Drawing generation system and drawing generation method

Abstract

To automatically generate a drawing including at least one of process information and sequence information.SOLUTION: The drawing generation system 1 includes a drawing generation unit 10 that receives an input of input data including a design specification that is a book in which design information of a plurality of devices is described and outputs output data including drawing data. The drawing generation unit 10 includes a device information generation unit 11 that generates device information on the devices based on the input design specification, and a connection information generation unit 12 that generates connection information indicating a connection relationship between the devices based on the input design specification, and the drawing generation unit 10 generates drawing data in which a plurality of figures including symbols representing the devices and lines connecting the devices are drawn in a frame of the drawing based on the device information and the connection information generated by the device information generation unit 11 and the connection information generation unit 12, and output the generated drawing data.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] FIELD Embodiments of the present invention relate to drawing generation techniques. [Background technology]

[0002] A plant is composed of numerous elements, including equipment, piping, valves, and electrical instrumentation. Constructing, operating, and maintaining a plant requires the creation of piping and instrumentation diagrams (P&IDs) and electrical and instrumentation drawings, which show the relationships between these elements. Examples of electrical and instrumentation drawings include interlock block diagrams (IBDs), single-line diagrams, and electrical wiring diagrams (ECWDs). To create these drawings, engineers with specialized knowledge and design experience extract and organize information from system design specifications and equipment design specifications, such as the system overview, design conditions (temperature, etc.), design principles (function, capability, capacity), and equipment specifications. Engineers then consider and design the configuration and connections of the piping, equipment, and instrumentation, and create the drawings using specialized CAD tools based on drawing rules. There is a need to reduce the time required for this process.

[0003] In recent years, large-scale language models (LLMs) that have been trained using large amounts of language data have been developed. Technologies that use these large-scale language models to support the generation of construction drawings are known. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent No. 7341581 Summary of the Invention [Problem to be solved by the invention]

[0005] There have been examples of using large-scale language models to improve the efficiency of drawing generation, but these were targeted at construction drawings and did not support the generation of drawings that handle process and sequence information such as P&ID, IBD, and ECWD. Large-scale language models can organize and extract the text content of design specifications, but they cannot handle the process and sequence information required to generate drawings, making it impossible to automatically generate drawings based on information such as design specifications.

[0006] The embodiments of the present invention have been made in consideration of the above circumstances, and have an object to automatically generate a drawing including at least one of process information and sequence information. [Means for solving the problem]

[0007] A drawing generation system according to an embodiment of the present invention includes a drawing generation unit that receives input data including a design specification, which is a document describing design information for a plurality of devices, and outputs output data including drawing data. The drawing generation unit includes: a device information generation unit that generates device information related to the devices based on the input design specification; and a connection information generation unit that generates connection information indicating the connection relationships between the devices based on the input design specification. The drawing generation unit is configured to generate the drawing data, based on the device information and connection information generated by the device information generation unit and the connection information generation unit, depicting a plurality of shapes including symbols representing the devices and lines connecting the devices within a drawing frame, and output the generated drawing data. [Effects of the Invention]

[0008] Embodiments of the present invention allow for the automatic generation of drawings that include process information and / or sequence information. [Brief explanation of the drawings]

[0009] [Figure 1] FIG. 1 is a block diagram showing a drawing generation system according to a first embodiment. [Figure 2] FIG. 2 is a block diagram showing a drawing generation unit according to the first embodiment. [Figure 3] 4 is a flowchart showing a drawing generation process according to the first embodiment. [Figure 4] 4 is a table showing device information according to the first embodiment. [Figure 5] 4 is a table showing connection information according to the first embodiment. [Figure 6] 3 is a table showing graphic symbols according to the first embodiment. [Figure 7] FIG. 2 is an explanatory diagram showing an example of the arrangement of graphic symbols according to the first embodiment. [Figure 8] FIG. 2 is an explanatory diagram showing an example of the arrangement of names and specifications according to the first embodiment. [Figure 9] FIG. 2 is an explanatory diagram showing an example of adding drawing information according to the first embodiment. [Figure 10] FIG. 10 is a block diagram showing a drawing generation unit according to a second embodiment. [Figure 11] 10 is a flowchart showing a drawing generation process according to a second embodiment. [Figure 12] FIG. 10 is a block diagram showing a drawing generation unit according to a third embodiment. [Figure 13] FIG. 11 is an explanatory diagram showing a prompt for generating device information according to the third embodiment. [Figure 14] FIG. 11 is an explanatory diagram showing a prompt for generating connection information according to the third embodiment. [Figure 15] 10 is a flowchart showing a drawing generation process according to a third embodiment. [Figure 16] FIG. 10 is a block diagram showing a drawing generation unit according to a fourth embodiment. [Figure 17] FIG. 13 is an explanatory diagram showing an input drawing prompt according to a fourth embodiment. [Figure 18] FIG. 13 is a block diagram showing a drawing generation unit according to a fifth embodiment. [Figure 19] FIG. 13 is an explanatory diagram showing a prompt according to the fifth embodiment. [Figure 20] 13 is a flowchart showing a prompt generation process according to the fifth embodiment. [Figure 21] FIG. 13 is a block diagram showing a drawing generation unit according to a sixth embodiment. [Figure 22] 13 is a flowchart showing a drawing generation process according to the sixth embodiment. [Figure 23] FIG. 13 is an explanatory diagram showing a graph based on the design specifications according to the sixth embodiment. [Figure 24] FIG. 13 is an explanatory diagram showing an example displayed on the display unit of the sixth embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0010] (First embodiment) DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of a drawing generation system and a drawing generation method will be described in detail with reference to the drawings. First, a first embodiment will be described with reference to FIGS.

[0011] 1, reference numeral 1 denotes a drawing generation system of this embodiment. This drawing generation system 1 is used to carry out a drawing generation method.

[0012] The drawing generation system 1 generates drawings from design specifications for a plant. The plant is, for example, a predetermined large-scale facility such as a power plant, a chemical plant, or a factory. The drawing generation system 1 may be applied not only to plants but also to other facilities.

