Reactor defect detection method and system based on 3D scanning technology
By using 3D scanning technology and three-dimensional voxel grid construction method in the defect detection of urea ammonia hydrolysis reactor, the efficiency and accuracy of large-scale point cloud data processing are solved, and more efficient and accurate defect detection is achieved.
Patent Information
- Application Number
- CN202411612696.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-13
- Publication Date
- 2025-02-11
AI Technical Summary
In the defect detection of urea ammonia hydrolysis reactors, the prior art has high computing resource consumption and time cost when processing large-scale point cloud data, and data noise and interference lead to missed or misdetection.
By 3D scanning of the urea ammonia hydrolysis reactor, a three-dimensional voxel grid is constructed, the data difference deviation in the voxel grid is determined, sampling point cloud data is extracted, defect point cloud domain is identified, and a three-dimensional defect calibration model is generated. Improve detection accuracy.
The data scale of three-dimensional point cloud data is reduced, the computing complexity and storage requirements are significantly reduced, the accuracy and efficiency of defect detection are improved, and misjudgment caused by data noise is avoided.
Smart Images
Figure CN119413969A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of defect detection technology, and more specifically, to a reactor defect detection method and system based on 3D scanning technology. Background Art
[0002] Reactor defect detection based on 3D scanning is an advanced non-contact detection technology that uses a high-precision 3D scanner to scan the inside of the reactor in detail and obtain its three-dimensional point cloud data. This process can not only capture the geometric features of the reactor surface, but also identify potential defects such as cracks, corrosion and deformation by analyzing this data.
[0003] Reactor defect detection based on 3D scanning is particularly important in the detection of urea ammonia hydrolysis reactors, because the reactor works under high temperature and high pressure environment, which is prone to material degradation and structural damage. By comparing the obtained point cloud data with the design model of the reactor, the defects can be accurately located and quantified, providing a basis for subsequent maintenance and safety assessment. However, in the prior art, when processing large-scale point cloud data, the consumption of computing resources and time costs may increase significantly, and there is noise and interference in the point cloud data with a large amount of data, which makes the detection of urea ammonia hydrolysis reactors miss detection or misdetection. Therefore, how to reduce the data scale of the three-dimensional point cloud data of the urea ammonia hydrolysis reactor and accurately extract the reactor defect features to improve the accuracy of urea ammonia hydrolysis reactor defect detection is a difficult problem faced by the industry. Summary of the invention
[0004] The present application provides a reactor defect detection method and system based on 3D scanning technology, which can reduce the data scale of three-dimensional point cloud data of a urea to ammonia hydrolysis reactor and accurately extract reactor defect features to improve the accuracy of defect detection in the urea to ammonia hydrolysis reactor.
[0005] In a first aspect, the present application provides a reactor defect detection method based on 3D scanning technology, the detection method comprising the following steps: Performing a 3D scan on a urea-ammonia hydrolysis reactor, and then obtaining original three-dimensional point cloud data of the urea-ammonia hydrolysis reactor according to the 3D scanning result; Constructing a three-dimensional voxel grid of the original three-dimensional point cloud data based on a set resolution, determining the data difference deviation corresponding to each point cloud data in the voxel grid for each voxel grid, determining the sampling point cloud data of the voxel grid according to all the data difference deviations, determining all defective point cloud domains of the urea to ammonia hydrolysis reactor from the sampling point cloud data of each voxel grid, and then determining the reactor defect features corresponding to each defective point cloud domain; Acquire geometric shape parameters of the urea-ammonia hydrolysis reactor, and generate a three-dimensional defect calibration model of the urea-ammonia hydrolysis reactor based on the geometric shape parameters, all sampling point cloud data and defect characteristics of each reactor; A defect detection report of the urea-to-ammonia hydrolysis reactor is generated according to the three-dimensional defect calibration model.
[0006] In this embodiment, a 3D scanning device with anti-rust and anti-corrosion functions is used to perform 3D scanning on the urea-to-ammonia hydrolysis reactor.
[0007] In this embodiment, the original three-dimensional point cloud data of the urea to ammonia hydrolysis reactor is obtained according to the 3D scanning result by converting the format of the three-dimensional data in the 3D scanning result to obtain the original three-dimensional point cloud data of the urea to ammonia hydrolysis reactor.
[0008] In this embodiment, constructing a three-dimensional voxel grid of the original three-dimensional point cloud data based on the set resolution specifically includes: Determine voxel grid parameters according to the set resolution; Dividing the three-dimensional space into a plurality of voxel grids according to the voxel grid parameters, thereby obtaining a voxel grid; The original three-dimensional point cloud data is mapped into the voxel grid to obtain the three-dimensional voxel grid of the original three-dimensional point cloud data.
