JP7121891B2 - Structural information display device, structural design support device, and structural design support model learning device - Google Patents

Description

Application of Article 30, Paragraph 2 of the Patent Act On November 17, 2017, http://www. takenaka. co. jp/news/2017/11/04/index. html? published on the website of

Application of Article 30, Paragraph 2 of the Patent Law Published on page 14 of the Nihon Keizai Shimbun dated November 17, 2017

Application of Article 30, Paragraph 2 of the Patent Law Published on November 20, 2017 Kensetsu Tsushin Shimbun, November 20, 2017 morning edition, page 3

Article 30, Paragraph 2 of the Patent Act applied Published on page 2 of the morning edition of the November 27, 2017 Nikkan Kensetsu Sangyo Shimbun on November 27, 2017

Application of Article 30, Paragraph 2 of the Patent Act Published on page 11 of the November 30, 2017 morning edition of Denki Shimbun on November 30, 2017

The present invention relates to a structural information display device, a structural design support device, and a structural design support model learning device.

Conventionally, there is known a building structural design support system that supports structural design in construction work (for example, Patent Literature 1).

JP-A-2009-26303

The structural design support system described in Patent Literature 1 supports structural design by determining the strength of a building. However, the technique described in Patent Document 1 does not take account of structural design performance information, so it may not be appropriate.

It is an object of the present invention to appropriately support structural design based on actual information on structural design of construction work in consideration of the above facts.

In order to achieve the above object, the structure information display device of the present invention includes: a reception unit that receives input information that is information about a building; corresponding to the input information and the display target information received by the reception unit from a database storing structural design data representing the information on the building and the actual information on the structure of the building for each of the buildings of An information acquisition unit that acquires track record information on a building, and a display unit that displays the track record information on the structure of the building acquired by the information acquisition unit. As a result, information on past construction work is displayed appropriately, thereby supporting structural design.

The information on the building of the structural information display device of the present invention includes at least one of structure type, building use, floor type, foundation type, and number of floors, and the performance information on the structure of the building includes information on the quantity of materials, At least one of calibration ratio, live load, seismic load, column coverage area, and column span may be included. As a result, information on past construction work is displayed appropriately, thereby supporting structural design.

The structural design support device of the present invention comprises an acquisition unit that acquires information on a building to be designed and information on the quantity of materials of the building to be designed; Based on the information on the quantity of materials of the building to be designed, the learned model machine-learned in advance from the learning data representing the information on the building for learning and the information on the quantity of materials of the building for learning, and an estimating unit that outputs information regarding whether the structural design of the building to be designed is appropriate. As a result, it is possible to appropriately support structural design based on the learned model in which past performance information of construction work is reflected.

The information relating to the quantity of materials for the learning building in the learning data of the present invention is information relating to the quantity of materials corresponding to each part of the learning building, and the estimating unit includes the acquiring unit Based on the information about the building to be designed acquired by and the learned model, information about whether or not the structural design of each part of the building to be designed is appropriate can be further output. As a result, based on the learned model that reflects past construction work performance information, information on whether or not the structural design of each part of the building is appropriate is output, so it is possible to appropriately support structural design. can be done.

Further, the structural design support apparatus of the present invention includes an acquisition unit that acquires initial conditions of a building to be designed, and the above-mentioned A simulation unit that performs at least one of a simulation related to building structural calculations and a simulation related to the quantity of building materials for the structural design of the building, and an evaluation of the structural design of the building according to the simulation results obtained by the simulation unit. a learning unit for obtaining a trained model for outputting the structural design of the building from the initial conditions of the building by performing reinforcement learning of a model for outputting the structural design of the building from the initial conditions of the building based on the values; a design output unit that outputs a structural design of the building to be designed based on the learned model obtained by the learning unit. Structural design can be appropriately supported by outputting a plurality of structural designs of the building to be designed based on the evaluation value for the structural design according to the simulation results.

The learning unit of the structural design support device of the present invention creates a model for outputting the next structural design of the building to be designed from the previously obtained structural design of the building to be designed based on the simulation result. Reinforcement learning is repeated to update the learned model, and the design output unit outputs the previously obtained structural design of the building to be designed based on the learned model updated by the learning unit. It is possible to output the next structural design of the building to be designed. Structural design candidates can be appropriately and efficiently obtained by using a trained model that is successively updated.

The design output unit of the structural design support device of the present invention reads out the performance information from a database storing performance information related to building structures, and extracts the performance information from each of the structural designs output based on the learned models. It is possible to output the structural design whose degree of similarity with information is equal to or greater than a threshold. By selecting structural design candidates that are similar to the track record information, it is possible to obtain structural design candidates that are similar to structural designs of past construction work.

The structural design support model learning device of the present invention acquires information on the quantity of building materials from information on the building based on learning data representing information on the building for learning and information on the quantity of materials of the building for learning. It comprises a learning unit that performs machine learning on a model to be output, and obtains a trained model that outputs information on the quantity of materials of the building from the information on the building. As a result, it is possible to obtain a trained model that outputs information about the quantity of building materials in accordance with the information about the building and the information about the quantity of building materials.

ADVANTAGE OF THE INVENTION According to this invention, the effect that a structural design can be supported appropriately based on the performance information regarding the structural design of construction work is acquired.

