Structure difference confirmation device and structure difference confirmation method

Abstract

To easily know whether there is a significant difference between an actual machine dimension and a design value without inputting a threshold for each actual machine.SOLUTION: A structure difference confirmation device comprises: a point group data acquisition part which acquires point group data as a result of imaging of a component; an object specification processing part which specifies the kind of a component from the acquired point group data; a common difference processing part which acquires a common difference corresponding to the specified kind of the component from a storage part stored with common differences by kinds of the component; a dimension processing part which extracts a difference between the dimension of the point group data on the component and a dimension of three-dimensional CAD data as design information on the component; and a determination processing part which determines whether the extracted common difference is within a range of the acquired common difference.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present invention relates to a structure difference confirmation device and a structure difference confirmation method. [Background technology]

[0002] There is a known technology that uses laser scanning to measure the actual structure of a building and recognizes the surface shape of the building using point cloud data, which is a collection of three-dimensional position data.

[0003] Patent Document 1 describes an as-built support device that includes "a design data storage unit that stores information including the shape and arrangement of the plant building, piping within the plant, and each piece of equipment as design information; a point cloud-to-2D image conversion unit that generates 2D image data from the point cloud data stored in the point cloud data storage unit; a 2D image recognition unit that compares the generated 2D image data with retained shape data to recognize the names and positions of plant parts; and a recognition information evaluation unit that extracts piping link information including the piping diameter, piping direction, and connection position from the point cloud data near the discovered object to narrow down the system, and further narrows down comparison candidates for CAD shapes related to the system from the piping link information based on the design information stored in the design data storage unit."

[0004] Patent Document 2 describes a structure difference extraction device comprising: a point cloud shadow portion determination unit that determines a first point cloud shadow portion, which is a point cloud in the range that is in the shadow of the second point cloud, among the first point cloud, based on a first point cloud obtained by scanning a first structure from one or more first measurement positions and a second point cloud obtained by scanning a second structure from one or more second measurement positions, and determines a second point cloud shadow portion, which is a point cloud in the range that is in the shadow of the first point cloud, among the second point cloud; and a point cloud comparison unit that extracts a difference between the first structure and the second structure by comparing a first deleted point cloud, which is a result of deleting the first point cloud shadow portion from the first point cloud, with a second deleted point cloud, which is a result of deleting the second point cloud shadow portion from the second point cloud.

[0005] Patent Document 3 describes that "the error detection means detects errors between the point cloud data and the three-dimensional CAD data compared by the comparison means, and the display means displays the errors as a color map. The errors are colored according to a predetermined threshold value, so that, for example, a convex error exceeding the predetermined threshold value is colored red, and a concave error below the predetermined threshold value is colored blue. In addition to displaying the color map, the display means displays information such as numerical values ​​of the error inspection results, and an output means provided in the computer outputs a report such as a spreadsheet file format." [Prior art documents] [Patent documents]

[0006] [Patent Document 1] JP 2022-19396 A [Patent Document 2] JP 2019-211264 A [Patent Document 3] JP 2008-256638 A Summary of the Invention [Problem to be solved by the invention]

[0007] Workers check each part on-site to determine whether the plant's equipment, piping, and other parts have been installed as designed. Checking a huge amount of parts on-site places a burden on workers, and in high-dose areas, it leads to increased man-hours of radiation exposure. Even if workers try to check visually, there is a risk that they may miss something. In fact, there have been reported cases in the past where defects occurred due to missing fireproofing material, which affected the construction schedule and required re-installation.

[0008] The as-built support device in Patent Document 1 automatically recognizes the names and positions of plant parts from measured 3D point cloud data. This device excels in adding attribute information to 3D point cloud data, but does not have a function to determine whether there is a significant difference between the recognized part dimensions and their design values.

[0009] The structural difference extraction device of Patent Document 2 is excellent in extracting the difference between any two structures, but does not have the function of determining whether or not there is a significant difference between the actual dimensions and the design values.

