Roughness detection method for aluminum alloy flat tube
By obtaining the three-dimensional morphology and two-dimensional roughness parameters of aluminum alloy flat tubes at the semi-finished stage, identifying and adjusting the detection parameters, the problem of inaccurate finished product detection results is solved, and efficient and accurate roughness detection is achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SHANDONG HONGYUAN METAL MATERIAL CO LTD
- Filing Date
- 2026-04-03
- Publication Date
- 2026-05-19
AI Technical Summary
Existing finished aluminum alloy flat tube inspection methods cannot determine the defect distribution in the semi-finished product stage, resulting in inaccurate roughness inspection results and difficulty in balancing inspection efficiency and accuracy.
In the semi-finished product stage, surface inspection data of three-dimensional morphology and two-dimensional roughness parameters are obtained to identify raised areas. Inspection parameters are dynamically adjusted to improve the inspection accuracy of raised areas in a targeted manner, and the inspection strategy is optimized by combining batch data.
It significantly improves the accuracy and reliability of surface roughness testing of finished aluminum alloy flat tubes, optimizes production quality management, and reduces testing costs.
Smart Images
Figure CN122062610A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of roughness testing technology, specifically to a roughness testing method for aluminum alloy flat tubes. Background Technology
[0002] As an important industrial material, the surface roughness of aluminum alloy flat tubes is a key indicator for evaluating product quality, directly affecting properties such as sealing performance, wear resistance, and fatigue strength. Currently, during the production process, roughness testing is typically performed at both the semi-finished and finished product stages to ensure that the final product meets quality standards.
[0003] However, the above methods cannot detect the defect distribution in the semi-finished product stage when inspecting finished products. They can only use uniform and fixed detection parameters to scan with average intensity. For potentially high-risk areas, small defects may be missed due to insufficient detection accuracy, resulting in inaccurate roughness detection results. Summary of the Invention
[0004] To address the technical problem of inaccurate roughness test results for finished aluminum alloy flat tubes, this application aims to provide a roughness test method for aluminum alloy flat tubes. The specific technical solution adopted is as follows: The roughness detection method for aluminum alloy flat tubes provided in this application includes: acquiring surface inspection data of the semi-finished aluminum alloy flat tube to be inspected, the surface inspection data including three-dimensional morphology information and two-dimensional roughness parameter detection values of the semi-finished aluminum alloy flat tube to be inspected; determining the probability of roughness non-compliance of each raised area based on the surface inspection data; adjusting the predetermined value of the detection parameter of each raised area based on the probability of roughness non-compliance of each raised area to obtain the target value of the detection parameter of each raised area; and performing roughness detection on the corresponding area of each raised area in the finished aluminum alloy flat tube based on the target value of the detection parameter of each raised area.
[0005] Optionally, the above-mentioned determination of the possibility that the roughness of each raised area is substandard based on the surface inspection data includes: dividing the surface of the semi-finished aluminum alloy flat tube to be inspected into multiple local inspection areas, and determining the deviation of the two-dimensional roughness parameter of each local inspection area based on the detected value of the two-dimensional roughness parameter of each local inspection area; determining the raised areas on the surface of the semi-finished aluminum alloy flat tube to be inspected and the protrusion degree of each raised area based on the three-dimensional topography data; and determining the possibility that the roughness of each raised area is substandard based on the deviation of the two-dimensional roughness parameter of each local inspection area and the protrusion degree of each raised area.
[0006] Optionally, based on the detected values of the two-dimensional roughness parameters of each local detection area, the deviation of the two-dimensional roughness parameters of each local detection area is determined, including: determining the individual deviation between the detected value of each two-dimensional roughness parameter of the first local detection area and the standard value; and determining the average of the individual deviations of all two-dimensional roughness parameters of the first local detection area as the deviation of the two-dimensional roughness parameters of the first local detection area.
[0007] Optionally, the aforementioned three-dimensional topography data includes the three-dimensional coordinates of multiple three-dimensional coordinate points. The determination of the protruding areas on the surface of the semi-finished aluminum alloy flat tube to be inspected, and the degree of protrusion of each protruding area, based on the three-dimensional topography data, includes: marking three-dimensional coordinate points in the three-dimensional topography data whose height exceeds a preset height threshold as suspected protrusion points; the height value of a three-dimensional coordinate point is the coordinate component of that three-dimensional coordinate point in the direction perpendicular to the inspection surface; clustering all suspected protrusion points to obtain multiple clusters; defining the area formed by the coordinates of the suspected protrusion points included in the same cluster as a protruding area; and determining the degree of protrusion of each protruding area based on the height of the suspected protrusion points included in each protruding area.