1 is a block diagram showing a schematic configuration of a construction information display device according to a first embodiment; FIG. It is a figure which shows an example of the information stored in a structural-design database. It is a figure which shows an example of the walk corresponding to each location of a building. FIG. 4 is an explanatory diagram for explaining input information and display target information; It is a figure which shows an example of the structural information display processing routine in 1st Embodiment. It is a block diagram showing a schematic configuration of a structural design support system according to a second embodiment. 8 is a block diagram showing a schematic configuration of a client terminal according to the second embodiment; FIG. It is a block diagram showing a schematic configuration of a structural design device according to a second embodiment. It is a block diagram showing a schematic configuration of a structural design support device according to a second embodiment. FIG. 10 is a diagram showing an example of a trained model according to the second embodiment; FIG. FIG. 10 is an explanatory diagram for explaining processing in the second embodiment; It is a figure which shows an example of the learning processing routine in 2nd Embodiment. It is a figure which shows an example of the estimation processing routine in 2nd Embodiment. FIG. 11 is a block diagram showing a schematic configuration of a structural design support system according to a third embodiment; FIG. It is a block diagram which shows schematic structure of the structural-design assistance apparatus which concerns on 3rd Embodiment. It is a figure which shows an example of the structural design support processing routine in 3rd Embodiment. FIG. 10 is a diagram showing another configuration example of a trained model;

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, embodiments of the present invention will be described in detail.

<System Configuration of Structural Information Display Device According to First Embodiment>

FIG. 1 is a block diagram showing an example of the configuration of the structural information display device according to the first embodiment. The structural information display device 10 can be functionally represented by a configuration including an operation device 12, a computer 20, and a display device 30, as shown in FIG.

In the first embodiment, a structural information display device that displays performance information regarding past structural designs will be described. By displaying performance information related to structural design, it is possible to support the design work of the structural designer.

The operation device 12 receives operation information input by the user of the structural information display device 10 .

The computer 20 includes a CPU (Central Processing Unit), a ROM (Read Only Memory) that stores programs and the like for realizing each processing routine, a RAM (Random Access Memory) that temporarily stores data, and a memory as storage means. , a network interface, etc. The computer 20 functionally includes a structural design database 22 , a reception section 24 , an information acquisition section 26 and a display section 28 .

The structural design database 22 stores structural design data representing information on the building and performance information on the structure of the building for each of a plurality of buildings.

FIG. 2 shows an example of information stored in the structural design database 22. As shown in FIG. As shown in FIG. 2A, the building table showing information about buildings includes structure type, building use, floor type, foundation type, and number of floors.

Further, as shown in FIG. 2B, the yield table showing actual information about the structure of the building includes the yield of each material. Yoke is the quantity of material per unit area or volume. For example, the steps table includes steps for concrete, steps for formwork, steps for rebar (R/A), and steps for rebar (R/C). It should be noted that the yield is an example of information relating to the quantity of the material of the present invention.

Further, as shown in FIG. 2(C), the structural information table showing the actual information about the structure of the building includes verification ratio, live load, seismic load, column governing area, and column span. The test ratio is the ratio (N/F) between the structural strength (F) of the building and the acting force (N) acting on the building.

The steps of each material in the steps table shown in FIG. 2(B) include steps corresponding to each part of the building. For example, as shown in FIG. 3, concrete steps for pillars, concrete steps for girders, concrete steps for small beams, concrete steps for walls, concrete steps for slabs, etc. are assigned. It is

The receiving unit 24 receives input information, which is information about buildings, and display target information, which is information designating performance information about building structures. For example, the user inputs information about a building similar to the building the user is designing. Also, the user inputs desired information about the building set as the input information as display target information.

FIG. 4 shows an explanatory diagram for explaining input information and display target information received by the receiving unit 24. As shown in FIG. For example, the user appropriately selects portions C and D on the display screen X shown in FIG. 4 to input input information. In the example shown in FIG. 4, the structure type, building use, floor type, foundation type, number of floors, etc. are input as the input information.

Also, the user appropriately selects portions A, A1, B, and B1 on the display screen X shown in FIG. 4 to input information to be displayed. In the example shown in FIG. 4, which of the step rate, test ratio, live load, seismic load, column governing area, and column span is selected is input as input information.

For example, when "gait" is selected at points A and B, the user selects concrete, formwork, reinforcing bar (R/A), reinforcing bar (R/C), and a steel frame or the like can be further selected. Also, the user can further select columns, girders, girders, braces, walls, and slabs at locations A1 and B1.

The information acquisition unit 26 acquires input information and display target information received by the reception unit 24 . Then, the information acquisition unit 26 acquires the actual building information corresponding to the input information and the display target information from the structural design database 22 .

The display unit 28 displays the performance information regarding the structure of the building acquired by the information acquisition unit 26 . For example, as indicated by Y in FIG. 4, the display unit 28 selects "gait" as the display target information A and B, and selects "concrete" and "total" as the display target information A1. In addition, when "reinforcing bar (R/A)" and "aggregation" are selected as the display target information B1, the display screen Y as shown in FIG. 4 is displayed.

This allows the user to grasp the approximate rate of construction of a building similar to the building he or she is designing. For example, if the building you are designing is an S structure, as shown in Figure 4 above, the concrete yield is approximately between 0.4 and 0.8, and the yield of reinforcing bars (R/A) is It can be seen that it is approximately between 0.1 and 0.8.

Further, for example, as shown in FIG. 4, the performance information is extracted from the structural design database 22, and the performance information is drawn on the display screen Y in an appropriate graph. Further, when a plot point E on the display screen Y is specified, details of the construction work corresponding to the plot point are displayed. As a result, the structural designer can refer to the performance information regarding the structural design of the building in the past, so that the structural design is properly supported.

