JP2022141269A - Design support device and design support program - Google Patents

Abstract

To provide a design support device and a design support program that can select appropriate positions and shapes of frame members, positions and types of connections of the frame members and positions and types of joints of the frame members from a total shape of a building.SOLUTION: A design support device machine-learns combinations of total shapes of buildings, positions and shapes of frame members, positions and types of connections, and positions and types of joints, and predetermined indexes of the buildings as teacher data; inputs a total shape of a building to be newly constructed into a learning result of the machine learning so as to determine positions and shapes of frame members, positions and types of connections, and positions and types of joints corresponding to the total shape of the building to be newly constructed; calculates a predetermined index of the newly constructed building on the basis of the positions and the shapes of the determined frame members, and the positions and the types of the connections and the joints; and selects the positions and the shapes of the frame members, the positions and the types of the connections, and the positions and the types of the joints. on the basis of the index.SELECTED DRAWING: Figure 2

Description

The present invention relates to a design support device and a design support program.

Conventionally, there is known a design system that uses structural members associated with identification information and shape information to model a frame and perform structural calculations for the modeled frame (see Patent Document 1).

Japanese Unexamined Patent Application Publication No. 2013-11961

However, with the technique described in Patent Document 1, the designer merely assembles the frame while confirming the fit, and is not sufficient to support the design of the building.

An object of the present invention is to make it possible to select appropriate positions and shapes of frame members, positions and types of joints of frame members, and positions and types of joints of frame members from the overall shape of a building. .

In order to achieve the above object, the design support device of the first aspect provides the overall shape of a building, the position and shape of structural members that constitute the building, the positions and types of joints of the structural members, and the structural members. Machine learning is performed using a combination of the joint positions and types and the predetermined index of the building as teacher data, and the overall shape of the building is input as the learning result of the machine learning. Determine the position and shape of the corresponding frame member, determine the position and type of the joint of the frame member and the position and type of the joint of the frame member from the position and shape of the frame member, and determine the determined frame Based on the position and shape of the member, the position and type of the joint of the frame member, and the position and type of the joint of the frame member, a predetermined index of the building to be newly constructed is calculated, and based on the index, the structure member The position and shape, the position and type of joints of the frame members, and the position and type of joints of the frame members are selected.

According to the first aspect, it is possible to select appropriate positions and shapes of frame members, positions and types of joints of frame members, and positions and types of joints of frame members from the overall shape of the building. can be done.

The design support device of the second aspect calculates the index by changing the joints of the selected frame members and the dimensions of the joints of the frame members.

According to the second aspect, it is possible to calculate the optimum dimensions of the joints and joints for the new building from the dimensions of the existing joints and joints.

In the design support device of the third aspect, the index is the strength or structural performance of the newly constructed building.

According to the third aspect, it is possible to design a building with predetermined strength or structural performance.

In the design support device of the fourth aspect, the frame members have a predetermined length, and the indicator is the effective usability of the combination of the frame members.

According to the fourth aspect, it is possible to design a building in which structural members can be efficiently used without waste.

In the design support device of the fifth aspect, the index is an evaluation of the layout of the space formed between the structural members of the building to be newly constructed.

According to the fifth aspect, it is possible to design a building with a predetermined spatiality.

The design support device according to the sixth aspect provides the combination of the overall shape of the new building, the determined position and type of joints of the frame members, and the position and type of joints of the frame members. In addition to data, said machine learning.

According to the sixth aspect, it is possible to optimize the learning result by increasing the amount of teacher data.

In the design support device of the seventh aspect, the structural member is made of wood, a combination of wood and steel frame, or a combination of wood and reinforced concrete.

According to the seventh aspect, it is possible to design a building using frame members made of various materials.

A design support program according to an eighth aspect causes a computer to function as each unit constituting the design support apparatus according to any one of the first to sixth aspects.

According to the eighth aspect, it is possible to select appropriate positions and shapes of frame members, positions and types of joints of frame members, and positions and types of joints of frame members from the overall shape of the building. can provide a design support program.

According to the present invention, it is possible to select appropriate positions and shapes of frame members, positions and types of joints of frame members, and positions and types of joints of frame members from the overall shape of a building. , an effect is obtained.

