Control panel design apparatus, control panel design method, and program

Abstract

This invention provides a panel design apparatus, panel design method, and program for efficiently designing high-quality control panels. [Solution] The panel design device comprises: an information management unit that stores data for an existing panel installed in a building, including a deployment diagram, an assembly diagram, an outline drawing, a floor plan and elevation drawing of the building portion where the panel is installed, and structural specifications based on the outline drawing; a learning processing unit that trains a learning model on the relationship between the data for the single-line diagram of the existing panel and the data for the deployment diagram, the relationship between the data for the floor plan and elevation drawing of the existing panel and the data for the structural specifications, and the relationship between the data for the deployment diagram and structural specifications and the data for the assembly diagram, outline drawing, floor plan and elevation drawing; and a drawing processing unit that uses the learning model to generate data for the assembly diagram and outline drawing of the new panel, as well as data for the floor plan and elevation drawing of the building portion where the panel is installed, from the data for the single-line diagram of the new panel and the data for the floor plan and elevation drawing of the building portion where the panel is installed, which are provided from an external source.

Description

Technical Field

[0004] , , ,

[0006] , , , , , , , ,

[0005] , , , , , ,

[0001] Embodiments of the present invention relate to a panel design device, a panel design method, and a program.

Background Art

[0002] For example, within the site of a plant such as a water supply station, there are several buildings, and in some buildings, various panels (for example, incoming switchboards, distribution boards, rectifier panels, etc.) that make up electrical equipment and the like are installed. When newly installing a panel or replacing an existing panel, the outer shape and mounting (mounting of circuit components etc. inside the panel) of the newly installed panel are determined in advance. Generally, the determination of the outer shape and mounting of a panel is carried out in the following procedure.

[0003] (1) Based on the conditions of the installation location (constraint conditions etc.) and the single-line connection diagram included in the customer's (for example, △△ Waterworks Bureau ×× Water Supply Station) order drawing or proposal document, the technical department assumes the mounting components of the panel and determines the dimensions and shape (door specifications etc.) of the panel.

[0004] (2) Receiving the transfer of the specifications from the technical department, the design department creates a developed connection diagram of the panel and determines the mounting components of the panel. Note that it takes about one to two months from the completion of step (1) to the completion of step (2).

[0005] (3) Determine the internal mounting of the panel from the mounting components, wire sizes, etc., and determine the final dimensions of the panel. Note that it takes about one to two months from the completion of step (2) to the completion of step (3).

Prior Art Documents

Patent Documents

[0006]

Patent Document 1

Patent Document 2

[0007] Several problems can arise when determining the external dimensions and mounting of the control panel.

[0008] For example, in the procedure described above, a difference arises between the dimensions of the board determined in procedure (1) and the dimensions of the board derived from procedures (2) and (3). If the latter dimensions are larger than the former, it results in excess costs; on the other hand, if they are smaller, it incurs costs associated with rework. If different people examine the dimensions of the board under the same conditions, differences in dimensions may occur (i.e., variations in quality may occur). Also, if the same person examines the dimensions of the board under the same conditions on different days and times, for example, differences in dimensions may occur (i.e., variations in quality may occur).

[0009] The problem that this invention aims to solve is to provide a panel design apparatus, a panel design method, and a program that enable the efficient design of high-quality panel designs. [Means for solving the problem]

[0010] The control panel design apparatus of the embodiment comprises: an information management unit that stores data of an unfolded connection diagram, an assembly diagram, and an outline drawing of an existing control panel installed in a building, as well as data of a plan view and an elevation view of the building portion where the control panel is installed, and data of structural specifications based on the outline drawing in a storage device; a learning processing unit that trains a learning model on the relationship between the single-line diagram data of the existing control panel and the unfolded connection diagram data, the relationship between the plan view and elevation view data of the existing control panel and the structural specifications data, and the relationship between the unfolded connection diagram and structural specifications data and the assembly diagram, outline drawing, plan view, and elevation drawing data; and a drawing processing unit that uses the learning model to generate data of an assembly diagram and an outline drawing of a new control panel, as well as data of a plan view and an elevation view of the building portion where the control panel is installed, from data of a single-line diagram of a new control panel provided from an external source, as well as data of a plan view and an elevation view of the building portion where the control panel is installed. [Brief explanation of the drawing]