[0008] Optionally, determining the protrusion degree of each protrusion region based on the height of the suspected protrusion points included in each protrusion region includes: determining a first local detection area and the total number of suspected protrusion points in the first local detection area, wherein the first local detection area is the local detection area to which the first protrusion region belongs, and the first protrusion region is any protrusion region on the surface of the semi-finished aluminum alloy flat tube to be inspected; and determining the protrusion degree of the first protrusion region based on the total number of protrusion regions on the surface of the semi-finished aluminum alloy flat tube to be inspected, the total number of suspected protrusion points in the first local detection area, the number of suspected protrusion points in the first protrusion region, and the average height of the suspected protrusion points in the first protrusion region.
[0009] Optionally, the above-mentioned determination of the roughness non-compliance probability of each raised region based on the deviation of the two-dimensional roughness parameter of each local detection region and the convexity of each raised region includes: determining the roughness non-compliance probability of the first local detection region based on the product of the deviation of the two-dimensional roughness parameter of the first local detection region and the convexity of each raised region.
[0010] Optionally, the above-mentioned adjustment of the predetermined value of the detection parameter of each raised area based on the probability of the roughness of each raised area not meeting the standard, to obtain the target value of the detection parameter of each raised area, further includes: acquiring the local detection area and raised area of other semi-finished aluminum alloy flat tubes belonging to the same batch as the semi-finished aluminum alloy flat tube to be tested; determining the abnormality ratio of the first local detection area based on the number of raised areas appearing in the first local detection area among the other semi-finished aluminum alloy flat tubes to be tested; determining the parameter adjustment degree of the first raised area based on the probability of the roughness of the first raised area not meeting the standard, the average probability of the roughness of all raised areas included in the first local detection area not meeting the standard, and the abnormality ratio of the first local detection area; adjusting the predetermined value of the detection parameter of the first raised area based on the parameter adjustment degree of the first raised area to obtain the target value of the detection parameter of the first raised area.
[0011] Optionally, the detection parameters include the spot size. The predetermined value of the detection parameters of the first protruding region is adjusted based on the parameter adjustment degree of the first protruding region to obtain the target value of the detection parameters of the first protruding region, which includes: determining the target adjustment amount based on the preset basic adjustment amount and the parameter adjustment degree of the first protruding region; adjusting the target adjustment amount downward based on the predetermined value of the spot size to obtain a candidate value of the spot size of the first protruding region; and determining the maximum value between the candidate value of the spot size and the minimum spot size as the target value of the spot size of the first protruding region.
[0012] Optionally, the roughness detection of the corresponding area of each protrusion in the finished aluminum alloy flat tube based on the target value of the detection parameters of each protrusion area includes: determining the corresponding area of each protrusion on the finished aluminum alloy flat tube based on a preset coordinate mapping relationship from semi-finished product to finished product; and performing roughness detection on the corresponding area of each protrusion based on the target value of the detection parameters of each protrusion.
[0013] Optionally, the method further includes: issuing a production anomaly warning message when the anomaly ratio in the first local detection area is greater than the anomaly ratio threshold, the production anomaly warning message being used to indicate a production anomaly in the same batch of semi-finished aluminum alloy flat tubes to be tested.
[0014] This application has the following beneficial effects: The roughness detection method for aluminum alloy flat tubes provided in this application acquires surface detection data containing three-dimensional morphology and two-dimensional roughness parameters at the semi-finished stage of the aluminum alloy flat tubes, and quantitatively assesses the possibility of roughness non-compliance in each raised area. It dynamically and differentially assigns appropriate target values of detection parameters (such as finer spot or slower scanning speed) to each raised area, and finally achieves differentiated and targeted accurate re-inspection. This significantly improves the accuracy of roughness detection and the reliability of judgment results for finished aluminum alloy flat tubes without significantly increasing the overall detection cost. Attached Figure Description
[0015] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0016] Figure 1 This is a flowchart illustrating a method for roughness testing of an aluminum alloy flat tube according to an embodiment of this application. Figure 2 This is a flowchart illustrating another method for roughness detection of aluminum alloy flat tubes provided in one embodiment of this application. Detailed Implementation
[0017] To further illustrate the technical means and effects adopted by this application to achieve the intended purpose of the invention, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a roughness detection method for aluminum alloy flat tubes proposed in this application. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.
[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.
[0019] Aluminum alloys, due to their lightweight, corrosion resistance, and ease of processing, are widely used in aerospace, automotive manufacturing, and architectural decoration. Their surface roughness directly affects key performance characteristics such as sealing and wear resistance, thus requiring precise testing. Currently, aluminum alloy roughness testing methods are mainly divided into two categories: contact and non-contact methods. These two methods complement each other to meet the testing needs of different scenarios, providing support for the quality control of aluminum alloy products.
[0020] Currently, aluminum alloy flat tubes undergo multiple stages in their production process, with surface roughness testing performed at both the semi-finished and finished product stages. However, existing methods test each stage independently. Since the semi-finished product stage is fundamental to the finished product's quality and its data guides the focus of finished product testing, the current methods fail to link the test results from each stage. This leads to a degree of inaccuracy in the surface roughness testing results of aluminum alloy flat tubes.
