A robotic ultrasonic thickness measurement method for thick-walled pressure-bearing equipment

Abstract

This invention discloses a robotic ultrasonic thickness measurement method for thick-walled pressure-bearing equipment. It utilizes a robot's end-effector 3D vision sensor to acquire point clouds of the equipment's outer surface. Through registration and fusion, target segmentation, and surface modeling, a sequence of thickness measurement points is generated to determine the area to be measured. The local surface normal of each measurement point is calculated, and the probe target pose is constructed. An angle threshold is dynamically adjusted based on local curvature to align the probe's incident direction with the normal. The robot's switching trajectory and normal approach trajectory are planned, and the end-effector is controlled to move to the measurement point. The actual normal contact force is acquired, and the target normal contact force and force control parameters are adaptively adjusted along the normal direction, making minor adjustments. When the actual normal contact force meets the stability criterion, ultrasonic thickness measurement is triggered. Echo signals are acquired to calculate the thickness value, and the validity of the measurement results is determined and recorded. This method improves the accuracy of thickness measurement point positioning, probe normal alignment capability, and contact consistency, thereby enhancing the accuracy and stability of the thickness measurement results.

Description

Technical Field

[0001] This invention relates to the technical fields of ultrasonic nondestructive testing, industrial robot control and three-dimensional visual perception, and in particular to a robotic ultrasonic thickness measurement method for ultrasonic thickness measurement of thick-walled pressure-bearing equipment. Background Technology

[0002] Thick-walled pressure-bearing equipment is widely used in gas storage and transportation, energy and chemical industries, special equipment, and high-pressure vessels. Typical examples include hot-spinning gas cylinders, pressure pipelines, end caps, and other thick-walled pressure-bearing components. During service, this type of equipment typically withstands high internal pressure, alternating loads, or complex operating conditions. Its wall thickness distribution directly affects the structural load-bearing capacity, service safety, and quality consistency. Therefore, it is necessary to conduct wall thickness testing on critical areas of thick-walled pressure-bearing equipment to obtain its thickness distribution information and identify risk areas such as localized thinning and abnormal wall thickness.

[0003] Ultrasonic thickness measurement offers advantages such as being non-destructive, rapid, and applicable to metal wall thickness testing, making it a commonly used technique for inspecting the wall thickness of thick-walled pressure equipment. Current ultrasonic thickness measurement methods often involve manual, point-by-point measurement using a handheld ultrasonic probe. During operation, a coupling agent is typically applied to the surface being measured, and the probe must maintain good contact with the workpiece surface to obtain effective echo signals and calculate the wall thickness. While this manual measurement method has relatively simple equipment, it suffers from low measurement efficiency, poor repeatability, and significant susceptibility to human experience in obtaining results. This is particularly problematic in curved transition areas, areas with significant local curvature changes, or areas with limited spatial accessibility, where the stability and consistency of manual measurements are difficult to guarantee.

[0004] In automated ultrasonic thickness measurement using robots, relying on liquid coupling agents for acoustic coupling not only increases the complexity of the agent application and replenishment process but also may affect the consistency of measurement conditions between different measurement points due to uneven agent distribution. To improve automation, existing technologies employ industrial robots equipped with ultrasonic thickness probes for automated thickness measurement. These methods typically plan the robot's motion path based on the workpiece's theoretical model, preset measurement points, or a manually taught trajectory, allowing the probe to sequentially reach each measurement position and complete the measurement. However, thick-walled pressure-bearing equipment is susceptible to factors such as forming errors, welding deformation, clamping deviations, and positioning errors during actual manufacturing, assembly, clamping, or on-site installation, often resulting in discrepancies between the actual surface morphology and the theoretical model. When the robot still performs measurements according to a preset model or fixed trajectory, inaccurate thickness point positioning and mismatches between the probe's posture and the actual surface normal can easily occur, affecting the accuracy of the thickness measurement results.

[0005] Furthermore, the outer surface of thick-walled pressure equipment often includes different structural regions such as cylinders, heads, shoulders, transition sections, or curved sections, and their local curvature characteristics may vary significantly. During ultrasonic thickness measurement, if the effective incident direction of the probe cannot be well aligned with the local surface normal at the measurement point, the ultrasonic beam will enter the workpiece surface in a way that deviates from the normal, easily leading to weakened echo signals, increased measurement errors, and consequently affecting the reliability of the wall thickness measurement results. Especially in areas with large curvature, areas with continuously changing curvature, or complex curved surfaces, relying solely on a preset pose or a rough trajectory is insufficient to ensure accurate tracking of the probe's normal at the measurement point.

[0006] Meanwhile, ultrasonic thickness measurement results are not only related to the probe's orientation but also closely related to the contact state between the probe and the workpiece surface. When a robot performs thickness measurement, inconsistent normal contact forces applied by the probe at different measurement points, or fluctuations in force, insufficient contact, and changes in coupling state during the contact process, can lead to increased dispersion in the measurement results, affecting the comparability and measurement stability between measurement points. Therefore, determining reasonable thickness measurement points for the actual outer surface of thick-walled pressure-bearing equipment, accurately aligning the ultrasonic probe with the local normal at each measurement point, and maintaining consistent and stable contact conditions at each measurement point has become a key problem to be solved in robotic ultrasonic thickness measurement. Summary of the Invention

[0007] The purpose of this invention is to solve the problems in the prior art and propose a robotic ultrasonic thickness measurement method for ultrasonic thickness measurement of thick-walled pressure-bearing equipment. This robotic ultrasonic thickness measurement method determines the thickness measurement points on the actual outer surface of the thick-walled pressure-bearing equipment, aligns the ultrasonic thickness measurement probe in the normal direction, maintains consistent contact conditions at each thickness measurement point, and completes ultrasonic thickness measurement and result judgment under stable contact conditions.