[0011] [Figure 1] Figure 1 shows an example of the basic configuration of a control panel design device according to an embodiment. [Figure 2] Figure 2 shows an example of the configuration of functions realized by the control panel design program 6. [Figure 3] Figure 3 shows the relationship between the learning model, explanatory variables, and target variable used by the learning processing unit 22 in Figure 2 during learning. [Figure 4] Figure 4 shows an example of the configuration of various functions of the learning processing unit 22 involved in controlling the operation of the "machine learning layer 1.1" model, the "machine learning layer 1.2" model, and the "machine learning layer 2" model. [Figure 5] Figure 5 shows a specific example of a floor plan used as an order drawing. [Figure 6] Figure 6 shows a specific example of an elevation drawing used as a commissioned drawing. [Figure 7] Figure 7 shows a specific example of a single-line diagram used as a customer order drawing. [Figure 8] Figure 8 shows a specific example of an outline drawing as an existing drawing. [Figure 9] FIG. 9 is a diagram showing a specific example of an implementation diagram as an existing drawing. [Figure 10] FIG. 10 is a diagram showing a specific example of information attached as part of the implementation diagram shown in FIG. 9. [Figure 11] FIG. 11 is a diagram showing a specific example of a single-line connection diagram as an existing drawing. [Figure 12] FIG. 12 is a diagram showing a specific example of a developed connection diagram as an existing drawing. [Figure 13] FIG. 13 is a diagram showing an example of the basic operation of the panel design. [Figure 14] FIG. 14 is a diagram showing the details of the learning process shown in FIG. 13. [Figure 15] FIG. 15 is a diagram showing the details of the drawing process shown in FIG. 13.

MODE FOR CARRYING OUT THE INVENTION

[0012] Hereinafter, embodiments will be described with reference to the drawings.

[0013] (Basic Configuration of the Device) FIG. 1 shows an example of the basic configuration of a panel design device according to an embodiment.

[0014] The panel design device shown in FIG. 1 is realized by a computer and includes an input device 1, an output device 2, a communication device 3, a memory 4, a storage device 5, a processor 10, and the like.

[0015] The input device 1 corresponds to a keyboard, a mouse, a touch pen, or a scanner, etc., and enables the input of various kinds of information.

[0016] The output device 2 corresponds to a display or a printer, etc., and outputs various kinds of information.

[0017] The communication device 3 communicates information with other information devices through a communication network such as the Internet.

[0018] Memory 4 corresponds to RAM (Random Access Memory) and temporarily stores various programs and data. Among these programs is the panel design program 6, which implements the functions of designing various types of control panels.

[0019] The storage device 5 corresponds to an HDD (Hard Disk Drive), SSD (Solid State Drive), ROM (Read Only Memory), etc., and stores various programs and data.

[0020] Processor 10 controls the overall operation of the computer and implements various functions by executing several programs in RAM. For example, processor 10 implements the function of designing a new control panel that is subject to a control panel design order by executing the control panel design program 6. Artificial Intelligence (AI), such as machine learning, is applied to this function.

[0021] This embodiment illustrates an example of designing a new control panel to replace some of the existing control panels located within a building on the premises of a plant such as a water supply station. In particular, the mounting components of the new control panel will be changed from those of the existing panel in order to meet customer requirements, which may result in differences in the dimensions and shape of each part of the new control panel. Furthermore, if it is required to install the new control panel while the existing control panel is still operational, the design of the new control panel will take into account the structural constraints and available installation area of ​​the new control panel within the building, so that it can be installed together with the existing control panel without any problems.

[0022] The input device 1 or communication device 3 described above takes in data from various drawings (existing drawings) related to existing panels installed in a building, and data from various drawings (ordered drawings) related to new panels that are subject to ordering panel design, in the form of CAD (Computer Aided Designed) data or image data, into the panel design device, or takes in the owner of the building in which the panel will be installed and building identification information to identify the building into the panel design device.

[0023] Building identification information is represented by the owner of the building where the panel will be installed, or the name of that building (the customer name of the client ordering the panel design, the building name, or a code or number to identify it). The reason for incorporating building identification information into the panel design device is to distinguish between the circuit components used and not used by different customers or buildings.

[0024] Output device 2 outputs data of various drawings (finished drawings) of the newly installed control panel, generated by the control panel design program 6 in response to an order for control panel design, for example, in the form of CAD data or image data. Building identification information may be added to the finished drawings.

[0025] (Functions of the control panel design program) Figure 2 shows an example of the configuration of functions realized by the control panel design program 6.

[0026] As shown in Figure 2, the functions realized by the control panel design program 6 include an input unit 11, a control panel design processing unit 12, and an output unit 13.

[0027] The control panel design program 6 handles the following order drawings, existing drawings, and completed drawings when designing control panels.

[0028] • Order drawings The order drawings include single-line diagrams, plan views, and elevation views. Each of these data sets is input from the input unit 11.

[0029] A single-line diagram is a simplified representation of the individual circuit elements and their electrical wiring to be installed in a newly constructed control panel that is the subject of a control panel design order. From a single-line diagram, it is possible to understand what kind of circuit elements are present, how many there are, and what function the control panel as a whole has.