[0021] During finished product inspection, it's impossible to know the product's inherent quality characteristics and potential defect distribution during the semi-finished stage. Therefore, only uniform, fixed inspection parameters (such as spot size and scanning speed) can be used for all areas with average intensity scanning. This creates a double problem: for areas already confirmed as good in the semi-finished stage, over-inspection leads to a waste of time and resources; for high-risk areas with potential defects in the semi-finished stage, insufficient inspection accuracy may prevent the effective identification of minute defects, resulting in missed detections.
[0022] Problem tracing is difficult: when a finished product is found to be non-conforming, it is difficult to accurately pinpoint whether the defect existed in the semi-finished product stage or was generated in subsequent processing, which is not conducive to closed-loop management of production quality and process optimization.
[0023] The contradiction between testing efficiency and accuracy: To improve the reliability of finished product testing, it is necessary to comprehensively improve testing accuracy (such as reducing the spot size and reducing the scanning speed), but this will lead to a significant increase in testing time and seriously affect production efficiency.
[0024] The specific scheme of the roughness detection method for aluminum alloy flat tubes provided in this application is described in detail below with reference to the accompanying drawings.
[0025] Please see Figure 1 The diagram illustrates a method flowchart for roughness detection of an aluminum alloy flat tube according to an embodiment of this application.
[0026] like Figure 1 As shown, the roughness detection method for this aluminum alloy flat tube includes S101-S104.
[0027] S101. Obtain the surface inspection data of the semi-finished aluminum alloy flat tube to be inspected.
[0028] The surface inspection data includes the three-dimensional morphology information and the measured values of the two-dimensional roughness parameters of the semi-finished aluminum alloy flat tube to be inspected.
[0029] It should be understood that three-dimensional topography information is used to characterize the microscopic undulations and texture structure of the surface to be inspected in three-dimensional space. It can be visualized as a three-dimensional digital model that reflects the distribution of microscopic peaks and valleys, texture direction, and defect morphology of the surface. This three-dimensional topography information should include the three-dimensional coordinates of multiple coordinate points on the surface of the semi-finished aluminum alloy flat tube to be inspected, where the Z-axis coordinate is used to characterize the height value of the coordinate point relative to the reference datum plane.
[0030] In one alternative implementation, the three-dimensional topography data can be acquired using non-contact three-dimensional topography measurement equipment such as a laser confocal microscope or a white light interferometer. The specific process is as follows: the semi-finished aluminum alloy flat tube to be inspected is stably clamped onto the inspection platform; the objective lens of the equipment is adjusted to be perpendicular to the surface to be measured (such as the main surface or the narrow side); the equipment scans the surface area point by point along a preset scanning path (e.g., along paths parallel and perpendicular to the extrusion direction of the aluminum alloy flat tube, respectively), recording the three-dimensional spatial coordinates of densely packed coordinate points on the surface.
[0031] It should be understood that two-dimensional roughness parameters are used to characterize the macroscopic roughness level of a surface. These two-dimensional roughness parameters include at least the profile arithmetic mean deviation (Ra), the maximum profile height (Rz), and the profile micro-irregularity spacing (Rsm).
[0032] Optionally, the detected value of the two-dimensional roughness parameter can be automatically calculated by the non-contact measuring device through its built-in analysis software while acquiring three-dimensional topographic information.
[0033] It should be understood that surface inspection data includes the main surface, narrow sidewalls, and the inner wall of the microchannels, and different inspection methods are used for surfaces with different characteristics.
[0034] Optionally, a portable laser confocal roughness meter can be used to inspect the main surface and narrow side surfaces. During the inspection, the sample is fixed and adjusted so that the inspection surface is perpendicular to the lens, and scanned once parallel to the extrusion direction and once perpendicular to the extrusion direction to fully capture the anisotropic surface features. For the inner wall of the microchannel, a benchtop laser confocal inspection instrument is required. The sample is fixed in the inspection chamber by a coaxial positioning bracket, and a high-magnification long working distance objective lens is selected to perform a three-dimensional scan at the channel entrance to obtain the two-dimensional roughness parameters and three-dimensional morphology data of the inner wall surface.
[0035] In one alternative implementation, when measuring three-dimensional topography data and two-dimensional roughness data, the above-mentioned measuring device can collect data in units of local detection areas, with one local detection area corresponding to a set of two-dimensional roughness data.
[0036] S102. Based on surface inspection data, determine the probability that the roughness of each raised area is substandard.
[0037] It should be understood that the surface of the semi-finished aluminum alloy flat tube to be inspected may have multiple raised areas, which are the main reason for the roughness of the surface of the semi-finished aluminum alloy flat tube to be inspected.
[0038] Understandably, the probability of a raised area failing to meet the roughness standard is used to characterize the likelihood that the roughness of the raised area exceeds the acceptable range.
[0039] In one alternative implementation, the maximum height value of each raised area can be extracted from the surface inspection data first, and the ratio between the maximum height value of each raised area and the historical maximum height value can be used to determine the probability of non-compliance with roughness standards.
