A method for detecting welding seams of roof waterproofing membranes

Abstract

The present invention belongs to the technical field of seam welding detection, and specifically discloses a method for detecting seam welding of roof waterproofing membranes, including: constructing a three-dimensional model of the welding area to be constructed, and marking the type of welding detection points; performing welding feature analysis of the welding area to obtain various welding feature indicators; confirming the number of welding detection points, and marking the welding detection points; confirming that each welding detection point corresponds to a standard passing threshold of each welding detection indicator; building a welding quality inspection model; performing quality inspections in sequence according to a set welding quality inspection plan, and recording quality inspection values; outputting welding quality score values, constructing a welding quality inspection report of waterproofing membranes, and providing feedback. The present invention effectively solves the shortcomings of the current welding detection points in terms of comprehensiveness of setting, fully combines the specific welding characteristics of waterproofing membranes, improves the adaptability of roof waterproofing membrane seam welding detection scenarios, and at the same time ensures the representativeness and rationality of welding detection results.

Description

Technical Field

[0001] The invention belongs to the technical field of seam welding detection, and in particular relates to a method for detecting seam welding of roof waterproofing coiled materials. Background Art

[0002] In construction projects, roof waterproofing is of vital importance. The quality of the seam welding of roof waterproofing membranes directly affects the waterproofing effect of the roof. Since the roof is exposed to the natural environment for a long time, it faces the test of various factors such as sunlight, rain, temperature changes, etc. If there are defects in the seam welding of waterproofing membranes, it is easy to cause leakage problems, which will not only affect the normal use of the building, but also threaten the structural safety of the building. Therefore, strict testing of the seam welding of roof waterproofing membranes is a key link to ensure the quality of roof waterproofing.

[0003] Prior art, such as a portable coil tester and coil testing method disclosed in the Chinese invention patent application with application number 201610009768.4, includes a driving device, a force measuring display device, a mobile clamping device and a fixed clamping device, and an instrument base. The driving device is arranged at one end of the instrument base and is connected to the mobile clamping device through the force measuring display device, and the fixed clamping device is fixedly connected to the other end of the instrument base. The mobile clamping device and the fixed clamping device are used to fix the coil test piece, the driving device is used to pull the mobile clamping device to move away from the fixed clamping device, and the force measuring display device is used to measure and display the peeling force value of the coil test piece. In addition, it can not only detect whether the overlapped edge is reliable or whether there is a problem of poor welding, which greatly improves the inspection efficiency after welding at the construction site, but also has a simple structure, small size and light weight, is easy to carry, and can perform peeling tests at the overlapped edge joints at any time.

[0004] Prior art, such as a method for testing the watertightness of the seams of waterproof membranes disclosed in a Chinese invention patent application with application number 202110109995.5, includes: sample preparation, sample installation, and sample testing. Under the premise of being able to test the watertightness of the lap joints, the pressure is further concentrated on the joints corresponding to the entire slit groove by covering the lap joints with the slit grooves, thereby ensuring that the pressure on the membrane at the lap joint to be tested is uniform, thereby improving the accuracy of the test results. At the same time, by eliminating the height difference of the sample after the overlap, and combining the support of the anti-sag bracket or pad, the test piece is positioned flat, further improving the accuracy of the test results.

[0005] Regarding the above two technical solutions, it is obvious that the current inspection of roof waterproofing membrane seam welding is focused on the inspection method level, and insufficient attention is paid to the preparatory work for the inspection. There are also the following aspects: 1. The comprehensiveness of welding inspection points is not enough: Currently, the number of welding inspection points is not confirmed in combination with the specific welding characteristics of the waterproofing membrane. The existing inspection methods may not be able to fully cover all welding inspection points. Especially for large roofing projects, the selection of inspection points may not be comprehensive enough, resulting in some welding inspection points not being detected, increasing the risk of missed inspections.

[0006] 2. There is still a certain deviation in the reliability of the test results. The current unified quality assessment standard for sampling does not incorporate information such as the specific location of the welding test points to revise the welding standards, resulting in insufficient representativeness and rationality of the welding test results. Summary of the invention

[0007] In view of this, in order to solve the problems raised in the above background technology, a method for detecting the welding seams of roof waterproofing membranes is now proposed.

[0008] The purpose of the present invention can be achieved through the following technical solutions: The present invention provides a method for detecting the welding seams of roof waterproofing membranes, the method comprising: S1, marking the type of welding detection points: performing a three-dimensional scan of the welding area of ​​the waterproofing membrane, constructing a three-dimensional model of the welding area to be constructed, and marking the type of welding detection points.