[0030] The floor plan is a top-down view of the entire floor within the building where the new and existing electrical panels, which are the subject of the panel design order, will be installed. The floor plan shows the planned installation area for the new panels and the installation area for the existing panels. It also shows the distance between the new panels and surrounding walls and obstacles (such as other panels).

[0031] An elevation drawing is a horizontal view (arrow view, cross-section, etc.) of the part of a building where the newly installed and existing panels shown in the floor plan will be installed. Elevation drawings may include views from more than one direction, or from different directions (for example, the four cardinal directions: east, west, north, and south). The elevation drawing should show the distance between the newly installed panel and its surrounding ceilings, walls, and obstacles (such as other panels).

[0032] • Existing drawings The existing drawings include single-line diagrams, unfolded connection diagrams, assembly diagrams, outline drawings, plan drawings, and elevation drawings of the existing panel. In the following, plan drawings and elevation drawings may be referred to as layout drawings. The data for the unfolded connection diagrams, assembly diagrams, outline drawings, plan drawings, and elevation drawings of the existing panel are stored in the database DB. Furthermore, data indicating structural specifications (dimensions of each part of the panel and door structure type (e.g., double-door, front access)) obtained from the outline drawing of the existing panel are also stored in the database DB. The data stored in the database DB will be used for learning, as described later.

[0033] The single-line diagrams, floor plans, and elevations of the existing panels will contain the same information as those described above, except that they do not include information about the newly installed panels.

[0034] A deployment diagram is a more detailed diagram of the circuit configuration than a single-line diagram, and it includes information such as the number of wires (number of cables) connecting the circuit elements and the specifications of each wire (cable specifications).

[0035] The implementation diagram is a diagram showing the arrangement of individual circuit elements to be mounted on the control panel and the specifications of each individual circuit element.

[0036] The outline drawing shows the external shape of the panel, and includes details such as the dimensions of each part of the panel and the specifications of the door.

[0037] • Completed drawings The completed drawings include the installation drawing, outline drawing, and layout drawing (plan view and elevation view) of the newly installed panel. Each of these data is output from the output unit 13.

[0038] The input unit 11 receives data from existing drawings or ordered drawings (single-line diagrams and floor plans / elevations) obtained through the aforementioned input device 1 or communication device 3, as well as building identification information (hereinafter abbreviated as "identification information"), and transmits it to the panel design processing unit 12.

[0039] The control panel design processing unit 12 includes an information management unit 21, a learning processing unit 22, and a drawing processing unit 23, and uses these to perform control panel design processing.

[0040] The information management unit 21 stores the data for the deployment and connection diagrams, implementation diagrams, and outline drawings of the existing control panel, as well as the data for the floor plans and elevations of the building portion where the control panel is installed, and the data for structural specifications based on the outline drawings, in a database DB. The database DB may be located in the storage device 5 mentioned above, or in another location, for example, in a storage device outside the control panel design device (for example, a cloud storage device accessible via the communication device 3). The various types of data managed by the information management unit 21 can be pre-entered from the input unit 11.

[0041] The learning processing unit 22 has the function of causing the machine learning model 20 (hereinafter abbreviated as "learning model 20") to perform machine learning. Specifically, the learning processing unit 22 causes the learning model 20 to learn the relationship between the data of the single-line diagram of the existing panel and the data of the unfolded connection diagram, the relationship between the data of the plan view and elevation view of the existing panel and the data of the structural specifications, and the relationship between the data of the data of the unfolded connection diagram and structural specifications of the existing panel and the data of the implementation drawing, outline drawing, plan view, and elevation view.

[0042] The drawing processing unit 23 has the function of performing drawing processing using a pre-trained learning model 20. Specifically, the drawing processing unit 23 uses the pre-trained learning model 20 to generate data for the installation drawing and outline drawing of the new panel, as well as data for the floor plan and elevation drawing (completion drawing) of the building portion where the panel will be installed, from the single-line diagram data of the new panel input from the input unit 11, and the floor plan and elevation drawing (order drawing) data of the building portion where the panel will be installed.

[0043] The information management unit 21, the learning processing unit 22, and the drawing processing unit 23 each use the aforementioned identification information in their respective processes. Specifically, the information management unit 21 stores each piece of data for each piece of identification information, the learning processing unit 22 trains the learning model 20 on the aforementioned relationships for each piece of identification information, and the drawing processing unit 23 generates data from the above relationships for each piece of identification information.

[0044] The output unit 13 outputs the implementation drawing, outline drawing, plan view, and elevation drawing data of the newly installed panel, which are generated by the drawing processing unit 23. This information is sent to the output device 2 mentioned above.