The display device 30 displays a display screen under the control of the display section 28 . The display device 30 is implemented by, for example, a display. For example, the display device 30 displays a display screen X, a display screen Y, etc., as shown in FIG.

<Action of structural information display device>

Next, the operation of the structural information display device 10 will be described. When structural design data is stored in the structural design database 22 and input information and display target information are input by the user via the operation device 12, the structural information display device 10 executes the structural information display processing routine shown in FIG. Run.

In step S100, the receiving unit 24 receives input information and display target information.

In step S102, the information acquisition unit 26 acquires the input information and display target information received in step S100. Then, the information acquisition unit 26 acquires the actual building information corresponding to the input information and the display target information from the structural design database 22 .

In step S104, the display unit 28 displays the performance information regarding the structure of the building obtained in step S102.

The display device 30 displays performance information regarding the structure of the building.

As described in detail above, in the first embodiment, actual building information corresponding to input information and display target information is displayed from a database storing structural design data for each of a plurality of buildings. As a result, past construction work performance information can be displayed appropriately, and structural design can be supported.

<Configuration of Structural Design Support System According to Second Embodiment>

In the second embodiment, past performance information of construction work is set as learning data, and a learned model is generated by machine learning based on the learning data. And it differs from the first embodiment in that structural design is supported using the generated learned model.

FIG. 6 is a block diagram showing an example configuration of a structural design support system according to the second embodiment. The structural design support system 210 can be functionally represented by a configuration including a client terminal 211, a structural design device 212, and a structural design support device 213, as shown in FIG. A client terminal 211 , a structural design device 212 and a structural design support device 213 are connected by a predetermined communication means 14 .

[Client terminal]

The client terminal 211 includes an operation device 220, a computer 230, and a display device 240, as shown in FIG.

Operation device 220 receives operation information input by a user. For example, it accepts design information about a given building input by a user who is a structural designer.

The computer 230 includes a CPU, a ROM that stores programs and the like for implementing each processing routine, a RAM that temporarily stores data, a memory as storage means, a network interface, and the like. The computer 230 functionally includes a reception section 232 , a design section 234 , a control section 236 and a transmission/reception section 238 .

Reception unit 232 receives operation information input from operation device 220 .

The design unit 234 designs a predetermined building according to the design information received by the reception unit 232 . For example, when a predetermined CAD (computer-aided design) system is installed in the computer 230, the design section 234 generates CAD data representing a predetermined building based on the design information input by the user. As a result, the user performs the design work of the building, and information on the building to be designed (hereinafter simply referred to as "designed building information") is generated. The designed building information includes, for example, the scale of the building to be designed, the type of structure, the usage, various design data, and the like.

The control unit 236 performs control processing according to the operation information received by the receiving unit 232 . For example, the control unit 236 controls the transmitting/receiving unit 238 so as to transmit design building information to the structural design device 212 and the structural design device 212 described later in accordance with the operation information received by the receiving unit 232 .

The transmitting/receiving unit 238 transmits/receives information to/from the structural design device 212 under the control of the control unit 236 . Further, the transmission/reception unit 238 transmits/receives information to/from the structural design support device 213 under the control of the control unit 236 .

The display device 240 displays a predetermined screen under the control of the control section 236 .

[Structural design equipment]

The structural design device 212 is composed of a computer including a CPU, a ROM that stores programs and the like for realizing each processing routine, a RAM that temporarily stores data, a memory as storage means, a network interface, and the like. The structural design device 212 functionally includes a transmission/reception section 250 and a yield calculation section 252, as shown in FIG.

The transmitting/receiving unit 250 transmits/receives information to/from the client terminal 211 .

The work rate calculation unit 252 calculates the work rate of the building to be designed according to the designed building information acquired by the transmission/reception unit 250 . For example, when the building to be designed is an RC structure, the step calculation unit 252 calculates the size of each pillar, the size of each beam, and the size of each floor of the building to be designed, which are included in the design building information. Calculate the concrete yield accordingly.

The transmission/reception unit 250 transmits the step calculated by the step calculation unit 252 to the client terminal 211 .

[Structural design support device]

The structural design support device 213 is composed of a computer including a CPU, a ROM that stores programs and the like for realizing each processing routine, a RAM that temporarily stores data, a memory as storage means, a network interface, and the like. . The structural design support device 213 functionally includes a structural design database 260, a learning unit 262, a trained model storage unit 264, a transmitting/receiving unit 266, and an estimating unit 268, as shown in FIG. ing. The transmission/reception unit 266 is an example of an acquisition unit of the present invention.

The structural design database 260 stores a building table showing information about the learning building and a construction table showing information about the construction of the learning building. Information stored in the structural design database 260 is used as learning data by a learning unit 262, which will be described later. For example, for one building, the building structure type, building use, floor type, foundation type, and number of floors, the building's concrete rate, formwork rate, reinforcing bar (R/A) rate, and The rate of rebar (R/C) is used as one learning data.

The learning unit 262 uses learning data for each of the plurality of buildings stored in the structural design database 260 to machine-learn the model to generate a trained model. Specifically, the learning unit 262 machine-learns a model for outputting information on the walking rate from the designed building information.