[0009] S2. Welding feature confirmation: Perform welding feature analysis on the welding area to obtain various welding feature indicators.

[0010] S3. Confirmation of the number of inspection points: confirming the number of welding inspection points and marking the welding inspection points on the three-dimensional model.

[0011] S4. Confirmation of welding inspection standards: Extract the 3D models and tracking records of each laid roof waterproofing membrane from the database, and import the meteorological tracking records of the area where the current roof is located, so as to confirm the standard passing threshold of each welding inspection point corresponding to each welding inspection indicator.

[0012] S5. Welding quality inspection setting: Build a welding quality inspection model based on the standard passing threshold of each welding inspection point corresponding to each welding inspection indicator.

[0013] S6. Welding quality inspection and record: Based on the location of each welding inspection point, quality inspection is carried out in sequence according to the set welding quality inspection plan, and the quality inspection value of each welding inspection point corresponding to each welding inspection indicator is recorded.

[0014] S7, report generation and feedback: import the data recorded in step S6 into the welding quality inspection model, output the welding quality score, and build the welding quality inspection report of the waterproof membrane based on the data recorded in step S6 and the welding quality score, and provide feedback.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) The present invention effectively solves the shortcomings of the current welding inspection points in terms of comprehensiveness by analyzing the welding characteristics of the welding area and confirming the number of welding inspection points. It fully combines the specific welding characteristics of the waterproof membrane, and can cover all welding inspection points as comprehensively as possible, thereby improving the adaptability of the roof waterproof membrane joint welding inspection scene. Especially for large roof projects, it can ensure the representativeness and reliability of the selection of inspection points, and minimize the probability of missed inspection and the risk of missed inspection of welding inspection points on the basis of ensuring inspection efficiency.

[0016] (2) The present invention combines welding characteristic indicators, welding inspection point location types, and the three-dimensional model and tracking records of the laid roof waterproofing membrane to confirm the standard passing threshold of each welding inspection point corresponding to each welding inspection indicator, thereby effectively compensating for the deviation in the reliability of current inspection results, breaking the limitations of current fixed and unified quality assessment standards, and thus ensuring the representativeness and rationality of welding inspection results.

[0017] (3) The present invention can standardize the detection process and realize flexible adjustment of the detection indicators by setting the detection setting compensation factor when confirming that each welding detection point corresponds to the standard passing threshold of each welding detection indicator, thereby ensuring the representativeness and rationality of the welding detection results.

[0018] (4) The present invention can realize efficient, accurate and reliable automatic detection of welding joints by building a primary quality inspection sub-model to evaluate the degree of quality inspection consistency and building a secondary quality inspection sub-model to output the score. This double-layer detection mechanism not only improves the accuracy and reliability of the detection, but also significantly improves the detection efficiency. It can effectively ensure the efficiency of the welding detection of the roof waterproofing membrane seams and the accuracy of the detection results. At the same time, by using the output of the primary quality inspection sub-model as the input of the secondary quality inspection sub-model, the continuity and coordination of the detection process are guaranteed. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0020] Figure 1 The present invention is a schematic flow chart of the steps for implementing the method.

[0021] Figure 2 This is a schematic diagram of the welding inspection point type labeling process of the present invention. DETAILED DESCRIPTION

[0022] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0023] See also Figure 1 As shown, the present invention provides a method for detecting welding seams of roof waterproofing membranes, the method comprising: S1, marking welding detection point types: performing a three-dimensional scan on the welding area of ​​the waterproofing membrane, constructing a three-dimensional model of the welding area to be constructed, and marking the welding detection point types.

[0024] S2. Welding feature confirmation: Perform welding feature analysis on the welding area to obtain various welding feature indicators.

[0025] S3. Confirmation of the number of inspection points: confirming the number of welding inspection points and marking the welding inspection points on the three-dimensional model.

[0026] S4. Confirmation of welding inspection standards: Extract the 3D models and tracking records of each laid roof waterproofing membrane from the database, and import the meteorological tracking records of the area where the current roof is located, so as to confirm the standard passing threshold of each welding inspection point corresponding to each welding inspection indicator.

[0027] S5. Welding quality inspection setting: Build a welding quality inspection model based on the standard passing threshold of each welding inspection point corresponding to each welding inspection indicator.

[0028] S6. Welding quality inspection and record: Based on the location of each welding inspection point, quality inspection is carried out in sequence according to the set welding quality inspection plan, and the quality inspection value of each welding inspection point corresponding to each welding inspection indicator is recorded.