[0013] The system configuration of the drawing creation system 1 will be described with reference to the block diagrams shown in Figures 1 and 2. Note that the drawing creation system 1 may include components other than those shown in these block diagrams, or some of the components may be omitted.

[0014] The drawing generation system 1 is configured as a computer having hardware resources such as a CPU (Central Processing Unit), GPU (Graphics Processing Unit), ROM (Read Only Memory), RAM (Random Access Memory), HDD (Hard Disk Drive), and SSD (Solid State Drive), and the CPU executes various programs to realize software-based information processing using the hardware resources. Furthermore, the drawing generation method is realized by having the computer execute the various programs.

[0015] Each component of the drawing creation system 1 does not necessarily have to be provided in one computer. For example, one drawing creation system 1 may be realized by a plurality of computers connected to each other via a network.

[0016] The configuration of the drawing generation system 1 may be realized as a cloud service. That is, the computer that constitutes the drawing generation system 1 may be a server on the cloud. For example, not only the configuration that performs storage processing but also the entire configuration that performs main processing may exist on the cloud, and the user may only configure the drawing generation system 1 and check input / output via an API (Application Programming Interface) or a web browser.

[0017] 1, the drawing generation system 1 includes a drawing generation computer 2. The drawing generation computer 2 includes a processing circuit 3, a storage unit 4, a communication unit 5, an input unit 6, and an output unit 7.

[0018] The processing circuit 3 is a circuit including, for example, a CPU, a GPU, or a dedicated or general-purpose processor. This processor realizes various functions by executing various programs stored in the storage unit 4. The processing circuit 3 may also be configured with hardware such as an FPGA (Field Programmable Gate Array) or an ASIC (Application Specific Integrated Circuit). Various functions can also be realized by such hardware. The processing circuit 3 can also realize various functions by combining software processing by a processor and a program with hardware processing.

[0019] The storage unit 4 stores predetermined programs and trained models executed by the processing circuit 3. The storage unit 4 also stores various information required when the drawing generation computer 2 generates drawings. For example, the storage unit 4 stores design specifications and drawing rules.

[0020] The storage unit 4 may have a predetermined database. A database is a collection of information stored in a memory, a HDD, or a cloud computing resource, and organized so that it can be searched or stored.

[0021] The communication unit 5 communicates with other computers via a communication line such as the Internet. Note that the drawing generation computer 2 and the other computers are connected to each other via the Internet, but other modes are also possible. For example, the drawing generation computer 2 and the other computers may be connected to each other via a LAN (Local Area Network), a WAN (Wide Area Network), or a mobile communication network. Furthermore, each device may be connected to each other via a bus.

[0022] Predetermined information is input to the input unit 6 in response to operations by a user of the drawing generation computer 2. The input unit 6 includes input devices such as a mouse, keyboard, and touch panel. In other words, predetermined information is input to the drawing generation computer 2 in response to operations of these input devices.

[0023] The output unit 7 outputs predetermined information. The output unit 7 includes a device for displaying images, such as a display that outputs the processing results. The display may be separate from the computer main body or may be integrated with the computer main body.

[0024] 2, the drawing generation computer 2 includes a drawing generation unit 10. The drawing generation unit 10 and its functions are realized by the processing circuit 3 executing a program and a trained model stored in the storage unit 4.

[0025] That is, the drawing generation system 1 includes a trained model that has been machine-learned in advance. Note that the drawing generation system 1 may include a plurality of trained models.

[0026] The trained model comprises an input layer, an intermediate layer, and an output layer. Input data is input to the input layer. The intermediate layer's parameters are machine-learned in advance using training data. The output layer outputs output data indicating the results of processing in the intermediate layer in response to input data input to the input layer. The trained model is machine-learned using training data in which at least one of the input data or data simulating the input data is input, and at least one of the output data or data simulating the input data is output.

[0027] The components of the drawing generation unit 10 as a trained model do not necessarily need to be provided on a single computer. For example, one drawing generation unit 10 may be realized by multiple computers connected to each other via a network. Note that the components of the drawing generation unit 10 may also be realized as a cloud service.

[0028] The trained model is, for example, a large language model (LLM). A large language model is a natural language processing model built using deep learning technology, and is designed to learn numerical representations of tokens (words and characters) and understand context. This enables the large language model to perform tasks such as sentence generation, semantic analysis, and question answering. In this embodiment, the large language model can generate (output) drawings. As a specific configuration example, a Transformer architecture is used to model dependencies between tokens and perform language understanding and generation.

[0029] Furthermore, a Vision-Language Model (VLM) capable of integrated processing of both visual and linguistic information may be used as a trained model. By learning the correspondence between images and text, a Vision-Language Model can simultaneously understand and process visual information such as drawings and linguistic information such as design specifications. This enables more accurate generation of drawings based on the contents of design specifications. An example of a VLM configuration is a multimodal transformer that combines an image encoder and a text encoder.

[0030] The drawing generation unit 10 of the drawing generation computer 2 inputs input data including a design specification, which is a document containing design information for multiple devices, into the trained model, and outputs output data including drawing data from the trained model.

[0031] The multiple devices are devices installed in a plant, such as pumps, check valves, manual valves, motor-operated valves, flow valves, piping, wiring, and cables. The devices may also include other items. The piping, wiring, and cables may also be configured as components for connecting other devices to each other.

[0032] The trained model generates device information about the devices and connection information indicating the connection relationships between the devices based on the input design specifications. Furthermore, the trained model generates drawing data depicting multiple figures, including symbols representing the devices and lines connecting the devices, within a drawing frame based on the generated device information and connection information. Furthermore, the trained model outputs the generated drawing data.

[0033] The processing circuit 3 may have some of the functions of the trained model. For example, the processing circuit 3 may temporarily store the device information and connection information generated by the trained model as output data in the storage unit 4, and input the device information and connection information temporarily stored in the storage unit 4 as input data to the trained model.

[0034] The drawing generation unit 10 of the drawing generation computer 2 inputs input data including drawing rules together with the design specifications into the trained model. The drawing rules include at least one of the line type used for drawing, the notation method for drawing, and the design rules for the equipment. The trained model generates drawing data based on the drawing rules.