[0008] To achieve the above objectives, this invention proposes a robotic ultrasonic thickness measurement method for ultrasonic thickness measurement of thick-walled pressure-bearing equipment, including S1, using a three-dimensional vision sensor installed at the end of an industrial robot to acquire point cloud data of the outer surface of the thick-walled pressure-bearing equipment under multiple observation poses, performing coordinate unification, registration and fusion, and target region segmentation on the point cloud data to establish a three-dimensional surface model of the actual outer surface of the equipment. S2, Based on the three-dimensional surface model, determine the thickness area to be measured of the thick-walled pressure-bearing equipment, and generate multiple thickness measurement points in the thickness area to be measured according to the preset thickness measurement rules to form a thickness measurement point sequence; S3, for each thickness measurement point, based on the three-dimensional surface model or the neighboring point of the thickness measurement point, the local surface normal at the thickness measurement point is calculated. The target thickness measurement pose of the ultrasonic thickness measurement probe is constructed according to the thickness measurement point position and the corresponding local surface normal, so that the effective incident direction of the ultrasonic thickness measurement probe is consistent with the local surface normal or the angle between the two is not greater than a preset angle threshold. The preset angle threshold is a fixed angle threshold or a dynamic angle threshold determined based on the local curvature index of the thickness measurement point. S4. Based on the thickness measurement point sequence and the target thickness measurement pose corresponding to each thickness measurement point, plan the switching trajectory of the industrial robot between adjacent thickness measurement points, and generate the approach trajectory along the local surface normal at the current thickness measurement point, so that the end of the industrial robot moves to the vicinity of each thickness measurement point in sequence and approaches the surface of the thick-walled pressure-bearing equipment along the local surface normal. S5, when the industrial robot moves to the vicinity of the current thickness measurement point, it controls the ultrasonic thickness measurement probe to establish contact with the surface of the thick-walled pressure-bearing equipment, obtains the actual normal contact force at the current thickness measurement point, and adaptively adjusts the target normal contact force and force control parameters based on the deviation between the actual normal contact force and the target normal contact force, combined with the local curvature index of the current thickness measurement point. It controls the end effector of the industrial robot to perform feedback adjustment along the local surface normal of the current thickness measurement point, so that the ultrasonic thickness measurement probe maintains a preset and consistent contact state at the current thickness measurement point. S6, determine the stability of the adjusted actual normal contact force. When the actual normal contact force meets the preset stability criterion and the ultrasonic thickness probe posture meets the angle threshold condition, trigger the ultrasonic thickness measurement operation at the current thickness measurement point, collect the ultrasonic echo signal and obtain the echo flight time, and calculate the thickness value of the current thickness measurement point based on the echo flight time. S7. Determine the validity of the thickness value of the current thickness measurement point. If the preset measurement validity conditions are met, record the thickness value of the current thickness measurement point and the corresponding thickness measurement point position information, and move to the next thickness measurement point. If the preset measurement validity conditions are not met, re-execute the contact adjustment and ultrasonic thickness measurement operation for the current thickness measurement point.

[0009] Preferably, the point cloud coordinate unification in step S1 includes: based on the pose information of the industrial robot end effector and the calibration relationship between the 3D vision sensor and the industrial robot end effector, uniformly converting the local point clouds acquired under each observation pose to the industrial robot base coordinate system; any first Point cloud points under observation pose The coordinates after transformation to the industrial robot base coordinate system satisfy: ;in, For the first The pose transformation matrix of the industrial robot's end effector coordinate system relative to the base coordinate system under each observed pose. This is the calibration transformation matrix of the 3D vision sensor coordinate system relative to the industrial robot end effector coordinate system.

[0010] Preferably, the thickness region to be measured in step S2 includes curved surface regions with different characteristics, and different thickness measurement point spacing, number of thickness measurement points, and arrangement of thickness measurement points are set according to the curvature characteristics, detection requirements, or risk levels of different regions to generate a thickness measurement point sequence; The coordinates of each thickness measurement point are represented as follows: ;in, For the first The spatial coordinates of each thickness measurement point in the industrial robot's base coordinate system, where T represents the transpose of the matrix, and the coordinates are represented by column vectors.

[0011] Preferably, the local surface normal of the thickness measurement point in step S3 is estimated through the local geometric relationship of the point cloud in the neighborhood of the thickness measurement point; let the first... The set of points in the neighborhood of each thickness measurement point is: The neighborhood center point is Then its covariance matrix satisfies: The eigenvector corresponding to the smallest eigenvalue of the covariance matrix is ​​taken as the first eigenvalue. Local surface normal at each thickness measurement point , The covariance matrix corresponding to the neighborhood point set of the i-th thickness measurement point.

[0012] Preferably, in step S3, according to the first Each thickness measurement point location and corresponding local surface normal The target thickness measurement pose of the ultrasonic thickness gauge is constructed, and the direction of the local surface normal is uniformly corrected so that the local surface normal at each thickness measurement point points towards a preset approach direction; the effective incident direction of the ultrasonic thickness gauge is determined. With the local surface normal satisfy: ; ;in The angle between the effective incident direction of the ultrasonic thickness gauge probe and the local surface normal. This is a preset angle threshold.

