Method of manufacturing architectural structure

Abstract

To provide a method of manufacturing an architectural structure so as to streamline a series of processes from architectural design to construction.SOLUTION: A method of manufacturing an architectural structure includes: an AI design generation step S1 of generating a design of an architectural structure to be manufactured; a 3D model transformation step S2 of generating 3D-printable 3D model data based on the design; a 3D model optimization step S3 of optimizing the 3D model data in line with output specifications of a predetermined 3D printer; a printer output step S4 of manufacturing a frame of the architectural structure to be manufactured, using materials output from a header of the predetermined 3D printer, based on the optimized 3D model data; and a frame assembling step S8 of assembling the manufactured frames in a predetermined place.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a method for manufacturing a building. [Background technology]

[0002] Conventionally, construction companies have manufactured buildings, particularly houses, using their own 3D printers (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2017-128073 Summary of the Invention [Problem to be solved by the invention]

[0004] However, even though buildings are manufactured using 3D printers, traditionally, designers often spend six months to a year or more designing the building, meaning that the benefits of 3D printers in shortening construction times are not fully realized.

[0005] The present invention has been made in consideration of the above circumstances, and aims to provide a manufacturing method for a building that can improve the efficiency of the entire process from design to construction. [Means for solving the problem]

[0006] In order to achieve the above object, a manufacturing method of a building according to one aspect of the present invention includes: An AI design generation step in which existing design or design image data, or linguistic information expressing the design concept of the building to be manufactured, is input into the AI ​​in combination with linguistic prompts, and the AI ​​generates a design of the building to be manufactured based on the output from the AI; a 3D model conversion step for generating 3D model data that can be 3D printed based on the design; a 3D model optimization step of optimizing the 3D model data to match the output specifications of a predetermined 3D printer; A printer output step of manufacturing the skeleton of the building to be manufactured using materials output from the header of the specified 3D printer based on the optimized 3D model data; a frame assembling step of assembling the manufactured frame at a predetermined location; Includes: [Effects of the Invention]

[0007] According to the present invention, it is possible to improve the efficiency of the entire process from design to construction. [Brief explanation of the drawings]

[0008] [Figure 1] 1A and 1B are diagrams illustrating an embodiment of a manufacturing method for a building of the present invention, in which (a) is a diagram of a 3D printer house as a building, and (b) is a flowchart relating to the manufacturing method. [Figure 2] FIG. 2 is a diagram illustrating a printer output step in FIG. [Figure 3] 2 is a diagram showing the manufacturing (printing) of a body using a 3D printer in the printer output step of FIG. 1. FIG. [Figure 4] 3 is a diagram illustrating an example of a configuration of an information processing system including the center server of FIG. 2. FIG. [Figure 5] FIG. 5 is a block diagram showing an example of a hardware configuration of a center server in the information processing system shown in FIG. 4. [Figure 6] 6 is a functional block diagram showing an example of a functional configuration of the center server of FIG. 5 in the information processing system of FIG. 4. [Figure 7] 2 is a flowchart showing details of the AI ​​design generation step in FIG. 1. [Figure 8] 2 is a flowchart of stacking judgment feedback in the 3D model optimization step of FIG. 1. [Figure 9] FIG. 2 is a diagram illustrating an example of the AI ​​design generation step in FIG. 1. [Figure 10] FIG. 10 is a diagram showing an example of the AI ​​design generation step following FIG. 9. [Figure 11] FIG. 11 is a diagram showing an example of the AI ​​design generation step following FIG. 10. [Figure 12] 2 illustrates an example of the 3D model transformation step of FIG. 1. [Figure 13] 2 is a diagram showing an example of the construction simulation steps of FIG. 1. FIG. [Figure 14] FIG. 2 illustrates an example of the transportation planning step of FIG. 1. [Figure 15] 2 is a diagram showing an example of the construction planning step of FIG. 1. FIG. [Figure 16] 2A to 2C are diagrams showing an example of a body assembly step in FIG. 1. [Figure 17] FIG. 1 is a diagram showing a timeline of the entire process of a building manufacturing method. [Figure 18] FIG. 1 is a diagram showing a comparison between the prior art and the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0009] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a diagram illustrating one embodiment of a manufacturing method for a building of the present invention, in which (a) is a diagram of a 3D printer house as a building, and (b) is a flowchart relating to the manufacturing method.

[0010] In Figure 1, a building (especially a residential building) manufactured using the 3D printing additive manufacturing method will be referred to as a "3D printer house" and will be given the reference number 100. The 3D printing additive manufacturing method is a manufacturing method in which at least a part of a building (here, the structure 5 described below) is produced by stacking concrete, mortar, or ceramic materials output from the header of a 3D printer (see Figure 3) described below in a predetermined direction based on the digital data of the building.

[0011] A "3D printer" is a device that forms one layer by outputting material onto a predetermined surface parallel to the XY plane based on a 3D model (the digital data described above, referred to here as 3D house data. The 3D house data also includes data related to the skeleton 5, which will be described later) prepared in advance for the manufacture of the 3D printed house 100, and then moves a header 6 (see Figure 3) in approximately the Z direction (approximately the vertical direction) to print the next layer by stacking it on top of the previous layer (3D printing additive manufacturing method). The 3D printer is designated by the reference numeral 3 (see Figures 2 and 3). By using the 3D printer 3, it becomes possible to manufacture (print) a 3D printer house 100 having a "spherical" wall surface, for example, as shown in FIG.