1 is a schematic block diagram of a design support device according to an embodiment of the present invention; FIG. 3 is a block diagram showing an example of functional configuration of ROM or storage of the design support device according to the embodiment of the present invention; FIG. FIG. 4 is an explanatory diagram for explaining an example of data modeled by a user based on an existing building according to the embodiment of the present invention; FIG. 4 is an explanatory diagram for explaining an example of data obtained by modeling the overall shape of a newly constructed building according to the embodiment of the present invention; FIG. 4 is an explanatory diagram showing an example of positions and shapes of structural members determined by a determining unit according to the embodiment of the present invention; It is explanatory drawing which shows an example of the position and kind of joint determined by the determination part which concerns on embodiment of this invention, and the position and kind of joint. FIG. 7 is an explanatory diagram for explaining an example of the position and type of joints and the position and type of joints shown in FIG. 6 according to the embodiment of the present invention. FIG. 4 is an explanatory diagram for explaining an example of an index according to the embodiment of the present invention; FIG. FIG. 9 is an explanatory diagram for explaining a modified example of the index shown in FIG. 8 according to the embodiment of the present invention; It is an explanatory view for explaining selection by the selection part concerning the embodiment of the present invention. 4 is a flow chart showing an example of the flow of processing in the information learning phase according to the embodiment of the present invention; 4 is a flow chart showing an example of the flow of processing in the operation phase according to the embodiment of the present invention; It is an explanatory view for explaining a modification of the present invention.

An example of an embodiment of the present disclosure will be described below with reference to the drawings. In each drawing, the same or equivalent components and portions are given the same reference numerals. Also, the dimensional ratios in the drawings are exaggerated for convenience of explanation, and may differ from the actual ratios.

FIG. 1 is a block diagram showing the hardware configuration of the design support device 100. As shown in FIG.

As shown in FIG. 1, the design support apparatus 100 includes a CPU (Central Processing Unit) 101, a ROM (Read Only Memory) 102, a RAM (Random Access Memory) 103, a storage 104, a communication interface 105, an input section 106, and a display section. 107. Each component is communicatively connected to each other via a bus 108 .

A CPU 101 is a central processing unit that executes various programs and controls each section. That is, the CPU 101 reads a program from the ROM 102 or the storage 104 and executes the program using the RAM 103 as a work area. The CPU 101 performs control of the above components and various arithmetic processing according to programs recorded in the ROM 102 or the storage 104 . In this embodiment, the ROM 102 or storage 104 stores programs or data.

The ROM 102 stores various programs and various data. The RAM 103 temporarily stores programs or data as a work area. The storage 104 is configured by a HDD (Hard Disk Drive) or an SSD (Solid State Drive), and stores various programs including an operating system and various data.

Further, in the present embodiment, the storage 104 manages joints and joints by assigning an ID to each type. A database storing data such as the shape, size, species of trees used, parts used in the existing building, and mechanical characteristics in the framework is stored for each type.
Here, a joint is a structure that combines and joins two or more members such as columns and beams that are structural members, and a joint is a structure that joins two members such as a column and a column. In addition, in the present embodiment, the structural members are made of wood, but are not limited to this, and may be made of materials other than wood, such as a combination of wood and a steel frame, or a combination of wood and reinforced concrete. including cases where

The communication interface 105 is an interface for communicating with other devices such as a server device (not shown). is used.

The input unit 106 includes a pointing device such as a mouse and a keyboard, and is used for various inputs.

Also, the user of the design support apparatus 100 uses the input unit 106 to input training data such as the overall shape of the building and the positions of structural members that make up the building. In addition, the overall shape of the new building and the like are input.

The display unit 107 is, for example, a liquid crystal display, and displays various information under the control of the CPU 101 . In this embodiment, the display unit 107 displays the positions and types of joints of structural members selected by the selection unit 150, which will be described later, and the positions and types of joints. Further, the display unit 107 may employ a touch panel system and function as the input unit 106 .

The design support device 100 implements various functions using the hardware resources described above. A functional configuration realized by the design support device 100 will be described.

FIG. 2 is a block diagram showing an example of the functional configuration of the design support device 100. As shown in FIG.

As shown in FIG. 2, the design support apparatus 100 includes a reception unit 110, a learning unit 120, a determination unit 130, an index calculation unit 140, and a selection unit 150 as functional configurations. Each functional configuration is realized by the CPU 201 reading and executing a program stored in the ROM 202 or storage unit (storage) 204 .