[0045] (Relationship between the learning model, explanatory variables, and target variable) Figure 3 shows the relationship between the learning model, explanatory variables, and target variable used by the learning processing unit 22 in Figure 2 during learning.

[0046] The learning model 20 uses the data of the existing control panel's single-line diagram, plan view, and elevation view, as well as the corresponding identification information, as explanatory variables 30. The learning model 20 also uses the data of the existing control panel's implementation drawing, outline drawing, plan view, and elevation view, stored in the database DB, as the target variable 40.

[0047] More specifically, the learning model 20 has a hierarchical structure consisting of multiple models, including the "machine learning hierarchy 1.1" model (the first model), the "machine learning hierarchy 1.2" model (the second model), and the "machine learning hierarchy 2" model (the third model). The operation of these models is controlled by the learning processing unit 22.

[0048] "Machine Learning Layer 1.1" takes data from the single-line diagram of the existing control panel and corresponding identification information as input, and performs training using data from the expanded connection diagram of the existing control panel as training data.

[0049] "Machine Learning Layer 1.2" takes the plan and elevation drawings of the existing control panel as input and performs training using the structural specifications data of the existing control panel as training data.

[0050] "Machine Learning Layer 2" takes data from the existing panel's deployment and connection diagrams and structural specifications, as well as corresponding identification information, as input and performs training using the existing panel's implementation drawings, outline drawings, plan drawings, and elevation drawings as training data.

[0051] (Various functions of the learning processing unit 22) Figure 4 shows an example of the configuration of various functions of the learning processing unit 22 involved in controlling the operation of the "Machine Learning Layer 1.1" model, the "Machine Learning Layer 1.2" model, and the "Machine Learning Layer 2" model. The following explanation will also refer to Figure 3.

[0052] As shown in Figure 4, the learning processing unit 22 includes a circuit component determination unit 51, a constraint condition determination unit 52, an installable area determination unit 53, a dimension determination unit 54, and a drawing generation unit 55. At least some of these functions are also applied to the drawing processing unit 23. In that case, it goes without saying that these functions are used for drawing processing, not for learning processing.

[0053] The following will explain the circuit equipment determination unit 51, the constraint condition determination unit 52, the installable area determination unit 53, the dimension determination unit 54, and the drawing generation unit 55 in order.

[0054] ·Circuit equipment determination section 51 The learning processing unit 22, using the circuit component determination unit 51, obtains information indicating individual circuit components and their connection relationships from the data of the single-line diagram of the existing panel. For example, it extracts information indicating the symbols of individual circuit components drawn in the single-line diagram of the existing panel, as well as information indicating the connection relationships of each circuit component, and obtains information indicating the number and type of each circuit component, and the type of circuit configuration (for example, Y-Δ starting, VVVF starting, Condorfa starting, reversible / irreversible information, etc.).

[0055] Furthermore, the learning processing unit 22 obtains information (first information) indicating individual circuit components and their connection relationships from the data of the deployed connection diagram of existing panels in the database DB, using the circuit component determination unit 51.

[0056] The learning processing unit 22 then inputs the identification information of the existing panel and the information indicating the individual circuit components and their connection relationships obtained from the single-line diagram data of the existing panel as explanatory variables into the "machine learning layer 1.1" model, and uses the information indicating the individual circuit components and their connection relationships obtained from the expanded connection diagram data of the existing panel in the database DB (first information) as the target variable to perform learning.

[0057] ·Constraint condition determination unit 52 The learning processing unit 22 obtains information from the constraint condition determination unit 52 indicating the distance between the existing panel and the surrounding walls, obstacles, and ceiling (for example, the distance to the walls, obstacles, and ceiling in the front, back, left, right, and height directions when the panel is viewed from the horizontal direction).

[0058] Furthermore, the learning processing unit 22 obtains information indicating the constraints of the panel (second information) from the structural specification data in the database DB using the constraint condition determination unit 52. The information indicating the constraints includes, for example, the maximum dimensions of each part of the panel and the maximum opening and closing range of the door.

[0059] The learning processing unit 22 then inputs information indicating the distance between the existing panel and the surrounding walls, obstacles, and ceiling, obtained from the plan and elevation drawings of the existing panel, as explanatory variables into the "machine learning layer 1.2" model, and uses information indicating constraints (second information) obtained from the structural specifications data of the existing panel in the database DB as the target variable to perform learning.