When structurally designing a building, it is necessary to appropriately perform the design work according to information about the building (structural type of building, building use, floor type, foundation type, and number of floors). However, since the structural design work is done manually, there are cases where omissions and omissions are included in the design content. In order to suppress this, a study is conducted by a plurality of structural designers. At this time, the information on the yield, which is an example of the information on the quantity of building materials, may be useful for examining the details of the structural design.

For example, based on the past record information of construction work, if the building scale is about "X", the construction rate of the building scale is about "Y". Therefore, if the building scale of the building to be designed is "X", the design content is inappropriate if the construction rate is significantly larger than "Y" or significantly smaller than "Y". can be determined to be highly probable.

Therefore, in the second embodiment, a learned model is generated using learning data in which information about a building and information about a walk are associated with each other. As a result, for example, if the building is of this scale, the information that the walking rate is of this level is reflected in the learned model. Therefore, the trained model can be used to assist in determining whether or not the structural design details of the building to be designed are appropriate.

For the trained model of the second embodiment, for example, a neural network as shown in FIG. 10 can be used as an example of the model, and deep learning can be used as an example of the learning algorithm. The learned model of the second embodiment is a learned model that outputs information about the walking distance from the designed building information.

For example, based on the learning data stored in the structural design database 260, the learning unit 262 selects the learning data with the design building information of "X" and the work rate of "A". The neural network is trained so that when "X" is input, the probability of the yield "A" increases.

The learning unit 262 then stores the obtained learned model in the learned model storage unit 264 .

The learned model storage unit 264 stores the learned model generated by the learning unit 262 .

The transmitting/receiving unit 266 receives the designed building information and the steps transmitted from the client terminal 211 .

The estimating unit 268 determines whether the structural design of the building to be designed is appropriate based on the designed building information and steps received by the transmitting/receiving unit 266 and the learned model stored in the learned model storage unit 264. information (hereinafter simply referred to as "inspection information") relating to whether the

For example, first, the estimation unit 268 inputs the designed building information received by the transmission/reception unit 266 to the learned model. By inputting the designed building information into the trained model, the probability of each step for each input designed building information is output as shown in FIG.

Then, as shown in FIG. 11, the estimating unit 268 compares the step “A” with the highest probability with the step “X” received by the transmitting/receiving unit 266 to determine whether the structural design of the building to be designed is correct. Generate inspection information that indicates whether or not it is appropriate.

For example, the estimation unit 268 outputs inspection information indicating that the structural design of the building to be designed is inappropriate when the difference between the walking rate "A" and the walking rate "X" is equal to or greater than a predetermined threshold. Generate. If the difference between the walking rate "A" and the walking rate "X" is less than a predetermined threshold value, the estimating unit 268 detects inspection information indicating that the structural design of the building to be designed is appropriate. to generate

The transmission/reception unit 266 transmits the examination information estimated by the estimation unit 268 to the client terminal 211 .

The transmission/reception unit 238 of the client terminal 211 receives the inspection information transmitted from the structural design support device 213 and outputs the inspection information to the control unit 236 . The control unit 236 causes the display device 240 to display the examination information.

A user operating the client terminal 211 confirms inspection information indicating whether or not the design building information of the building he or she is designing is appropriate. This allows the user to know whether the structural design of the building he or she is designing is appropriate.

<Action of structural information display device>

Next, the operation of the structural design support system 210 will be described. First, in the structural design support device 213, a learning processing routine shown in FIG. 12 is executed.

<Learning processing routine>

In step S<b>200 , the learning unit 262 reads each piece of data for learning stored in the structural design database 260 .

In step S202, the learning unit 262 learns a neural network, which is an example of a model, by deep learning based on the learning data read out in step S200 to obtain a trained model.

In step S<b>204 , the learning unit 262 stores the learned model obtained in step S<b>202 in the learned model storage unit 264 .

Next, a user who operates the client terminal 211 of the structural design support system 210 operates the client terminal 211 to create design building information of the building to be designed. The user then transmits the created designed building information to the structural design device 212 .

The transmitter/receiver 250 of the structural design device 212 receives the designed building information transmitted from the client terminal 211 . Then, the work rate calculator 252 of the structural design device 212 calculates the work rate of the designed building according to the designed building information received by the transmitter/receiver 250 . The transmission/reception section 250 of the structural design device 212 transmits the construction rate of the building to be designed calculated by the construction rate calculation section 252 to the client terminal 211 .

The transmitter/receiver 238 of the client terminal 211 receives the steps transmitted from the transmitter/receiver 250 of the structural design device 212 . Then, the user operates the operation device 220 of the client terminal 211 to transmit the designed building information obtained by the design section 234 and the steps transmitted from the structural design device 212 to the structural design support device 213 .

When the transmitter/receiver 266 of the structural design support device 213 receives the designed building information and the steps transmitted from the client terminal 211, the structural design support device 213 executes the estimation processing routine shown in FIG.

In step S<b>250 , the estimation unit 268 acquires the designed building information and the rate received by the transmission/reception unit 266 .

In step S<b>252 , the estimation unit 268 reads the learned model stored in the learned model storage unit 264 .

In step S254, the estimation unit 268 inputs the designed building information acquired in step S250 to the learned model read out in step S252. As a result, the probability of each step corresponding to the designed building information is output from the learned model.

In step S256, the estimation unit 268 identifies the step with the highest probability among the probabilities of the steps obtained in step S254.

In step S258, the estimation unit 268 compares the walking rate acquired in step S250 with the walking rate specified in step S256. For example, the estimation unit 268 calculates the difference between the walking rate acquired in step S250 and the walking rate specified in step S256.