[0029] S7, report generation and feedback: import the data recorded in step S6 into the welding quality inspection model, output the welding quality score value, and build the welding quality inspection report of the waterproof membrane based on the data recorded in step S6 and the welding quality score value, and provide feedback.

[0030] Understandably, regarding step S1, it should be added that the three-dimensional scanning of the welding area of ​​the waterproof membrane can be performed using a high-precision three-dimensional scanner, such as a handheld laser scanner or a structured light scanner, and the welding area of ​​the waterproof membrane can be scanned from multiple angles to ensure that all key parts are covered. At the same time, three-dimensional modeling software can be used to splice scanning data from different perspectives into a complete three-dimensional model. By way of example, the three-dimensional modeling software is specifically AutoCAD, SolidWorks, Blender, etc., and the scanning equipment and three-dimensional modeling software are not specifically limited here.

[0031] In a specific embodiment, a specific supplementary example of the three-dimensional modeling step is as follows: 1) Importing scan data: importing the point cloud data generated by the scanner into the three-dimensional modeling software.

[0032] 2) Data processing: De-noise and smooth the point cloud data to remove unnecessary data points.

[0033] 3) Model reconstruction: Generate a three-dimensional model based on point cloud data.

[0034] It should be added that building a three-dimensional model in an actual scene also includes optimizing the model. Since the present invention adopts the existing relatively mature three-dimensional model building means, the specific construction process and optimization process of the three-dimensional model are omitted here.

[0035] It should also be added that the connection point type marking in the area to be welded is mainly carried out through 3D marking software such as Revit, Navisworks, etc.

[0036] Understandably, see Figure 2 As shown, the welding inspection point type marking in step S1 includes: identifying the welding starting point position, each corner welding inspection point position, each lap welding inspection point position, each structural welding inspection point position and each edge welding inspection point position of the welding area from the three-dimensional model, and making corresponding markings in the three-dimensional model, and adding Class I welding inspection point type at the marked position.

[0037] The positions of the plane welding inspection points and the center welding inspection point are identified from the three-dimensional model, and corresponding annotations are made in the three-dimensional model. At the same time, the type of Class II welding inspection point is added at the annotated position.

[0038] It should be added that the plane welding point refers to the welding performed in a flat area without special structures, the edge welding point refers to the welding point between the edge of the coil and other coils or substrates, the starting welding point refers to the starting point of the waterproof coil laying, and the corner welding point refers to the welding point of the coil at the corner. The lap welding point refers to the welding point at the overlap between two coils, the structural welding point refers to the welding point of the coil in special structures such as drains, vents, wall pipes, etc., the center welding point refers to the welding point in the middle part of the coil, and the plane welding point refers to the welding performed in a flat area without special structures.

[0039] It should also be added that the edge welding point is the key part of the waterproofing membrane laying, which is easily affected by the external environment, such as rain, wind pressure, etc. The welding quality of the starting welding point directly affects the subsequent construction quality and the overall waterproofing effect. The corner welding point is a stress concentration area, which is easily affected by mechanical damage and water flow impact. The lap welding point is the main connection part of the waterproofing membrane laying, and its welding quality directly affects the overall continuity of the waterproof layer. These locations of structural welding points are prone to become weak links for leakage, while the center welding point is usually located in the middle part of the membrane, especially in large flat areas. Although it is not as important as the edge welding point and the corner welding point, it is also necessary to ensure the welding quality to prevent water penetration. The stress and environmental impact of the plane welding point is relatively small, but it is still an important part of the waterproofing system. In summary, the type I welding point type and the type II welding point type are distinguished in terms of the degree of influence, and the degree of influence of the type I welding point type is higher than that of the type II welding point type.

[0040] It can be understood that the welding feature analysis of the welding area described in step S2 includes: S21, identifying the edge line, corner position and overlap area in the three-dimensional model, counting the curvature of each edge welding detection point and each corner welding detection point, and at the same time counting the thickness of each overlap area, and calculating the welding complexity of the welding area, which is recorded as .

[0041] In a specific embodiment, the three-dimensional model of the welding area can be obtained by using Revit software.

[0042] S22, identifying the positions of the boundary points of the welding area in the three-dimensional model, identifying the three-dimensional contour volume of the welding area in the three-dimensional model, and calculating the minimum circumscribed rectangular volume and the minimum circumscribed spherical volume of the welding area.

[0043] S23, the three-dimensional contour volume, the minimum circumscribed rectangular volume and the minimum circumscribed spherical volume of the welding area are respectively recorded as , and , calculate the degree of irregularity of the welding area , , To set the volume overlap ratio.