[0040] S103. Based on the possibility that the roughness of each raised area is not up to standard, adjust the predetermined value of the detection parameter of each raised area to obtain the target value of the detection parameter of each raised area.
[0041] It should be understood that since the probability of roughness non-compliance varies for each raised area, different testing parameters should be used for different raised areas when conducting roughness testing on the finished aluminum alloy flat tubes. The greater the probability of roughness non-compliance, the higher the testing accuracy should be.
[0042] Optionally, there can be multiple detection parameters, such as spot size and scanning speed.
[0043] It should be understood that the smaller the spot size and the lower the scanning speed, the higher the acquisition accuracy.
[0044] In one optional implementation, the local inspection area and protrusion area of other semi-finished aluminum alloy flat tubes belonging to the same batch as the semi-finished aluminum alloy flat tube to be inspected can be obtained; based on the number of protrusion areas appearing in the first local inspection area among the other semi-finished aluminum alloy flat tubes, the abnormality ratio of the first local inspection area is determined; based on the probability of the roughness of the first protrusion area not meeting the standard, the average probability of the roughness of all protrusion areas included in the first local inspection area not meeting the standard, and the abnormality ratio of the first local inspection area, the parameter adjustment degree of the first protrusion area is determined; based on the parameter adjustment degree of the first protrusion area, the predetermined value of the inspection parameter of the first protrusion area is adjusted to obtain the target value of the inspection parameter of the first protrusion area.
[0045] The first local detection area is any local detection area included in the semi-finished aluminum alloy flat tube to be tested, and the first protruding area is any protruding area on the surface of the semi-finished aluminum alloy flat tube to be tested.
[0046] Optionally, other semi-finished aluminum alloy flat tubes are semi-finished aluminum alloy flat tubes that have already been inspected. Therefore, the inspection data of other semi-finished aluminum alloy flat tubes, namely the local inspection area and the protrusion area, can be obtained.
[0047] Optionally, the detection data may also include the number of raised areas.
[0048] Optionally, the ratio between the number of other semi-finished aluminum alloy flat tubes with raised areas in the first local detection area and the total number of other semi-finished aluminum alloy flat tubes can be determined as the abnormality ratio of the first local detection area.
[0049] Optionally, the default anomaly rate can be set to 0 for the first sample in a batch.
[0050] Optionally, the parameter adjustment degree of a raised region satisfies the following formula: in, Indicates the first The first local detection region The parameter adjustment degree of each raised area Indicates the first The first local detection region The possibility that the roughness of a raised area is not up to standard. Indicates the first The possibility that the average roughness of all raised areas in a local detection region is substandard. Indicates the first The number of other semi-finished aluminum alloy flat tubes with raised areas in the local inspection area. This indicates the total quantity of other semi-finished aluminum alloy flat tubes. Indicates the first The proportion of abnormalities in a local detection area. and This is represented by the preset weighting coefficient.
[0051] Based on this formula, it should be understood that This represents the relative risk of an individual, i.e., the first... The first raised area in the The relative risk prominence of a local detection area; the larger this ratio, the greater the parameter adjustment should be. This indicates a batch-related potential hazard. The higher the ratio of the percentage of localized detection areas across the entire production batch, the more pronounced the risk prevalence of that specific area. Abnormalities in a localized detection area are likely related to systemic factors such as mold wear and process parameter drift.
[0052] This formula, through weighted multiplication, achieves a "risk synergistic amplification" effect on the final adjustment degree. This effect ensures that a higher parameter adjustment degree is generated only when both the conditions of "prominent individual relative risk" and "batch-related hidden danger at this location" are met simultaneously, thereby triggering significant optimization of detection parameters (such as a substantial reduction in spot size). Conversely, when there are minor traces that frequently occur in a batch but have very low individual risk, or a very prominent but accidental and isolated defect, the system will not overreact, achieving an optimal balance between detection efficiency and detection reliability.
[0053] Weighting coefficient and To balance the relative importance of the two risk dimensions in decision-making, a default value of 0.5 can be set in the absence of pre-defined preferences. This is the product of the time term weights. Reaching the maximum value indicates the formula has the highest overall response sensitivity, reflecting a quality control philosophy that places equal emphasis on individual quality and batch stability. In practical applications, the sensitivity can be adjusted based on the specific characteristics of the production line (e.g., whether more attention is paid to occasional defects or systemic fluctuations) while maintaining... Under certain conditions, the weighting coefficients can be fine-tuned. Changes in the weighting coefficients will linearly affect the contribution of each risk factor, but will not change the core logic of "risk resonance amplification" inherent in the product form.
[0054] This formula, while taking into account the degree of risk prominence in local detection areas, further dynamically amplifies or reduces the adjustment range based on whether the location is a batch-related problem, ensuring that high-precision detection resources are prioritized and reasonably allocated to high-risk areas that are both prominent in individual cases and prevalent in batches.