[0035] The trained model generates attribute information including at least one of the device name and specifications based on the input design specifications, and assigns the generated attribute information to the drawing data in association with the figures.

[0036] 2, the drawing generating unit 10 includes various functional units. For example, the drawing generating unit 10 includes a device information generating unit 11, a connection information generating unit 12, and a layout unit 13.

[0037] The design specification and the drawing rules are input to the drawing generation unit 10 in response to an input operation by the user. That is, the drawing generation unit 10 receives input data including the drawing rules from the user. Note that it is sufficient that at least the design specification is input to the drawing generation unit 10.

[0038] A design specification is a document that describes design information for plant systems and equipment used to draw drawings. Design specifications include, for example, system design specifications and equipment design specifications.

[0039] The drawing rules describe design rules such as line types and notation methods when drawing. They also include design rules such as the need to place a protection circuit immediately before a specific device when using that device.

[0040] In addition to text, document files such as Word and PDF can also be input to the drawing generation unit 10. Also, image files can be input and the characters and images in the files can be extracted.

[0041] The equipment information generation unit 11 (FIG. 2) extracts equipment information about the equipment described in the input design specifications and drawing rules. The equipment information includes information about piping and wiring. Furthermore, the equipment information generation unit 11 organizes the equipment information to generate design information. The following explanation uses a piping and instrumentation diagram (P&ID) as an example.

[0042] The term "extraction" does not simply mean extracting information, but also means extracting related information from multiple design specifications and organizing this information. In other words, the aspect of extracting device information or connection information from design specifications includes the aspect of organizing the extracted information.

[0043] Figure 4 shows an example of equipment information extracted from a system design specification. This equipment information includes the equipment name as the primary key, along with information such as the equipment type, equipment specifications, and the system name to which the equipment belongs.

[0044] The connection information generating unit 12 (FIG. 2) extracts connection information relating to the connection relationship between the equipment described in the design specifications and the drawing rules and the piping or wiring.

[0045] Figure 5 shows an example of connection information extracted from design specifications and drawing rules. This connection information registers information such as connection points and connection destinations, with the device name as the primary key. In this example, for each device, the device and its connection destination are registered. When describing this connection relationship, it is possible for one device to have multiple connection points. For example, a device may have an upstream connection point and a downstream connection point. Figure 5 expresses the connection points as "Port_a" or "Port_b," and records which connection point of the destination device each device is connected to.

[0046] The layout unit 13 (FIG. 2) creates drawings based on device information, connection information, and drawing rules. The layout unit 13 is compatible with any drawing tool and can create drawings in a format such as 2D CAD. The layout unit 13 also includes drawing templates (drawing frames) to be used within the drawing tool.

[0047] The layout unit 13 includes, for example, a symbol database (not shown) that holds drawing symbols used in depiction. Note that the user can register any drawing symbol in the symbol database.

[0048] The layout unit 13 also has a function of referencing a symbol database and organizing the correspondence between equipment information and drawing symbols. The layout unit 13 also has a function of arranging and depicting the organized drawing symbols within a drawing frame. The drawing symbols include symbols indicating piping and wiring.

[0049] Figure 6 shows an example of a graphic symbol used in a piping and instrumentation diagram (P&ID). This graphic symbol has information such as the graphic symbol registered with the equipment type as the primary key.

[0050] FIG. 7 shows an example in which the layout unit 13 has drawn drawing symbols (G) for pumps, valves, etc., and lines (L) for piping, etc., within a drawing frame (F).

[0051] FIG. 8 is an example in which the layout unit 13 writes the name (N) and specifications (S) of the device on the drawing.

[0052] The layout unit 13 has a function for assigning information such as the name (N) and specifications (S) of each drawing symbol (G) representing equipment and the line (L) representing piping. This attribute information may be explicitly written on the drawing, or may be stored as supplementary information for the drawing symbol (G) even if it is not explicitly written.

[0053] Figure 9 shows an example of adding the system name (SN), drawing name (DN), and drawing date (D) to the drawing frame.

[0054] The layout unit 13 has a function of writing drawing information including information about the drawing such as a system name (SN) and a drawing name (DN) in a drawing frame.

[0055] The drawing generation unit 10 (FIG. 2) has a function of outputting, as drawing data, the drawing created by the layout unit 13. For example, the drawing generation unit 10 outputs the drawing data in a file format compatible with any of the above-mentioned drawing tools.

[0056] Next, the drawing generation process executed by using the drawing generation system 1 will be described with reference to the flowchart in Fig. 3. Note that the drawings mentioned above may be referenced.

[0057] First, in step S1, a design specification and drawing rules are input by a user to a drawing generation computer 2 (FIG. 1). That is, the drawing generation computer 2 receives input data including the design specification and drawing rules from the user. Then, the processing circuit 3 inputs the design specification and drawing rules to a drawing generation unit 10 (FIG. 2). For example, the processing circuit 3 extracts text information from the design specification and drawing rules and inputs the extracted text information to the drawing generation unit 10.

[0058] In the next step S2, the device information generation unit 11 generates device information about the devices described in the drawings based on the design specifications. For example, the device information generation unit 11 generates device information (FIG. 4) about the devices, piping, cables, etc. described in the design specifications. The device information generation unit 11 inputs the device information to the connection information generation unit 12.

[0059] In the next step S3, the connection information generation unit 12 generates connection information indicating the mutual connection relationships between devices, pipes, cables, etc., based on the design specifications and drawing rules. The connection information generation unit 12 inputs the device information and connection information to the layout unit 13.

[0060] In the next step S4, the layout unit 13 draws graphic symbols on the drawing based on the device information and connection information. For example, the layout unit 13 places graphic symbols representing devices within a drawing frame. The layout unit 13 also draws piping or cable lines between the devices. The layout unit 13 draws the graphic symbols by referring to a symbol database (not shown).

[0061] In the next step S5, the layout unit 13 assigns attribute information such as names and specifications to the graphic symbols in the drawing. For example, the layout unit 13 assigns names and specification information to the equipment, piping, and cables depicted in the drawing frame.

[0062] In the next step S6, the layout unit 13 assigns drawing information including information about the drawing such as a system name and a drawing name to the drawing.