[0044] In a specific embodiment, The specific value can be 0.5.

[0045] S24, identifying the contour of the welding area in the three-dimensional model, and counting the contour perimeter of the welding area, recorded as ,Will , and As an indicator of each welding characteristic.

[0046] Furthermore, the welding complexity of the welding area is calculated in step S21, including: B1, calculating the mean of the curvature of each edge welding detection point, and calculating the average curvature of each edge welding detection point to obtain the average edge curvature.

[0047] B2. Compare the curvature of each edge welding detection point with the set interference curvature, count the number of edge welding detection points whose curvature is greater than the set interference curvature, and divide it by the number of edge welding detection points to obtain the ratio of the curvature exceeding the number of edge welding detection points.

[0048] B3. Normalize the average edge curvature and the ratio of the number of welding detection points beyond the edge, import the normalized result into the Sigmoid function, and use the output result as the edge complexity, which is recorded as .

[0049] It should be added that the area with large curvature is prone to become a stress concentration point, so the larger the curvature, the more complex the geometry of the welding area.

[0050] It should also be added that the normalization processing can be specifically performed by using a maximum-minimum normalization processing method.

[0051] B4. Calculate the mean value of the curvature of each corner welding detection point, and calculate the average curvature of each corner welding detection point to obtain the average corner curvature.

[0052] B5. Compare the curvature of each corner welding detection point with the set interference curvature. Define the ratio of the number of welding detection points exceeding the corner in the same way as the ratio of the number of curvature exceeding the edge welding detection points. Calculate the corner complexity in the same way, which is recorded as .

[0053] B6. Calculate the average thickness of each overlapping area to obtain the average thickness of the overlapping area, which is recorded as , calculate the overlap area complexity, denoted as , , Indicates the reference thickness, is a natural constant.

[0054] It should be added that thick materials require longer welding time. At the same time, thick materials may produce greater internal stress during the welding process. In addition, increased thickness will lead to increased concealment of internal defects, and the probability of welding problems will increase. Therefore, the thicker the overlap area, the more complicated it will be.

[0055] B7. Set welding complexity compensation factor , and calculate the welding complexity of the welding area , .

[0056] It should be noted that the maximum value is used to count the welding complexity of the welding area because there is interference among the complexities of the three. If the sum of the three is used as the welding complexity of the welding area, the welding complexity of the welding area will be magnified at the numerical level. At the same time, taking the average will weaken the complexity of a certain part. Therefore, the maximum value is selected and the final value is guaranteed not to exceed 1 to facilitate subsequent calculations. Directly selecting the maximum value as the welding complexity of the welding area ignores the influence of other types of welding complexity. Therefore, the mean value and the number of items exceeding the reference complexity are used to set the welding complexity compensation factor to ensure the representativeness and reliability of the welding complexity calculation results of the welding area.

[0057] The welding complexity compensation factor is set in step B7, including: comparing the edge complexity, corner complexity and overlap area complexity with the preset reference structure complexity. For comparison, count the number of items that exceed the reference complexity, recorded as .

[0058] Set welding complexity compensation factor , .

[0059] In a specific embodiment, The specific value can be 0.5.

[0060] The embodiment of the present invention can standardize the detection process and achieve flexible adjustment of the detection indicators by setting the detection setting compensation factor when confirming that each welding detection point corresponds to the standard passing threshold of each welding detection indicator, thereby ensuring the representativeness and rationality of the welding detection results.

[0061] Understandably, the step S3 of confirming the number of welding inspection points includes: S31, extracting from each welding characteristic index and and .

[0062] S32, set the detection density compensation factor , , and Respectively represent the weights of welding complexity and welding profile, , and Set the reference welding complexity and welding profile respectively.

[0063] It should be added that structures with high welding complexity usually contain multiple welds and complex welding paths. This complexity increases the uncertainty and potential defect risks in the welding process, such as lack of fusion, pores, cracks, etc. At the same time, complex welded structures may require different welding parameters, such as current, voltage, speed, etc., which increases the difficulty and uncertainty of the welding process. Structures with high welding irregularity will indeed increase the difficulty of detection, because the irregular shape may make it difficult for the detection equipment to accurately capture the position and state of the weld. However, welding irregularity is a local influence, that is, welding irregularity mainly affects the detection effect of the local area, while welding complexity affects the quality of the entire welded structure. Therefore, the weight of welding complexity is set higher than the weight of welding irregularity. For example, The specific value can be 0.6. The specific value can be 0.4.

[0064] In a specific embodiment, The specific value can be 0.5. The specific value can be 0.6.