[0010] The product inspection device of Patent Document 3 is excellent in that it makes it easy to visually check whether the difference between the point cloud data of the inspection object and the 3D CAD data is equal to or greater than a predetermined threshold value. However, it is necessary to set the threshold value individually for each inspection object, which is time-consuming.

[0011] Therefore, an object of the present invention is to easily know whether or not there is a significant difference between the actual dimensions and the design values, without having to input a threshold value for each actual device. [Means for solving the problem]

[0012] The structural difference confirmation device of the present invention is characterized by comprising a point cloud data acquisition unit that acquires point cloud data resulting from imaging a part, an object identification processing unit that identifies the type of part from the acquired point cloud data, a tolerance processing unit that acquires a tolerance corresponding to the identified type of part from a memory unit that stores tolerances for each type of part, a dimension processing unit that extracts the difference between the dimensions in the point cloud data of the part and the dimensions in 3D CAD data as design information for the part, and a judgment processing unit that judges whether the extracted difference is within the range of the acquired tolerance. Other means will be described in the description of the preferred embodiment of the invention. Effect of the Invention

[0013] According to the present invention, it is possible to easily know whether there is a significant difference between the actual dimensions and the design values ​​without inputting a threshold value for each actual device. That is, the present invention makes it possible to automatically determine whether there is a difference between the actual dimensions and the design values, thereby reducing the number of confirmation oversights. In addition, the simplification of the confirmation work by the automatic determination reduces the burden on on-site workers and the labor hours of exposure. Furthermore, if it is determined that the difference between the actual dimensions and the design values ​​is outside the tolerance range, the output of the report clearly indicates the location, enabling early action to be taken. [Brief description of the drawings]

[0014] [Figure 1] FIG. 2 is a diagram for explaining the configuration of a structure difference confirmation device. [Diagram 2] FIG. [Diagram 3] 1 is an example of tolerance data. [Figure 4] 1 is an example of a form. [Diagram 5] 13 is a flowchart of a processing procedure. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0015] Hereinafter, an embodiment of the present invention (referred to as the present embodiment) will be described in detail with reference to the drawings. The present embodiment is an example of judging whether or not the difference between the actual dimensions and the design values ​​of piping etc. installed at the site of various plants such as nuclear plants and chemical plants is within the tolerance range. However, the present invention is generally applicable to the case of manufacturing and constructing a structure based on design information of 3D CAD (Computer Aided Design) data, or the case of inspecting the degree of deformation of a manufactured and constructed structure.

[0016] 1 is a diagram for explaining the configuration of a structure difference confirmation device 1. The structure difference confirmation device 1 is a general computer. The structure difference confirmation device 1 includes a central control device 11, an input device 12 such as a keyboard or a mouse, an output device 13 such as a display or a printer, a main memory device 14, an auxiliary memory device 15, and a communication device 16.

[0017] The point cloud data acquisition unit 21, object identification processing unit 22, tolerance processing unit 23, dimension processing unit 24, judgment processing unit 25, report output unit 26, and site display processing unit 27 in the main storage device 14 are programs and are the main units of each process described below. Hereinafter, when "XX unit" is written, it means that the central control device 11 reads each program from the auxiliary storage device 15, loads it into the main storage device 14, and then executes each process. The auxiliary storage device 15 stores tolerance data 31, point cloud data 32, 3D CAD data 33, and reports 34. For example, the tolerance data 31, etc. may be stored externally via the network 2.

[0018] A designer 4 uses the designer terminal device 3 to create three-dimensional CAD data 33 as design information for manufacturing and constructing an actual machine. The designer terminal device 3 transmits the three-dimensional CAD data 33 to the structure difference confirmation device 1 via the network 2.

