with intelligent parameter comparison function according to claim 1, characterized in that, The coordinate mapping comparison in step S3 specifically involves: extracting the actual coordinate values ​​from the workpiece's real-time coordinate system. Through formula Perform inverse transformation calculation, where T is the real-time pose deviation transformation matrix established in step S1, to obtain the compensated coordinate values ​​under the theoretical design reference after eliminating placement errors. Then The final comparison deviation value is obtained by calculating the difference between the theoretical coordinate values ​​in the standard parameter model and the actual coordinate values.

8. The fully automatic detection station with intelligent parameter comparison function according to claim 1, characterized in that: The monitor is a touch-screen industrial monitor, whose display interface includes an image display area, a deviation data area, and a status indicator area. The image display area displays a partial image of the workpiece captured by the camera in real time, and draws the theoretical outline corresponding to the standard parameter model on the image with a highlight color. The deviation data area displays the measured size, theoretical size, and final comparison deviation value of each rectangular groove in a list format.

9. A fully automatic detection station with intelligent parameter comparison function according to claim 1, characterized in that: The electrical control cabinet (1) is fixed to the ground by multiple height-adjustable support legs at the bottom; the flatness of the marble platform (2) is calibrated to 0.01mm within the overall detection area, providing a unified physical reference surface for the moving shooting of the camera (9) and the flatness detection of the workpiece.

10. A fully automatic detection station with intelligent parameter comparison function according to any one of claims 1-9, characterized in that: The workpiece to be inspected is an aluminum heat pipe, which has a rectangular shape when viewed from above and an I-shaped shape when viewed from the side. The feature to be inspected is a plurality of rectangular grooves formed by cutting one side of the workpiece. During inspection, the workpiece is placed with the end with the rectangular groove facing the operating side and the end without the groove attached to the second limiting baffle (11). The controller controls the camera (9) to perform intelligent parameter comparison of the position, groove length and groove width of each rectangular groove in sequence.

Images