[0013] Preferably, in step S3, according to the first Fit a local surface to the point cloud of a thickness measurement point and calculate the principal curvature at that thickness measurement point. and Using local curvature index Characterizing the curvature features at the thickness measurement point, the local curvature index satisfies The angle threshold between the effective incident direction of the ultrasonic thickness gauge and the local surface normal is related to the local curvature index. Related dynamic angle thresholds .

[0014] Preferably, in step S4, a normal proximity point is set outside each thickness measurement point. Normal proximity point corresponding to each thickness measurement point Represented as: ;in, For the first The coordinates of the normal approximation point corresponding to each thickness measurement point The local surface normal of the thickness measurement point. To determine the preset normal approach distance, the industrial robot's end effector first moves to the normal approach point, and then moves along the first normal approach distance. The local surface normal of a thickness measurement point is close to the surface of a thick-walled pressure-bearing device.

[0015] Preferably, the actual normal contact force in step S5 is acquired by a force sensor installed at the end of the industrial robot or estimated by the joint drive torque of the industrial robot; the industrial robot maintains the target thickness measurement pose in the tangential position direction and the probe posture direction, and performs pose fine-tuning based on normal contact force feedback in the normal direction of the local surface at the thickness measurement point, constituting force-position hybrid control; let the target normal contact force be... , No. The normal contact force error for each control cycle is: ;in, For the first Actual normal contact force for each control cycle.

[0016] Preferably, the normal displacement correction in step S5 is controlled by incremental proportional-integral control, satisfying the following: The normal correction position command of the industrial robot end effector at the current thickness measurement point satisfies: ;in, For the first Normal displacement correction per control cycle For proportional gain, For integral gain, To control the cycle, This is the current thickness measurement point location. The local surface normal of the current thickness measurement point is set; and a limiting constraint and a stable dead zone are set for the normal displacement correction amount.

[0017] Preferably, the proportional gain and integral gain in step S5 are related to the local curvature index of the current thickness measurement point. The relevant dynamic control parameters are denoted as follows: and The and Based on local curvature index Adaptive adjustment is used to improve the stability of normal contact force adjustment in different curvature regions.

[0018] Preferably, the stability criterion in S6 includes: And continue to satisfy: ;in, This is the normal contact force error threshold. For the current stable duration, A preset stable duration threshold is defined; or, the stability criterion satisfies: ; and continuously meet the preset stable time threshold; when the stability criterion is met and the ultrasonic thickness measurement probe posture meets the requirements. When this occurs, the ultrasonic thickness measurement operation at the current thickness measurement point is triggered.

[0019] Preferably, in step S6, after the stable contact criterion is met, the ultrasonic echo signal of the current thickness measurement point is acquired, and the effective echo round-trip flight time is obtained. Let the ultrasonic propagation velocity in the material of the thick-walled pressure-bearing equipment be... The thickness value of the current thickness measurement point satisfy: Wherein, the ultrasonic propagation speed The results are obtained through standard test block calibration or pre-set according to the material category of thick-walled pressure equipment.

[0020] Preferably, in step S7, the same thickness measurement point is continuously acquired. Secondary thickness value When the following conditions are met: The thickness value of the current thickness measurement point is determined to be valid, where, This is the thickness deviation threshold.

[0021] Preferably, the measurement validity condition further includes that the ultrasonic echo quality parameter meets a preset threshold condition, wherein the ultrasonic echo quality parameter includes at least one of echo amplitude, signal-to-noise ratio, or echo correlation parameter.

[0022] Preferably, the ultrasonic thickness measuring probe is a dry-coupled ultrasonic thickness measuring probe with a flexible membrane layer at the front end. The flexible membrane layer is bonded to the outer surface of the thick-walled pressure-bearing equipment to form a stable acoustic coupling, so as to reduce the influence of surface unevenness or roughness on the stability of ultrasonic thickness measurement.

[0023] The beneficial effects of the robotic ultrasonic thickness measurement method for thick-walled pressure-bearing equipment of the present invention are as follows:

[0024] 1. This invention improves the accuracy of thickness measurement point determination for thick-walled pressure-bearing equipment. It acquires multi-view point cloud data of the actual outer surface of the equipment using a three-dimensional vision sensor, and establishes a three-dimensional surface model based on this data. Based on this model, the area to be measured is determined, and a sequence of thickness measurement points is generated. Compared to measurement methods relying on theoretical models, fixed measurement points, or manual teaching, this invention better adapts to morphological deviations in thick-walled pressure-bearing equipment under actual manufacturing, assembly, clamping, or on-site installation conditions, thereby improving the accuracy of thickness measurement point positioning and the rationality of the thickness measurement arrangement.

[0025] 2. This invention improves the alignment capability of ultrasonic thickness gauge probes with the local normal of the thickness measurement point. By calculating the local surface normal at each thickness measurement point based on the neighborhood point cloud or a 3D surface model, and constructing the target thickness measurement pose of the ultrasonic thickness gauge probe accordingly, the effective incident direction of the ultrasonic thickness gauge probe is consistent with or approximately consistent with the local surface normal of the thickness measurement point. This method helps reduce echo attenuation and measurement errors caused by the ultrasonic incident direction deviating from the normal, and is particularly suitable for ultrasonic thickness measurement of end caps, shoulders, transition sections, or other areas with continuously changing curvature, thereby improving the accuracy of the thickness measurement results. By introducing a dynamic angle threshold based on local curvature, the adaptability of ultrasonic thickness measurement pose constraints in high-curvature areas can be improved, further enhancing the normal alignment accuracy in complex curved surface areas.