[0019] A site 5 exists at a location separate from the structure difference confirmation device 1. At the site 5, a worker 6 and the actual structure are present. The actual structure here is equipment 41 and pipes 42-44 that have been installed at the site 5. The worker 6 wears a wearable device 7 on his head. The wearable device 7 has the functions of a three-dimensional laser scanner and smart glasses (glasses portion). The three-dimensional laser scanner of the wearable device 7 captures images of the equipment 41 and pipes 42-44, and transmits the point cloud data 32 acquired as a result to the structure difference confirmation device 1 via the network 2. The structure difference confirmation device 1 can transmit any information to the wearable device 7 via the network 2, and project the information on the smart glasses as AR (Augmented Reality).

[0020] There may be a fixed or mobile scanner 8 at the site 5. The scanner 8 is also a three-dimensional laser scanner, which captures images of the equipment 41 and the pipes 42-44, and transmits the point cloud data 32 acquired as a result to the structure difference confirmation device 1 via the network 2. Here, the configuration for capturing an image of the subject as a "scanner" is merely one example. The configuration for capturing an image of the subject does not have to be one that scans the subject, and may be anything that can ultimately acquire point cloud data of the subject.

[0021] (Point cloud data and 3D CAD data) The point cloud data 32 is a collection (raster data) of many points that make up the surface of a structure. Such points are stored in association with three-dimensional coordinates. The three-dimensional coordinates are calculated based on the time and direction that it takes for a laser beam emitted from one or more three-dimensional laser scanners to be reflected by the surface of a structure in real space and return.

[0022] The three-dimensional CAD data 33 is a collection of lines and surfaces (vector data) drawn by a computer in a virtual space. For example, a line is represented by the three-dimensional coordinates of two end points, and a plane is represented by a three-dimensional equation that indicates the plane.

[0023] FIG. 2 is a diagram for explaining the dimension comparison. First, attention is paid to the upper diagram of FIG. 2. First, the structure difference confirmation device 1 specifies the start point and end point of the dimension to be compared, and calculates the dimension. The structure difference confirmation device 1 may accept the user to specify the start point and end point. The figure with the heading "CAD" is the three-dimensional CAD data 33 of the equipment 41 and the pipes 42 and 43. According to the three-dimensional CAD data 33, the dimension between the left end (start point) of the pipe 42 and the lower right end (end point) of the equipment 41 is "300". The dimension between the upper end (start point) of the pipe 43 and the lower end (end point) of the pipe 43 is "1000".

[0024] The figure labeled "Point Cloud" is point cloud data 32 of equipment 41 and pipes 42 and 43. According to point cloud data 32, the dimension between the left end (starting point) of pipe 42 and the lower right end (end point) of equipment 41 is "300". The dimension between the upper end (starting point) of pipe 43 and the lower end (end point) of pipe 43 is "1300".

[0025] Next, attention is paid to the lower diagram of FIG. 2. Secondly, the structure difference confirmation device 1 compares dimensions and extracts the difference. The dimension between the left end of the pipe 42 and the lower right end of the equipment 41 is "300" in the three-dimensional CAD data 33 and "300" in the point cloud data 32, and the difference between the two is "0". The dimension between the upper end of the pipe 43 and the lower end of the pipe 43 is "1000" in the three-dimensional CAD data 33 and "1300" in the point cloud data 32, and the difference between the two is "300". In this way, it is actually common for the dimensions of the actual equipment at the site to deviate from the design values. However, the question is whether the difference is acceptable or not.

[0026] Third, the structure difference confirmation device 1 allocates (acquires) a tolerance according to the type of part, and compares the tolerance with the difference. In the tolerance data 31, the tolerance of the part "piping" is within "200". The dimensional difference "300" between the left end of the pipe 42 and the lower right end of the equipment 41 is larger than this tolerance "200". In other words, this difference is not permissible.

[0027] Fourth, the structure difference confirmation device 1 outputs the result of the dimensional comparison on a form and / or displays it in AR. The contents output and / or displayed here are that the difference in the dimension between the left end of the pipe 42 and the lower right end of the equipment 41 is "OK", and that the difference in the dimension between the upper end of the pipe 43 and the lower end of the pipe 43 is "NG".