[0065] S33, the contour perimeter of the welding area The area perimeters corresponding to the set reference detection densities are matched and compared to obtain the matched reference detection density, which is used as the preliminary detection density and recorded as ,Will As the target detection density ,Will As the number of welding inspection points, Indicates the round-up symbol.

[0066] The embodiment of the present invention effectively solves the shortcomings of the current welding inspection points in terms of comprehensiveness of setting by analyzing the welding characteristics of the welding area and confirming the number of welding inspection points. It fully combines the specific welding characteristics of the waterproof membrane, and can cover all welding inspection points as comprehensively as possible, thereby improving the adaptability of the roof waterproof membrane joint welding inspection scene. Especially for large-scale roof projects, it can ensure the representativeness and reliability of the selection of inspection points, and reduce the missed detection probability and the risk of missed detection of welding inspection points as much as possible on the basis of ensuring the detection efficiency.

[0067] Understandably, the step S4 of confirming that each welding inspection point corresponds to each welding inspection index standard passing threshold includes: S41, based on the meteorological tracking record of the area where the current roof is located, calculating the meteorological state interference factor of the welding area, recorded as .

[0068] S42. Based on the three-dimensional model and tracking records of each laid roof waterproofing membrane, set the detection setting compensation factor of each welding detection point corresponding to each welding detection index, which is recorded as , Indicates the welding inspection point number, , Indicates the welding inspection index number, .

[0069] S43. If a welding inspection point is marked as a Class I welding inspection point type, the welding position correction factor of the welding inspection point is recorded as If a welding inspection point is marked as a Class II welding inspection point, the welding position correction factor of the welding inspection point is recorded as , and the welding position correction factor of each welding detection point is obtained, which is recorded as , The value is or , .

[0070] It should be added that The specific value can be 0.6. The specific value can be 0.4.

[0071] S44, extracting the initial passing threshold of each welding inspection index corresponding to each welding inspection point from the database, recorded as , calculate the standard passing threshold of each welding inspection index corresponding to each welding inspection point, and record it as , , , and They represent the proportion of meteorological interference factor, detection setting compensation factor and welding position correction factor respectively. .

[0072] It should be added that welding detection indicators include but are not limited to the number of cracks, maximum crack width, maximum crack depth and number of pores. The values ​​of these detection indicators tend to be as small as possible. Therefore, when setting the passing threshold, the smaller the better. Therefore, the initial value is selected as the benchmark value, and the corresponding reduction is made on the basis of the benchmark value. When it comes to the size detection of the welding gap, in order to maintain the stability and consistency of the subsequent model, the difference between the actual value of the corresponding size of the welding gap and the initial setting value can be selected for the passing threshold analysis.

[0073] It should also be added that the meteorological interference factor mainly considers the impact of environmental factors such as temperature, humidity, wind speed, etc. on the welding test results. The detection setting compensation factor mainly considers the subsequent welding tracking of the same type of waterproofing membrane, while the welding position correction factor considers the complex situation of the location of the welding test point itself. Because the weather is changeable and the controllability is relatively small, the meteorological interference factor is set to have the smallest proportion, which can be set to 0.2. The subsequent welding tracking of the same type of waterproofing membrane directly displays and predicts the subsequent stability of the welding test point. Therefore, the detection setting compensation factor has the largest proportion, which can be set to 0.5. The detection setting compensation factor also directly affects the setting of the standard passing threshold of the welding test index, so the detection setting compensation factor has the second largest proportion, which can be set to 0.3.

[0074] The embodiment of the present invention confirms the standard passing threshold of each welding inspection point corresponding to each welding inspection indicator by combining welding characteristic indicators, welding inspection point location type, and the three-dimensional model and tracking records of the laid roof waterproofing membrane, thereby effectively compensating for the deviation of current inspection results at the reliability level, breaking the limitations of current fixed and unified quality assessment standards, and thus ensuring the representativeness and rationality of welding inspection results.

[0075] Furthermore, the meteorological state interference factor of the welding area is calculated in step S41, including: extracting the average monitored temperature, average monitored humidity and average wind speed of each monitoring month from the meteorological tracking record data.

[0076] The interference temperature range, interference humidity range and interference wind speed range of the waterproof membrane joints are extracted from the database.

[0077] The average monitored temperature of each monitoring month is compared with the interference temperature range. The monitoring months within the interference temperature range are recorded as temperature interference monitoring months. The number of temperature interference monitoring months is counted and compared with the number of monitoring months to obtain the temperature interference monitoring ratio.