The reception unit 110 receives a combination of the overall shape of the building, the positions and shapes of the structural members that make up the building, the positions and types of joints of the structural members, the positions and types of the joints of the structural members, and a predetermined index of the building. , accept.
Specifically, input the data that models the overall shape of the building in 3D, the position and shape of the structural members that make up the modeled building, the position and type of joints, and the position and type of joints into the data. The building data that has been obtained and the index of the building data, which will be described later, are received as training data for machine learning.

In addition to being input by the user using the input unit 106, the teacher data includes the overall shape of the new building determined by the determination unit 130, which will be described later, and the positions and types of joints of the structural members determined. , and combinations of joint locations and types of structural members. That is, the combination of the overall shape of the new building determined by the determining unit, the determined positions and types of the joints of the structural members, and the determined positions and types of the joints of the structural members is calculated by the index calculation unit 140. It is also possible to add such indexes as training data. As a result, the teacher data is updated and accumulated, and by increasing the teacher data, the learning results are optimized.

Here, the teacher data input by the user is mainly data modeled by the user based on an existing building, as shown in FIG. That is, as shown in FIG. 3(A), buildings of different shapes (buildings A to D) built in the past are modeled in 3D as shown in FIG. 3(B). Then, based on the modeled data, an index, which will be described later, is calculated to create teacher data.

The data is not limited to data modeled based on an existing building, and may be data modeled from a fictitious building. In addition, the data modeled based on the existing building is created by the user using modeling software, but is not limited to this, and is a known method of modeling the floor plan of the existing building by machine learning. may be created using

The reception unit 110 also receives data on the overall shape of a new building to be input to the learning result of machine learning created by the user of the design support apparatus 100 .
Specifically, as shown in FIG. 4, the overall shape data of the new building is data obtained by modeling the rough appearance of the new building in 3D. Here, the data of the overall shape of the building includes the positions of structural members such as columns and beams outside the building and connection information between the structural members. It is also desirable to include information about the structural members, such as cross-sectional dimensions, material types, and length information.

The learning unit 120 creates a trained model by machine-learning a plurality of teacher data received by the receiving unit 110 . A trained model is configured using, for example, a neural network. A trained model is configured using, for example, a deep neural network (DNN), which is a multi-layered neural network targeted for deep learning. As the DNN, for example, a convolutional neural network (CNN: Convolutional Neural Network) targeting images is used.

The determination unit 130 determines the position and shape of the structural members corresponding to the overall shape of the building by inputting the overall shape of the newly constructed building by the user of the design support apparatus 100 into the learning result of the machine learning.
Specifically, from the overall shape of the newly constructed building shown in FIG. 4, a plurality of patterns of positions and shapes of frame members corresponding to the overall shape as shown in FIG. 5 are obtained. Here, the position of the frame member is where the frame member is placed in the building, and the shape is the shape of the combination of frame members such as beams and columns when they are placed in the building.

Further, the determination unit 130 determines the position and type of joints and the position and type of joints from the determined positions and shapes of the structural members.
Specifically, the positions and types of joints and the positions and types of joints of multiple patterns, which are obtained by adding the positions and types of joints and the positions and types of joints as shown in FIG. A position and type are obtained.

For example, in the example shown in FIG. 6, joints and joints are obtained at 20 positions. Specifically, as shown in FIG. 7, the connection or joint ID is "001" at position "1", the connection or joint ID is "002" at position "2", and the connection or joint ID is at position "3". "001", connection or joint ID "001" at position "4", connection or joint ID "001" at position "5", connection or joint ID "001" at position "6", position "7" Connection or joint ID "002" at position "8" Connection or joint ID "002" at position "9" Connection or joint ID "001" at position "10" ", connection or joint ID "001" at position "11", connection or joint ID "001" at position "12", connection or joint ID "002" at position "13", and connection or joint ID "002" at position "14". Connection or joint ID '003', connection or joint ID '001' at position '15', connection or joint ID '002' at position '16', connection or joint ID '001' at position '17', A combination of a connection or joint ID "001" at position "18", a connection or joint ID "003" at position "19", and a connection or joint ID "004" at position "20". be. The types of joints or joints are not limited to IDs "001" to "004". Also, the number of joints and joints is not limited to 20, but varies depending on the pattern obtained.