[0028] (Tolerance Data) 3 is an example of the tolerance data 31. In the tolerance data 31, tolerances are stored in a tolerance field 102 in association with parts stored in a part field 101. The parts in the parts column 101 indicate the types of parts that make up the structure. The tolerances in the Tolerance column are the maximum allowable difference between the actual dimensions of a structure and its design dimensions. The tolerances here are absolute values ​​with the positive and negative signs omitted.

[0029] (Report) 4 is an example of the form 34. The structure difference confirmation device 1 outputs and / or displays the form 34 to the output device 13 and / or the wearable device 7 (worker 6). In the form 34, in association with a name stored in a name column 111, a point cloud dimension column 112 stores a point cloud dimension, a CAD dimension column 113 stores a CAD dimension, a tolerance column 114 stores a tolerance, a tolerance judgment column 115 stores a tolerance judgment, and a book number column 116 stores a book number.

[0030] The name in the name column 111 is an identifier that uniquely identifies an individual part that constitutes a structure. The name includes the type of the part. The point cloud dimensions in the point cloud dimension field 112 are the three-dimensional (length, width, depth) dimensions of the part in the point cloud data 32. The CAD dimensions in the CAD dimension column 113 are the three-dimensional dimensions of the part in the three-dimensional CAD data 33.

[0031] The tolerances in the tolerance column 114 are the same as the tolerances in Figure 3. The tolerances are also the thresholds applied to the difference between the point cloud dimensions and the CAD dimensions. The tolerance judgment in the tolerance judgment column 115 is either "OK" or "NG." "NG" indicates that at least one dimension of the difference between the point cloud dimensions of the part and the CAD dimensions is larger than the tolerance. "OK" indicates that all dimensions of the difference between the point cloud dimensions of the part and the CAD dimensions are within the tolerance. The document number in the document number column 116 is a number that uniquely identifies a document in which design information including the 3D CAD data 33 of the part is described, or is link information to that document.

[0032] FIG. 5 is a flowchart of the processing procedure. In step S201, the point cloud data acquisition unit 21 acquires point cloud data. Specifically, the point cloud data acquisition unit 21 acquires point cloud data 32 of the actual structure from the auxiliary storage device 15, and divides the data into a plurality of units that can be recognized as parts of the structure. Then, one of the divided units is acquired as a "target part." The point cloud data 32 is previously captured by the wearable device 7 or the scanner 8 and then stored in the auxiliary storage device 15.

[0033] In step S202, the object identification processing unit 22 identifies the type of the part. Specifically, the object identification processing unit 22 identifies the type of the target part acquired in step S201, for example, by comparing the type of the target part with a shape sample (not shown) prepared in advance for each type of part.

[0034] In step S203, the tolerance processing unit 23 acquires the tolerance of the part. Specifically, the tolerance processing unit 23 acquires the tolerance corresponding to the target part by referring to the tolerance data 31. This process is part of the process described as “third” in FIG. 2.

[0035] In step S204, the dimension processing unit 24 identifies the start point and the end point and extracts the difference. Specifically, the dimension processing unit 24 executes the processes described as "firstly" and "secondly" in FIG.

[0036] In step S205, the dimension processing unit 24 compares the difference with the tolerance. Specifically, the dimension processing unit 24 executes the remaining part of the process described as "third" in FIG.

[0037] In step S206, the judgment processing unit 25 judges whether or not the difference is within the tolerance. Specifically, if the comparison result in step S205 is "the difference is within the tolerance" (step S206 "Yes"), the judgment processing unit 25 proceeds to step S207, and otherwise (step S206 "No"), the judgment processing unit 25 proceeds to step S209.

[0038] In step S207, the report output unit 26 outputs the report as OK. Specifically, the report output unit 26 executes the report output among the processes described as "fourth" in Fig. 2. That is, the report output unit 26 outputs the report 34 to the output device 13. The report 34 output here has a record indicating that the tolerance judgment of the target part is "OK" among the report 34 in Fig. 4.