[0009] In this embodiment, determining the sampling point cloud data of the voxel grid according to all data difference deviations specifically includes: Determine the data difference deviation threshold through all data difference deviations; Point cloud data with a data difference deviation greater than the data difference deviation threshold are eliminated, thereby obtaining the sampling point cloud data of the voxel grid.
[0010] In this embodiment, all defect point cloud domains of the urea-to-ammonia hydrolysis reactor are determined by sampling point cloud data of each voxel grid, specifically including: Establishing three-dimensional sampling point cloud data of the urea-to-ammonia hydrolysis reactor according to the sampling point cloud data of each voxel grid; Performing defect comparison between the three-dimensional sampling point cloud data and pre-set three-dimensional point cloud data of a reactor to obtain a plurality of defect comparison domains; Determine the defect confidence coefficient for each defect comparison domain; All defect point cloud domains of the urea to ammonia hydrolysis reactor are determined based on the respective defect confidence coefficients.
[0011] In this embodiment, the reactor defect features corresponding to each defect point cloud domain are determined by extracting features from each defect point cloud domain respectively, thereby obtaining the reactor defect features corresponding to each defect point cloud domain.
[0012] In this embodiment, generating the three-dimensional defect calibration model of the urea-to-ammonia hydrolysis reactor according to the geometric shape parameters, all the sampling point cloud data and the defect characteristics of each reactor specifically includes: Generate a basic three-dimensional model of the urea-to-ammonia hydrolysis reactor using the geometric shape parameters and all the sampled point cloud data; The defect features of each reactor are calibrated in the basic three-dimensional model, thereby generating a three-dimensional defect calibration model of the urea-to-ammonia hydrolysis reactor.
[0013] In this embodiment, the defect detection report of the urea-to-ammonia hydrolysis reactor includes basic information of the urea-to-ammonia hydrolysis reactor, reactor defect location, defect type and defect response strategy.
[0014] In a second aspect, the present application provides a reactor defect detection system based on 3D scanning technology, which is used to perform a reactor defect detection method based on 3D scanning technology, and the detection system includes: A 3D scanning module, used to perform 3D scanning on the urea-ammonia hydrolysis reactor, and then obtain original three-dimensional point cloud data of the urea-ammonia hydrolysis reactor according to the 3D scanning result; a defect feature determination module, for constructing a three-dimensional voxel grid of the original three-dimensional point cloud data based on a set resolution, determining, for each voxel grid in the three-dimensional voxel grid, a data difference deviation corresponding to each point cloud data in the voxel grid, determining the sampling point cloud data of the voxel grid according to all the data difference deviations, determining all defect point cloud domains of the urea to ammonia hydrolysis reactor from the sampling point cloud data of each voxel grid, and further determining the reactor defect feature corresponding to each defect point cloud domain; A three-dimensional model generation module, used to obtain geometric shape parameters of the urea-ammonia hydrolysis reactor, and generate a three-dimensional defect calibration model of the urea-ammonia hydrolysis reactor based on the geometric shape parameters, all sampling point cloud data and each reactor defect feature; A defect report generation module is used to generate a defect detection report of the urea-to-ammonia hydrolysis reactor according to the three-dimensional defect calibration model.
[0015] The technical solution provided by the embodiments disclosed in this application has the following beneficial effects: A urea ammonia hydrolysis reactor is 3D scanned, and then the original 3D point cloud data of the urea ammonia hydrolysis reactor is obtained according to the 3D scanning result; a 3D voxel grid of the original 3D point cloud data is constructed based on a set resolution, and for each voxel grid in the 3D voxel grid, the data difference deviation corresponding to each point cloud data in the voxel grid is determined, and the sampling point cloud data of the voxel grid is determined according to all the data difference deviations, and all defect point cloud domains of the urea ammonia hydrolysis reactor are determined according to the sampling point cloud data of each voxel grid, and then the reactor defect features corresponding to each defect point cloud domain are determined; the geometric shape parameters of the urea ammonia hydrolysis reactor are obtained, and a 3D defect calibration model of the urea ammonia hydrolysis reactor is generated according to the geometric shape parameters, all the sampling point cloud data and each reactor defect feature; a defect detection report of the urea ammonia hydrolysis reactor is generated according to the 3D defect calibration model.