[0063] In the next step S7, the drawing generating unit 10 outputs drawing data including the generated drawing, and the process is completed.

[0064] According to the first embodiment, the processing circuitry 3 extracts device information and connection information, which are information required for drawings, from the design specifications as text information, and thereby allows drawings containing at least one of process information and sequence information to be handled by the program and the trained model. Therefore, it is possible to automate the work that previously required manual drawing.

[0065] (Second embodiment) Next, a second embodiment will be described with reference to Figures 10 and 11. The aforementioned drawings may be referenced. Components identical to those shown in the aforementioned embodiment are designated by the same reference numerals, and redundant description will be omitted.

[0066] The second embodiment is configured so that the device information generation unit 11, the connection information generation unit 12, and the layout unit 13 can supplement and generate drawings even when they lack knowledge about the drawings, by referring to past drawings that have already been accumulated.

[0067] In a second embodiment, the trained model extracts drawing information from similar design specifications similar to the design specification and similar drawings related to the similar design specifications. This drawing information includes at least one of similar device information similar to the device information, similar connection information similar to the connection information, and similar layout information indicating the layout of the similar drawings. The trained model then generates drawing data based on the extracted drawing information.

[0068] In addition, the trained model detects differences between similar design specifications and the design specification, and generates drawing data based on the detected differences.

[0069] As shown in Fig. 10, there are provided a design specification database 20 and a drawing database 21. These databases may be provided in the drawing generation computer 2 (Fig. 1) or in another computer connected via a network.

[0070] The design specification database 20 stores similar design specifications, which are past design specifications similar to the design specification of the input data. The drawing database 21 stores similar drawings, which are past drawings similar to the drawing to be generated. The drawing database 21 stores similar drawings linked to similar design specifications.

[0071] The drawing generation unit 10 includes a device information generation unit 11, a connection information generation unit 12, a layout unit 13, a similar design specification acquisition unit 14, a similar drawing acquisition unit 15, a difference detection unit 16, and a drawing information extraction unit 17.

[0072] The similar design specification acquisition unit 14 acquires similar design specifications from the design specification database 20. The similar drawing acquisition unit 15 acquires similar drawings from the drawing database 21.

[0073] The difference detection unit 16 has a function of detecting the difference between the design specification of the input data and the similar design specification.

[0074] The drawing information extraction unit 17 has a function of extracting drawing information including similar device information, similar connection information, and similar layout information from similar drawings acquired by the similar drawing acquisition unit 15 from the drawing database 21. For example, if the similar drawings are CAD format data, drawing information including similar device information, similar connection information, and similar layout information may be extracted from the text in the CAD format data. Also, if the similar drawings are paper or PDF format data, they may be converted into an image format, and then image processing such as an object detection algorithm, optical character recognition, or line segment detection may be used to detect symbols and text on the drawings, as well as wiring indicating the connection status of each symbol, to extract drawing information including similar device information, similar connection information, and similar layout information.

[0075] Next, the drawing generation process executed by using the drawing generation system 1 will be described with reference to the flowchart in Fig. 11. Note that the drawings mentioned above may be referenced.

[0076] First, in step S11, the processing circuit 3 (FIG. 1) inputs the design specifications and drawing rules to the drawing generating unit 10 (FIG. 10).

[0077] In the next step S12, the similar design specification acquisition unit 14 acquires, from the design specification database 20, similar design specifications that are similar to the input design specification.

[0078] In the next step S13, the difference detection unit 16 compares the input design specification with the similar design specification, and detects differences where the design specification differs from the similar design specification.

[0079] In the next step S14, the similar drawing acquisition unit 15 acquires, from the drawing database 21, a similar drawing that is related to the similar design specification and is linked to the similar design specification.

[0080] In the next step S15, the drawing information extraction unit 17 extracts drawing information including similar device information, similar connection information, and similar layout information from the similar drawings. When extracting the drawing information, the difference parts are referred to as appropriate.

[0081] In the next step S16, the drawing generating unit 10 inputs the similar device information, similar connection information, and similar layout information of the similar drawing to the device information generating unit 11, the connection information generating unit 12, and the layout unit 13.

[0082] In the next step S17, the equipment information generating unit 11 refers to the equipment information and connection information of the similar drawings and generates equipment information (FIG. 4) of the equipment, piping, cables, etc. described in the drawings from the design specifications.

[0083] In the next step S18, the connection information generating unit 12 refers to the equipment information and connection information of the similar drawings and generates connection information indicating the mutual connection relationships between the equipment, piping, cables, etc. described in the drawings from the design specifications.

[0084] In the next step S19, the layout unit 13 draws the graphic symbols on the drawing based on the device information and connection information and by referring to the similar layout information of the similar drawing.

[0085] In the next step S20, the layout unit 13 assigns attribute information such as names and specifications to the graphic symbols on the drawing.

[0086] In the next step S21, the layout unit 13 assigns drawing information including information about the drawing such as a system name and a drawing name to the drawing.

[0087] In the next step S22, the drawing generating unit 10 outputs drawing data including the generated drawing, and the process is completed.

[0088] According to the second embodiment, even when the trained model lacks knowledge about drawings, the device information generation unit 11, the connection information generation unit 12, and the layout unit 13 can compensate for the lack of knowledge. That is, the missing information is compensated for by the device information, connection information, and layout information described in similar design specifications and similar drawings. Therefore, the trained model can automatically generate drawings. Furthermore, the trained model may be retrained or additionally trained when the missing information is to be compensated for.

[0089] (Third embodiment) Next, a third embodiment will be described with reference to Figures 12 to 15. The aforementioned drawings may be referenced. Components identical to those shown in the aforementioned embodiments are designated by the same reference numerals, and redundant description will be omitted.

[0090] The third embodiment uses a large-scale language model that requires prompt input, where the prompt contains at least design information.

[0091] In the third embodiment, the trained model generates a prompt that instructs the generation of device information and connection information, and generates drawing data based on the generated prompt.

[0092] 12, the drawing generation unit 10 includes a device information generation unit 11, a connection information generation unit 12, and a layout unit 13, as well as a prompt generation unit 18. The prompt generation unit 18 generates prompts to be input to the trained model based on the design specifications and drawing rules.