That is, by inputting the overall shape of the new building to the learned model, the determining unit 130 determines the positions and shapes of multiple patterns of structural members, the positions and types of multiple patterns of joints, and joints of structural members. determine the location and type of .
In addition to the case where the position and type of joints and the position and type of joints of frame members are determined after determining the position and shape of frame members, the position and type of joints are determined while determining the position and shape of frame members. and type, and the location and type of structural member joints.

The index calculation unit 140 calculates a predetermined index of the newly constructed building based on the position and shape of the structural members and the positions and types of joints and joints determined by the determination unit 130 .

Next, specific examples of indices will be described.

The indicator is the strength or structural performance of the building. The index calculation unit 140 calculates the strength or structural performance for each position and shape of the structural member determined by the determination unit 130 and the positions and types of joints and joints. Then, the evaluation is raised for those with high strength or structural performance, and the evaluation is lowered for those with low strength or structural performance. It should be noted that the higher the strength or structural performance, the higher the evaluation is not limited to, and the best evaluation may be given if it exceeds a predetermined threshold value.

Here, the strength of a building includes seismic performance (bending rigidity, shear rigidity, strength), wind pressure resistance, shear force, etc., which must be satisfied in terms of regulations and usage. In addition, the structural performance of a building has a positive effect on spatiality, such as when it is possible to secure a large space without columns and high ceiling height by improving the performance of beams and columns. Note that it is not necessary to strictly separate the strength and structural performance of a building;

Also, the indicator is the effective utilization of the combination of frame members.
The effective utilization of the combination of frame members is an index for evaluating whether or not waste is eliminated and the right materials are placed in the right places for the types and lengths of materials that can be used for the frame members.
In other words, the columns and beams that are the materials of the structural members are often set to a predetermined length, and the combination of structural members with lengths other than the predetermined length is evaluated as wasting the materials of the structural members. If it is possible to combine structural members with a predetermined length, the evaluation will be raised as effective use of the materials for the structural members.

Also, the index is the layout of the space formed between the structural members of the newly constructed building.
The layout of the space is, for example, the size of the room R, and if a large space can be secured, the evaluation is raised, and if the space that can be secured is small, the evaluation is lowered. In the examples shown in FIGS. 8A and 8B, a large room R is formed and a cohesive space is secured, so the evaluation is high, and FIG. The evaluation is low because it is not possible to secure a coherent space.

Moreover, the layout of the space is not limited to the mere size of the room R, and includes the case where the index is evaluated from factors other than the size of the room R. In this case, a known method of evaluating the floor plan of a building in consideration of elements other than the size of the rooms R, such as the connection between the rooms R ``Architectural Institute of Japan, Journal of Planning and Planning, No. 480'', Go Hasegawa, Yoshiharu Tsukamoto, Masahiro Tanaka, ``Arrangement of rooms comparing the size of space: Sameness and difference in the spatial composition of contemporary architectural works (2)'' "Architectural Institute of Japan, Journal of Planning, Vol. 79, No. 699, etc.) may be used.

In addition, the layout of the space may include not only the size of the room R but also whether or not the line of flow P, which is the path along which people move within the building, is optimal. For example, as shown in FIG. 9(A), when the line of flow P is simple, that is, when the line of flow P from the opening G such as the entrance or the door is short and there are few turns, the evaluation is high. As shown in B), when the line of flow P is complicated, that is, when the line of flow P is long or has many turns, the evaluation is low. In such a case, the teacher data includes information on the opening G such as the position of the entrance and the door for learning, and the position of the structural member includes the information on the opening G such as the position of the entrance and the door. In addition, whether the flow line P is optimal is not limited to the case where the flow line P is simply evaluated based on whether it is simple or not. It may be calculated by using

Also, the indicators are not limited to those described above. For example, the size of the gradual decrease rate of the exterior of the building, the golden ratio of the combination of structural members, whether the ventilation is good or bad as estimated from the position of the opening G, whether the sunshine is good as estimated based on the direction Other indicators, such as bad, may be employed. In addition, when the joints and joints are visible, the cleanliness of the joints and joints is converted into points in advance, and the total points are counted according to the determined joints and joints, and the total points are calculated. The position and type of joints may be calculated.