[0016] It can be seen that in the present application, the data scale of the three-dimensional point cloud data of the urea ammonia hydrolysis reactor can be reduced, and the defect characteristics of the reactor can be accurately extracted; wherein, by constructing a three-dimensional voxel grid, a large amount of point cloud data can be aggregated into a smaller data set, significantly reducing the computational complexity and storage requirements, and by analyzing the voxel grid, the anomalies in the local area can be effectively identified, the detection rate of defects can be improved, and small but important defects can be avoided from being missed; then, by determining the data difference deviation, the defect area can be more clearly defined, which helps to more accurately extract and quantify the defect characteristics of the reactor and avoid misjudgment caused by data noise; further, a spatial reference frame is provided for the point cloud data based on the geometric shape parameters, which can help to accurately locate the defect position in the urea ammonia hydrolysis reactor, and the defects are calibrated in the basic three-dimensional model, so that each defect can be associated with a specific geometric parameter, and the size, depth, morphology and other characteristics of the defect can be further quantified, which can help to identify and quantify these complex defects; finally, a defect detection report for the urea ammonia hydrolysis reactor is generated based on the three-dimensional defect calibration model, which helps to improve the accuracy of defect detection in the urea ammonia hydrolysis reactor.
[0017] In summary, the technical solution adopted in the present application can reduce the data scale of the three-dimensional point cloud data of the urea to ammonia hydrolysis reactor and accurately extract the reactor defect features to improve the accuracy of defect detection in the urea to ammonia hydrolysis reactor. BRIEF DESCRIPTION OF THE DRAWINGS
[0018] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only the embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.
[0019] Figure 1 is a flow chart of a reactor defect detection method based on 3D scanning technology provided in the present application; Figure 2 is an exemplary flow chart for determining the deviation of data difference provided by the present application; Figure 3 is an exemplary flow chart for generating a three-dimensional defect calibration model according to the present application; Figure 4 This is a module structure diagram of a reactor defect detection system based on 3D scanning technology provided in this application. DETAILED DESCRIPTION
[0020] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0021] The embodiment of the present application provides a reactor defect detection method and system based on 3D scanning technology, the core of which is to perform 3D scanning on a urea to ammonia hydrolysis reactor, and then obtain original 3D point cloud data of the urea to ammonia hydrolysis reactor based on the 3D scanning result; construct a 3D voxel grid of the original 3D point cloud data based on a set resolution, and for each voxel grid in the 3D voxel grid, determine the data difference deviation corresponding to each point cloud data in the voxel grid, determine the sampling point cloud data of the voxel grid based on all the data difference deviations, determine all defect point cloud domains of the urea to ammonia hydrolysis reactor from the sampling point cloud data of each voxel grid, and then determine the reactor defect features corresponding to each defect point cloud domain; obtain the geometric shape parameters of the urea to ammonia hydrolysis reactor, and generate a 3D defect calibration model of the urea to ammonia hydrolysis reactor based on the geometric shape parameters, all the sampling point cloud data and each reactor defect feature; and generate a defect detection report of the urea to ammonia hydrolysis reactor based on the 3D defect calibration model. The above scheme can reduce the data size of the three-dimensional point cloud data of the urea to ammonia hydrolysis reactor and accurately extract the reactor defect features to improve the accuracy of defect detection of the urea to ammonia hydrolysis reactor. Example
[0022] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods. Figure 1As shown in FIG. 1 , this figure is an exemplary flow chart of a reactor defect detection method based on 3D scanning technology according to this embodiment of the present application, and the detection method includes the following steps: In step S1, a urea-ammonia hydrolysis reactor is 3D scanned, and then original three-dimensional point cloud data of the urea-ammonia hydrolysis reactor is obtained according to the 3D scanning result.
[0023] In the specific implementation, first, according to the size and material of the urea to ammonia hydrolysis reactor, a suitable 3D scanner is selected, and it is ensured that the 3D scanner has anti-rust and anti-corrosion functions and can work stably in the working environment of the urea to ammonia hydrolysis reactor; then, the urea to ammonia hydrolysis reactor is 3D scanned using a 3D scanning device with anti-rust and anti-corrosion functions, that is, the 3D scanner is fixed at multiple positions around the urea to ammonia hydrolysis reactor to perform an all-round scan to ensure that all key areas of the urea to ammonia hydrolysis reactor are covered, and the scan can be performed multiple times to obtain a more complete 3D scanning result, which includes three-dimensional data of the urea to ammonia hydrolysis reactor.