In step S260, the estimating unit 268 generates inspection information indicating whether or not the structural design of the building to be designed is appropriate, based on the difference between the rates obtained in step S258 and a preset threshold value. to generate

The transmission/reception unit 266 of the structural design support device 213 transmits the inspection information obtained in step S260 to the client terminal 211. FIG.

The transmission/reception unit 238 of the client terminal 211 receives the inspection information transmitted from the structural design support device 213 and causes the display device 240 to display the inspection information.

A user operating the client terminal 211 confirms inspection information indicating whether or not the design building information of the building he or she is designing is appropriate. This allows the user to know whether the structural design of the building he or she is designing is appropriate.

As described in detail above, in the second embodiment, based on the learning building information and the learning data representing the walking rate of the learning building, the information relating to the walking rate is output from the information relating to the building. Machine learning is performed on the model for to obtain a trained model. As a result, by using machine learning, it is possible to obtain a trained model that outputs information on steps in accordance with information on buildings and information on steps.

In addition, in the second embodiment, the structural design of the building to be designed is appropriately based on the design building information, the information on the progress of the building to be designed, and the learned model that has been machine-learned in advance from the learning data. Outputs information about whether or not As a result, it is possible to appropriately support structural design based on the learned model in which past performance information is reflected.

<Configuration of Structural Design Support System According to Third Embodiment>

In the third embodiment, simulation is performed according to initial conditions regarding structural design. Then, in the third embodiment, a learned model is generated by reinforcement learning based on the simulation results. Then, it differs from the first and second embodiments in that the obtained learned model is used to support the structural design.

FIG. 14 is a block diagram showing an example configuration of a structural design support system according to the third embodiment. The structural design support system 310 can be functionally represented by a configuration including a client terminal 211 and a structural design support device 213, as shown in FIG.

[Client terminal]

The user operates the operation device 220 of the client terminal 211 to set the initial conditions of the building to be designed. Then, the user transmits the initial conditions of the building to be designed to the structural design support device 213 via the transmitter/receiver 238 . As the initial conditions of the building to be designed, for example, CAD data in which the initial conditions of the building are set can be used.

[Structural design support device]

The structural design support device 213 is composed of a computer including a CPU, a ROM that stores programs and the like for realizing each processing routine, a RAM that temporarily stores data, a memory as storage means, a network interface, and the like. . As shown in FIG. 15, the structural design support device 213 functionally includes a transmission/reception unit 320, a design output unit 321, a simulation unit 322, a learning unit 324, a trained model storage unit 326, a structure A design database 330 is provided. The transmission/reception unit 320 is an example of an acquisition unit of the present invention.

The transmitting/receiving unit 320 acquires the initial conditions of the building to be designed transmitted from the client terminal 211 .

The design output unit 321 changes the structure of the building to be designed corresponding to the initial conditions acquired by the transmitting/receiving unit 320 to generate structural design candidates for the building to be designed.

In the third embodiment, a trained model is generated by reinforcement learning. Therefore, the structure of the building to be designed that is changed by the design output unit 321 is assumed to be the environment s in reinforcement learning. Action a is to change the size of columns, beams, floors, and the like, which are structural members of the building to be designed. An evaluation value according to the state when the structural members of the building to be designed are changed is set as a reward r. Therefore, in this embodiment, the environment s is the structure of the building to be designed, the action a is to change the structure of the building, and the reward r is an evaluation according to the state when the structural members of the building are changed. value.

First, the design output unit 321 changes the CAD data indicating the building structure corresponding to the initial conditions or the candidate for the building structure design generated by the previous process.

For example, as _a predetermined action a1, the design output unit 321 removes a predetermined column from a plurality of columns included in the CAD data of the building to be designed. Alternatively, the design output unit 321 adds a pillar to a predetermined location of the CAD data of the building to be designed as a predetermined action _a2 . Alternatively, the design output unit 321 thickens the pillars included in the CAD data of the building to be designed as a predetermined action _a3 . Alternatively, the design output unit 321 thins the pillars included in the CAD data of the building to be designed as a predetermined action _a3 . As a result, structural design candidates for the building to be designed are generated.

The simulation unit 322 evaluates each piece of CAD data obtained by changing the CAD data of the building to be designed. Specifically, the simulation unit 322 performs a simulation regarding the structural calculation of the building for the structural design candidate generated by the design output unit 321, and evaluates the structural strength of the structural design candidate. Further, the simulation unit 322 performs a yield calculation, which is a simulation related to the quantity of building materials, for the structural design candidate, and evaluates the cost of the structural design candidate. Then, the simulation unit 322 sets the reward r according to the simulation result.

For example, the simulation unit 322 evaluates each of the CAD data obtained by predetermined actions a ₁ to a ₄ using a predetermined evaluation function regarding structural design and yield. For example, with respect to CAD data obtained by _a predetermined action a1, if the structural strength is low, the reward r1 representing the evaluation will be small even if the _gait is small. Further, regarding the CAD data obtained by the predetermined action _a2 , the reward r2 representing the evaluation becomes large when the gait is small and the structural strength is _high .

Based on the simulation results obtained by the simulation unit 322, the learning unit 324 performs reinforcement learning on a model for outputting the structural design of the building from the initial conditions of the building, and obtains a trained model. Specifically, the learning unit 324 reinforces and learns a model for outputting the next structural design from the previously obtained structural design candidate based on the evaluation value for the structural design of the building among the simulation results. , iteratively updating the trained model.