[0026] 3. This invention improves the consistency of contact conditions and measurement stability between different thickness measurement points. After the industrial robot reaches the vicinity of the thickness measurement point according to the planned trajectory, the actual normal contact force at the current thickness measurement point is obtained. Based on the deviation between the actual normal contact force and the target normal contact force, the end effector of the industrial robot is controlled to make feedback adjustments along the local surface normal of the current thickness measurement point, so that the ultrasonic thickness measuring probe maintains a preset and consistent contact state at each thickness measurement point. This method helps to reduce the measurement dispersion caused by contact fluctuations, insufficient adhesion, or inconsistent coupling states, thereby improving the stability and repeatability of the ultrasonic thickness measurement process for thick-walled pressure-bearing equipment. Preferably, a dry-coupled ultrasonic thickness measuring probe with a flexible membrane layer at the front end is used, which helps to reduce the measurement fluctuations caused by uneven medium distribution in the liquid coupling method during automatic thickness measurement.

[0027] 4. This invention facilitates coordinated planning and control during thickness measurement. It organically combines thickness measurement point sequence planning, target thickness measurement pose construction, normal approach trajectory generation, and normal contact force feedback adjustment at the thickness measurement points. This enables the industrial robot to not only sequentially reach the vicinity of each thickness measurement point but also to make minor adjustments along the local surface normal after contact is established. Compared to methods that only perform trajectory planning or rely solely on fixed contact actions, this invention improves the robot's adaptability and reliability when performing ultrasonic thickness measurement tasks on complex curved surfaces in thick-walled pressure-bearing equipment. By introducing dynamic force control parameters based on local curvature, the stability of normal contact force adjustment in high-curvature regions can be improved, reducing overshoot and oscillation during contact adjustment.

[0028] 5. This invention improves the effectiveness and reliability of thickness measurement results. This invention triggers ultrasonic thickness measurement when the actual normal contact force meets a preset stability criterion and the probe posture meets an angle threshold condition, and then determines and records the validity of the thickness value at the current measurement point. Compared to methods that measure directly after contact is established, this invention allows ultrasonic thickness measurement to be performed under more stable contact conditions, and improves the reliability of measurement results through thickness value consistency or echo quality criteria.

[0029] 6. This invention can improve the automation level and engineering application value of ultrasonic thickness measurement for thick-walled pressure-bearing equipment. This invention organically combines point cloud sensing, thickness measurement point generation, normal pose guidance, normal contact force control, and ultrasonic thickness measurement result judgment to form a robotic ultrasonic thickness measurement method for thick-walled pressure-bearing equipment. This method reduces the reliance on operational experience for manual point-by-point measurement, improves thickness measurement efficiency, consistency, and repeatability, and is applicable to wall thickness detection and related automated quality inspection scenarios for hot-spinning gas cylinders, pressure pipelines, and other thick-walled pressure-bearing components.

[0030] The features and advantages of the present invention will be described in detail through embodiments and in conjunction with the accompanying drawings. Attached Figure Description

[0031] Figure 1 This is a schematic diagram of a robotic ultrasonic thickness measurement experimental device for a hot-spinning gas cylinder end cap, as described in an embodiment of the present invention.

[0032] Figure 2 This is a flowchart of the overall process for the robotic ultrasonic thickness measurement method for thick-walled pressure-bearing equipment according to the present invention.

[0033] Figure 3 This is a schematic diagram illustrating the point cloud modeling, thickness measurement area division, and thickness measurement point generation of the outer surface of the thermally spun gas cylinder head in an embodiment of the present invention.

[0034] Figure 4 This is a schematic diagram of the local surface normal of the thickness measurement point, the target thickness measurement pose, and the dynamic angle threshold constraint in an embodiment of the present invention.

[0035] Figure 5 This is a schematic diagram illustrating normal approach, force-position hybrid control, stable contact determination, and ultrasonic thickness measurement triggering in an embodiment of the present invention. Detailed Implementation

[0036] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. However, it should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of the invention. Furthermore, descriptions of well-known structures and technologies are omitted in the following description to avoid unnecessarily obscuring the concept of the invention.

[0037] In the description of this invention, it should be noted that when an element is referred to as being "fixed to" or "set on" another element, it can be directly on or indirectly on the other element. When an element is referred to as being "connected to" another element, it can be directly connected to or indirectly connected to the other element.

[0038] In the description of this invention, it should be noted that the terms "center," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship commonly used when the product of this invention is in use. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first," "second," and "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified. "Several" means one or more, unless otherwise explicitly specified.

[0039] In the description of this invention, it should also be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," "connect," and "link" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0040] Example 1:

[0041] In this embodiment, a hot-spinning gas cylinder end cap is used as a specific implementation object of a thick-walled pressure-bearing device to illustrate the robotic ultrasonic thickness measurement method of the present invention. In this application, "thick wall" is defined as a wall thickness of 30-60 mm. The thickness measurement system includes an industrial robot, a three-dimensional vision sensor installed at the end of the industrial robot, an ultrasonic thickness measurement probe, a force sensor, a controller, and an ultrasonic thickness gauge. The industrial robot can be a six-axis industrial robot, such as the UR5e six-axis industrial robot; the three-dimensional vision sensor is used to acquire point cloud data of the actual outer surface of the hot-spinning gas cylinder end cap; the ultrasonic thickness measurement probe is preferably a dry-coupled ultrasonic thickness measurement probe with a flexible membrane layer at its front end, used to emit and receive ultrasonic signals after establishing stable acoustic coupling contact with the outer surface of the hot-spinning gas cylinder end cap to obtain the thickness value of the current measurement point; the force sensor is used to acquire the actual normal contact force between the ultrasonic thickness measurement probe and the outer surface of the hot-spinning gas cylinder end cap; the controller is used to complete point cloud modeling, thickness point generation, normal pose construction, robot trajectory planning, normal contact force control, and thickness measurement result determination.