[0078] The humidity interference monitoring ratio and wind speed interference monitoring ratio are defined in the same way as the temperature interference monitoring ratio, and the average of the temperature interference monitoring ratio, the humidity interference monitoring ratio and the wind speed interference monitoring ratio is taken as the meteorological condition interference factor of the welding area.

[0079] Exemplarily, the meteorological state interference factor=(temperature interference monitoring ratio+humidity interference monitoring ratio+wind speed interference monitoring ratio) / 3.

[0080] It should be added that, in a specific embodiment, example values ​​of the interference temperature range may be below 10°C and above 30°C, example values ​​of the interference humidity range may be 80% to 95%, and example values ​​of the interference wind speed range may be 5 meters per second to 8 meters per second.

[0081] Furthermore, in step S42, a compensation factor is set for each welding inspection point corresponding to each welding inspection index, including: calculating the similarity between the three-dimensional model of each laid roofing waterproofing membrane and the three-dimensional model of the welding area to be built through a similarity measurement algorithm, and obtaining the similarity between each laid roofing waterproofing membrane and the three-dimensional model corresponding to the welding area to be built.

[0082] The three-dimensional model of the laid roofing waterproofing membrane with a similarity greater than the set reference similarity is recorded as a similar model.

[0083] The tracking quality inspection values ​​of each welding inspection point corresponding to each welding inspection index at each tracking are extracted from the tracking records of each similar model.

[0084] With the tracking order as the horizontal axis and the tracking quality inspection value as the vertical axis, a tracking quality inspection value change curve of each welding inspection point corresponding to each welding inspection index in each similar model is constructed, and the slope and amplitude are extracted from it as the quality inspection change rate and quality inspection change value, respectively.

[0085] The quality inspection change rate and quality inspection change value are respectively averaged to obtain the average quality inspection change rate and average quality inspection change value of each welding inspection index corresponding to each welding inspection point, which are recorded as and .

[0086] Set the compensation factor for each welding inspection point corresponding to each welding inspection index , , and The set The reference quality inspection change rate and reference quality inspection change value of each welding inspection indicator.

[0087] In a specific embodiment, the similarity measurement algorithm described above can specifically select one or more similarity measurement algorithms such as Euclidean distance or cosine similarity. When selecting Euclidean distance or cosine similarity, the corresponding points between the laid roof waterproofing membrane and the three-dimensional model corresponding to the welding area to be built can be found by using feature matching algorithms such as RANSAC, ICP, etc., and a feature vector is constructed. Based on the constructed feature vector, it is then imported into the selected similarity measurement algorithm, and finally the similarity between each laid roof waterproofing membrane and the three-dimensional model corresponding to the welding area to be built is output.

[0088] In another specific embodiment, the reference similarity may be set to a value of 0.8.

[0089] It can be understood that the welding quality inspection model described in step S5 is composed of a primary quality inspection sub-model and a secondary quality inspection sub-model, and the output of the primary quality inspection sub-model is the input of the secondary quality inspection sub-model.

[0090] Among them, the expression formula of a quality inspection sub-model is as follows: , Indicates the floor symbol, Indicates The welding inspection point corresponds to The quality inspection value of each welding inspection index, It represents the quality inspection consistency index. Indicates the number of welding inspection points. Indicates the number of welding inspection indicators.

[0091] The expression formula of the secondary quality inspection sub-model is as follows: , is the set rated welding quality score, It is the output welding quality score value.

[0092] It is also understandable that what needs to be added in step S7 is that the specific output process of outputting the welding quality score is: importing the quality inspection value of each welding inspection index corresponding to each welding inspection point into the primary quality inspection sub-model, and then using the output of the primary quality inspection sub-model as the input of the secondary quality inspection sub-model, and then outputting the welding quality score.

[0093] The embodiment of the present invention can realize efficient, accurate and reliable automatic detection of welding joints by building a primary quality inspection sub-model to evaluate the degree of quality inspection consistency and building a secondary quality inspection sub-model to output the score. This double-layer detection mechanism not only improves the accuracy and reliability of detection, but also significantly improves the detection efficiency, which can effectively ensure the efficiency of roof waterproofing membrane seam welding detection and the accuracy of detection results. At the same time, by using the output of the primary quality inspection sub-model as the input of the secondary quality inspection sub-model, the continuity and coordination of the detection process are guaranteed.

[0094] The above contents are merely examples and explanations of the concept of the present invention. Those skilled in the art may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, they shall all fall within the protection scope of the present invention.