For example, the learning unit 324 learns the action-value function Q(s, a) according to the following equation (1) based on the simulation result (Internet (URL: http://www.sist.ac.jp/ ~kanakubo/research/reinforcement_learning.html).).

（１）

(1)

where s _t represents the environment at time t. Also, a _t represents the action at time t. Action a _t changes the environment to s _t+1 . Also, r _t+1 is a reward that is given according to changes in the environment.

The term with max in the above equation (1) is obtained by multiplying the Q value when the action a with the highest Q value (known at that time) is selected under the environment s _t+1 by γ. γ is a parameter satisfying 0<γ≦1 and is called a discount rate. α is a learning coefficient and is in the range of 0<α≦1.

The action-value function Q(s, a) as a model is learned according to the update formula shown in the above formula (1), and the action-value function Q(s, a) as a learned model is generated. The action value function Q(s, a) is sequentially updated according to the evaluation value regarding the structural strength of the structural design candidate and the evaluation value regarding the cost of the structural design candidate.

The learned model storage unit 326 stores the action value function Q(s, a) obtained by the learning unit 324 . The action value function Q(s, a) is represented by, for example, a Q table storing a Q value for each combination of state s and action a. Therefore, according to the action-value function Q(s, a) and the current building structure s, the next action a (for example, action to change the size of pillars, beams, floors, etc.) is determined. be.

When the design output unit 321 generates the next structural design candidate from the structural design candidate obtained last time, the action value function Q(s, a ), the structural design candidates for the building to be designed are output. By using the action value function Q(s, a), the design output unit 321 can generate candidates with high evaluation values regarding structural strength and cost. Therefore, when generating the next candidate for structural design, it is easier to generate a more suitable candidate than the previous one.

Generation of structural design candidates by the design output unit 321, evaluation of structural design candidates by the simulation unit 322, and learning processing by the learning unit 324 are repeated a predetermined number of times. This generates multiple structural design candidates.

In this embodiment, each time a structural design candidate is generated, the action value function Q(s, a) is sequentially updated based on the reward, which is the evaluation value of the structural design candidate obtained by the simulation unit 322. . As a result, structural design candidates are generated such that the more the learning process by the learning unit 324 is repeated, the higher the evaluation value. Therefore, candidates suitable for structural design can be efficiently generated.

Next, after the processing of the design output unit 321, the processing of the simulation unit 322, and the processing of the learning unit 324 are repeated a predetermined number of times, the design output unit 321 selects a plurality of structural design candidates generated so far. Sort out. Specifically, the design output unit 321 selects a plurality of structural design candidates generated so far based on the structural design performance information stored in the structural design database 330 . For example, in the structural design database 330, information such as that shown in FIG. 2 is stored as performance information.

By using the action value function Q (s, a), while candidates with high evaluation values regarding structural strength and evaluation values regarding cost are generated, the structural design of the newly designed building is similar to the past structural design. It may be preferable to be similar to For this reason, in the present embodiment, from among the plurality of structural design candidates generated using the action value function Q(s, a), a candidate that is as close as possible to past structural design performance information is selected as a structural design candidate to be output. Select to leave as

Specifically, the design output unit 321 outputs the structural design candidates and the results stored in the structural design database 330 for each of the plurality of structural design candidates generated using the action-value function Q(s, a). Calculate the similarity between information. For example, the degree of similarity is calculated according to the structure type, building use, floor type, foundation type, number of floors, etc., and the verification ratio, loading load, seismic load, column governing area, column span, and the like. The degree of similarity increases as the structural design candidates and the performance information of buildings constructed in the past are more similar.

Next, the design output unit 321 selects a structural design candidate whose degree of similarity is greater than a preset threshold as a structural design candidate to be output. Then, the design output unit 321 outputs a plurality of structural design candidates selected according to the performance information as a result.

The transmitting/receiving unit 320 transmits the plurality of structural design candidates output by the design output unit 321 to the client terminal 211 .

The transmitting/receiving unit 238 of the client terminal 211 receives a plurality of structural design candidates transmitted from the structural design support device 213 . Then, the control unit 236 of the client terminal 211 displays a plurality of structural design candidates on the display device 240 .

The user selects the most appropriate structural design candidate as the final proposal from among the multiple structural design candidates displayed on the display device 240 . As a result, the structural design of the building to be designed is obtained by setting only the initial conditions. The user then operates the operation device 220 of the client terminal 211 to transmit the structural design selected as the final plan to the structural design support device 213 .

The transmission/reception unit 320 of the structural design support device 213 stores the structural design of the final plan transmitted from the client terminal 211 in the structural design database 330 . When the final structural design is stored in the structural design database 330, weight information may be assigned to the final structural design before storing. Thereby, for example, structural design candidates can be selected according to the weight information.

<Operation of Structural Design Support System>

Next, the operation of the structural design support system 310 will be described. When the initial conditions of the building to be designed are transmitted from the client terminal 211 to the structural design support device 213, the structural design support device 213 executes the structural design support processing routine shown in FIG.

<Structural design support processing routine>

In step S<b>300 , the transmission/reception unit 320 acquires the initial conditions of the building to be designed, which are transmitted from the client terminal 211 .