[0024] In this embodiment, the original three-dimensional point cloud data of the urea to ammonia hydrolysis reactor is obtained according to the 3D scanning results by converting the format of the three-dimensional data in the 3D scanning results, thereby obtaining the original three-dimensional point cloud data of the urea to ammonia hydrolysis reactor; in specific implementation, point cloud processing software (such as CloudCompare, MeshLab, etc.) can be used to convert the format of the three-dimensional data in the 3D scanning results. These tools support input and output in multiple formats. In the point cloud processing software, the three-dimensional data to be converted is selected and exported to the required original three-dimensional point cloud data format to obtain the original three-dimensional point cloud data of the urea to ammonia hydrolysis reactor.
[0025] In step S2, a three-dimensional voxel grid of the original three-dimensional point cloud data is constructed based on a set resolution, and for each voxel grid in the three-dimensional voxel grid, the data difference deviation corresponding to each point cloud data in the voxel grid is determined, and the sampling point cloud data of the voxel grid is determined according to all the data difference deviations, and all defective point cloud domains of the urea to ammonia hydrolysis reactor are determined from the sampling point cloud data of each voxel grid, and then the reactor defect characteristics corresponding to each defective point cloud domain are determined.
[0026] In this embodiment, the three-dimensional voxel grid of the original three-dimensional point cloud data can be constructed based on the set resolution in the following manner, namely: Determine voxel grid parameters according to the set resolution; Dividing the three-dimensional space into a plurality of voxel grids according to the voxel grid parameters, thereby obtaining a voxel grid; The original three-dimensional point cloud data is mapped into the voxel grid to obtain the three-dimensional voxel grid of the original three-dimensional point cloud data.
[0027] In the specific implementation, first, the resolution can be set according to historical experiments. The resolution determines the parameters of each voxel grid. The voxel grid parameters in this application refer to the length of the voxel grid, for example, the length of the voxel grid (such as 1mm, 0.5mm, etc.). The voxel grid in this application is a cube, and the side length of the voxel grid is equal to the set resolution; then, the three-dimensional space is divided into multiple voxel grids according to the voxel grid parameters, that is, the minimum and maximum coordinate values of the three-dimensional space can be determined according to the original three-dimensional point cloud data to define the range of the voxel grid, and the voxel grid in each dimension is calculated according to the three-dimensional space boundary and the voxel grid parameters. In actual implementation, the number of voxel grids in each dimension can be determined using the following formula: the number of voxel grids in the i-th dimension = (the maximum coordinate in the i-th dimension - the minimum coordinate in the i-th dimension) / voxel grid parameter. The number of voxel grids in the three-dimensional space can be obtained in the above manner. According to the calculated number of voxel grids, a plurality of voxel grids are divided according to the set voxel grid parameters to form a three-dimensional voxel grid. Finally, the original three-dimensional point cloud data is mapped to the voxel grid, that is, the original three-dimensional point cloud data is traversed to obtain the coordinate information (x, y, z) of each point cloud data, and the voxel grid position of each point cloud data is calculated according to the coordinates of each point cloud data. According to the calculated voxel grid position, each point cloud data is mapped to the corresponding voxel grid, and then the states of all voxel grids are integrated into a three-dimensional data structure, that is, the three-dimensional voxel grid of the original three-dimensional point cloud data.
[0028] Preferably, in this embodiment, reference Figure 2 As shown, this figure is an exemplary flow chart of determining the data difference deviation in an embodiment of the present application. In this embodiment, determining the data difference deviation corresponding to each point cloud data in the voxel grid can be implemented by the following steps: First, in step S21, the grid centroid of the voxel grid is determined; Then, in step S22, the deviation between each point cloud data in the voxel grid and the grid centroid is determined; Finally, in step S23, the data difference deviation corresponding to each point cloud data in the voxel grid is determined according to all the deviations.