Claims

1. A method for detecting welding seams of roof waterproofing membranes, characterized in that: include: S1. Welding inspection point type marking: perform 3D scanning on the welding area of ​​the waterproof membrane, build a 3D model of the welding area to be built, and mark the welding inspection point type; S2. Welding feature confirmation: Welding feature analysis of the welding area is performed to obtain various welding feature indicators; each welding feature indicator includes the welding complexity of the welding area, the irregularity of the welding area, and the contour perimeter of the welding area, which are recorded as , , ; S3. Confirmation of the number of inspection points: confirming the number of welding inspection points and marking the welding inspection points on the three-dimensional model; The confirmation of the number of welding inspection points includes: Extract from each welding characteristic index and and ; Set the detection density compensation factor , , and Respectively represent the weights of welding complexity and welding profile, , and Set the reference welding complexity and welding profile respectively; The perimeter of the weld area The area perimeters corresponding to the set reference detection densities are matched and compared to obtain the matched reference detection density, which is used as the preliminary detection density and recorded as ,Will As the target detection density ,Will As the number of welding inspection points, Indicates the rounding up symbol; S4. Confirmation of welding inspection standards: extract the three-dimensional model and tracking records of each laid roof waterproofing membrane from the database, and import the meteorological tracking records of the area where the current roof is located, and then confirm the standard passing threshold of each welding inspection point corresponding to each welding inspection index; S5. Welding quality inspection setting: Build a welding quality inspection model based on the standard passing threshold of each welding inspection point corresponding to each welding inspection index; S6. Welding quality inspection and record: Based on the location of each welding inspection point, quality inspection is carried out in sequence according to the set welding quality inspection plan, and the quality inspection value of each welding inspection point corresponding to each welding inspection index is recorded; S7, report generation and feedback: import the data recorded in step S6 into the welding quality inspection model, output the welding quality score value, and build the welding quality inspection report of the waterproof membrane based on the data recorded in step S6 and the welding quality score value, and provide feedback.

2. A method for detecting welding seams of roof waterproofing membranes according to claim 1, characterized in that: The welding inspection point type marking includes: Identify the welding starting point position, each corner welding detection point position, each lap welding detection point position, each structural welding detection point position and each edge welding detection point position of the welding area from the three-dimensional model, and make corresponding annotations in the three-dimensional model, and add the type of Class I welding detection point at the annotated position; The positions of the plane welding inspection points and the center welding inspection point are identified from the three-dimensional model, and corresponding annotations are made in the three-dimensional model. At the same time, the type of Class II welding inspection point is added at the annotated position.

3. A method for detecting welding seams of roof waterproofing membranes according to claim 1, characterized in that: The welding characteristic analysis of the welding area includes: Identify the edge line, corner position and overlap area in the three-dimensional model, count the curvature of each edge welding detection point and each corner welding detection point, count the thickness of each overlap area, and calculate the welding complexity of the welding area, which is recorded as ; Identify the positions of each boundary point of the welding area in the three-dimensional model, identify the three-dimensional contour volume of the welding area in the three-dimensional model, and calculate the minimum circumscribed rectangular volume and the minimum circumscribed spherical volume of the welding area; The three-dimensional contour volume, the minimum circumscribed rectangular volume, and the minimum circumscribed spherical volume of the welding area are respectively denoted as , and , calculate the degree of irregularity of the welding area , , To set the volume overlap ratio; Identify the contour of the welding area in the three-dimensional model, and count the contour perimeter of the welding area, which is recorded as ,Will , and As an indicator of each welding characteristic.

4. A method for detecting welding seams of roof waterproofing membranes according to claim 3, characterized in that: The calculation of the welding complexity of the welding area includes: The curvature of each edge welding detection point is averaged and the average curvature of each edge welding detection point is calculated to obtain the average edge curvature; The curvature of each edge welding detection point is compared with the set interference curvature, the number of edge welding detection points whose curvature is greater than the set interference curvature is counted, and the number is divided by the number of edge welding detection points to obtain the ratio of the curvature exceeding the number of edge welding detection points; The average edge curvature and the ratio of the number of welding detection points beyond the edge are normalized, and the normalized results are introduced into the Sigmoid function. The output result is used as the edge complexity and recorded as ; The curvature of each corner welding detection point is averaged and the average curvature of each corner welding detection point is calculated to obtain the average corner curvature; The curvature of each corner welding detection point is compared with the set interference curvature. The ratio of the number of welding detection points exceeding the corner is defined in the same way as the ratio of the number of curvature exceeding the edge welding detection points. The corner complexity is calculated in the same way, which is recorded as ; The thickness of each overlapping area is averaged and the average thickness of the overlapping area is obtained, which is recorded as , calculate the overlap area complexity, denoted as , , Indicates the reference thickness, is a natural constant; Set welding complexity compensation factor , and calculate the welding complexity of the welding area , .