[0060] ·Installation area determination unit 53 The learning processing unit 22, using the installable area determination unit 53, obtains information from the plan view and elevation view data of the existing panel, indicating the distance between the panel and the surrounding walls, obstacles, and ceiling, similar to the case of the constraint condition determination unit 52. Furthermore, from this information, it obtains information (third information) indicating the installable area for the new panel. The installable area is defined as an area where there is sufficient distance from walls, obstacles, and ceiling, and where at least the minimum size of the panel (width, depth, and height) can be secured (for example, W600 x D740 x H2300).

[0061] The learning processing unit 22 provides the model of "machine learning layer 2" with information indicating the installable area of ​​the newly installed panel (third information) as an explanatory variable.

[0062] • Dimension determination unit 54 The learning processing unit 22, using the dimension determination unit 54, obtains information (fourth information) indicating the dimensions, structure, and placement area of ​​each part of the existing panel from the implementation drawing, outline drawing, plan drawing, and elevation drawing data of the existing panel in the database DB.

[0063] The learning processing unit 22 then inputs the following as explanatory variables into the "machine learning layer 2" model: identification information of the existing panel, information indicating individual circuit components and their connection relationships obtained from the data of the existing panel's unfolded connection diagram (first information), information indicating constraints obtained from the data of the existing panel's structural specifications (second information), and information indicating the area in which the newly installed panel can be installed (third information). The learning processing unit 22 then uses information indicating the dimensions, structure, and placement area within the building of each part of the panel, obtained from the data of the existing panel's implementation diagram, outline drawing, plan drawing, and elevation drawing in the database DB (fourth information), as the target variable and performs learning.

[0064] • Drawing generation unit 55 The learning processing unit 22 can use the drawing generation unit 55 to generate data for completed drawings of new control panels from data for ordered drawings of new control panels, using the learned learning models 20 (i.e., the learned "machine learning hierarchy 1.1" model, the "machine learning hierarchy 1.2" model, and the "machine learning hierarchy 2" model). This function allows for evaluation of the learning results by inputting test data into the learning model 20 and checking the output data, but the learning processing unit 22 does not necessarily need to have the functionality of the drawing generation unit 55. On the other hand, the drawing processing unit 23 has the functionality of the drawing generation unit 55.

[0065] (Various functions of the drawing processing unit 23) The drawing processing unit 23 has functions equivalent to those of the learning processing unit 22 (such as the circuit equipment determination unit 51, constraint condition determination unit 52, and drawing generation unit 55 mentioned above). With these functions, the drawing processing unit 23 inputs the identification information of the newly installed panel input from the input unit 11, as well as the single-line diagram, plan view, and elevation view (order drawings) data of the newly installed panel provided by the client, into the trained learning model 20. This allows the trained "machine learning layer 1.1" model to generate data for the unfolded connection diagram of the newly installed panel, the trained "machine learning layer 1.2" model to generate data for the structural specifications of the newly installed panel, and the trained "machine learning layer 2" model to output data (CA data or image data) for the implementation drawing, outline drawing, plan view, and elevation view (finished drawing) of the newly installed panel. Furthermore, since the system can generate data for the deployment and connection diagram of a newly installed control panel from a pre-trained "machine learning layer 1.1" model, it can also output the data for that deployment and connection diagram (CA data or image data).

[0066] (Specific examples of order drawings, existing drawings, and completed drawings) The following provides specific examples of order drawings, existing drawings, and completed drawings.

[0067] • Order drawings Figures 5, 6, and 7 show specific examples of plan views, elevation views, and single-line diagrams used as order drawings, respectively.

[0068] Figure 5 shows the floor plan of the "□th floor of the ○× water supply station".

[0069] The floor plan in Figure 5 shows the installation locations of various panels to be placed on the floor, with numbers indicating their positions. The floor is divided into unit areas according to classification symbols 1-6 and A-D. Of the multiple panels, existing panels are those numbered 1-6, 21-22, and 31-36, while new panels for which panel design is to be ordered are those numbered 11-17. New panels are marked with decorations (e.g., hatching) to indicate the construction area. The legend contains information such as the correspondence between panel numbers and panel names. Although not shown in Figure 5, the floor plan in Figure 5 also includes information such as distances between objects (e.g., walls) and dimensions (length, width, etc.), which allows for the determination of the distance between new panels and surrounding walls and obstacles (other panels, etc.).

[0070] Figure 6 shows an elevation view corresponding to the arrow-direction view of section AA in Figure 5. Only one elevation view is shown here as an example, but there may be multiple elevation views corresponding to the arrow-direction views of parts other than section AA in Figure 5.

[0071] In the elevation view of Figure 6, similar to the plan view of Figure 5, the newly installed panel is decorated with hatching to indicate the construction area. The elevation view of Figure 6 includes information such as the height dimensions of each part, allowing the distance between the newly installed panel and the ceiling to be determined from this information.