[0039] In step S208, the site display processing unit 27 displays OK on the site. Specifically, the site display processing unit 27 executes the AR display among the processes described as "fourth" in Fig. 2. That is, the site display processing unit 27 displays the form 34 on the wearable device 7. The form 34 displayed here has a record indicating that the tolerance judgment of the target part is "OK" among the form 34 in Fig. 4. Thereafter, the processing procedure ends.

[0040] In step S209, the report output unit 26 outputs a report as NG. Specifically, the report output unit 26 executes the report output among the processes described as "fourth" in Fig. 2. That is, the report output unit 26 outputs the report 34 to the output device 13. The report 34 output here has a record indicating that the tolerance judgment of the target part is "NG" among the report 34 in Fig. 4.

[0041] In step S210, the site display processing unit 27 displays NG on the site. Specifically, the site display processing unit 27 executes the AR display among the processes described as "fourth" in Fig. 2. That is, the site display processing unit 27 displays the form 34 on the wearable device 7. The form 34 displayed here has a record indicating that the tolerance judgment of the target part is "NG" among the forms 34 in Fig. 4. Then, the processing procedure ends. The processing of steps S202 to S210 is repeated for each target part.

[0042] (Variation 1) In step S202, the object identification processing unit 22 may identify the type of the part by using a part type inference model. The part type inference model is a function that receives point cloud data 32 of the target part as input and outputs the type of the target part. The object identification processing unit 22 performs machine learning on the part type inference model using supervised learning data, and optimizes parameters of the part type inference model.

[0043] (Variation 2) The tolerance data 31 may store three-dimensional (length, width, depth) tolerances in association with parts. Similarly, the form 34 may store three-dimensional tolerances and three-dimensional tolerance judgments in association with names. In this case, the tolerance judgment is "NG" if the difference in that dimension is larger than the tolerance in that dimension, and "OK" if the difference in that dimension is within the tolerance in that dimension.

[0044] (Variation 3) In step S206, the determination processing unit 25 may determine whether or not the difference is within the tolerance range for each dimension.

[0045] (Variation 4) In step S207 or S209, the report output unit 26 may output "OK" or "NG" as the tolerance judgment for each dimension. At this time, the tolerance judgment output is, for example, "(length, width, depth) = (OK, NG, OK)". In step S208 or S210, the site display processing unit 27 may display "OK" or "NG" as the tolerance judgment for each dimension. At this time, the tolerance judgment displayed is, for example, "(length, width, depth) = (OK, NG, OK)".

[0046] (Variation 5) In step S207 or S209, the report output unit 26 may output "OK" or "NG" as the tolerance judgment for each dimension and for the magnitude relationship between the point cloud dimension and the CAD dimension. In this case, the tolerance judgment to be output is, for example, "(length, width, depth) = (OK, NG [large], NG [small])". "Large" indicates that the point cloud dimension is larger than the CAD dimension. "Small" indicates that the point cloud dimension is smaller than the CAD dimension. In step S208 or S210, the site display processing unit 27 may display "OK" or "NG" as the tolerance judgment for each dimension and for the magnitude relationship between the point cloud dimension and the CAD dimension. In this case, the tolerance judgment to be displayed is, for example, "(length, width, depth) = (OK, NG [large], NG [small])".

[0047] (Variation 6) In step S206, if the comparison result in step S205 is "the difference is within the tolerance" (step S206 "Yes"), the determination processing unit 25 may end the processing procedure.

[0048] (Variation 7) The processing of steps S202 to S206 may be repeated multiple times for each target component, and when these processing have been completed for all target components, the processing of steps S207 to S210 may be executed all at once.