[0055] It should be understood that when the detection parameter is smaller and the accuracy is greater, the value of the detection parameter can be decreased based on the adjustment degree of the parameter. Conversely, when the detection parameter is larger and the accuracy is greater, the value of the detection parameter can be increased based on the adjustment degree of the parameter.
[0056] In one alternative implementation, taking the spot size as an example, a target adjustment amount can be determined based on a preset basic adjustment amount and the parameter adjustment degree of the first protruding region; based on the predetermined value of the spot size, the target adjustment amount is adjusted downward to obtain a candidate value of the spot size of the first protruding region; the maximum value between the candidate value of the spot size and the minimum spot size is determined as the target value of the spot size of the first protruding region.
[0057] Optionally, the target adjustment amount can be determined by multiplying the preset basic adjustment amount and the parameter adjustment degree of the first protrusion area.
[0058] Optionally, the target adjustment amount can be subtracted from the predetermined value of the spot size to obtain the candidate value of the spot size of the first protrusion region.
[0059] For example, the preset basic adjustment amount can be determined based on the minimum adjustable accuracy of the detection equipment and the accuracy requirements under normal detection scenarios. For instance, it can be set to 5 times the minimum adjustable accuracy.
[0060] It should be understood that the minimum spot size is the accuracy limit of the detection equipment. If the selected value exceeds the limit of the detection equipment, it cannot be used for actual detection. Therefore, the maximum value between the selected spot size and the minimum spot size can be determined as the target value of the spot size of the first protrusion area.
[0061] In this embodiment, by statistically analyzing the abnormal proportions of other semi-finished aluminum alloy flat tubes in the same batch within the same local inspection area, common and systematic process problems (such as mold wear) can be identified. Thus, when determining the parameter adjustment degree of the local inspection area, individual risk (the possibility of the roughness not meeting the standard), local relative risk (the relative risk of each area within the local inspection area), and batch systemic risk (abnormal proportion) are considered simultaneously. This ensures that inspection resources are more accurately tilted towards high-risk areas indicated by batch and systemic problems, significantly improving the robustness and adaptability of the inspection strategy.
[0062] In one alternative implementation, if the abnormality ratio in the first local detection area is greater than the abnormality ratio threshold, a production abnormality warning message can be issued. This production abnormality warning message is used to indicate that the production of the same batch of semi-finished aluminum alloy flat tubes to be inspected is abnormal.
[0063] Understandably, if the abnormal proportion in the first local detection area exceeds the abnormal proportion threshold, it indicates that there may be systemic process problems or equipment hazards affecting the overall quality of the batch. In this case, staff should be promptly reminded to carry out maintenance and repairs, and professional personnel should conduct quality inspections on all semi-finished aluminum alloy flat tubes in that batch.
[0064] Optionally, the re-inspection results can be used for diversion processing. For semi-finished aluminum alloy flat tubes that pass the inspection, they are allowed to flow into the subsequent processing steps to continue production into finished products. For semi-finished aluminum alloy flat tubes that fail the inspection, they are isolated, marked, and finally scrapped or reworked according to the preset process to prevent defects from flowing into the finished product stage.
[0065] For example, the abnormality ratio threshold can be determined based on the quality control target of the semi-finished aluminum alloy flat tube, and an example value can be 0.5.
[0066] S104. Based on the target value of the detection parameters of each raised area, roughness detection is performed on the corresponding area of each raised area in the finished aluminum alloy flat tube.
[0067] Specifically, the target value of the detection parameters for each raised area can be sent to the roughness detection equipment (such as a laser confocal microscope). Then, the roughness detection equipment is controlled to use the target value of the detection parameters to perform a fine scan of the corresponding position of the raised area on the finished aluminum alloy flat tube. After the equipment completes the scan, it outputs the final roughness measurement result of the raised area in the finished product state.
[0068] It should be understood that areas other than the raised areas can continue to be scanned using the predetermined values of the detection parameters.
[0069] In one alternative implementation, the corresponding region of each raised area on the finished aluminum alloy flat tube can be determined based on a preset coordinate mapping relationship from semi-finished product to finished product; and roughness detection is performed on the corresponding region of each raised area based on the target value of the detection parameters of each raised area.
[0070] It should be understood that during the process of transforming a semi-finished product into a finished product, operations such as stretching and deformation may occur. Therefore, it is not possible to directly determine the coordinates of the protruding areas in the finished aluminum alloy flat tube based on the coordinates of the protruding areas in the semi-finished product.
[0071] It can be understood that this coordinate mapping relationship is a mapping relationship between coordinate points, specifically a mapping relationship between coordinate points in the semi-finished aluminum alloy flat tube and coordinate points in the finished aluminum alloy flat tube. Based on this mapping relationship, the coordinates of each coordinate of the protruding area in the semi-finished aluminum alloy flat tube in the finished aluminum alloy flat tube can be obtained. Then, based on all the corresponding coordinates in the finished aluminum alloy flat tube, the corresponding area can be obtained.