The selection unit 150 selects the position and shape of the frame member, the position and type of the joint of the frame member, and the position and type of the joint of the frame member based on the index.
That is, the index calculated by the index calculation unit 140 from among the position and shape of the structural member, the position and type of joint, and the position and type of joint determined by the determination unit 130 satisfies the predetermined condition. Select the position and shape of frame members, the position and type of joints, and the position and type of joints.

Such selection is performed by performing predetermined filtering (weighting). For example, as shown in FIGS. 10(A) to 10(F), the determination unit 130 determines the position and shape of the structural member, the position and type of joints, and the position and type of joints for six patterns. A case will be described. In this example, index 1 is the strength or structural performance of the building, index 2 is the effective use of the combination of frame members, and index 3 is the layout of the space formed between the frame members. As a result of calculation by the index calculation unit 140, the pattern in FIG. 10A is that index 1 is superior, index 2 is superior, and index 3 is acceptable. The pattern in FIG. 10C is that index 1 is good, the index 2 is good, and the index 3 is good, and the pattern in FIG. 1 is acceptable, index 2 is excellent, and index 3 is acceptable. The pattern in FIG. The pattern is that index 1 is excellent, index 2 is fair, and index 3 is excellent. In such a case, in this example, weighting is performed on the index 1 and the index 3, and patterns in which the index 1 and the index 3 are superior (patterns in FIGS. 10B and 10F) are selected. be done.

It should be noted that the selection is not limited to a combination of indices, and selection may be made using only one index.

The selection by the selection unit 150 is based on the position and shape of the frame member, the position and type of the joint determined by the determination unit 130, and the position and shape of the frame member and the position of the joint selected from the position and type of the joint. and type, and the position and type of the joint are extracted and displayed on the display unit 107 to indicate to the user, or the position and shape of all or many structural members determined by the determination unit 130, the position and type of the joint While displaying the type and the position and type of the joint on the display unit 107, the position and shape of the optimum frame member, the position and type of the joint, and the position and type of the joint are conspicuously displayed. The display unit 107 may display the positions and shapes of all or a large number of structural members, the positions and types of joints, and the positions and types of joints determined by the unit 130 in order of optimum ranking for the user.

Next, the operation of the design support device 100 will be described.

FIG. 11 is a flow chart showing the flow of the learning phase by the design support device 100. As shown in FIG. The learning phase is performed by CPU 101 reading a program from ROM 102 or storage 104, developing it in RAM 103, and executing it.

First, in step S100, CPU 101 (accepting unit 110) accepts teacher data from a user. Then, the process proceeds to the next step S102.

In step S102, CPU 101 (learning unit 120) performs machine learning to create a trained model. Then, the process proceeds to the next step S104.

In step S<b>104 , CPU 101 (learning unit 120 ) stores the trained model created in step S<b>102 in storage 104 . Then, the process ends.

FIG. 12 is a flow chart showing the operation phase flow of the design support device 100 . The operation phase is performed by CPU 101 reading a program from ROM 102 or storage 104, developing it in RAM 103, and executing it.

First, in step S200, CPU 101 (receiving unit 110) receives the overall shape of a new building. Then, the process proceeds to the next step S202.

In step S202, CPU 101 (determining unit 130) inputs the overall shape of the new building received in step S200 to the learned model stored in storage 104 in step S104 shown in FIG. Then, the process proceeds to the next step S204.

In step S204, the CPU 101 (determination unit 130) determines positions and shapes of multiple patterns of structural members corresponding to the overall shape of the newly constructed building. Then, the process proceeds to the next step S206.

In step S206, the CPU 101 (determining unit 130) determines the positions and types of joints of a plurality of patterns and the positions and types of joints of the frame members from the positions and shapes of the frame members determined in step S204. Then, the process proceeds to the next step S208.

In step S208, the CPU 101 (index calculation unit 140) determines the positions and shapes of the multiple patterns of structural members determined in steps S204 and S206, the positions and types of the multiple patterns of joints, and the positions of the joints of the structural members. and type to calculate the index of the new building. Then, the process proceeds to the next step S210.

In step S210, the CPU 101 (selection unit 150) determines the position and shape of the frame member, the position and type of the joint of the frame member, and the position and type of the joint of the frame member based on the index calculated in step S208. to elect. Then, the process ends.