[0093] In addition, the processing circuit 3 may temporarily store the prompt generated by the prompt generation unit 18 in the memory unit 4 as output data, and input the prompt temporarily stored in the memory unit 4 as input data to the trained model.

[0094] Figures 13 and 14 are examples of prompts. Figure 13 is a prompt that instructs the generation of equipment information. Figure 14 is a prompt that instructs the generation of connection information. These are prompts for generating equipment information and connection information for P&ID. These prompts contain the equipment information and connection information extracted from the design specifications, the rules extracted from the drawing rules, and instructions for extracting the information.

[0095] Next, the drawing generation process executed by using the drawing generation system 1 will be described with reference to the flowchart in Fig. 15. Note that the drawings mentioned above may be referenced.

[0096] First, in step S31, the processing circuit 3 (FIG. 1) inputs the design specifications and drawing rules to the drawing generating unit 10 (FIG. 12).

[0097] In the next step S32, the prompt generating unit 18 extracts design information from the design specifications, and further extracts rules relating to drawing from the drawing rules.

[0098] In the next step S33, the prompt generating unit 18 generates a device information generating prompt (FIG. 13) that describes the design information and drawing rules and instructs the generation of device information.

[0099] In the next step S34, the prompt generating unit 18 generates a connection information generating prompt (FIG. 14) that describes the design information and drawing rules and instructs the generation of connection information.

[0100] In the next step S35, the prompt generating unit 18 inputs a device information generating prompt (FIG. 13) to the device information generating unit 11.

[0101] In the next step S36, the prompt generating unit 18 inputs a connection information generating prompt (FIG. 14) to the connection information generating unit 12.

[0102] In the next step S37, the device information generating unit 11 and the connection information generating unit 12 input the device information and connection information to the layout unit 13.

[0103] In the next step S38, the layout unit 13 draws the graphic symbols on the drawing based on the device information and connection information and by referring to the similar layout information of the similar drawing.

[0104] In the next step S39, the layout unit 13 assigns attribute information such as names and specifications to the graphic symbols on the drawing.

[0105] In the next step S40, the layout unit 13 assigns drawing information including information about the drawing such as the system name and drawing name to the drawing.

[0106] In the next step S41, the drawing generating unit 10 outputs drawing data including the generated drawing, and the process is completed.

[0107] According to the third embodiment, even when a large-scale language model that requires prompt input is used, the prompt can be automatically generated, and a drawing can be automatically generated.

[0108] (Fourth embodiment) Next, a fourth embodiment will be described with reference to Figures 16 and 17. The drawings mentioned above may be referenced. Components that are the same as those shown in the above-mentioned embodiments are given the same reference numerals, and duplicated descriptions will be omitted.

[0109] The fourth embodiment uses a large-scale language model that requires prompt input, where the prompt contains at least design information.

[0110] 16, the drawing generation unit 10 includes a device information generation unit 11, a connection information generation unit 12, and a layout unit 13. Input data including a design specification, drawing rules, and input drawing prompts is input to the drawing generation unit 10.

[0111] 17 shows an example of an input drawing prompt. The input drawing prompt contains instructions for driving the drawing generator 10. The input drawing prompt may be entered by a user or may be read from a predetermined database containing prompt templates.

[0112] According to the fourth embodiment, the drawing generation unit 10 can be appropriately driven by creating an input drawing prompt. When a user creates an input drawing prompt, the drawing generation unit 10 can be driven as intended by the user. When a prompt is read from a predetermined database, the user does not need to take the time to create a prompt.

[0113] (Fifth embodiment) Next, a fifth embodiment will be described with reference to Figures 18 to 20. The drawings mentioned above may be referenced. Components that are the same as those shown in the above-mentioned embodiments are given the same reference numerals, and duplicated descriptions will be omitted.

[0114] The fifth embodiment uses a large-scale language model that requires prompt input, where the prompt contains at least design information.

[0115] In the fifth embodiment, the trained model generates a prompt including at least one of similar device information, similar connection information, and similar layout information, and generates drawing data based on the generated prompt. The trained model also generates a prompt including a difference, and generates drawing data based on the generated prompt.

[0116] As shown in FIG. 18, a design specification database 20 and a drawing database 21 are provided.

[0117] The drawing generation unit 10 includes a similar design specification acquisition unit 14, a similar drawing acquisition unit 15, a difference detection unit 16, a drawing information extraction unit 17, and a prompt generation unit 18.

[0118] An example of a prompt is shown in Figure 19. This prompt contains information about drawing instructions, design specification information, drawing rules, as well as differences with similar design specifications, similar drawings, and equipment information.

[0119] Next, the prompt generation process executed by using the drawing generation system 1 will be described with reference to the flowchart in Fig. 20. Note that the aforementioned drawings may be referenced.

[0120] First, in step S51, the processing circuit 3 (FIG. 1) inputs the design specifications and drawing rules to the drawing generating unit 10 (FIG. 18).

[0121] In the next step S52, the similar design specification acquisition unit 14 acquires from the design specification database 20 similar design specifications that are similar to the input design specification.

[0122] In the next step S53, the difference detection unit 16 compares the input design specification with the similar design specification, and detects differences where the design specification differs from the similar design specification.

[0123] In the next step S54, the drawing generating unit 10 inputs the similar design specifications and the differences to the prompt generating unit 18.

[0124] In the next step S55, the similar drawing acquisition unit 15 acquires, from the drawing database 21, a similar drawing that is related to the similar design specification and is linked to the similar design specification.

[0125] In the next step S56, the device information generation unit 11 references the device information and connection information of the similar drawings and generates device information from the design specifications. Also, the connection information generation unit 12 references the device information and connection information of the similar drawings and generates connection information from the design specifications.

[0126] In the next step S57, the device information generating unit 11 and the connection information generating unit 12 input the device information and connection information to the prompt generating unit .

[0127] In the next step S58, the prompt generator 18 extracts design information from the design specifications.

[0128] In the next step S59, the prompt generating unit 18 generates a prompt (FIG. 19) that describes the design information, device information, connection information, and difference locations, and instructs the generation of device information and connection information, and then completes the process.