[0072] Then, based on the target value of the detection parameters of each raised area, roughness detection can be performed on the corresponding area of each raised area. The detection method will not be described in detail here.
[0073] The methods provided in S101-S104 above acquire surface inspection data containing three-dimensional morphology and two-dimensional roughness parameters at the semi-finished stage of aluminum alloy flat tubes, quantify the possibility of roughness non-compliance in each raised area, dynamically and differentially assign appropriate target values of inspection parameters (such as finer spot or slower scanning speed) to each raised area, and finally achieve differentiated and targeted accurate re-inspection. This significantly improves the accuracy of roughness detection and the reliability of judgment results for finished aluminum alloy flat tubes without significantly increasing the overall inspection cost.
[0074] Combination Figure 1 ,like Figure 2 As shown, in one implementation of this application embodiment, the above-mentioned S102 can be specifically implemented by S201-S203.
[0075] S201. Divide the surface of the semi-finished aluminum alloy flat tube to be inspected into multiple local inspection areas, and determine the deviation of the two-dimensional roughness parameter of each local inspection area based on the detected value of the two-dimensional roughness parameter of each local inspection area.
[0076] Optionally, the length and width of the surface of the semi-finished aluminum alloy flat tube to be inspected can be determined at fixed intervals (e.g., 10mm). Divide the grid into 10mm sections, with the center of each grid serving as a local detection area.
[0077] In one alternative implementation, the individual deviation between the detected value and the standard value of each two-dimensional roughness parameter in the first local detection region can be determined; the average of the individual deviations of all two-dimensional roughness parameters in the first local detection region is determined as the deviation of the two-dimensional roughness parameters in the first local detection region.
[0078] Based on the description of the above embodiments, it should be understood that the two-dimensional roughness parameters include multiple parameters, such as the arithmetic mean deviation of the profile, the maximum height of the profile, and the spacing of the micro-irregularities of the profile. Each parameter corresponds to a standard value, which can be the upper limit of tolerance specified in the product drawing, the limit specified in the industry standard, or the internal control threshold preset according to the process capability.
[0079] Optionally, the difference between the detected value and the standard value of each two-dimensional roughness parameter can be calculated first. When the difference is less than or equal to 0, it means that the detected value of the two-dimensional roughness parameter is better than or equal to the standard value and is within the normal range. At this time, the deviation of the two-dimensional roughness parameter can be set to a very small positive number, such as 0.01.
[0080] Optionally, when the difference is greater than 0, the individual deviation of a two-dimensional roughness parameter satisfies the following formula: in, Indicates the first Two-dimensional roughness parameters of a local detection region The degree of deviation of a single item Indicates the first Two-dimensional roughness parameters of a local detection region The detected value, Indicates the first Two-dimensional roughness parameters of a local detection region The standard value.
[0081] In this formula, Characterizing two-dimensional roughness parameters The deviation of the detected value from its standard value. This is used to construct a normalized benchmark that is related to the magnitude of the parameter itself, eliminating the influence of dimensions and achieving normalization.
[0082] It should be understood that by first calculating the individual deviations of each two-dimensional roughness parameter and then taking the average, the overall deviation of a local inspection area from the standard in all key quality dimensions can be reflected in a balanced manner.
[0083] S202. Based on three-dimensional topographic data, determine the raised areas on the surface of the semi-finished aluminum alloy flat tube to be inspected and the degree of protrusion of each raised area.
[0084] In one alternative implementation, three-dimensional coordinate points in the three-dimensional topography data whose height exceeds a preset height threshold can be marked as suspected convex points. All suspected convex points are clustered to obtain multiple clusters. The region formed by the coordinates of the suspected convex points included in the same cluster is determined as a convex region. Based on the height of the suspected convex points included in each convex region, the convexity of each convex region is determined.
[0085] The height value of a three-dimensional coordinate point is the coordinate component of that three-dimensional coordinate point in the direction perpendicular to the detection surface, which is generally the Z-axis coordinate.
[0086] It is understandable that the preset height threshold can be determined based on the statistical values of the surface height distribution of historical qualified samples. Since the surface of a qualified semi-finished aluminum alloy flat tube should be smooth, an exemplary value of the preset height threshold can be 0.
[0087] Alternatively, the k-means clustering method can be used to cluster all suspected convex points, the number of clusters can be determined by the silhouette coefficient method, and the distance between different suspected convex points can be determined by the Euclidean distance of the two-dimensional coordinates of different suspected convex points.
[0088] It should be understood that the two-dimensional coordinates of a suspected convex point refer to the two coordinates other than the height in the three-dimensional coordinate system. The area covered by the two-dimensional coordinates of all suspected convex points in each cluster can be identified as a convex region. The suspected convex points included in a cluster are suspected convex points that are close to each other.
[0089] Understandably, the degree of protrusion of a raised area is used to characterize the severity of the protrusion in that area.
[0090] In one alternative implementation, the average height of suspected protrusion points in a protruding region can be determined as the protrusion degree of that region.