[0012] The 3D printer 3 (see FIG. 2, collectively referring to 3D printers 3-1 to 3-n (n is an integer value of 1 or more)) has different data formats depending on, for example, the printer manufacturer or printer type. In other words, the form of the digital data is different. For this reason, the center server 1, which will be described later in FIG. 2, is provided with a function that enables conversion of the format of the digital data to suit the 3D printer 3 (output 3D printer) to be used.

[0013] The 3D printer house 100 is constructed by assembly on, for example, a foundation F, and is composed of a house main body 110 consisting of multiple skeletons 5 described below, a roof 120 also consisting of skeletons 5, and multiple windows M (doors and fixtures). A 3D printer house 100 configured in this manner is manufactured through various steps including, for example, an AI design generation step S1, a 3D model conversion step S2, a 3D model optimization step S3, a printer output step S4, a construction simulation step S5, a transportation planning step S6, a construction planning step S7, and a frame assembly step S8.

[0014] The method for manufacturing a building in this embodiment includes a series of manufacturing processes from design using AI to manufacturing and assembly of the structure 5 using a 3D printer 3. The manufacturing method for a building in this embodiment includes five main steps and three optional steps. The former main steps are the above-mentioned AI design generation step S1, 3D model conversion step S2, 3D model optimization step S3, printer output step S4, and body assembly step S8, and the latter optional steps are the construction simulation step S5, transportation planning step S6, and construction planning step S7.

[0015] Details will be explained in the explanation of this embodiment, but by going through five main steps, automation of design using AI is achieved, shortening the design process, which previously took six months to a year or more, to just a few days. Furthermore, by combining this with technology using a 3D printer, it is possible to complete the entire process from printing to construction within 24 hours, significantly shortening the construction period and reducing costs.

[0016] At any step, construction efficiency and safety can be improved by running AI-based construction simulations in advance. In addition, automatic generation of transportation plans using AI enables safe and efficient transportation of the structure 5, reducing the risk of damage during transportation while improving the efficiency of the transportation process. In addition, the automatic generation of construction plans using AI can optimize the assembly procedures for the structure 5, the necessary equipment, work schedules, etc., reducing human error and improving the efficiency and quality of the construction process.

[0017] Although the order of the steps may vary, first, of the main steps, the printer output step S4 will be explained. FIG. 2 is a diagram for explaining the printer output step in FIG.

[0018] The information processing system for manufacturing the 3D printer house 100, specifically the manufacturing of the skeleton 5, can manufacture the skeleton 5 by controlling the execution of the printer output step S4 in the following manner. First, the user operates the user terminal 4 to input order information, and the input order information is transmitted from the user terminal 4 to the center server 1. This order information includes the conditions desired by the user, that is, the user conditions (the user conditions are omitted).

[0019] Next, the center server 1 receives the order information sent from the user terminal 4. The center server 1 determines judgment factors for determining the 3D printer 3 that will manufacture at least a part of the 3D printed house 100 based on the user conditions included in the order information. After the decision factors have been determined, the center server 1 extracts one or more suitable candidates for use from among a plurality of pre-registered 3D printers 3 (registered 3D printers) based on the decided decision factors.

[0020] When the order information is received, the center server 1 also executes the above-mentioned AI design generation step S1, 3D model conversion step S2, and 3D model optimization step S3, which are major steps (each of these steps will be described later).

[0021] After one or more suitable candidates for use are extracted from among multiple 3D printers 3, the center server 1 determines the 3D printer 3 (output 3D printer) to be used to manufacture the body 5 from among the extracted one or more candidates. Once the 3D printer 3 is determined, the center server 1 converts the 3D house data (digital data) to match the format of the 3D printer 3 that has been determined to be used. Specifically, the format of the 3D house data is converted into a digital data format that can be handled by the 3D printer 3 by a slicer for the 3D printer 3.

[0022] After the data is converted into a format that can be handled by the 3D printer 3 (output 3D printer) used to manufacture the body 5, the center server 1 sends the 3D house data to the print server 2 that manages the 3D printer 3. The print server 2 that manages the 3D printer 3 receives the 3D house data sent from the center server 1.

[0023] After receiving the 3D house data, the print server 2 that manages the 3D printer 3 (output 3D printer) transmits the 3D house data to the 3D printer 3 that is used to manufacture the skeleton 5. After receiving the 3D house data, the 3D printer 3 (output 3D printer) starts manufacturing the body 5.

[0024] One or more skeletons 5 that make up the 3D printer house 100 are manufactured by the information processing system according to the above-described flow. In the printer output step S4, one or more frameworks 5 that make up the 3D printer house 100 are manufactured using the 3D printer 3, which can contribute to shortening the construction period and reducing costs.