[0072] Figure 7 shows a portion of the single-line diagram of a newly installed panel (a portion of the single-line diagram of the "System 1 Control Power Distribution Panel" numbered 11 in the plan view of Figure 5). Here, only the single-line diagram of the "System 1 Control Power Distribution Panel" is shown as an example, among the three types of distribution panels numbered 11 to 13 ("System 1 Control Power Distribution Panel," "Uninterruptible Power Distribution Panel," and "System 2 Control Power Distribution Panel"), but it is assumed that single-line diagrams of the other newly installed panels (various panels numbered 12 to 13 and 14 to 17 in the plan view of Figure 5) also exist. In the single-line diagram, the parts showing newly installed equipment are decorated (e.g., hatching) to indicate the scope of construction.

[0073] In the example single-line diagram in Figure 7, symbols for individual circuit components such as fuses, lamps, and miniature circuit breakers (MCCBs) installed in the "System 1 control power distribution panel," along with information about their associated components and connection relationships, are simply depicted.

[0074] • Existing drawings Here, we will show some specific examples of existing drawings, including single-line diagrams, unfolded connection diagrams, assembly diagrams, outline drawings, plan drawings, and elevation drawings.

[0075] Figure 8 shows a specific example of an outline drawing as an existing drawing.

[0076] Figure 8 shows the external outlines of the three types of distribution boards mentioned above ("System 1 Control Power Distribution Board," "Uninterruptible Power Distribution Board," and "System 2 Control Power Distribution Board"). The external outlines in Figure 8 include, for example, a front view, a rear view, and a right side view, but may also include a top view and a left side view.

[0077] Figure 8 shows that the product names for the "System 1 Control Power Distribution Panel," "Uninterruptible Power Distribution Panel," and "System 2 Control Power Distribution Panel" are represented by <1DC1>, <1AC>, and <1DC2>, respectively. Figure 8 also includes the dimensions of each part of the panel (including the position of the nameplate and door handle), as well as the specifications of the nameplate and door handle.

[0078] Figure 9 shows a specific example of an implementation drawing as an existing drawing.

[0079] The implementation diagram in Figure 9 shows the arrangement of individual circuit elements to be implemented in the "System 1 Control Power Distribution Panel <1DC1>". The implementation diagram in Figure 9 includes, for example, a front view, a view of the BB section shown in the front view, and a rear view, but there may also be views of parts other than the BB section.

[0080] Figure 9 shows the arrangement of the individual circuit elements implemented in the "System 1 Control Power Distribution Panel <1DC1>", as well as the numbers assigned to each circuit element.

[0081] Figure 10 shows a specific example of the information included as part of the implementation diagram shown in Figure 9.

[0082] Figure 10 shows the specifications of individual circuit elements as part of the implementation diagram shown in Figure 9. These specifications include the circuit element number (NO), part name, model, rating, equipment identification number (DEV.NO), application name, size (SIZE), etc.

[0083] Figure 11 shows a specific example of a single-line diagram as an existing drawing.

[0084] The single-line diagram in Figure 11 shows a circuit diagram centered on the "System 1 Control Power Distribution Panel <1DC1>", and also shows the relationship with other panels and equipment connected to the "System 1 Control Power Distribution Panel <1DC1>". Compared to the single-line diagram in the aforementioned order drawing, the single-line diagram in Figure 11 contains more detailed information (detailed specifications of individual circuit elements, and the circuit configuration of other panels and equipment connected to it).

[0085] Figure 12 shows a specific example of a development and connection diagram as an existing drawing.

[0086] The unfolded connection diagram in Figure 12, targeting the "System 1 Control Power Distribution Panel <1DC1>", shows a more detailed circuit configuration than the single-line diagram in Figure 11. Specifically, it includes the number of wires (number of cables) connecting the circuit elements and the specifications of each wire (cable specifications). The cable specifications include the indicator (number) assigned to each cable in the circuit diagram, the cable code, the number of cores, the connection source, and the connection destination.

[0087] While specific examples of existing floor plans and elevations are not provided here, these drawings are equivalent to or contain more detailed information (such as more detailed dimensions and structure of each part) than the aforementioned floor plans and elevations used as ordering drawings.

[0088] (Example of operation) The following describes the operation of the panel design, which is realized when the processor 10 executes the panel design program 6, with reference to Figures 13 to 15.

[0089] Figure 13 shows an example of the basic operation of control panel design.

[0090] In step S1, the processor 10 has previously stored in the database DB, data of the deployment and connection diagram, mounting diagram, and outline drawing of the existing panel, as well as data of the floor plan and elevation drawing of the building portion where the panel will be installed, and data of structural specifications based on the outline drawing, through the information management unit 21.