[0029] In specific implementation, first, the grid centroid of the voxel grid is determined, and the coordinates (x, y, z) of all point cloud data in the voxel grid can be extracted, so as to calculate the grid centroid of the voxel grid according to the coordinates of all point cloud data. In actual implementation, the grid centroid C of the voxel grid can be determined according to the following formula:
[0030] Where n represents the total number of point cloud data in the voxel grid, Represents the coordinates of the i-th point cloud data in the voxel grid; then, for each point cloud data in the voxel grid, determine the deviation between the point cloud data and the grid centroid. The deviation represents the degree of deviation between the point cloud data and the grid centroid of the voxel grid in which it is located. In actual implementation, the deviation can be determined according to the following formula:
[0031] in, Indicates the deviation between the i-th point cloud data and the grid centroid, indicating Represents the coordinates of the i-th point cloud data in the voxel grid, The coordinates of the grid centroid representing the voxel grid can be obtained by the above method to obtain the deviation between each point cloud data in the voxel grid and the grid centroid; finally, the data difference deviation corresponding to each point cloud data can be determined according to the deviation between the point cloud data and the grid centroid, that is, for each point cloud data in the voxel grid, the ratio of the deviation between the point cloud data and the grid centroid to the standard deviation of all deviations can be used as the data difference deviation corresponding to the point cloud data, and the data difference deviation is used to represent the overall difference between the point cloud data and other point cloud data in the voxel grid. The data difference deviation corresponding to each point cloud data in the voxel grid can be obtained by the above method.
[0032] In this embodiment, the sampling point cloud data of the voxel grid may be determined according to all data difference deviations in the following manner, namely: Determine the data difference deviation threshold through all data difference deviations; Point cloud data with a data difference deviation greater than the data difference deviation threshold are eliminated, thereby obtaining the sampling point cloud data of the voxel grid.
[0033] In the specific implementation, first, the data difference deviation threshold can be determined through all the data difference deviations, that is, the smallest data difference deviation among all the data difference deviations is used as the data difference deviation threshold; then, the point cloud data with a data difference deviation greater than the data difference deviation threshold can be eliminated, that is, the point cloud data with the smallest data difference deviation in the voxel grid is retained, and other point cloud data are eliminated, so that the point cloud data with the smallest data difference deviation is used as the sampling point cloud data of the voxel grid. The sampling point cloud data of each voxel grid can be obtained in the above manner.
[0034] In this embodiment, all defective point cloud domains of the urea-to-ammonia hydrolysis reactor can be determined by the sampling point cloud data of each voxel grid in the following manner, namely: Establishing three-dimensional sampling point cloud data of the urea-to-ammonia hydrolysis reactor according to the sampling point cloud data of each voxel grid; Performing defect comparison between the three-dimensional sampling point cloud data and pre-set three-dimensional point cloud data of a reactor to obtain a plurality of defect comparison domains; Determine the defect confidence coefficient for each defect comparison domain; All defect point cloud domains of the urea to ammonia hydrolysis reactor are determined based on the respective defect confidence coefficients.
[0035] In the specific implementation, first, the three-dimensional sampling point cloud data of the urea ammonia hydrolysis reactor can be established according to the sampling point cloud data of each voxel grid, that is, the sampling point cloud data of each voxel grid can be integrated into a unified three-dimensional sampling point cloud data set, so as to obtain the three-dimensional sampling point cloud data of the urea ammonia hydrolysis reactor, and the three-dimensional sampling point cloud data describes the three-dimensional shape of the urea ammonia hydrolysis reactor; then, the pre-set reactor three-dimensional point cloud data is loaded, that is, the standard three-dimensional point cloud data (ideal model) of the urea ammonia hydrolysis reactor is obtained, and the reactor three-dimensional point cloud data is used as a benchmark for defect comparison, so that the three-dimensional sampling point cloud data can be compared with the pre-set reactor three-dimensional point cloud data using a point cloud comparison algorithm (such as ICP algorithm or Hausdorff distance). The defect comparison domains are compared based on the data, and this process will help identify areas with differences in shape, size and position, thereby marking multiple defect comparison domains; then, for each identified defect comparison domain, its defect confidence coefficient is calculated. In actual implementation, the shape similarity between the shape of the defect comparison domain and the standard model can be calculated and compared, and the shape similarity is used as the defect confidence coefficient of the defect comparison domain. The defect confidence coefficient indicates the degree of confidence that there are defects in the defect comparison domain. The defect confidence coefficient of each defect comparison domain can be obtained in the above manner; finally, according to the preset confidence coefficient threshold, the defect comparison domains with defect confidence coefficients higher than the confidence coefficient threshold are screened out and used as the defect point cloud domain of the urea to ammonia hydrolysis reactor, and the defect point cloud domain contains the identified reactor defect information.
[0036] In this embodiment, the reactor defect features corresponding to each defect point cloud domain are determined by performing feature extraction on each defect point cloud domain respectively, and then obtaining the reactor defect features corresponding to each defect point cloud domain; in specific implementation, feature extraction is performed on each defect point cloud domain, that is, the geometric features and distribution features of the defect point cloud domain are extracted. For example, the volume of the defect area is estimated and the total area of the defect surface is calculated through the number and spatial distribution of point cloud data in the defect point cloud domain. The boundary generation algorithm of the point (such as Alpha shape) is usually used to organize the extracted geometric features and distribution features into structured data, that is, the reactor defect features corresponding to each defect point cloud domain, and the reactor defect features include information such as the type, size, shape, and position of each defect.