In step S302, the design output unit 321 changes the structure of the building to be designed corresponding to the initial conditions acquired in step S300 or the structural design candidate generated in the previous step S302, Generate structural design candidates for buildings. When generating structural design candidates by changing the structure of the building to be designed corresponding to the initial conditions, the design output unit 321 generates structural design candidates by randomly changing the structure of the building, for example. . In addition, when the design output unit 321 generates the next structural design candidate from the structural design candidate obtained in the previous step S302, the learned model stored in the learned model storage unit 326 is used. Based on the action-value function Q(s, a), the next structural design candidate is output.

In step S304, the simulation unit 322 performs a simulation related to building structural calculation for the structural design candidate generated in step S302, and evaluates the structural strength of the structural design candidate. Further, the simulation unit 322 performs a commission calculation, which is a simulation of the cost of the building, on the structural design candidate generated in step S302, and evaluates the cost of the structural design candidate. Then, the simulation unit 322 sets the reward r according to the simulation result.

In step S306, the learning unit 324 performs reinforcement learning of a model for outputting the structural design of the building from the initial conditions of the building or the previously generated structural design candidates based on the simulation results obtained in step S304. to obtain an action-value function Q(s, a) as a trained model.

In step S308, the learning unit 324 stores the action value function Q(s, a), which is the learned model obtained in step S306, in the learned model storage unit 326 to update the learned model.

In step S310, it is determined whether or not the processes of steps S302 to S308 have been repeated a predetermined number of times. If the processing from step S302 to step S308 has been repeated a predetermined number of times, the process proceeds to step S312. On the other hand, if the processes from step S302 to step S308 have not been repeated a predetermined number of times, the process returns to step S302.

In step S<b>312 , the design output unit 321 reads the performance information stored in the structural design database 330 .

In step S314, the design output unit 321 calculates the degree of similarity between the plurality of structural design candidates generated in step S302 and the performance information read out in step S312. Then, the design output unit 321 selects a structural design candidate whose degree of similarity is greater than a preset threshold as a structural design candidate to be output.

In step S<b>316 , the transmitting/receiving unit 320 transmits the plurality of structural design candidates obtained in step S<b>314 to the client terminal 211 .

The transmitting/receiving unit 238 of the client terminal 211 receives a plurality of structural design candidates transmitted from the structural design support device 213 . Then, the control unit 236 of the client terminal 211 displays a plurality of structural design candidates on the display device 240 .

The user selects the most appropriate structural design candidate as the final proposal from among the multiple structural design candidates displayed on the display device 240 . The user then operates the operation device 220 of the client terminal 211 to transmit the structural design selected as the final plan to the structural design support device 213 .

The transmission/reception unit 320 of the structural design support device 213 stores the structural design of the final plan transmitted from the client terminal 211 in the structural design database 330 .

As described in detail above, in the third embodiment, in the structural design of a building when the structure of the building to be designed is changed according to the initial conditions of the building to be designed, Conduct simulations and simulations on building material quantities. In the third embodiment, a model for outputting a structural design of a building from the initial conditions of the building is subjected to reinforcement learning based on the evaluation value for the structural design of the building according to the simulation result. Then, in the third embodiment, the structural design of the building to be designed is output based on the obtained learned model. By using reinforcement learning, it is possible to obtain a trained model that outputs a building structural design from each building structural design candidate. Further, by outputting a plurality of structural design candidates based on the learned model obtained by reinforcement learning, structural design can be appropriately supported.

In addition, it is possible to obtain structural design candidates for the building to be designed simply by inputting the initial conditions of the building to be designed. In addition, since a trained model is generated based on the evaluation values related to structural calculation and the evaluation values related to costs among the simulation results, and the trained model outputs candidates for structural design, the evaluation values related to structural calculation and costs related to Candidates for structural design that are considered appropriate from the viewpoint of evaluation value can be obtained.

In addition, in the third embodiment, based on the simulation results, a model for outputting the structural design of the building to be designed next time is subjected to reinforcement learning based on the structural design of the building to be designed which was obtained last time. Iterate to update the model. Therefore, by using a trained model that is successively updated, structural design candidates can be appropriately and efficiently obtained.

In addition, since the learned model obtained by reinforcement learning is also used in the structural design of other buildings to be designed, the learned model is updated each time processing is performed. Therefore, candidates for structural design can be appropriately and efficiently obtained.

In addition, in the third embodiment, structural designs whose similarity to actual information is equal to or greater than a threshold are output from each structural design output based on a learned model obtained by reinforcement learning. As a result, structural design candidates that are similar to the actual performance information are selected, so that structural design candidates that are similar to structural designs of past construction work can be obtained.

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

For example, in the above-described embodiment, the case of using gaikake as an example of information on the quantity of building materials has been described, but the present invention is not limited to this. information may be used.

Further, in the above-described second embodiment, the case of outputting information regarding whether or not the structural design of the building to be designed is appropriate is taken as an example, using the gaikake, which is an example of information regarding the quantity of building materials. Illustrated, but not limited to. For example, using information different from the information on the quantity of building materials, information on whether or not the structural design of the building to be designed is appropriate may be output. In this case, for example, building scale, number of floors, floor area, structure type, structural calculation route, load table, span, floor height, wall quantity, brace ratio, material strength, material end conditions, loading load, finishing load, Additional load, etc., seismic load, wind load, soil and water pressure load, building drawing recognition results, building usage, column axial force, column pitch (interval), building material, building designer, year of design, construction site, estimate Information regarding whether or not the structural design of the building to be designed is appropriate may be output using the results, test ratio, floor type, roof type, presence/absence of braces, and the like.