[0129] According to the fifth embodiment, even when a large-scale language model that requires prompt input is used, it is possible to automatically generate prompts and generate drawings. Furthermore, by adding the contents of similar design specifications, device information and connection information of similar drawings to the prompt, it is possible to generate drawings even when using a large-scale language model that lacks drawing knowledge.

[0130] (Sixth embodiment) The sixth embodiment will be described with reference to Figures 21 to 23. The aforementioned drawings may be referenced. The same components as those shown in the aforementioned embodiments are designated by the same reference numerals, and redundant description will be omitted.

[0131] The sixth embodiment is configured so that the device information generation unit 11 and the connection information generation unit 12 can appropriately reflect similar device information and similar connection information of similar drawings even when device information and connection information are insufficient in the design specifications.

[0132] In the sixth embodiment, the drawing generation system 1 includes a display unit 25 that displays drawings, graphs, and similar drawings generated by the drawing generation unit 10. The drawing generation unit 10 is connected to the display unit 25 and includes a device information generation unit 11, a connection information generation unit 12, a layout unit 13, a similar design specification acquisition unit 14, a similar drawing acquisition unit 15, a difference detection unit 16, a drawing information extraction unit 17, a graph generation unit 22, a similarity determination unit 23, and a correction unit 24.

[0133] The graph generation unit 22 generates a graph in which the device information is a node and the connection information is an edge, based on the device information and connection information generated based on the design specification. The graph generation unit 22 also generates a graph of the similar drawing based on the similar device information and similar connection information generated based on the similar design specification or the similar drawing corresponding to the similar design specification.

[0134] The graph generated by this graph generation unit 22 is either a directed graph or an undirected graph based on the connection direction included in the connection information. Here, a directed graph is a graph in which the connections between devices have directionality, and an undirected graph is a graph in which the connections between devices have no directionality.

[0135] Fig. 23 shows an example of a graph generated by the graph generation unit 22. For example, when the types of device information and connection information shown in Fig. 4 and Fig. 5 are expressed as a directed graph, devices such as pumps, check valves, manual valves, motorized valves, and flow meters are configured as nodes, and pipes are configured as edges, as shown in Fig. 23(A). Alternatively, as shown in Fig. 23(B), it is also possible to include pipes in the nodes and represent only the connection information between the pipes and devices as edges.

[0136] The similarity determination unit 23 compares a graph generated based on the design specifications with a graph of a similar drawing generated based on a similar design specification, and determines from the graph of the similar drawing similar nodes (similar nodes) and similar edges (similar edges) to the nodes and edges included in the graph based on the design specifications, based on attribute information such as name, specifications, equipment type, etc. Furthermore, the similarity determination unit 23 may determine similar nodes and similar edges from the graph of the similar drawing by taking into account the structural features of the nodes or edges adjacent to these nodes in addition to the attribute information of the nodes.

[0137] Here, a similar node refers to a node whose name or specifications match or correspond in the device information of the design specification and the similar drawing, for example. Also, a similar edge refers to an edge whose connection status between devices match or correspond in the connection information of the design specification and the similar drawing.

[0138] Furthermore, the similarity determination unit 23 may estimate connection candidates that do not exist in the graph of the design specification based on the connection patterns in the graph of the similar drawing. For example, when using a link prediction method that predicts the connection relationships between devices on a drawing, it is possible to probabilistically estimate the nodes in the design specification by taking into consideration the connection relationships between nodes that frequently appear in the similar drawing, similarities based on the attributes (names, specifications) of the nodes, and the adjacent structure of the nodes.

[0139] The similarity determination unit 23 may also utilize an AI method using a trained model. For example, by using a machine learning model or a graph neural network (GNN) that has learned past drawing data and connection patterns, it becomes possible to estimate with high accuracy the connectivity between unconnected nodes in the design specification.

[0140] The display unit 25 displays the drawing generated by the layout unit 13 of the drawing generation unit 10, and the graph generated by the graph generation unit 22 based on the device information and connection information. The display unit 25 visually displays each node and edge in the graph in association with each symbol (graphic symbol) and line (connection line) in the drawing. This clarifies the relationship between the drawing and the graph structure, allowing the designer to intuitively grasp the connection state and device configuration.

[0141] The display unit 25 may display, together with the drawing and graph, attribute information (such as name, specifications, connection status, etc.) of each node and each edge in the graph generated based on the equipment information and connection information, as well as basis information in the drawing generation unit 10 (for example, at least one of the judgment result by the similarity judgment unit 23, the estimated basis, and the correction history by the correction unit 24).

[0142] 24 shows an example of a drawing or graph displayed by display unit 25. For example, display unit 25 includes functional unit 25-1 that uploads design specifications, functional unit 25-2 that uploads similar drawings, functional unit 25-3 that generates drawings, functional unit 25-4 that performs similarity determination, functional unit 25-5 that saves the generated drawings, functional unit 25-6 that displays drawings generated from the design specifications, functional unit 25-7 that displays graphs generated from the design specifications, and functional unit 25-8 that displays the basis for similarity determination with the similar drawings. Among these functional units, functional unit 25-8 that displays the basis for similarity determination with the similar drawings displays the basis for similarity determination with the similar drawings in text, for example, based on the contents of the similar design specifications corresponding to the similar drawings.

[0143] The modification unit 24 enables the addition and deletion of nodes and edges, the modification of attributes, and the modification of connection relationships to the graphs and drawings displayed by the display unit 25. Furthermore, the modification unit 24 enables the designer to modify the drawings and graphs based on the state of the internal processing of the drawing generation unit 10 (for example, at least one of the estimation result and the judgment result by the similarity judgment unit 23).

[0144] Next, the drawing generation process executed by using the drawing generation system 1 will be described with reference to the flowchart in Fig. 22. Note that the drawings mentioned above may be referenced.

[0145] First, in step S61, the processing circuitry 3 inputs the design specifications and drawing rules to the drawing generation unit 10.

[0146] In the next step S62, the equipment information generating unit 11 generates equipment information about the equipment described in the drawing based on the design specifications. Also, the connection information generating unit 12 generates connection information indicating the mutual connection relationships between the equipment, pipes, cables, etc. based on the design specifications.