[0037] It should be noted that by constructing a three-dimensional voxel grid, a large amount of point cloud data can be aggregated into a smaller data set, significantly reducing computational complexity and storage requirements. By analyzing the voxel grid, anomalies in local areas can be effectively identified, the detection rate of defects can be improved, and small but important defects can be avoided from being missed. By determining the data difference deviation, the defect area can be more clearly defined, helping to more accurately extract and quantify the defect characteristics of the reactor and avoid misjudgment caused by data noise.
[0038] In step S3, the geometric shape parameters of the urea to ammonia hydrolysis reactor are obtained, and a three-dimensional defect calibration model of the urea to ammonia hydrolysis reactor is generated based on the geometric shape parameters, all sampling point cloud data and each reactor defect feature.
[0039] In a specific implementation, the engineering design drawings or CAD models of the urea-ammonia hydrolysis reactor can be obtained. These documents usually contain detailed geometric parameters, such as size, shape, wall thickness, etc., so that the geometric shape parameters of the urea-ammonia hydrolysis reactor can be obtained.
[0040] Preferably, in this embodiment, reference Figure 3 As shown in FIG. 1 , this figure is an exemplary flow chart of generating a three-dimensional defect calibration model in an embodiment of the present application. In this embodiment, the three-dimensional defect calibration model of the urea-to-ammonia hydrolysis reactor is generated according to the geometric shape parameters, all the sampling point cloud data and the defect characteristics of each reactor, which can be specifically implemented by the following steps: First, in step S31, the basic three-dimensional model of the urea-to-ammonia hydrolysis reactor is generated using the geometric shape parameters and all the sampling point cloud data; Then, in step S32, the defect features of each reactor are calibrated in the basic three-dimensional model, thereby generating a three-dimensional defect calibration model of the urea to ammonia hydrolysis reactor.
[0041] In the specific implementation, first, according to the obtained geometric shape parameters (such as length, diameter, wall thickness, etc.) and all the sampling point cloud data, the basic three-dimensional model of the urea ammonia hydrolysis reactor is created in the modeling software (such as SolidWorks, AutoCAD, CATIA, etc.), and the actual structural characteristics of the urea ammonia hydrolysis reactor are considered in the model, including the inlet and outlet pipes, support structures and other auxiliary components, to ensure that the basic model accurately reflects the geometric shape of the urea ammonia hydrolysis reactor; then, the defect characteristics of each reactor are calibrated in the basic three-dimensional model, that is, the corresponding defect information (such as cracks, corrosion, deformation, etc.) is applied to the basic three-dimensional model, and a corresponding three-dimensional shape is created for each defect to ensure that it truly shows the characteristics of the defect in the model. For example, cracks can be represented by an elongated spatial volume, while corrosion areas can be represented by irregular shapes, and it is ensured that all calibrated defect information is seamlessly combined with the basic three-dimensional model, thereby generating a three-dimensional defect calibration model of the urea ammonia hydrolysis reactor.
[0042] It should be noted that a spatial reference frame is provided for the point cloud data based on the geometric shape parameters, which can help accurately locate the defect positions in the urea to ammonia hydrolysis reactor, calibrate the defects in the basic three-dimensional model, so that each defect can be associated with specific geometric parameters, and further quantify the size, depth, morphology and other characteristics of the defects, which can help identify and quantify these complex defects.
[0043] In step S4, a defect detection report of the urea-to-ammonia hydrolysis reactor is generated according to the three-dimensional defect calibration model.
[0044] It should be noted that, in the present application, the defect detection report of the urea to ammonia hydrolysis reactor includes basic information of the urea to ammonia hydrolysis reactor, reactor defect location, defect type and defect response strategy.
[0045] In the specific implementation, the generated three-dimensional defect calibration model is imported to ensure that all defect information is contained in the three-dimensional defect calibration model, and the specific parameters of the defects, including the reactor defect location and defect type, are extracted from the three-dimensional defect calibration model, and the basic information of the urea ammonia hydrolysis reactor is obtained. A defect response strategy is generated according to the defect handling manual, and then a defect detection report of the urea ammonia hydrolysis reactor is generated according to the basic information of the urea ammonia hydrolysis reactor, the reactor defect location, the defect type and the defect response strategy, and the defect detection report is sent to the management center.