[0091] In another alternative implementation, a first local detection area and the total number of suspected protrusions in the first local detection area can be determined; then, based on the total number of protrusions on the surface of the semi-finished aluminum alloy flat tube to be inspected, the total number of suspected protrusions in the first local detection area, the number of suspected protrusions in the first protrusion area, and the average height of the suspected protrusions in the first protrusion area, the protrusion degree of the first protrusion area can be determined.
[0092] Optionally, the two-dimensional coordinates of each suspected protrusion point in the first protrusion region are compared with the region range of each local detection region to determine the local detection region to which each suspected protrusion point belongs, and the local detection region with the highest proportion of suspected protrusion points is determined as the local detection region to which the first protrusion region belongs.
[0093] Optionally, the convexity of a raised region satisfies the following formula: in, Indicates the first The first local detection region The degree of protrusion of each raised area This indicates the total number of raised areas on the surface of the semi-finished aluminum alloy flat tube to be inspected. Indicates the first The total number of suspected protrusions in each local detection area Indicates the first The number of suspected protrusions in each protruding area Indicates the first The first local detection region The average height of the suspected protrusions in the protruding areas This represents a normalization function, such as max-min normalization, used to normalize... Map to the interval [0, 1].
[0094] In this formula, This represents a moderating factor based on the overall background. This reflects an abnormal global universality. This reflects an abnormal local concentration, when It's very big. When the value is very small, it indicates that the anomaly is widespread globally but sparse locally, and the assessment of the protrusion depends more on the relative density ratio of each protruding region within the detection point (i.e., Conversely, when Very small When the value is very large, it indicates that the anomaly is globally sparse but locally concentrated. In this case, the assessment of the protrusion degree depends more on the absolute height characteristics of the protruding region itself (i.e., ); Characterizing the first The average severity of each raised area in the height dimension. The larger the value, the more likely it is to be the first. The more pronounced the protrusion of a raised area, the more obvious the physical anomaly.
[0095] The above method comprehensively evaluates the convexity from three dimensions: height, local density, and global distribution. This results in a convex region with a large number of suspected convex points, a high average height, and a sparse distribution having a higher convexity, which is more in line with the judgment logic of the severity of defects in actual production and improves the effectiveness of convexity determination.
[0096] S203. Based on the deviation of the two-dimensional roughness parameter of each local detection area and the protrusion of each protrusion area, determine the possibility that the roughness of each protrusion area is not up to standard.
[0097] In one alternative implementation, the probability that the roughness of the first local detection area is substandard can be determined based on the product of the deviation of the two-dimensional roughness parameter of the first local detection area and the protrusion of each protruding area.
[0098] It is understandable that since the deviation and convexity of the two-dimensional roughness parameters are both less than 1, their product may be too small. Therefore, the product can be normalized again (e.g., by using the maximum normalization method based on historical experience) to obtain the probability that the roughness does not meet the standard.
[0099] The methods provided in S201-S203 above first identify specific protruding regions and evaluate their protrusion degree. Then, based on the two-dimensional roughness parameters, they determine the deviation of the two-dimensional roughness parameters of the local detection region. Features are extracted from two complementary dimensions: "local microscopic anomalies" and "regional macroscopic deviations". Finally, the features of these two dimensions are correlated and fused to determine the possibility of non-compliance. This makes the prediction results take into account both the specific physical morphological anomalies and the compliance status of the overall roughness parameters of the region. This makes the determined possibility of roughness non-compliance more scientific and comprehensive, and the resulting probability index more convincing and instructive.
[0100] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0101] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.
Claims
1. A method for detecting the roughness of an aluminum alloy flat tube, characterized in that, include: Obtain surface inspection data of the semi-finished aluminum alloy flat tube to be inspected. The surface inspection data includes the three-dimensional morphology information and the detected values of the two-dimensional roughness parameters of the semi-finished aluminum alloy flat tube to be inspected. Based on the surface detection data, determine the probability that the roughness of each raised area is substandard; Based on the probability that the roughness of each raised region is substandard, the predetermined value of the detection parameter of each raised region is adjusted to obtain the target value of the detection parameter of each raised region. Roughness detection is performed on the corresponding area of each protrusion in the finished aluminum alloy flat tube based on the target value of the detection parameters of each protrusion area.
2. The roughness detection method for aluminum alloy flat tubes according to claim 1, characterized in that, The determination of the probability that the roughness of each raised area is substandard based on the surface detection data includes: The surface of the semi-finished aluminum alloy flat tube to be inspected is divided into multiple local inspection areas, and the deviation of the two-dimensional roughness parameter of each local inspection area is determined based on the detected value of the two-dimensional roughness parameter of each local inspection area. Based on the three-dimensional topography data, the raised areas on the surface of the semi-finished aluminum alloy flat tube to be tested and the protrusion degree of each raised area are determined. Based on the deviation of the two-dimensional roughness parameter of each local detection area and the protrusion of each protrusion area, the probability of the roughness of each protrusion area failing to meet the standard is determined.