[0042] In this embodiment, the 3D vision sensor and the ultrasonic thickness gauge probe are mounted on the same end effector, or separately mounted on the end effector of the industrial robot and have a pre-calibrated relative pose relationship. For ease of explanation, the industrial robot's base coordinate system is assumed to be... The coordinate system of the industrial robot end effector is The coordinate system of the 3D vision sensor is The coordinate system of the ultrasonic thickness gauge probe is .

[0043] Combination Figures 1 to 5 This embodiment takes the hot-spinning gas cylinder end cap as an example to illustrate the robotic ultrasonic thickness measurement method for thick-walled pressure equipment of the present invention, which specifically includes the following steps.

[0044] Step S1: Acquisition of point cloud and surface modeling of the outer surface of the thermally spun gas cylinder head.

[0045] First, an industrial robot carrying a 3D vision sensor is controlled to move to multiple observation poses to perform multi-view scanning of the outer surface of the hot-spinning gas cylinder head, acquiring multiple local point cloud data. Preferably, multiple observation poses, such as more than six, can be set around the curved surface area of ​​the hot-spinning gas cylinder head and its adjacent transition areas to ensure that the area to be measured on the hot-spinning gas cylinder head is effectively observed. Let the first... Any point acquired by the 3D vision sensor under the observation pose is... Then the coordinates of this point after transformation to the industrial robot base coordinate system satisfy:

[0046]

[0047] in, For the first The pose transformation matrix of the industrial robot's end effector coordinate system relative to the base coordinate system under each observed pose. This is the calibration transformation matrix of the 3D vision sensor coordinate system relative to the industrial robot end effector coordinate system.

[0048] After uniformly transforming the local point clouds acquired from multiple observation poses to the industrial robot base coordinate system, the local point clouds are registered and fused to obtain a global point cloud representing the true outer surface morphology of the hot-spinning gas cylinder head. Further, background removal, target region segmentation, and surface reconstruction are performed on the global point cloud to establish a three-dimensional surface model of the true outer surface of the hot-spinning gas cylinder head. The three-dimensional surface model can be a triangular mesh model, a parametric surface model, or other surface representation that can reflect the geometric features of the true outer surface of the hot-spinning gas cylinder head.

[0049] Step S2: Determine the thickness measurement area of ​​the end cap and generate the thickness measurement point.

[0050] After obtaining a three-dimensional surface model of the outer surface of the hot-spinning gas cylinder head, the area to be measured is determined based on the structural characteristics of the head and the thickness measurement requirements. Preferably, the area to be measured can be divided into the curved surface area of ​​the head and adjacent transition areas, and different thickness measurement point arrangement rules can be set according to the degree of curvature change and detection requirements of different areas. For example, a more regular grid layout can be used in areas with gentle curvature changes, while the spacing between thickness measurement points can be appropriately reduced in areas with large curvature changes and adjacent transition areas to improve the thickness measurement coverage of areas with curvature changes.

[0051] Let the first The spatial coordinates of each thickness measurement point in the industrial robot's base coordinate system are: Then we have:

[0052]

[0053] In this embodiment, the thickness measurement points can be generated according to preset axial and circumferential spacing, or the points can be densely distributed based on the local curvature of the hot-spinning gas cylinder head, historical weak areas, or risk level, thereby forming a thickness measurement point sequence suitable for ultrasonic thickness measurement of hot-spinning gas cylinder heads. .

[0054] Step S3: Calculate the local normal of the thickness measurement point and construct the target thickness measurement pose.

[0055] For any thickness measurement point generated in step S2 The local surface normal is estimated by the local geometric relationship of the point cloud in the neighborhood of the thickness measurement point. Let the first... The set of points in the neighborhood of each thickness measurement point is: Its neighborhood center point is The corresponding covariance matrix satisfies:

[0056]

[0057] For covariance matrix Perform eigenvalue decomposition and take the eigenvector corresponding to the smallest eigenvalue as the eigenvector of the first eigenvalue. Local surface normal at each thickness measurement point Preferably, the number of neighborhood points A value of around 20 can be chosen to balance the stability and locality of the normal estimation.

[0058] In obtaining the first After determining the local surface normal at each thickness measurement point, the principal curvature at that thickness measurement point is further calculated by fitting a local surface based on the point cloud of the neighborhood of that thickness measurement point. and and take .

[0059] This serves as a local curvature index for the thickness measurement point. For the head surface region with significant local curvature and adjacent transition regions, stricter attitude constraints are employed to improve the tracking accuracy of the ultrasonic thickness gauge probe on the local surface normal of the thickness measurement point. Specifically, the first... Dynamic angle threshold at each thickness measurement point satisfy:

[0060]

[0061] Furthermore, to ensure the industrial robot maintains a consistent approach direction when measuring thickness on the outer surface, the normal direction of the local surface is corrected to align with a preset direction from the outside to the inside of the ultrasonic thickness measuring probe as it approaches the outer surface of the hot-spinning gas cylinder head. Subsequently, based on the thickness measurement point position... and local surface normal Construct the target thickness measurement pose of the ultrasonic thickness gauge probe. Let the effective incident direction of the ultrasonic thickness gauge probe be... Then it should satisfy:

[0062]

[0063]

[0064] in, The angle between the effective incident direction of the ultrasonic thickness gauge probe and the normal to the local surface of the thickness measurement point. This is a preset angle threshold. By making... Not greater than the dynamic angle threshold This allows the ultrasonic beam to be incident on the workpiece along the local surface normal of the thickness measurement point as much as possible in different curvature regions, thereby improving the accuracy of thickness measurement.