[0049] (Variation 8) A user may attach any marker to actual parts that need to be inspected repeatedly in advance. The appearance of the marker (e.g., shape, color, etc.) may differ depending on the inspection cycle. The scanner 8 or the like captures images of the parts with markers attached together with parts without markers. Then, in step S201, the point cloud data acquisition unit 21 acquires point cloud data 32 of the actual structure from the auxiliary storage device 15, and divides the data into a plurality of units that can be recognized as parts of the structure. Then, one of the divided units that has a marker of any appearance captured thereon may be acquired as the "target part."

[0050] (Variation 9) The term "dimension" refers to the concept that includes the angle of a specific part. The term "tolerance" refers to the maximum allowable difference between the angle of a specific part of an actual machine and the angle in the design.

[0051] The present invention is not limited to the above-described embodiment, but includes various modified examples. For example, the above-described embodiment has been described in detail to easily explain the present invention, and is not necessarily limited to those including all of the configurations described. It is also possible to replace a part of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add the configuration of another embodiment to the configuration of one embodiment. It is also possible to add, delete, or replace a part of the configuration of each embodiment with another configuration.

[0052] The effects of this embodiment are as follows. (1) The structural difference confirmation device enables the user to know whether the difference between the actual dimensions and the design values ​​is within the tolerances, without being aware of the tolerances for each type of part. (2) The structural difference confirmation device can store point cloud data and 3D CAD data of parts. (3) The structural difference confirmation device can output a report indicating whether the difference between the actual dimensions and the design values ​​is within the tolerance. (4) The structural difference confirmation device can display, as augmented reality, to on-site workers whether the difference between the actual dimensions and the design values ​​is within the tolerances. [Explanation of symbols]

[0053] 1 Structure difference confirmation device 2 Network 3. Designer terminal device 4 Designer 5 On-site 6. Workers 7. Wearable Devices 8. Scanner 11 Central control unit 12 Input Devices 13 Output Devices 14 Main memory 15 Auxiliary storage 16 Communication Equipment 21 Point cloud data acquisition section 22 Item specific processing section 23 Tolerance Processing Section 24 Dimension Processing Section 25 Judgment processing unit 26 Report Output Section 27 On-site display processing section 31 Tolerance Data 32 Point Cloud Data 33 3D CAD data 34 Reports 41 Equipment 42, 43, 44 Piping

Claims

1. a point cloud data acquisition unit that acquires point cloud data resulting from imaging the part; an object identification processing unit that identifies a type of part from the acquired point cloud data; a tolerance processing unit that acquires a tolerance corresponding to the identified type of the part from a storage unit that stores a tolerance for each type of the part; a dimension processing unit that extracts a difference between a dimension in the point cloud data of the part and a dimension in three-dimensional CAD data as design information of the part; a determination processing unit that determines whether or not the extracted difference is within the acquired tolerance range; A structure difference confirmation device comprising:

2. The storage unit is storing the point cloud data and the three-dimensional CAD data; 2. The structure difference confirmation device according to claim 1,

3. a report output unit that outputs, as a report, whether or not the extracted difference is within the range of the acquired tolerance; 2. The structure difference confirmation device according to claim 1,

4. a site display processing unit that displays, as an augmented reality, whether or not the extracted difference is within the range of the acquired tolerance; 2. The structure difference confirmation device according to claim 1,

5. The point cloud data acquisition unit of the structure difference confirmation device includes: Obtain point cloud data that is the result of imaging the part, The object identification processing unit of the structure difference confirmation device includes: Identifying a type of part from the acquired point cloud data; The tolerance processing unit of the structure difference confirmation device includes: Obtaining a tolerance corresponding to the identified type of part from a storage unit that stores tolerances for each type of part; The dimension processing unit of the structure difference confirmation device includes: Extracting a difference between a dimension in the point cloud data of the part and a dimension in three-dimensional CAD data as design information of the part; The determination processing unit of the structural difference confirmation device includes: determining whether the extracted difference is within the obtained tolerance range; A method for confirming differences in structures, comprising:

Images