[0091] In step S2, the processor 10 uses the learning processing unit 22 to train the learning model 20 on the relationships between the data of the single-line diagram of the existing panel and the data of the unfolded connection diagram, the relationships between the data of the plan view and elevation view of the existing panel and the data of the structural specifications, and the relationships between the data of the data of the unfolded connection diagram and structural specifications of the existing panel and the data of the implementation drawing, outline drawing, plan view, and elevation view.

[0092] In step S3, the processor 10, using the drawing processing unit 23, generates implementation drawings and outline drawings of the new panel, as well as floor plans and elevation drawings (finished drawings) of the building portion where the panel will be installed, from the single-line diagram data of the new panel input from the input unit 11, and the floor plans and elevation drawings (order drawings) of the building portion where the panel will be installed.

[0093] Figure 14 shows the details of the learning process shown in Figure 13.

[0094] In step S11, the learning processing unit 22 obtains data from the database DB stored by the information management unit 21, including data for the deployment and connection diagram, the implementation diagram, and the outline drawing of the existing panel, as well as data for the floor plan and elevation of the building portion where the panel is installed, and data for the structural specifications based on the outline drawing.

[0095] In step S12, the learning processing unit 22 inputs the identification information of the existing panel and the information indicating the individual circuit components and their connection relationships obtained from the single-line diagram data of the existing panel as explanatory variables into the "machine learning layer 1.1" model, and uses the information indicating the individual circuit components and their connection relationships obtained from the expanded connection diagram data of the existing panel (first information) as the target variable to perform learning.

[0096] Simultaneously, the learning processing unit 22 inputs information indicating the distance between the existing panel and the surrounding walls, obstacles, and ceiling, obtained from the plan and elevation drawings of the existing panel, as explanatory variables into the "machine learning layer 1.2" model, and uses information indicating constraints (second information) obtained from the structural specifications of the existing panel as the target variable to perform learning. Furthermore, the learning processing unit 22 provides the obtained information indicating the area in which the new panel can be installed (third information) as an explanatory variable into the "machine learning layer 2" model.

[0097] In step S13, the learning processing unit 22 inputs the following as explanatory variables into the "machine learning layer 2" model: identification information of the existing panel, information indicating individual circuit components and their connection relationships obtained from the data of the existing panel's unfolded connection diagram (first information), information indicating constraints obtained from the data of the existing panel's structural specifications (second information), and information indicating the area in which the newly installed panel can be installed (third information). The learning processing unit then uses information indicating the dimensions, structure, and placement area within the building of each part of the panel, obtained from the data of the existing panel's implementation drawing, outline drawing, plan drawing, and elevation drawing (fourth information), as the target variable and performs learning.

[0098] Figure 15 shows the details of the drawing process shown in Figure 13.

[0099] In step S21, the drawing processing unit 23 inputs the identification information of the newly installed panel, as well as the data of the single-line diagram, plan view, and elevation view (order drawings) of the newly installed panel provided by the ordering party.

[0100] In step S22, the drawing processing unit 23 determines information indicating individual circuit components and their connection relationships from the single-line diagram data of the new panel, determines information indicating constraints from the structural specifications data of the new panel, and determines information indicating the installation area of ​​the new panel from the plan view and elevation view data.

[0101] In step S23, the drawing processing unit 23 inputs the information determined in step S22, along with the identification information of the newly installed panel, into the learning model 20, and causes the learning model 20 to generate data for the installation drawing and outline drawing of the newly installed panel, as well as data for the floor plan and elevation (finished drawing) of the building portion where the panel will be installed.

[0102] As described in detail above, according to the embodiment, it is possible to provide a panel design apparatus, a panel design method, and a program that enable efficient design of high-quality panel designs.

[0103] In particular, according to this embodiment, since panel design is performed using machine learning rather than by humans, dimensional differences like those in the past are eliminated, and the time required for panel design can be significantly reduced. In other words, variations in quality like those in the past are eliminated, and excessive costs and costs associated with rework can be reduced.

[0104] While several embodiments of the present invention have been described, these embodiments are presented as examples only and are not intended to limit the scope of the invention. These novel embodiments can be carried out in a variety of other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their variations are included in the scope and spirit of the invention, as well as in the claims of the invention and its equivalents. [Explanation of Symbols]

[0105] 1...Input device, 2...Output device, 3...Communication device, 4...Memory, 5...Storage device, 6...Panel design program, 10...Processor, 11...Input unit, 12...Panel design processing unit, 13...Output unit, 20...Learning model, 21...Information management unit, 22...Learning processing unit, 23...Drawing processing unit, 30...Explanatory variables, 40...Target variable, 51...Circuit equipment determination unit, 52...Constraint condition determination unit, 53...Installable area determination unit, 54...Dimension determination unit, 55...Drawing generation unit, DB...Database.