[0065] Step S4: Thickness measurement point pose planning and normal approach trajectory generation.

[0066] After obtaining the thickness measurement point sequence and its corresponding target thickness measurement pose, the switching trajectory of the industrial robot between adjacent thickness measurement points is planned based on the thickness measurement point sequence. To avoid direct contact between the industrial robot's end effector and the workpiece surface in complex curved areas, a normal approach point is set on the outside of each thickness measurement point. Normal proximity point corresponding to each thickness measurement point Represented as:

[0067]

[0068] in, The preset normal approach distance is used. The end effector of the industrial robot first moves along the switching trajectory between adjacent thickness measurement points to the normal approach point, and then approaches the outer surface of the hot-spinning gas cylinder head along the local surface normal of the thickness measurement point. By combining the inter-point switching trajectory with the intra-point normal approach trajectory, the industrial robot can not only sequentially reach the vicinity of each thickness measurement point, but also establish contact along the normal of the thickness measurement point in a reasonable posture during the approach phase.

[0069] Step S5: Establishment of normal contact and force-position hybrid control.

[0070] Industrial robot end-effector reaches the first After the normal approach point corresponding to each thickness measurement point, the ultrasonic thickness measuring probe is controlled to slowly approach the outer surface of the hot-spinning gas cylinder head along the local surface normal. The probe's flexible membrane layer at the tip adheres to the outer surface of the hot-spinning gas cylinder head, forming a stable dry-coupled acoustic contact condition. Preferably, when the force sensor detects that the initial contact force reaches a preset contact threshold, it is determined that the ultrasonic thickness measuring probe has established initial contact with the outer surface of the hot-spinning gas cylinder head, and then the force-position hybrid control stage based on normal contact force feedback begins.

[0071] In this embodiment, the industrial robot maintains target thickness measurement pose tracking in the tangential position direction and probe posture direction, and performs pose fine-tuning based on normal contact force feedback in the local surface normal direction at the thickness measurement point. Let the target normal contact force be... , No. The actual normal contact force for each control cycle is Then the normal contact force error satisfies:

[0072]

[0073] To improve the stability of normal contact force control and reduce steady-state error, this embodiment uses incremental proportional-integral control to calculate the normal displacement correction. It satisfies:

[0074]

[0075] The normal correction position command of the industrial robot end effector at the current thickness measurement point satisfies:

[0076]

[0077] in, For the first Normal displacement correction per control cycle For proportional gain, For integral gain, To control the cycle, This is the current thickness measurement point location. This represents the local surface normal at the current thickness measurement point.

[0078] In this embodiment, to improve the adaptability of normal contact force adjustment in regions with different curvatures, the proportional gain and integral gain are adjusted according to the local curvature index of the current thickness measurement point. Adaptive adjustment, satisfying the following conditions respectively:

[0079]

[0080]

[0081] in, and As the reference control parameters, and This is the curvature adjustment coefficient. For the head surface region with large local curvature and the adjacent transition region, the proportional gain and integral gain are appropriately reduced to reduce overshoot and oscillation during the contact adjustment process and improve the stability of the normal contact adjustment.

[0082] Furthermore, to avoid contact impact due to excessive single correction or probe jitter due to force signal fluctuations, this embodiment sets amplitude limits and a stability dead zone for the normal displacement correction. When the absolute value of the normal contact force error is less than the preset dead zone threshold, the current normal displacement is maintained and no further adjustment is made. Through this method, the ultrasonic thickness gauge probe maintains a preset and consistent contact state at each thickness measurement point. Preferably, the target normal contact force... It can also be based on local curvature index Make adaptive adjustments.

[0083] Step S6: Stable contact determination and ultrasonic thickness measurement trigger.

[0084] During the normal contact force feedback adjustment process, the stability of the actual normal contact force at the current thickness measurement point is determined. Preferably, the actual normal contact force satisfies the following formula:

[0085]

[0086] And continue to satisfy:

[0087]

[0088] Then it is determined that the contact state at the current thickness measurement point has reached a stable state. Among them, This is the normal contact force error threshold. For the current stable duration, This is a preset stable duration threshold.

[0089] As another method for determining stability, the following range criterion can also be used:

[0090]

[0091] And it continuously meets the preset stable time threshold.

[0092] Under the condition that the stable contact criterion is met and the ultrasonic thickness gauge probe posture is satisfied When the ultrasonic thickness measurement operation at the current measurement point is triggered, the ultrasonic thickness gauge collects the echo signal from the current measurement point and calculates the ultrasonic echo flight time. Let the ultrasonic propagation speed in the thermally spun gas cylinder end cap material be... The effective round-trip flight time of one echo at the current thickness measurement point is The thickness value of the current thickness measurement point satisfy:

[0093]

[0094] Wherein, the ultrasonic propagation speed It can be obtained through standard test block calibration, or preset according to the material type of thick-walled pressure equipment and the current measurement conditions.

[0095] Step S7: Determine and record the validity of the thickness measurement results.

[0096] To improve the reliability of thickness measurement results, this embodiment continuously acquires data from the same thickness measurement point. Secondary thickness value When the following conditions are met:

[0097]

[0098] The thickness value of the current thickness measurement point is determined to be valid, whereby... This is the thickness deviation threshold. Preferably, Three or more measurements can be taken to improve the consistency of the measurement results.