[0046] It can be seen that in the present application, the data scale of the three-dimensional point cloud data of the urea ammonia hydrolysis reactor can be reduced, and the defect characteristics of the reactor can be accurately extracted; wherein, by constructing a three-dimensional voxel grid, a large amount of point cloud data can be aggregated into a smaller data set, significantly reducing the computational complexity and storage requirements, and by analyzing the voxel grid, the anomalies in the local area can be effectively identified, the detection rate of defects can be improved, and small but important defects can be avoided from being missed; then, by determining the data difference deviation, the defect area can be more clearly defined, which helps to more accurately extract and quantify the defect characteristics of the reactor and avoid misjudgment caused by data noise; further, a spatial reference frame is provided for the point cloud data based on the geometric shape parameters, which can help to accurately locate the defect position in the urea ammonia hydrolysis reactor, and the defects are calibrated in the basic three-dimensional model, so that each defect can be associated with a specific geometric parameter, and the size, depth, morphology and other characteristics of the defect can be further quantified, which can help to identify and quantify these complex defects; finally, a defect detection report for the urea ammonia hydrolysis reactor is generated based on the three-dimensional defect calibration model, which helps to improve the accuracy of defect detection in the urea ammonia hydrolysis reactor.
[0047] In summary, the technical solution adopted in the present application can reduce the data scale of the three-dimensional point cloud data of the urea to ammonia hydrolysis reactor and accurately extract the reactor defect features to improve the accuracy of defect detection in the urea to ammonia hydrolysis reactor. Example
[0048] This application provides a reactor defect detection system based on 3D scanning technology, referring to Figure 4 As shown, this figure is a schematic diagram of a detection system according to this embodiment of the present application, and the detection system includes: The 3D scanning module 100 is used to perform a 3D scan on the urea-ammonia hydrolysis reactor, and then obtain original three-dimensional point cloud data of the urea-ammonia hydrolysis reactor according to the 3D scanning result; The defect feature determination module 200 is used to construct a three-dimensional voxel grid of the original three-dimensional point cloud data based on a set resolution, determine the data difference deviation corresponding to each point cloud data in the voxel grid for each voxel grid in the three-dimensional voxel grid, determine the sampling point cloud data of the voxel grid according to all the data difference deviations, determine all defect point cloud domains of the urea to ammonia hydrolysis reactor from the sampling point cloud data of each voxel grid, and then determine the reactor defect features corresponding to each defect point cloud domain; A three-dimensional model generation module 300 is used to obtain geometric shape parameters of the urea-ammonia hydrolysis reactor, and generate a three-dimensional defect calibration model of the urea-ammonia hydrolysis reactor based on the geometric shape parameters, all sampling point cloud data and each reactor defect feature; The defect report generating module 400 is used to generate a defect detection report of the urea to ammonia hydrolysis reactor according to the three-dimensional defect calibration model.
[0049] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0050] A person skilled in the art may understand that all or part of the steps in the various methods of the above embodiments may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, the storage medium including a read-only memory (ROM), a random access memory (RAM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), a one-time programmable read-only memory (OTPROM), an electronically-erasable programmable read-only memory (EEPROM), a compact disc (CD-ROM) or other optical disc storage, magnetic disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.
[0051] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.
Claims
1. A reactor defect detection method based on 3D scanning technology, characterized in that: The detection method comprises the following steps: Performing a 3D scan on a urea-ammonia hydrolysis reactor, and then obtaining original three-dimensional point cloud data of the urea-ammonia hydrolysis reactor according to the 3D scanning result; Constructing a three-dimensional voxel grid of the original three-dimensional point cloud data based on a set resolution, determining the data difference deviation corresponding to each point cloud data in the voxel grid for each voxel grid, determining the sampling point cloud data of the voxel grid according to all the data difference deviations, determining all defective point cloud domains of the urea to ammonia hydrolysis reactor from the sampling point cloud data of each voxel grid, and then determining the reactor defect features corresponding to each defective point cloud domain; Acquire geometric shape parameters of the urea-ammonia hydrolysis reactor, and generate a three-dimensional defect calibration model of the urea-ammonia hydrolysis reactor based on the geometric shape parameters, all sampling point cloud data and defect characteristics of each reactor; A defect detection report of the urea-to-ammonia hydrolysis reactor is generated according to the three-dimensional defect calibration model.