[0147] In the next step S63, the similar drawing acquisition unit 15 inputs, from the drawing database 21, a similar drawing that is related to the similar design specification and is linked to the similar design specification.

[0148] Next, in step S64, the drawing information extraction unit 17 extracts similar device information and similar connection information from the similar drawings.

[0149] In the next step S65, the graph generation unit 22 generates a graph based on the design specifications, with the device information as nodes and the connection information as edges, based on the device information and connection information. Note that the nodes and edges in this graph can be set based on any definition, such as device name, specifications, system, etc., in addition to device type.

[0150] In the next step S66, the graph generation unit 22 generates a graph of similar diagrams based on the similar device information and similar connection information, with the similar device information as nodes and the similar connection information as edges. This graph may be, for example, either a directed graph or an undirected graph.

[0151] In the next step S67, the similarity determination unit 23 determines similar nodes and edges in the graph of the similar drawing that are similar to each node and edge of the graph of the design specification, based on the graph based on the design specification and the graph of the similar drawing. In determining similar nodes and edges, the determination may be made not only based on information about each node or edge alone, but also on structural features of the nodes and edges adjacent to them.

[0152] In the next step S68, the similarity determination unit 23 reflects the results of the determination of similar nodes and similar edges by adding the similar nodes and similar edges to the graph of the design specification.

[0153] In the next step S69, the display unit 25 displays a graph of the similar drawing or a graph based on the design specification. The display unit 25 may visually display the attribute information (such as name, specifications, and connection state) of each node and each edge in the graph, as well as the state and basis information of the internal processing of the drawing generation unit 10, such as the determination result by the similarity determination unit 23.

[0154] In the next step S70, the correction unit 24 enables operations such as adding or deleting nodes or edges and changing attributes of the displayed graph, thereby correcting the graph based on the design specification or the graph of a similar drawing. The correction unit 24 may enable the designer to correct the graph while checking the estimation result or the judgment result based on the state of the internal processing of the drawing generation unit 10 visualized by the display unit 25 (at least one of the estimation result and the judgment result by the similarity judgment unit 23).

[0155] In the next step S71, the layout unit 13 acquires (inputs) device information and connection information from a graph based on the design specifications.

[0156] In the next step S72, the layout unit 13 draws the graphic symbols on the drawing based on the device information and connection information.

[0157] In the next step S73, the layout unit 13 assigns attribute information such as names and specifications to the graphic symbols on the drawing.

[0158] In the next step S74, the layout unit 13 assigns drawing information including information about the drawing such as the system name and drawing name to the drawing.

[0159] In the next step S75, the display unit 25 displays the drawing generated by the drawing generation unit 10. In addition to the graphic symbols and attribute information, the display unit 25 may visually display, for example, each graphic symbol on the drawing in association with a node of a graph.

[0160] In the next step S76, the modifying unit 24 modifies the drawing displayed on the display unit 25. Based on the drawing displayed on the display unit 25 and its generation process, the modifying unit 24 can add, delete, or move graphic symbols, change attribute information, modify connection relationships, and the like.

[0161] In the next step S77, the drawing generating unit 10 outputs drawing data including the generated drawing, and the process is completed.

[0162] According to the sixth embodiment, even if the design specifications lack device information and connection information, by utilizing the estimation results based on the graph structure of the similar drawings and link prediction, the similar device information and similar connection information of the similar drawings can be appropriately reflected in the drawings generated based on the design specifications.

[0163] The present invention has been described above based on the first to fifth embodiments, but the configuration applied in any of the embodiments may be applied to other embodiments, and the configurations applied in each embodiment may be combined.

[0164] Note that the arrows in the above-mentioned functional block diagram are merely examples of the process flow, and other process flows may be used. Furthermore, the order of each process is not necessarily fixed, and the order of some processes may be reversed. Furthermore, some processes may be executed in parallel with other processes.

[0165] In the above-described flowchart, an example is shown in which the steps are executed serially, but the order of the steps is not necessarily fixed, and the order of some steps may be reversed. Also, some steps may be executed in parallel with other steps. Furthermore, the above-described flowchart may show at least some of the steps, and other steps may be included in the above-described flowchart.

[0166] The above-mentioned system includes a control device, a storage device, an output device, an input device, and a communication interface. Here, the control device includes a highly integrated processor such as a central processing unit (CPU), a graphics processing unit (GPU), a field programmable gate array (FPGA), or a dedicated chip. The storage device includes a read-only memory (ROM), a random access memory (RAM), a hard disk drive (HDD), or a solid-state drive (SSD). The output device includes a display panel, a head-mounted display, a projector, a printer, etc. The input device includes a mouse, a keyboard, a touch panel, etc. This system can be realized with a hardware configuration using a normal computer.

[0167] The program or trained model executed by the aforementioned system is provided in advance as a ROM or the like. Additionally or alternatively, the program or trained model is provided as a file stored in a computer-readable non-transitory storage medium in an installable or executable format. This storage medium includes CD-ROMs, CD-Rs, memory cards, DVDs, flexible disks (FDs), etc.

[0168] The program or trained model executed by this system may be stored on a computer connected to a network such as the Internet and provided by downloading it via the network. That is, the program or trained model may be provided from cloud computing resources. Alternatively, a server on the cloud may execute the program or trained model, and only the processing results may be provided via the cloud. This system may also be configured by combining separate modules that independently perform the functions of the components and interconnect them via a network or dedicated lines.

[0169] Although the above-described embodiment illustrates the generation of a piping and instrumentation diagram (P&ID), other aspects may be used. For example, an interlock block diagram (IBD), a single-line diagram, or an exploded wiring diagram (ECWD) may be generated.

[0170] Note that some of the processes executed by the trained model in the above-described embodiment may be executed in other ways. For example, some of the processes may be executed based on a predetermined rule. Specifically, when performing layout processing using CAD, the processes may be executed using a CAD creation command.

[0171] According to at least one of the embodiments described above, the diagram generating unit 10 generates device information about devices and connection information indicating the connection relationships between the devices based on the input design specifications. This makes it possible to automatically generate diagrams that include at least one of process information and sequence information.