[0025] Furthermore, according to the information processing system, first, multiple 3D printers 3, both domestically and internationally, including those owned by other companies, are registered in advance, and then the 3D printer 3 to be used to manufacture the main body 5 is selected from among them, and 3D house data is sent and received.In this case, for example, an administrator of the information processing system, or even if the company itself does not increase the number of 3D printers it owns (or even if the company does not own any 3D printers 3), can manufacture many 3D printed houses 100 in various locations in response to user orders. In other words, by adopting the above-mentioned information processing system (or the services provided using the above-mentioned information processing system), it is possible to increase the production volume and popularize 3D printer houses 100 by using a method similar to remote control, even if the company does not increase the number of 3D printers it owns.

[0026] Next, the manufacturing (printing) of the body 5 using the 3D printer 3 will be described with reference to FIG. FIG. 3 is a diagram showing an example of manufacturing (printing) a body using a 3D printer in the printer output step of FIG.

[0027] A 3D printer 3 that performs additive manufacturing by printing is installed in a factory (not shown). The 3D printer 3 can temporarily store concrete, mortar, or ceramic material in a storage section by passing it through, for example, two hoses, and then output (discharge) it from the header 6. The 3D printer 3 forms one layer by outputting material onto a specific surface parallel to the XY plane based on the 3D house data, and then moves the header 6 approximately in the Z direction (approximately vertical direction) to print the next layer by stacking it on top of the previous layer. The 3D printer 3 can print a wall-like skeleton 5 of the 3D printer house 100, for example, as shown in FIG. 3, by stacking multiple layers in the approximate Z direction (approximately the vertical direction).

[0028] Next, the configuration of the above-mentioned information processing system will be described with reference to FIG. FIG. 4 is a diagram illustrating an example of the configuration of an information processing system including the center server of FIG.

[0029] The information processing system shown in FIG. 4 is configured to include a center server 1, a print server 2, a 3D printer 3, and a user terminal 4. The center server 1, the print server 2, and the user terminal 4 are connected to each other via a predetermined network N such as the Internet. The 3D printer 3 is directly connected to the corresponding print server 2 (this is an example, and the connection may be via the above-mentioned predetermined network N or another network).

[0030] The center server 1 is, for example, an information processing device managed by an administrator of the information processing system. The center server 1 executes various processes while communicating with the print server 2 and the user terminal 4 as needed.

[0031] The print server 2 is managed by, for example, a printer manager who manages the 3D printers 3 at various locations, or by, for example, an administrator of an information processing system. The print server 2 is an information processing device for controlling the 3D printers 3. The print server 2 communicates with the center server 1 and the 3D printer 3 as needed and executes various processes to realize this service. It is assumed here that there are a plurality of print servers 2, and they are designated by the reference numerals 2-1 to 2-n (n is an integer value of 1 or more). When there is no need to distinguish between them, they will be referred to as print servers 2.

[0032] As explained in Figures 2 and 3, the 3D printer 3 is a device that forms one layer by outputting material onto a specific surface parallel to the XY plane based on a 3D model prepared in advance for the production of the 3D printer house 100 (for the production of one or more frames 5), and then moves the header 6 approximately in the Z direction (approximately vertical direction) to print the next layer by stacking it on top of the previous layer. There is one or more 3D printers 3 for each of the print servers 2-1 to 2-n, and for example, the print server 2-1 is designated by the reference numerals 3-1-1 to 3-1-m (m is an integer value of 1 or more). Also, for example, the print servers 2-n are designated by the reference numerals 3-n-1 to 3-1-p (p is an integer value of 1 or more) (when it is not necessary to distinguish between them, they will be referred to as 3D printers 3 as described above).

[0033] The user terminal 4 is an information processing device that is managed and operated by a user (not shown). The user terminal 4 is configured as a personal computer, a tablet, a smartphone, or the like. Since there are multiple users, there are multiple user terminals 4. Here, they are designated by the reference numerals 4-1 to 4-k (k is an integer value of 1 or greater), and when there is no need to distinguish between them, they will be referred to as user terminals 4.

[0034] Next, an example of the hardware configuration of the center server 1 in the information processing system shown in FIG. 4 will be described with reference to FIG. FIG. 5 is a block diagram showing an example of a hardware configuration of a center server in the information processing system shown in FIG.

[0035] The center server 1 includes a CPU (Central Processing Unit) 11, a ROM (Read Only Memory) 12, a RAM (Random Access Memory) 13, a bus 14, an input / output interface 15, an input unit 16, an output unit 17, a memory unit 18, a communication unit 19, and a drive 20.

[0036] The CPU 11 executes various processes according to a program recorded in the ROM 12 or a program loaded from the storage unit 18 into the RAM 13 . The RAM 13 also stores data and the like necessary for the CPU 11 to execute various processes.

[0037] The CPU 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output interface 15 is also connected to this bus 14. An input unit 16, an output unit 17, a storage unit 18, a communication unit 19, and a drive 20 are connected to the input / output interface 15.

[0038] The input unit 16 is configured with, for example, a keyboard, a touch panel, etc., and accepts input of various information. The output unit 17 is configured with a display such as a liquid crystal display, a speaker, etc., and outputs various information as images and sounds. The storage unit 18 is configured with a DRAM (Dynamic Random Access Memory) or the like, and stores various data. The communication unit 19 communicates with other devices (for example, the print server 2 and the user terminal 4 in FIG. 2) via a network N including the Internet.