Claims

1. An information management unit stores data such as the unfolded connection diagram, assembly diagram, and outline drawing of an existing panel installed in a building, as well as data such as the floor plan and elevation drawing of the building portion in which the panel is installed, and data of structural specifications based on the outline drawing, in a storage device. A learning processing unit that trains a learning model on the relationship between the data of the single-line diagram of the existing panel and the data of the unfolded connection diagram, the relationship between the data of the plan view and elevation view of the existing panel and the data of the structural specifications, and the relationship between the data of the unfolded connection diagram and structural specifications and the data of the implementation drawing, outline drawing, plan view, and elevation view. A drawing processing unit generates implementation drawings and outline drawings of the newly installed panel, as well as plan and elevation drawings of the building portion where the panel is installed, from data of a single-line diagram of the newly installed panel provided externally, and data of a plan and elevation drawings of the building portion where the panel is installed, using the aforementioned learning model. A control panel design device equipped with the following features.

2. The aforementioned information management unit stores each piece of data for each piece of identification information that identifies the building. The learning processing unit uses the identification information provided from an external source to train the learning model on each of the relationships, The drawing processing unit generates data from the relationships corresponding to the identification information provided from an external source. The control panel design apparatus according to claim 1.

3. The learning processing unit applies the first model among the learning models, The identification information provided from an external source, and the information indicating the individual circuit components and their connection relationships obtained from the single-line diagram of the existing panel, are input as explanatory variables. The first piece of information, which shows the individual circuit components and their connection relationships obtained from the aforementioned unfolded connection diagram of the existing panel, is used as the target variable for learning. The control panel design apparatus according to claim 2.

4. The learning processing unit then applies the second model in the learning model to: Information indicating the distances between the existing panel and the surrounding walls, obstacles, and ceiling, obtained from the plan view and elevation view of the existing panel provided externally, is input as explanatory variables. The second piece of information, which represents the constraints obtained from the aforementioned structural specifications, is used as the target variable for learning. The control panel design apparatus according to claim 3.

5. The aforementioned learning processing unit, Based on information obtained from the plan view and elevation view of the existing panel, which shows the distances between the panel and the surrounding walls, obstacles, and ceiling, a third piece of information is obtained that indicates the area in which the new panel can be installed. The control panel design apparatus according to claim 4.

6. The learning processing unit then selects the third model among the learning models, The first information, the second information, and the third information of the existing panel, as well as the identification information, are input as explanatory variables. The learning process is performed using a fourth set of information as the target variable: the dimensions, structure, and placement area within the building of each part of the existing panel, obtained from the aforementioned implementation drawing, outline drawing, plan drawing, and elevation drawing of the panel. The control panel design apparatus according to claim 5.

7. The data of the unfolded connection diagram, assembly diagram, and outline drawing of the existing panel installed in the building, as well as the data of the floor plan and elevation drawing of the building portion in which the panel is installed, and the data of the structural specifications based on the outline drawing are stored in a storage device. The relationship between the data of the single-line diagram of the existing panel and the data of the unfolded connection diagram, the relationship between the data of the plan view and elevation view of the existing panel and the data of the structural specifications, and the relationship between the data of the unfolded connection diagram and structural specifications and the data of the implementation drawing, outline drawing, plan view, and elevation view are to be taught to the learning model. Using the aforementioned learning model, data is generated from externally provided single-line diagram data of the newly installed panel, and floor plan and elevation data of the building portion where the panel is installed, to generate implementation drawings and outline drawings of the newly installed panel, and floor plan and elevation data of the building portion where the panel is installed. A control panel design method, including the following.

8. On the computer, The system has a function to store in a storage device data such as the unfolded connection diagram, assembly diagram, and outline drawing of an existing panel installed in a building, as well as the floor plan and elevation drawing of the building portion where the panel is installed, and structural specification data based on the outline drawing. A function to train a learning model on the relationship between the data of the single-line diagram of the existing panel and the data of the unfolded connection diagram, the relationship between the data of the plan view and elevation view of the existing panel and the data of the structural specifications, and the relationship between the data of the unfolded connection diagram and structural specifications and the data of the implementation drawing, outline drawing, plan view, and elevation view. Using the aforementioned learning model, the system generates implementation drawings and outline drawings of the newly installed panel, as well as floor plans and elevation drawings of the building portion where the panel is installed, from data of a single-line diagram of the newly installed panel provided externally, and data of floor plans and elevation drawings of the building portion where the panel is installed. A program to achieve this.

Images