[0099] Furthermore, the measurement validity condition may also include the ultrasonic echo quality parameter meeting a preset threshold condition. The ultrasonic echo quality parameter includes at least one of echo amplitude, signal-to-noise ratio, or echo correlation parameter. When the repeated measurement results meet the consistency criterion and the echo quality parameter meets the preset threshold condition, the thickness value of the current thickness measurement point, its corresponding thickness measurement point location information, region number, and measurement status information are recorded, and the measurement is moved to the next thickness measurement point to continue. If the thickness value of the current thickness measurement point does not meet the validity criterion, the contact adjustment and ultrasonic thickness measurement operation are re-executed for the current thickness measurement point.

[0100] In this embodiment, the industrial robot repeatedly executes the processes of "point cloud modeling, thickness point determination, normal pose guidance, normal approach, normal contact force control, stable contact determination, ultrasonic thickness measurement triggering, and result validity determination" according to the thickness measurement point sequence, thereby obtaining a set of thickness data for different regions of the hot-spinning gas cylinder head. This thickness data set can be further used to form the wall thickness distribution results of the hot-spinning gas cylinder head and to identify locally thinned areas or areas with abnormal wall thickness.

[0101] In this embodiment, a hot-spinning gas cylinder end cap is used as the specific implementation object of a thick-walled pressure-bearing device to illustrate the robotic ultrasonic thickness measurement method of the present invention. However, the present invention is not limited to the specific implementation described above. Other equivalent schemes that use three-dimensional point cloud perception to determine the thickness measurement point and the local surface normal of the thickness measurement point, and combine them with industrial robot normal contact force feedback control to achieve stable ultrasonic thickness measurement, should all fall within the protection scope of the present invention.

[0102] It should be noted that although the above embodiments have been described herein, this does not limit the scope of patent protection for this invention. Therefore, any changes and modifications made to the embodiments described herein based on the innovative concept of this invention, or equivalent structural or procedural transformations made using the description and drawings of this invention, directly or indirectly applying the above technical solutions to other related technical fields, are all included within the scope of protection of this invention.

Claims

1. A robotized ultrasonic thickness measurement method for thick-walled pressure equipment, characterized in that, Includes the following steps: S1. Use a 3D vision sensor installed at the end of an industrial robot to acquire point cloud data of the outer surface of the equipment, and process the point cloud data to establish a 3D surface model of the outer surface of the equipment. S2, Based on the three-dimensional surface model, determine the thickness area to be measured in the device, and generate multiple thickness measurement points in the thickness area to be measured to form a thickness measurement point sequence; S3, calculate the local surface normal at the thickness measurement point, construct the target thickness measurement pose of the ultrasonic thickness measurement probe based on the thickness measurement point position and the corresponding local surface normal, so that the angle between the effective incident direction of the probe and the local surface normal is not greater than the preset angle threshold. S4. Based on the thickness measurement point sequence and the target thickness measurement pose corresponding to each thickness measurement point, plan the robot's motion trajectory so that the robot's end effector moves sequentially to the vicinity of each thickness measurement point and approaches the equipment surface along the local surface normal. S5, when the robot moves to the vicinity of the current thickness measurement point, control the probe to establish contact with the equipment surface, and control the robot end effector to adjust the feedback based on the deviation between the actual normal contact force and the target normal contact force, so that the probe maintains the preset contact state; S6. When the contact state meets the preset stability condition and the probe posture meets the angle threshold condition, the ultrasonic thickness measurement operation of the thickness measurement point is triggered to obtain the thickness value of the point. S7, determine the validity of the current thickness value at the measurement point. If the preset validity conditions are met, record the thickness value and corresponding position information of the point, and proceed to the next thickness measurement point; if not, re-execute contact adjustment and ultrasonic thickness measurement operations for that point. The actual normal contact force in step S5 is obtained by a force sensor installed at the end of the industrial robot or estimated by the joint drive torque of the industrial robot. The industrial robot maintains the target thickness measurement posture in the tangential position direction and the probe posture direction, and performs posture fine-tuning based on normal contact force feedback in the normal direction of the local surface at the thickness measurement point, constituting force-position hybrid control. The normal displacement correction in step S5 adopts incremental proportional-integral control. To improve the adaptability of normal contact force adjustment in different curvature regions, the proportional gain is... and integral gain Based on the current local curvature index of the thickness measurement point Adaptive adjustment, satisfying the following conditions respectively: ; ;in, and As the reference control parameters, and For the curvature adjustment coefficient, the proportional gain and integral gain are appropriately reduced for the local curvature of the head surface area and the adjacent transition area to reduce overshoot and oscillation during the contact adjustment process and improve the stability of the normal contact adjustment.

2. The robotic ultrasonic thickness measurement method for thick-walled pressure-bearing equipment as described in claim 1, characterized in that, In step S1, a three-dimensional vision sensor installed at the end of an industrial robot is used to acquire point cloud data of the outer surface of the thick-walled pressure-bearing equipment under multiple observation poses. The process of establishing a three-dimensional surface model includes coordinate unification, registration and fusion, and target region segmentation of the point cloud data. In step S3, the local surface normal at the thickness measurement point is calculated based on the three-dimensional surface model or the neighborhood point cloud of the thickness measurement point. The preset angle threshold is a fixed angle threshold or a dynamic angle threshold determined based on the local curvature index of the thickness measurement point.