[0039] Removable media 30, such as a magnetic disk, optical disk, magneto-optical disk, or semiconductor memory, is appropriately attached to the drive 20. A program read from the removable media 30 by the drive 20 is installed in the storage unit 18 as needed. Furthermore, the removable medium 30 can also store various data stored in the storage unit 18 in the same manner as the storage unit 18.

[0040] Although not shown, the print server 2 and the user terminal 4 in Fig. 4 can also have basically the same hardware configuration as that shown in Fig. 5. Therefore, a description of the hardware configuration of the print server 2 and the user terminal 4 of the information processing system will be omitted.

[0041] The center server 1 can execute various processes by cooperation of various hardware and various software that constitute the information processing system of FIG. 4 including the center server 1 of FIG.

[0042] The functional configuration executed in the center server 1 that constitutes the information processing system will be described below. FIG. 6 is a functional block diagram showing an example of the functional configuration of the center server of FIG. 5 in the information processing system of FIG.

[0043] As shown in Figure 5, the CPU 11 of the center server 1 functions as an AI design generation unit 51, a 3D model conversion unit 52, a 3D model optimization unit 53, a printer output unit 54, a construction simulation unit 55, a transportation planning unit 56, a construction planning unit 57, and a body assembly unit 58. In addition, the storage unit 18 of the center server 1 is provided with an order information DB71, a 3D printer information DB72, and a 3D structure data DB73.

[0044] The AI ​​design generation unit 51 inputs existing design or design image data, or linguistic information expressing the design concept of the building to be manufactured, into the AI ​​in combination with linguistic prompts, and executes control to generate the design of the building to be manufactured based on the output from the AI. In addition, the AI ​​design generation unit 51 executes control to generate a design that takes structural strength into consideration through physical simulation. In addition, the AI ​​design generation unit 51 generates multiple design proposals using AI and performs control to generate a final design based on a proposal selected by the user from the multiple design proposals.

[0045] More specifically, the AI ​​design generation unit 51 has the function of inputting existing architectural designs, image data of designs, or linguistic information expressing design concepts into the AI ​​in combination with prompts, and generating architectural designs for buildings. The AI ​​design generation unit 51 includes sub-modules that perform input data formatting processing, processing using an AI model, and physical simulation as necessary. The AI ​​design generation unit 51 analyzes the structural strength against external forces such as self-weight, wind pressure, and earthquake force, and generates multiple design proposals that take into account both design and structural strength. In addition, a user selection function allows the user to select one of the multiple generated design proposals and then finalize the final design. The AI ​​design generation unit 51 is characterized by its processing efficiency, which allows design work that previously took more than six months to be completed in just a few days.

[0046] The physical simulation mentioned above involves numerically analyzing the stress distribution and deformation of the structure in response to external forces acting on the building, such as its own weight, wind pressure, snow load, and seismic force, to confirm the adequacy of the structural strength. These calculations, which structural designers previously performed using dedicated software and took a lot of time, can now be performed simply and quickly using LLM, making it possible to generate design proposals that take structural feasibility into account from the conceptual design stage.

[0047] The 3D model conversion unit 52 executes control for generating 3D model data (3D house data) that can be 3D printed based on the architectural design. Furthermore, the 3D model conversion unit 52 executes control for generating 3D model data in a programming language using an AI language model.

[0048] More specifically, the 3D model conversion unit 52 performs a process of converting the architectural design data created by the AI ​​design generation unit 51 into 3D model data (3D house data) that can be 3D printed. This 3D model conversion unit 52 combines an AI language model and a programming language, analyzes the design data, and converts it into a format that can be output by the 3D printer 3. The function of the 3D model conversion unit 52 makes it possible to generate a 3D model that has actual unevenness representation, rather than texture representation, which has been common in conventional 2D to 3D model generation. The 3D model conversion unit 52 generates a complete 3D model including the model's geometry information, material properties, and structural properties, and outputs it in a data format suitable for the optimization process in the next step.

[0049] The 3D model optimization unit 53 executes control to optimize the 3D model data in accordance with the output specifications of a predetermined 3D printer 3. In addition, the 3D model optimization unit 53 executes control to optimize the model to match the output specifications of a specified 3D printer 3 through AI iteration using a feedback function that determines stackability.

[0050] More specifically, the 3D model optimization unit 53 is responsible for optimizing the 3D model data (3D house data) generated by the 3D model conversion unit 52 to match the output specifications of a predetermined 3D printer 3. The 3D model optimization unit 53 decomposes the 3D model into multiple parts using a model decomposition process, and evaluates each part using a feedback function that determines whether it can be stacked. For parts that are determined by the 3D model optimization unit 53 to be unstackable, corrections are made repeatedly through AI iterations, and optimization is continued until all parts can be stacked. The 3D model optimization unit 53 performs optimization processing that takes into account the performance characteristics of the 3D printer 3, material characteristics, layering direction, the need for support structures, etc., and has the function of minimizing the risk of print failure.

[0051] The printer output unit 54 executes control to manufacture the structure 5 of the building to be manufactured (3D printer house 100) using materials output from the header 6 of a specified 3D printer 3 based on the optimized 3D model data (3D house data).