2. A reactor defect detection method based on 3D scanning technology as claimed in claim 1, characterized in that: The urea-to-ammonia hydrolysis reactor was 3D scanned using a 3D scanning device with anti-rust and anti-corrosion functions.
3. The reactor defect detection method based on 3D scanning technology according to claim 1, characterized in that: The original three-dimensional point cloud data of the urea-ammonia hydrolysis reactor is obtained according to the 3D scanning result by converting the format of the three-dimensional data in the 3D scanning result to obtain the original three-dimensional point cloud data of the urea-ammonia hydrolysis reactor.
4. The reactor defect detection method based on 3D scanning technology according to claim 1, characterized in that: Constructing the three-dimensional voxel grid of the original three-dimensional point cloud data based on the set resolution specifically includes: Determine voxel grid parameters according to the set resolution; Dividing the three-dimensional space into a plurality of voxel grids according to the voxel grid parameters, thereby obtaining a voxel grid; The original three-dimensional point cloud data is mapped into the voxel grid to obtain the three-dimensional voxel grid of the original three-dimensional point cloud data.
5. The method for detecting reactor defects based on 3D scanning technology according to claim 1, characterized in that: Determining the sampling point cloud data of the voxel grid according to all data difference deviations specifically includes: Determine the data difference deviation threshold through all data difference deviations; Point cloud data with a data difference deviation greater than the data difference deviation threshold are eliminated, thereby obtaining the sampling point cloud data of the voxel grid.
6. The method for detecting reactor defects based on 3D scanning technology according to claim 1, characterized in that: The sampling point cloud data of each voxel grid is used to determine all defect point cloud domains of the urea-ammonia hydrolysis reactor, including: Establishing three-dimensional sampling point cloud data of the urea-to-ammonia hydrolysis reactor according to the sampling point cloud data of each voxel grid; Performing defect comparison between the three-dimensional sampling point cloud data and pre-set three-dimensional point cloud data of a reactor to obtain a plurality of defect comparison domains; Determine the defect confidence coefficient for each defect comparison domain; All defect point cloud domains of the urea to ammonia hydrolysis reactor are determined based on the respective defect confidence coefficients.
7. The method for detecting reactor defects based on 3D scanning technology according to claim 1, characterized in that: Determining the reactor defect features corresponding to each defect point cloud domain is to extract features from each defect point cloud domain respectively, and then obtain the reactor defect features corresponding to each defect point cloud domain.
8. The method for detecting reactor defects based on 3D scanning technology according to claim 1, characterized in that: Generating the three-dimensional defect calibration model of the urea-to-ammonia hydrolysis reactor according to the geometric shape parameters, all the sampling point cloud data and the defect characteristics of each reactor specifically includes: Generate a basic three-dimensional model of the urea-to-ammonia hydrolysis reactor using the geometric shape parameters and all the sampled point cloud data; The defect features of each reactor are calibrated in the basic three-dimensional model, thereby generating a three-dimensional defect calibration model of the urea-to-ammonia hydrolysis reactor.
9. The reactor defect detection method based on 3D scanning technology according to claim 1, characterized in that: The defect detection report of the urea to ammonia hydrolysis reactor includes basic information of the urea to ammonia hydrolysis reactor, reactor defect location, defect type and defect response strategy.
10. A reactor defect detection system based on 3D scanning technology, used to execute a reactor defect detection method based on 3D scanning technology as claimed in any one of claims 1 to 9, characterized in that: The detection system comprises: A 3D scanning module, used to perform 3D scanning on the urea-ammonia hydrolysis reactor, and then obtain original three-dimensional point cloud data of the urea-ammonia hydrolysis reactor according to the 3D scanning result; a defect feature determination module, for constructing a three-dimensional voxel grid of the original three-dimensional point cloud data based on a set resolution, determining, for each voxel grid in the three-dimensional voxel grid, a data difference deviation corresponding to each point cloud data in the voxel grid, determining the sampling point cloud data of the voxel grid according to all the data difference deviations, determining all defect point cloud domains of the urea to ammonia hydrolysis reactor from the sampling point cloud data of each voxel grid, and further determining the reactor defect feature corresponding to each defect point cloud domain; A three-dimensional model generation module, used to obtain geometric shape parameters of the urea-ammonia hydrolysis reactor, and generate a three-dimensional defect calibration model of the urea-ammonia hydrolysis reactor based on the geometric shape parameters, all sampling point cloud data and each reactor defect feature; A defect report generation module is used to generate a defect detection report of the urea-to-ammonia hydrolysis reactor according to the three-dimensional defect calibration model.