3. The robotic ultrasonic thickness measurement method for thick-walled pressure-bearing equipment as described in claim 1, characterized in that, The point cloud coordinate unification in step S1 includes: based on the pose information of the industrial robot end effector and the calibration relationship between the 3D vision sensor and the industrial robot end effector, uniformly transforming the local point clouds acquired under each observation pose to the industrial robot base coordinate system; any first Point cloud points under observation pose The coordinates after transformation to the industrial robot base coordinate system satisfy: ;in, For the first The pose transformation matrix of the industrial robot's end effector coordinate system relative to the base coordinate system under each observed pose. This is the calibration transformation matrix of the 3D vision sensor coordinate system relative to the industrial robot end effector coordinate system.

4. The robotic ultrasonic thickness measurement method for thick-walled pressure-bearing equipment as described in claim 1, characterized in that, The thickness region to be measured in step S2 includes curved surface regions with different characteristics. Different thickness measurement point spacing, number of thickness measurement points, and arrangement methods are set according to the curvature characteristics, detection requirements, or risk levels of different regions to generate a thickness measurement point sequence. The coordinates of each thickness measurement point are represented as follows: ;in, For the first The spatial coordinates of the thickness measurement points in the industrial robot's base coordinate system, where T represents the transpose of the matrix.

5. The robotic ultrasonic thickness measurement method for thick-walled pressure-bearing equipment as described in claim 1, characterized in that, The local surface normal of the thickness measurement point in step S3 is estimated through the local geometric relationship of the point cloud in the neighborhood of the thickness measurement point; let the first... The set of points in the neighborhood of each thickness measurement point is: The neighborhood center point is Then its covariance matrix satisfies: ; Take the eigenvector corresponding to the smallest eigenvalue of the covariance matrix as the first... Local surface normal at each thickness measurement point , Let the covariance matrix be the neighborhood point set corresponding to the i-th thickness measurement point; according to the... Each thickness measurement point location and corresponding local surface normal The target thickness measurement pose of the ultrasonic thickness gauge is constructed, and the direction of the local surface normal is uniformly corrected so that the local surface normal at each thickness measurement point points towards a preset approach direction; the effective incident direction of the ultrasonic thickness gauge is determined. With the local surface normal satisfy: ; ;in The angle between the effective incident direction of the ultrasonic thickness gauge probe and the local surface normal. This is a preset angle threshold.

6. The robotic ultrasonic thickness measurement method for thick-walled pressure-bearing equipment as described in claim 5, characterized in that: In step S3, according to the first Fit a local surface to the point cloud of a thickness measurement point and calculate the principal curvature at that thickness measurement point. and Using local curvature index Characterizing the curvature features at the thickness measurement point, the local curvature index satisfies The angle threshold between the effective incident direction of the ultrasonic thickness gauge and the local surface normal is related to the local curvature index. Related dynamic angle thresholds .

7. The robotic ultrasonic thickness measurement method for thick-walled pressure-bearing equipment as described in claim 1, characterized in that, In step S4, a normal proximity point is set outside each thickness measurement point. Normal proximity point corresponding to each thickness measurement point Represented as: ;in, For the first The coordinates of the normal approximation point corresponding to each thickness measurement point The local surface normal of the thickness measurement point. To determine the preset normal approach distance, the industrial robot's end effector first moves to the normal approach point, and then moves along the first normal approach distance. The local surface normal of a thickness measurement point is close to the surface of a thick-walled pressure-bearing device.

8. The robotic ultrasonic thickness measurement method for thick-walled pressure-bearing equipment as described in claim 1, characterized in that, In step S5, the target normal contact force is set to be... , No. The normal contact force error for each control cycle is: ;in, For the first The actual normal contact force for each control cycle; the normal displacement correction in step S5 adopts incremental proportional-integral control, satisfying... The normal correction position command of the industrial robot end effector at the current thickness measurement point satisfies: ;in, For the first Normal displacement correction per control cycle For proportional gain, For integral gain, To control the cycle, This is the current thickness measurement point location. The local surface normal at the current thickness measurement point is defined; and a limiting constraint and a stable dead zone are set for the normal displacement correction amount; the proportional gain and integral gain are related to the local curvature index of the current thickness measurement point. The relevant dynamic control parameters are denoted as follows: and The and Based on local curvature index Adaptive adjustment.

9. The robotic ultrasonic thickness measurement method for thick-walled pressure-bearing equipment as described in claim 1, characterized in that, The stability criterion in S6 includes: And continue to satisfy: ;in, This is the normal contact force error threshold. For the current stable duration, A preset stable duration threshold is defined; or, the stability criterion satisfies: ; and continuously meet the preset stable time threshold; when the stability criterion is met and the ultrasonic thickness measurement probe posture meets the requirements. When this occurs, the ultrasonic thickness measurement operation at the current thickness measurement point is triggered.

10. The robotic ultrasonic thickness measurement method for thick-walled pressure-bearing equipment as described in claim 1, characterized in that, In step S6, after the stable contact criterion is met, the ultrasonic echo signal of the current thickness measurement point is acquired, and the round-trip time of one effective echo is obtained. ; Let the ultrasonic propagation velocity in the material of thick-walled pressure equipment be... The thickness value of the current thickness measurement point satisfy: Wherein, the ultrasonic propagation speed The results are obtained through standard test block calibration or pre-set according to the material category of thick-walled pressure equipment.

11. The robotic ultrasonic thickness measurement method for thick-walled pressure-bearing equipment according to claim 1, characterized in that, In step S7, the same thickness measurement point is continuously acquired. Secondary thickness value When the following conditions are met: The thickness value of the current thickness measurement point is determined to be valid, where, The thickness deviation threshold is used as the measurement validity condition. The measurement validity condition also includes that the ultrasonic echo quality parameter meets the preset threshold condition. The ultrasonic echo quality parameter includes at least one of echo amplitude, signal-to-noise ratio, or echo correlation parameter.

Images