[0052] More specifically, the printer output unit 54 controls the 3D printer 3 based on the optimized 3D model data (3D house data) and has the function of manufacturing the skeleton 5 of the building to be manufactured (3D printer house 100). The printer output unit 54 adjusts various parameters of the 3D printer 3, such as header control, material supply control, and temperature control, to produce a high-quality body 5. The printer output section 54 includes algorithms that optimize factors that affect output quality, such as material selection, layer thickness setting, and print speed adjustment. The printer output unit 54 has a function for monitoring the printing status in real time, which makes it possible to respond immediately when an abnormality occurs. The printer output unit 54 is characterized by highly efficient output control that normally completes the construction of the frame in approximately six hours.

[0053] The construction simulation unit 55 executes control to optimize the manufacturing process by executing a construction simulation using AI before the printer output unit 54 starts functioning.

[0054] More specifically, the construction simulation unit 55 has the function of executing a construction simulation to optimize the manufacturing process of the skeleton 5 before printing. The construction simulation unit 55 implements a simulation algorithm that predicts potential problems such as spalling due to lateral pressure when pouring concrete in advance and takes appropriate measures. The construction simulation unit 55 optimizes the manufacturing process through a comprehensive simulation that takes into account factors such as material properties, manufacturing environment, and construction procedures. The construction simulation unit 55 has a function of automatically generating a schedule, and has a function of creating a construction plan that optimizes the required time for each process, the required resources, the work sequence, etc.

[0055] The transportation planning unit 56 executes control for generating plans for carrying out, transporting, and carrying in the manufactured body 5 using AI.

[0056] More specifically, the transportation planning unit 56 has a function of generating a plan for carrying out, transporting, and carrying in the manufactured body 5 by the printer output unit 54 using AI. The transportation planning unit 56 implements an algorithm that analyzes characteristics of the structure 5, such as its shape, weight, and vulnerability, and determines the optimal transportation method, route, and equipment. The transportation planning unit 56 has a function of analyzing stress during transportation, and by using this function, predicts the force acting on the skeleton 5 during transportation and identifies areas that require reinforcement. In addition, the transportation planning unit 56 has the function of formulating a comprehensive transportation plan that takes into account external factors such as weather conditions, traffic conditions, and legal regulations, and this function enables safe and efficient transportation in approximately 12 hours.

[0057] The construction planning unit 57 uses AI to execute control for generating implementation and construction plans.

[0058] More specifically, the construction planning unit 57 has the function of generating an implementation and construction plan using AI before the skeleton 5 is assembled. The construction planning unit 57 considers the conditions of the construction site, available equipment, and the skill sets of the workers to plan the optimal assembly procedures and division of labor. The construction planning unit 57 has a function of creating a 3D model of the construction site, and by using this function, it simulates the layout of the skeleton 5 and aims to optimize the work space and equipment layout. The construction planning unit 57 also has a function of analyzing dependencies between processes and identifying critical paths, thereby formulating a construction plan that minimizes the overall construction time.

[0059] The frame assembly unit 58 executes control for assembling the frame 5 manufactured by the printer output unit 54 at a predetermined location.

[0060] More specifically, the body assembly unit 58 has the function of controlling the process of assembling the body 5 manufactured by the printer output unit 54 at a predetermined location. The frame assembly department 58 comprehensively manages various processes related to assembly work, such as controlling heavy machinery such as cranes, managing positioning accuracy, and processing joints. The skeleton assembly unit 58 works in conjunction with a real-time position measurement system (not shown), which ensures the accuracy of the installation position of the skeleton 5 and improves assembly precision. Furthermore, the frame assembly unit 58 implements an algorithm for selecting the optimum joining method and conditions for processing the joints according to the material properties. The frame assembly unit 58 has an assembly progress monitoring function that detects discrepancies between the plan and the actual results and re-optimizes the construction plan as necessary, which allows the assembly work to be completed in a short time, usually around six hours.

[0061] The order information DB 71 stores order information from the user terminal 4. The 3D printer information DB 72 stores information about the 3D printer 3 (various information such as manufacturer, type, installation location, and operating status) and information required for transportation (for example, transportation cost and transportation method). The 3D structure data DB 73 stores digital data such as 3D house data.

[0062] Next, the AI ​​design generation step S1 in FIG. 1 will be described with reference to FIG. FIG. 7 is a flowchart showing the details of the AI ​​design generation step in FIG.

[0063] In the AI ​​design generation step S1, in process S101 shown in Figure 7, image data of an existing architectural design or design, or linguistic information expressing the design concept of the building to be manufactured, is prepared as input data. These input data are combined with linguistic prompts and input to the AI.

[0064] In the process S102 of combining input data and prompts, the data is formatted in a format that can be understood by the AI, and then the process proceeds to S103, where the data is processed by the AI. In the AI ​​processing step S103, a lightweight physical simulation step S104 using a large-scale language model (LLM) is performed as needed to generate a design that takes structural strength into consideration. This physical simulation step S104 using an LLM enables efficient design iteration.

[0065] As a result of the AI ​​processing S103 (execution of physical simulation S104), multiple design proposals S105 are generated. These proposals are presented to the user (client, etc.), and the final design is decided after the user makes a selection S106.