[0172] In particular, the computer converts the sequence information of the drawing into device information and connection information, which are then handled as text information. This allows a large-scale language model to handle the drawing. Then, by using a large-scale language model that has added or learned the relationship between the design specification and the device information and connection information, it is possible to generate device information and connection information from the design specification. Drawings containing sequence information can be automatically generated by drawing on a drawing frame based on this information.

[0173] Although several embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, substitutions, modifications, and combinations can be made without departing from the spirit of the invention. These embodiments and modifications thereof are intended to be included within the scope and spirit of the invention, as well as within the scope of the invention described in the claims and their equivalents. [Explanation of symbols]

[0174] 1...Drawing generation system, 2...Drawing generation computer, 3...Processing circuit, 4...Memory unit, 5...Communication unit, 6...Input unit, 7...Output unit, 10...Drawing generation unit, 11...Equipment information generation unit, 12...Connection information generation unit, 13...Layout unit, 14...Similar design specification acquisition unit, 15...Similar drawing acquisition unit, 16...Difference detection unit, 17...Drawing information extraction unit, 18...Prompt generation unit, 20...Design specification database, 21...Drawing database.

Claims

1. a drawing generation unit that receives input data including a design specification that is a document describing design information for a plurality of devices, and outputs output data including drawing data; The drawing generation unit a device information generating unit that generates device information related to the device based on the input design specification; a connection information generating unit that generates connection information indicating a connection relationship between the devices based on the input design specification; Including, The drawing generation unit generating the drawing data depicting a plurality of figures including symbols representing the devices and lines connecting the devices within a frame of a drawing based on the device information and the connection information generated by the device information generation unit and the connection information generation unit; outputting the generated drawing data; It is configured as follows: Drawing generation system.

2. The drawing generation unit Accepting input of the input data including drawing rules; outputting the drawing data based on the drawing rules; It is structured as follows: The drawing rules include at least one of a line type used in drawing, a notation method for drawing, and a design rule for the device. The drawing generation system according to claim 1 .

3. The drawing generation unit generating attribute information including at least one of the name and specifications of the device based on the input design specification; the generated attribute information is assigned to the drawing data in association with the graphic; It is configured as follows:

3. The drawing generation system according to claim 1.

4. the drawing generation unit includes a prompt generation unit that generates a prompt that instructs generation of the device information and the connection information; the prompt includes at least the design information; the drawing generation unit is configured to output the drawing data based on the prompt generated by the prompt generation unit.

3. The drawing generation system according to claim 1.

5. the drawing generation unit includes a drawing information extraction unit that extracts drawing information including at least one of similar device information similar to the device information, similar connection information similar to the connection information, and similar layout information indicating a layout of the similar drawing from a similar design specification similar to the design specification and a similar drawing related to the similar design specification, the drawing generation unit is configured to output the drawing data based on the drawing information extracted by the drawing information extraction unit.

3. The drawing generation system according to claim 1.

6. the drawing generation unit includes a prompt generation unit that generates a prompt including at least one of the similar device information, the similar connection information, and the similar layout information; the drawing generation unit is configured to output the drawing data based on the prompt generated by the prompt generation unit. The drawing generation system according to claim 5 .

7. the drawing generation unit includes a difference detection unit that detects a difference between the similar design specification and the design specification, the drawing generation unit is configured to output the drawing data based on the difference detected by the difference detection unit. The drawing generation system according to claim 5 .

8. The drawing generation unit generating a prompt containing the difference; outputting the drawing data based on the generated prompt; It is configured as follows: The drawing generation system according to claim 7 .

9. the drawing generation unit includes a graph generation unit that generates a graph in which the device information is a node and the connection information is an edge, based on the device information and the connection information generated based on the design specification; The drawing generation system according to claim 1 or 2.

10. the graph generation unit generates either a directed graph or an undirected graph based on a connection direction included in the connection information. The drawing generation system according to claim 9.

11. The drawing generation unit includes a graph generation unit that generates a graph in which the device information is a node and the connection information is an edge, based on the device information and the connection information generated based on the design specification. the graph generation unit generates a graph of the similar drawing based on similar device information and similar connection information generated based on either a similar design specification or a similar drawing; The drawing generation system according to claim 5 .

12. the drawing generation unit includes a similarity determination unit that compares the graph generated based on the design specification with a graph of the similar drawing generated based on either the similar design specification or the similar drawing, and determines nodes and edges similar to nodes and edges included in the graph based on the design specification from the graph of the similar drawing based on attribute information of the nodes including names, specifications, and equipment types; The drawing generation system according to claim 11.

13. the similarity determination unit determines similar nodes by taking into consideration structural features of nodes or edges adjacent to the node in addition to attribute information of the node; The drawing generation system according to claim 12.

14. a display unit is connected to the drawing generation unit to display the drawing generated by the drawing generation unit and a graph generated based on the device information and the connection information; the display unit visually displays each node and each edge in the graph in association with each symbol and each line in the drawing; The drawing generation system according to claim 9.

15. the diagram generation unit includes a modification unit that enables addition and deletion of nodes and edges, change of attributes, and modification of connection relationships to the graph and the diagram displayed by the display unit; the modification unit enables a designer to modify the drawings and graphs based on at least one of an estimation result and a determination result, which are states of internal processing of the drawing generation unit; The drawing generation system according to claim 14.

16. the display unit visually displays at least one of an estimation basis, a determination result, and a revision history, which are states of the internal processing of the drawing generation unit, together with the drawing or the graph; The drawing generation system according to claim 14.

17. A method in which a drawing generation unit receives input data including a design specification, which is a document describing design information for a plurality of devices, and executes a process of outputting output data including drawing data, The drawing generation unit a device information generating unit that generates device information related to the device based on the input design specification; a connection information generating unit that generates connection information indicating a connection relationship between the devices based on the input design specification; Including, The drawing generation unit generating the drawing data depicting a plurality of figures including symbols representing the devices and lines connecting the devices within a frame of a drawing based on the device information and the connection information generated by the device information generation unit and the connection information generation unit; outputting the generated drawing data; Drawing generation method.

Images