5. A method for detecting welding seams of roof waterproofing membranes according to claim 4, characterized in that: The setting of the welding complexity compensation factor includes: The edge complexity, corner complexity and overlap area complexity are compared with the preset reference structure complexity. For comparison, count the number of items that exceed the reference complexity, recorded as ; Set welding complexity compensation factor , .

6. A method for detecting welding seams of roof waterproofing membranes according to claim 1, characterized in that: The confirmation of the standard passing threshold value of each welding inspection point corresponding to each welding inspection index includes: Based on the meteorological tracking records of the current roof area, the meteorological interference factor of the welding area is calculated and recorded as ; Based on the three-dimensional model and tracking records of each laid roof waterproofing membrane, the detection setting compensation factor of each welding detection point corresponding to each welding detection index is set, which is recorded as , Indicates the welding inspection point number, , Indicates the welding inspection index number, ; If a welding inspection point is marked as a Class I welding inspection point, the welding position correction factor of the welding inspection point is recorded as If a welding inspection point is marked as a Class II welding inspection point, the welding position correction factor of the welding inspection point is recorded as , and the welding position correction factor of each welding detection point is obtained, which is recorded as , The value is or , ; The initial passing threshold of each welding inspection index corresponding to each welding inspection point is extracted from the database and recorded as , calculate the standard passing threshold of each welding inspection index corresponding to each welding inspection point, and record it as , , , and They represent the proportion of meteorological interference factor, detection setting compensation factor and welding position correction factor respectively. .

7. A method for detecting welding seams of roof waterproofing membranes according to claim 6, characterized in that: The calculation of the meteorological state interference factor of the welding area includes: Extracting the average monitored temperature, average monitored humidity and average wind speed of each monitored month from the meteorological tracking record data; Extract the interference temperature range, interference humidity range and interference wind speed range of the waterproof membrane joint from the database; The average monitoring temperature of each monitoring month is compared with the interference temperature interval, and the monitoring months in the interference temperature interval are recorded as temperature interference monitoring months. The number of temperature interference monitoring months is counted and compared with the number of monitoring months to obtain the temperature interference monitoring ratio; The humidity interference monitoring ratio and wind speed interference monitoring ratio are defined in the same way as the temperature interference monitoring ratio, and the average of the temperature interference monitoring ratio, the humidity interference monitoring ratio and the wind speed interference monitoring ratio is taken as the meteorological condition interference factor of the welding area.

8. A method for detecting welding seams of roof waterproofing membranes according to claim 6, characterized in that: The setting of the detection compensation factor of each welding detection point corresponding to each welding detection index includes: The three-dimensional model of each laid roof waterproofing membrane and the three-dimensional model of the welding area to be built are calculated by a similarity measurement algorithm to obtain the similarity between the three-dimensional model of each laid roof waterproofing membrane and the welding area to be built; The three-dimensional model of the laid roof waterproofing membrane with a similarity greater than the set reference similarity is recorded as a similar model; Extract the tracking quality inspection value of each welding inspection point corresponding to each welding inspection index at each tracking from the tracking records of each similar model; With the tracking order as the horizontal coordinate and the tracking quality inspection value as the vertical coordinate, a tracking quality inspection value change curve corresponding to each welding inspection index of each welding inspection point in each similar model is constructed, and the slope and amplitude are extracted from it as the quality inspection change rate and quality inspection change value respectively; The quality inspection change rate and quality inspection change value are respectively averaged to obtain the average quality inspection change rate and average quality inspection change value of each welding inspection index corresponding to each welding inspection point, which are recorded as and ; Set the compensation factor for each welding inspection point corresponding to each welding inspection index , , and The set The reference quality inspection change rate and reference quality inspection change value of each welding inspection indicator.

9. A method for detecting welding seams of roof waterproofing membranes according to claim 6, characterized in that: The welding quality inspection model consists of a primary quality inspection sub-model and a secondary quality inspection sub-model, and the output of the primary quality inspection sub-model is the input of the secondary quality inspection sub-model; Among them, the expression formula of a quality inspection sub-model is as follows: , Indicates the floor symbol, Indicates The welding inspection point corresponds to The quality inspection value of each welding inspection index, It represents the quality inspection consistency index. Indicates the number of welding inspection points. Indicates the number of welding inspection indicators; The expression formula of the secondary quality inspection sub-model is as follows: , is the set rated welding quality score, The output welding quality score value.

Images