[0066] Next, the 3D model optimization step S3 in FIG. 1 will be described with reference to FIG. FIG. 8 is a flowchart of stacking judgment feedback in the 3D model optimization step of FIG.

[0067] In the 3D model optimization step S3, the 3D model data generated in the 3D model conversion step S2 is optimized to match the output specifications of a predetermined 3D printer. 8, the initial 3D model is decomposed into multiple parts (not shown) by a model decomposition process S201. These multiple parts are evaluated by stackability determination feedback S202, and it is determined whether each part can be additively manufactured by a 3D printer (S203).

[0068] If the result of S203 is NG, the 3D model is modified by AI iteration S204 to generate an optimized 3D model S205. This process is repeated as necessary until all parts are determined to be stackable. If the determination result in S203 is OK, the process proceeds to printer output step S4. AI iterations with a feedback function to determine buildability ensure that models are designed for 3D printing.

[0069] Hereinafter, the description will be centered on the examples. First, an example of the AI ​​design generation step S1 will be described with reference to FIGS. FIG. 9 is a diagram illustrating an example of the AI ​​design generation step in FIG. FIG. 10 is a diagram showing an example of the AI ​​design generation step following FIG. FIG. 11 is a diagram showing an example of the AI ​​design generation step following FIG.

[0070] As shown in Figure 9, data extracted from the designer's past designs is input into the AI, and the AI ​​automatically generates multiple design proposals (e.g., more than 1,800 images in half a day). The client selects their preferred design from the generated images, and the final architectural design is completed based on that selection, as shown in Figure 10. This method can shorten the architectural design process, which previously took six months to a year or more, to just three days, as shown in Figure 11.

[0071] Next, an example of the 3D model conversion step S2 will be described with reference to FIG. FIG. 12 illustrates an example of a 3D model transformation step.

[0072] In FIG. 12, in the 3D model conversion step S2, 3D model data that can be 3D printed is generated based on the design and structural design drawings created in the AI ​​design generation step S1. In this embodiment, it is possible to generate 3D model data using a programming language using an AI language model. Conventional methods for generating 3D models from 2D often rely on texture representation, but here, by generating a 3D model with actual unevenness representation, output data suitable for 3D printing can be efficiently generated.

[0073] The 3D model optimization step S3 is as shown in FIG. 8, and the printer output step S4 is as shown in FIG. 2, so a description thereof will be omitted here.

[0074] Next, an example of the construction simulation step S5 will be described with reference to FIG. FIG. 13 is a diagram showing an example of the construction simulation steps of FIG.

[0075] The construction simulation step S5 is an optional step, and is executed before or after the printer output step S4. In the construction simulation step S5, a construction simulation using AI is performed to optimize the manufacturing process. As shown in Figure 13, potential problems such as spalling due to lateral pressure when pouring concrete can be simulated in advance and countermeasures can be taken. It is also possible to automatically generate a schedule using AI.

[0076] Next, an example of the transportation planning step S6 will be described with reference to FIG. FIG. 14 is a diagram illustrating an example of the transportation planning step of FIG.

[0077] The transportation planning step S6 is an optional step, and is executed between the printer output step S4 and the frame assembly step S8. In the transportation planning step S6, a plan for carrying out, transporting, and carrying in the manufactured body 5 is generated using AI. As shown in FIG. 14, a safe and efficient method for transporting the body 5 is planned.

[0078] Next, an example of the construction planning step S7 will be described with reference to FIG. FIG. 15 is a diagram showing an example of the construction planning step of FIG.

[0079] The construction planning step S7 is an optional step and is executed before the skeleton assembly step S8. In the construction planning step S7, an implementation and construction plan is generated using AI. As shown in FIG. 15, the assembly procedure for the skeleton 5, the necessary equipment, the work schedule, etc. are planned in detail.

[0080] Next, an example of the body assembling step S8 will be described with reference to FIG. FIG. 16 is a diagram showing an example of the body assembly step of FIG.

[0081] In the skeleton assembly step S8, the manufactured skeleton 5 is assembled at a predetermined location. As shown in FIG. 16, the 3D printer house 100 is manufactured by installing and assembling the skeleton 5 in a predetermined position using a crane or the like. The process of the body assembly step S8 is usually completed in about six hours.

[0082] FIG. 17 is a diagram showing a timeline of all steps in a manufacturing method for a building. FIG. 18 is a diagram showing a comparison between the prior art and the present invention.

[0083] In the manufacturing method of the present invention, by robotizing the manufacturing process and utilizing the 3D printer 3, it is possible to complete the manufacturing of a building (3D printed house 100) in a total of 24 hours, with approximately 6 hours for the output process, approximately 12 hours for the transportation process, and approximately 6 hours for the construction process, as shown in Figure 17. This achieves a significant reduction in the construction period compared to conventional techniques. As shown in Figure 18, while conventional manual design required 180 to 365 days (six months to more than a year), the AI-based method of the present invention can complete design in approximately three days, which significantly reduces the overall construction period.