3. The roughness detection method for aluminum alloy flat tubes according to claim 2, characterized in that, Based on the detected values of the two-dimensional roughness parameters in each local detection region, the deviation of the two-dimensional roughness parameters in each local detection region is determined, including: Determine the individual deviation between the detected value and the standard value of each two-dimensional roughness parameter in the first local detection area; The average value of the individual deviations of all two-dimensional roughness parameters in the first local detection area is determined as the deviation of the two-dimensional roughness parameters in the first local detection area.
4. The roughness detection method for aluminum alloy flat tubes according to claim 2, characterized in that, The three-dimensional topography data includes the three-dimensional coordinates of multiple three-dimensional coordinate points. Based on the three-dimensional topography data, determining the raised areas on the surface of the semi-finished aluminum alloy flat tube to be inspected and the protrusion degree of each raised area includes: Three-dimensional coordinate points in the three-dimensional topography data whose height exceeds the preset height threshold are marked as suspected convex points. The height value of a three-dimensional coordinate point is the coordinate component of the three-dimensional coordinate point in the direction perpendicular to the detection surface. All suspected convex points are clustered to obtain multiple clusters; The region defined by the coordinates of suspected convex points within the same cluster is identified as a convex region. The degree of protrusion of each protrusion region is determined based on the height of the suspected protrusion points included in each protrusion region.
5. The roughness detection method for aluminum alloy flat tubes according to claim 4, characterized in that, The determination of the protrusion degree of each protrusion region based on the height of the suspected protrusion points included in each protrusion region includes: Determine the first local detection area and the total number of suspected protrusions in the first local detection area. The first local detection area is the local detection area to which the first protrusion area belongs. The first protrusion area is any protrusion area on the surface of the semi-finished aluminum alloy flat tube to be inspected. Based on the total number of raised areas on the surface of the semi-finished aluminum alloy flat tube to be tested, the total number of suspected raised points in the first local detection area, the number of suspected raised points in the first raised area, and the average height of the suspected raised points in the first raised area, the protrusion degree of the first raised area is determined.
6. The roughness detection method for aluminum alloy flat tubes according to claim 2, characterized in that, The determination of the probability that the roughness of each raised area is substandard, based on the deviation of the two-dimensional roughness parameter of each local detection area and the raisedness of each raised area, includes: The probability that the roughness of the first local detection area is substandard is determined by multiplying the deviation of the two-dimensional roughness parameter of the first local detection area with the protrusion of each protruding area.
7. The roughness detection method for aluminum alloy flat tubes according to claim 1, characterized in that, The step of adjusting the predetermined value of the detection parameter for each raised region based on the probability that the roughness of each raised region is substandard, to obtain the target value of the detection parameter for each raised region, further includes: Obtain the local detection area and protrusion area of other semi-finished aluminum alloy flat tubes belonging to the same batch as the semi-finished aluminum alloy flat tube to be tested; Based on the number of raised areas appearing in the first local detection area among other semi-finished aluminum alloy flat tubes to be inspected, the abnormality ratio of the first local detection area is determined. Based on the probability of the roughness of the first raised region not meeting the standard, the average probability of the roughness of all raised regions included in the first local detection region not meeting the standard, and the abnormality ratio of the first local detection region, the parameter adjustment degree of the first raised region is determined. Based on the parameter adjustment degree of the first protrusion region, the predetermined value of the detection parameter of the first protrusion region is adjusted to obtain the target value of the detection parameter of the first protrusion region.
8. The roughness detection method for aluminum alloy flat tubes according to claim 7, characterized in that, The detection parameters include the spot size. The step of adjusting the predetermined value of the detection parameters of the first protruding region based on the parameter adjustment degree of the first protruding region to obtain the target value of the detection parameters of the first protruding region includes: The target adjustment amount is determined based on the preset basic adjustment amount and the parameter adjustment degree of the first protrusion area; Based on a predetermined value for the spot size, the target adjustment amount is adjusted downwards to obtain a candidate value for the spot size of the first protruding region; The maximum value between the candidate value of the light spot size and the minimum light spot size is determined as the target value of the light spot size in the first protrusion region.
9. The roughness detection method for aluminum alloy flat tubes according to claim 1, characterized in that, The roughness detection of the corresponding area of each raised region in the finished aluminum alloy flat tube based on the target value of the detection parameters of each raised region includes: Based on the preset coordinate mapping relationship from semi-finished product to finished product, the corresponding area of each of the protruding areas on the finished aluminum alloy flat tube is determined. Based on the target value of the detection parameters for each raised region, roughness detection is performed on the corresponding region of each raised region.
10. The roughness detection method for aluminum alloy flat tubes according to claim 7, characterized in that, The method further includes: If the abnormality ratio in the first local detection area is greater than the abnormality ratio threshold, a production abnormality warning message is issued. The production abnormality warning message is used to indicate that the production of the semi-finished aluminum alloy flat tubes to be tested in the same batch is abnormal.