Further, in the above-described second embodiment, the case where a neural network model as an example of a learning model is learned by deep learning has been described as an example, but the present invention is not limited to this. For example, a model different from the neural network model may be learned by a learning method different from deep learning.

Further, in the above-described second embodiment, according to the probability of each step output from the learned model, the step with the highest probability is compared with the step calculated by the structural design device 212 as an example. , but it is not limited to this. For example, as shown in FIG. 17, the designed building information and the rate calculated by the structural design device 212 are input to the learned model, and the probability that the combination is appropriate (the probability that it is OK) and the The trained model may be configured to output the probability that the combination is not appropriate (probability of NG).

In addition, in the above-described second embodiment, the case of simply outputting information regarding whether or not the structure of the building to be designed is appropriate according to the design building information and the rate of construction has been described as an example. It is not limited.

For example, the information regarding the walking rate of the learning building in the learning data may be set as the information regarding the walking rate corresponding to each part of the learning building. In this case, the learned model is configured to calculate the rate of progress of each part of the building to be designed, and the estimation unit 268 determines whether the structural design of each part of the building to be designed is appropriate. You may make it output further the information regarding. As a result, for example, it is automatically pointed out that the cross section of a beam or column at a specific location is designed to be small, so that it is possible to automatically assist the structural designer in structural design. For example, according to the design building information, by further outputting information indicating that the construction of the X-floor beam is about A for a building of this scale, It is possible to grasp whether the part of is inappropriate. Also, the trained model may be configured to further advise on building steps based on information about which locations are inappropriate. For example, it may be possible to advise that the quantity of concrete in a location determined to be inappropriate is insufficient.

Also, the trained model may be configured to give advice on structural design in general. In this case, for example, advice on analysis condition settings (for example, compared to similar cases, it tells you which items have been changed from the default with a high probability), advice on trends in setting values (for example, A If it is changed, B is also changed, etc.), advice on the adequacy of the amount of reinforcement (for example, columns, large beams, small beams, and earthquake-resistant walls, etc.), advice on load setting (for example, leakage compared with similar cases) pointing out, etc.), advice on the adequacy of the set cross section (beam size), and advice on the adequacy of the set cross section (column size) may be configured to output the model. In addition, as a past case example search, it may be configured so as to be able to search for cases of past construction work, which is actual record information.

Advice on setting analysis conditions and advice on trends in set values are based on factors such as building scale, number of floors, floor area, structure type, and structural calculation route.

In addition, regarding the advice on the adequacy of the amount of reinforcement, the load table, span, story height, wall quantity, brace ratio, number of floors, material strength, and material end conditions are assumed to be factors, so these information Advice will be given based on

In addition, with regard to advice on load settings, factors such as live load, finish load, additional load, etc., seismic load, wind load, earth and water pressure load, building drawing recognition results, building usage, and building scale are assumed as factors. Advice is given based on this information.

In addition, advice on the appropriateness of the set cross section (beam size) is based on the loading load, finish load, span, number of floors, structure type, and end conditions, etc. will be

Advice on the adequacy of the set cross section (column size) is given based on such information as column axial force, number of floors, column pitch (interval), structure type, etc. are assumed to be factors.

Further, regarding the past case case search, for example, the search is configured to be performed according to the building scale, the number of floors, the floor area, the structure type, and the drawing recognition result of the building.

When giving advice on structural design, for example, by clustering a plurality of buildings corresponding to actual performance information, a plurality of buildings are classified in advance, and according to the results of comparison between the classification result and the building to be designed, the structure Advice on design may be given. As an example of a model used when clustering a plurality of buildings, for example, the k-means method can be used.

In the above description, the program is pre-stored (installed) in a storage unit (not shown), but the program may be recorded on any recording medium such as a CD-ROM, DVD-ROM, micro SD card, or the like. It is also possible to provide it in the form provided.

10 Structural information display devices 12, 220 Operating devices 20, 230 Computers 22, 260, 330 Structural design databases 24, 232 Reception unit 26 Information acquisition unit 28 Display units 30, 240 Display devices 210 and 310 Structural design support system 211 Client terminal 212 Structural design device 213 Structural design support device 234 Design unit 236 Control units 238, 250, 266, 320 Transmission/reception unit 252 Percentage calculation units 262, 324 Learning units 264, 326 Trained model storage unit 268 Estimation unit 321 Design output unit 322 Simulation Department

Claims (2)

a reception unit that receives input information, which is information about a building, and display target information, which is information specifying performance information about the structure of the building;
Corresponding to the input information and the display target information received by the reception unit from a database storing structural design data representing information on the building and performance information on the structure of the building for each of a plurality of buildings, an information acquisition unit that acquires performance information of the building;
a display unit for displaying performance information related to the structure of the building acquired by the information acquisition unit;
including
The display unit displays a numerical value represented by the performance information about the structure of the building on each axis of a scatter diagram in which one building is one plot point, and displays so that each plot point is drawn for each information about the building. death,
The type of performance information related to the structure of the building and the type of information related to the building are displayed so as to be selectable by the user,
When the plot point is selected by the user, construction details of the building corresponding to the plot point are displayed;
Structural information display device. The information about the building includes at least one of structure type, building use, floor type, foundation type, and number of floors,
The performance information on the structure of the building includes at least one of information on the quantity of materials, verification ratio, dead load, seismic load, column governing area, and column span,
The structural information display device according to claim 1.

Images