[0084] Although one embodiment of the present invention has been described above, the present invention is not limited to the above-described embodiment, and modifications, improvements, etc. within the scope of achieving the object of the present invention are included in the present invention. In the above-described embodiment, the printing target of the 3D printer 3 is the 3D printer house 100 (skeleton 5), but this is not particularly limited, and any building will suffice.

[0085] Furthermore, in the above-described embodiment, a plurality of 3D printers 3 are managed by the print server 2, but this is not intended to be limiting.

[0086] Furthermore, in the above-described embodiment, one 3D printer 3 is determined as the output 3D printer, but this is not particularly limited to this. For example, a 3D printer 3 in prefecture X and a 3D printer 3 in prefecture Y may be determined as the output 3D printers, and the 3D printer house 100 (structure 5) may be manufactured by division of labor. Furthermore, in the above-described embodiment, the 3D printer house 100 is manufactured in, for example, prefecture X, but this is not limited to this, and multiple 3D printer houses 100 may be manufactured simultaneously in, for example, multiple countries or multiple regions within a country.

[0087] Furthermore, the system configuration shown in FIG. 4 and the hardware configuration of each device of the center server 1 shown in FIG. 5 are merely examples for achieving the object of the present invention, and are not particularly limited.

[0088] Furthermore, the functional block diagram shown in Fig. 6 is merely an example and is not particularly limited. That is, it is sufficient if the information processing system in Fig. 4 is provided with a function that can execute the various processes described above as a whole, and the functional blocks and databases used to realize this function are not particularly limited to the example in Fig. 6.

[0089] Furthermore, the locations of the function blocks, models, and databases are not limited to those shown in FIG. 6 and may be arbitrary. For example, at least a part of the function blocks, models, or databases arranged on the center server 1 side may be provided in the print server 2, the user terminal 4, or another information processing device (not shown).

[0090] The above-described series of processes can be executed by hardware or software. Furthermore, one functional block may be configured as a single piece of hardware, a single piece of software, or a combination thereof.

[0091] When a series of processes is executed by software, the programs that make up the software are installed into a computer or the like from a network or a recording medium. The computer may be a computer built on dedicated hardware. The computer may also be a computer capable of executing various functions by installing various programs, such as a server, a general-purpose smartphone, or a personal computer.

[0092] The recording medium containing such a program may be composed not only of a removable medium (not shown) that is distributed separately from the device main body in order to provide the program to the user, but also of a recording medium that is provided to the user in a state that is pre-installed in the device main body.

[0093] In this specification, the steps describing the program to be recorded on the recording medium include not only processes that are performed chronologically in accordance with the order, but also processes that are not necessarily performed chronologically but are performed in parallel or individually.

[0094] To sum up, the manufacturing method of a building to which the present invention is applied is sufficient as long as it has the following configuration, and can take on a variety of different embodiments.

[0095] That is, the manufacturing method of a building to which the present invention is applied (for example, a manufacturing method based on the main steps from the AI ​​design generation step S1 to the skeleton assembly step S8 in FIG. 1) is as follows: An AI design generation step (e.g., AI design generation step S1 in Figure 1, etc.; also, AI design generation step S1 shown in the flowchart of Figure 7 and the examples of Figures 9 to 11) in which existing design or design image data, or linguistic information expressing the design concept of the building to be manufactured, is input into the AI ​​in combination with linguistic prompts, and the AI ​​generates a design of the building to be manufactured based on the output from the AI; A 3D model conversion step (e.g., the 3D model conversion step S2 in FIG. 1, etc., or the 3D model conversion step S2 shown in the embodiment of FIG. 12) that generates 3D model data (e.g., the above-mentioned 3D house data) that can be 3D printed based on the design; A 3D model optimization step (e.g., the 3D model optimization step S3 in FIG. 1, or the flowchart in FIG. 8) that optimizes the 3D model data according to the output specifications of a predetermined 3D printer (e.g., the 3D printer 3 in FIGS. 2 and 3); A printer output step (e.g., printer output step S4 in FIG. 1) for manufacturing the skeleton of the building to be manufactured (e.g., skeleton 5 in FIGS. 2 and 3) using materials output from the header (e.g., header 6 in FIG. 3) of the specified 3D printer based on the optimized 3D model data; a frame assembly step (e.g., frame assembly step S8 in FIG. 1) of assembling the manufactured frame at a predetermined location; It is sufficient to include

[0096] By automating design using AI in this way, the design process, which previously took six months to a year or more, can be shortened to just a few days. In addition, by combining it with 3D printer technology, the entire process from printing to construction can be completed within 24 hours, significantly shortening construction time and reducing costs.

[0097] In addition, in the manufacturing method of a building to which the present invention is applied, The 3D model conversion step includes: generating the 3D model data in a programming language using an AI language model (see, for example, the embodiment of FIG. 12 ); It is possible.

[0098] By combining an AI language model with a programming language in this way, it is possible to efficiently generate 3D models with three-dimensional relief representation without relying on texture, and to generate output data suitable for 3D printing.

[0099] In addition, in the manufacturing method of a building to which the present invention is applied, The 3D model optimization step includes: The AI ​​is iterated using a feedback function that determines stackability, and the AI ​​is optimized to match the output specifications of the predetermined 3D printer (see, for example, the flowchart of FIG. 8 ); It is possible.