It should be noted that step S204 and step S206 may not be processed as separate steps, but may be processed in the same step.

Also, the learning phase shown in FIG. 11 and the operation phase shown in FIG. 12 have been described as being executed by the same design support apparatus 100, but the present invention is not limited to this. For example, it may be divided into a device that executes the learning phase and a device that executes the operation phase.

Various processors other than the CPU 101 may execute the processing executed by the CPU 101 by reading the software (program) in the above embodiment. The processor in this case is a PLD (Programmable Logic Device) whose circuit configuration can be changed after manufacturing such as an FPGA (Field-Programmable Gate Array), and an ASIC (Application Specific Integrated Circuit) for executing specific processing. A dedicated electric circuit or the like, which is a processor having a specially designed circuit configuration, is exemplified. Also, each process may be executed by one of these various processors, or a combination of two or more processors of the same or different type (for example, a plurality of FPGAs, a combination of a CPU and an FPGA, etc.). ) can be run. More specifically, the hardware structure of these various processors is an electric circuit in which circuit elements such as semiconductor elements are combined.

Further, in the above-described embodiment, the program for each process has been pre-stored (installed) in the ROM 102 or the storage 104, but the present invention is not limited to this. The program may be provided in a form recorded on a recording medium such as a CD-ROM (Compact Disk Read Only Memory), a DVD-ROM (Digital Versatile Disk Read Only Memory), and a USB (Universal Serial Bus) memory. Also, the program may be downloaded from an external device via a network.

(Modification)
The present invention is not limited to the embodiments described above, and various modifications and applications are possible without departing from the gist of the present invention.

The indices may be calculated by changing the joints of the frame members selected by the selection unit 150 and the dimensions of the joints of the frame members. That is, based on the joints and joints selected by the selection unit 150, the dimensions are changed within a predetermined range, for example, by several millimeters to several centimeters, and the index is also calculated for the adjusted pattern. can be
For example, as shown in FIG. 13, the index is calculated by changing the vertical length of the joining portion of the joint from M1 to M2 and changing the horizontal length from L1 to L2. In addition, the dimension to be changed is not limited to the vertical direction or the horizontal direction, and other dimensions such as the angle of the joining portion of the joint may be used.

100 design support device 101 CPU
102 ROMs
103 RAM
104 storage 105 communication interface 106 input unit 107 display unit 108 bus 110 reception unit 120 learning unit 130 determination unit 140 index calculation unit 150 selection unit

Claims (8)

A combination of the overall shape of the building, the position and shape of the structural members that make up the building, the position and type of joints of the structural members, the position and type of joints of the structural members, and a predetermined index of the building. machine learning as training data,
By inputting the overall shape of the building to be newly constructed as the learning result of the machine learning, the position and shape of the structural members corresponding to the overall shape of the building are determined,
determining the position and type of the joint of the frame member and the position and type of the joint of the frame member from the position and shape of the frame member;
calculating a predetermined index of the building to be newly constructed based on the determined positions and shapes of the frame members, positions and types of joints of the frame members, and positions and types of joints of the frame members;
A design support device that selects the position and shape of the frame member, the position and type of the joint of the frame member, and the position and type of the joint of the frame member based on the index. 2. The design support device according to claim 1, wherein the index is calculated by changing the joints of the selected frame members and the dimensions of the joints of the frame members. 3. The design support device according to claim 1, wherein said index is strength or structural performance of said newly constructed building. The frame member has a predetermined length,
4. The design support device according to any one of claims 1 to 3, wherein said index is effective utilization of a combination of said structural members. 5. The design support device according to any one of claims 1 to 4, wherein said index is an evaluation of the arrangement of spaces formed between said structural members of said building to be newly constructed. The combination of the overall shape of the newly constructed building, the determined position and type of joints of the frame members, and the position and type of joints of the frame members are added to the teacher data, and the machine learning is performed. 6. The design support device according to any one of 1 to 5. 7. The design support device according to any one of claims 1 to 6, wherein the frame members are made of wood, a combination of wood and steel frames, or a combination of wood and reinforced concrete. A design support program for causing a computer to function as each unit constituting the design support apparatus according to any one of claims 1 to 6.

Images