[0100] This AI iteration with a feedback function that determines stackability allows the design of only models that can be reliably 3D printed, preventing failures in the manufacturing process.

[0101] In addition, in the manufacturing method of a building to which the present invention is applied, The AI ​​design generation step includes: A step of generating the design by physical simulation, taking structural strength into consideration (see, for example, the flowchart in FIG. 7 ); It is possible.

[0102] By using physical simulations to consider structural strength from the design stage, it is possible to efficiently design buildings that are not only aesthetically pleasing but also practical, and to avoid strength problems in advance at the construction stage.

[0103] In addition, in the manufacturing method of a building to which the present invention is applied, The AI ​​design generation step includes: The method includes a step of generating a plurality of design proposals as the design proposals using the AI, and generating the final design based on a selection proposal selected by a user from the plurality of design proposals (see, for example, the flowchart of FIG. 7 ). It is possible.

[0104] In this way, AI can generate a large number of design proposals in a short time (for example, more than 1,800 images in half a day) and refine them based on user selection, shortening the traditional design process to a few days and increasing user satisfaction.

[0105] In addition, the manufacturing method of a building to which the present invention is applied is as follows: Before or after the printer output step, a construction simulation step (see, for example, construction simulation step S5 of FIG. 1 or the example of FIG. 13) is further included, in which a construction simulation using an AI that is the same as or different from the AI ​​is performed to optimize the manufacturing process. It is possible.

[0106] By running such AI-based construction simulations in advance, potential problems such as spalling due to lateral pressure during concrete injection can be predicted and avoided, improving construction efficiency and safety.

[0107] In addition, the manufacturing method of a building to which the present invention is applied is as follows: Between the printer output step and the body assembly step, a transportation planning step is further included in which a plan for carrying out, transporting, and carrying in the manufactured body is generated using an AI that is the same as or different from the AI ​​(see, for example, transportation planning step S6 in FIG. 1 or the example in FIG. 14). It is possible.

[0108] This automatic generation of transportation plans using AI enables the safe and efficient transportation of the structure, reducing the risk of damage during transport while improving the efficiency of the transportation process.

[0109] In addition, the manufacturing method of a building to which the present invention is applied is as follows: Before the frame assembly step, a construction planning step (see, for example, construction planning step S7 in FIG. 1 or the example in FIG. 15) is further included, in which an implementation and construction plan is generated using the same or different AI as the AI. It is possible.

[0110] This automatic generation of construction plans using AI can optimize the assembly procedures for the structure, the necessary equipment, and work schedules, reducing human error and improving the efficiency and quality of the construction process. [Explanation of symbols]

[0111] 1 Center server, 2 Print server, 3 3D printer, 4 User terminal, 5 Building frame, 6 Header, 51 AI design generation unit, 52 3D model conversion unit, 53 3D model optimization unit, 54 Printer output unit, 55 Construction simulation unit, 56 Transportation planning unit, 57 Construction planning unit, 58 Building frame assembly unit, 71 Order information DB ,72···3D printer information DB, 73···3D structure data DB, 100···3D printed house, S1···AI design generation step, S2···3D model conversion step, S3···3D model optimization step, S4···Printer output step, S5···Construction simulation step, S6···Transportation planning step, S7···Construction planning step, S8···Frame assembly step

Claims

1. An AI design generation step in which existing design or design image data, or linguistic information expressing the design concept of the building to be manufactured, is input to the AI ​​in combination with linguistic prompts, and the AI ​​generates a design of the building to be manufactured based on the output from the AI; a 3D model conversion step for generating 3D model data that can be 3D printed based on the design; a 3D model optimization step of optimizing the 3D model data to match the output specifications of a predetermined 3D printer; A printer output step of manufacturing the skeleton of the building to be manufactured using materials output from the header of the specified 3D printer based on the optimized 3D model data; a frame assembling step of assembling the manufactured frame at a predetermined location; A method for manufacturing a building, including:

2. The 3D model conversion step includes: generating the 3D model data in a programming language using an AI language model; A method for manufacturing the building according to claim 1.

3. The 3D model optimization step includes: Optimizing the AI ​​iterations using a feedback function that determines buildability to fit the output specifications of the predetermined 3D printer; A method for manufacturing the building according to claim 1.

4. The AI ​​design generation step includes: generating the design by physical simulation, taking structural strength into consideration; A method for manufacturing the building according to claim 1.

5. The AI ​​design generation step includes: generating a plurality of design proposals as the design proposals using the AI, and generating a final design based on a selection proposal selected by a user from the plurality of design proposals; A method for manufacturing the building according to claim 1.

6. Further included is a construction simulation step of executing a construction simulation using an AI that is the same as or different from the AI ​​before or after the printer output step to optimize the manufacturing process. A method for manufacturing the building according to claim 1.

7. The method further includes, between the printer output step and the body assembly step, a transportation planning step of generating a plan for carrying out, transporting, and carrying in the manufactured body using an AI that is the same as or different from the AI. A method for manufacturing the building according to claim 1.

8. Before the frame assembly step, a construction planning step is further included in which an implementation and construction plan is generated using the same or different AI as the AI; A method for manufacturing the building according to claim 1.

Images