JP2021105874A - Frame data generation device, frame data generation method, and frame data generation program - Google Patents

Abstract

To support creating design data that is used as teacher data in machine learning.SOLUTION: A frame data generation device comprises: an initial condition acquisition unit for acquiring a planar shape, height and a number of floors of a building; a frame shape generation unit for changing span division of a main frame including columns and girders within a range of the acquired planar shape and the number of floors to generate a plurality of frame shapes; a simulation unit for calculating, for each of the generated frame shapes, at least three of an evaluation value that is a guideline for earthquake resistance, an evaluation value that is a guideline for workability, an evaluation value that is a guideline for material cost, and an evaluation value that is a guideline for balance of the building, and obtaining a Pareto solution set using the calculated evaluation values as an objective function to store it in a storage device.SELECTED DRAWING: Figure 1

Description

The present technology relates to a frame data generation device, a frame data generation method, and a frame data generation program.

Conventionally, a computer equipped with an input operation means for inputting by a designer, a display means for displaying various screens, and an arithmetic processing device for receiving an input signal by the input operation means, processing the signal, and outputting the signal to the display means. A system has been proposed (eg, Patent Document 1). This computer system has a database that structurally calculates the size and structure of the basic components of pillars, beams, and floors according to the number of floors of the building, determines the standard in advance, and records it. Basic structure data setting means for setting basic structural data such as the number of X spans, the number of Y spans, the number of floors, the length of each span, the height of each floor, the type of structure, and the purpose of the building by operating the input operation means. Operation of the basic component data setting means that searches from the database based on the number of floors and usage of the building set by the basic structure data setting means and sets the standard of the size and structure of the basic component, and the input operation means of the designer. The shape, pillar, and beam of the building based on the change data setting means that sets the change data that changed the basic structure data and the basic component data, and the set basic structure data, the basic component data, and the set change data. The structural model constructing means for constructing the structural model by arranging the data, and the image processing means for displaying the structural model diagram for each floor and each span of the structural model configured by the structural model constructing means on the display means. There is.

In addition, a method for supporting structural planning by obtaining a Pareto solution set with the steel material cost and the maximum interlayer deformation angle as objective functions using a multipurpose genetic algorithm has also been proposed (for example, Non-Patent Document 1).

Japanese Unexamined Patent Publication No. 2003-343006

Naoto Tamura, "Study on Design Support in Structural Planning of Steel Structures Using Multi-Objective Optimization Method and Its Practical Use", Doctoral Dissertation, Nagoya University, 2012

Conventionally, a technique has been proposed in which a standard of size and structure of basic components is set based on a rule based on the set number of floors and usage of a building. Further, in the case of machine learning such as supervised learning, there is a problem that proper learning cannot be performed unless sufficient design data is accumulated.

An object of the present invention is to support the creation of design data used as teacher data in machine learning.

The frame data generation device according to the present invention has an initial condition acquisition unit for acquiring the plane shape, height and number of floors of a building, and a span division of a main frame including columns and girders within the range of the acquired plane shape and number of floors. For each of the frame shape generator that generates multiple frame shapes and the generated frame shape, the evaluation value that is a guideline for earthquake resistance, the evaluation value that is a guideline for workability, and the guideline for material cost It is provided with a simulation unit that calculates at least three of the evaluation values and the evaluation values that serve as a guideline for the balance of the building, obtains the Pareto solution set using the calculated evaluation values as the objective function, and stores them in the storage device.

In this way, it is possible to obtain a Pareto solution set based on a sufficient number of evaluation values for a plurality of frame shapes having different span divisions of the main frame. Further, for example, the designer evaluates the Pareto solution set and uses this as teacher data, which can be used for machine learning. Further, since a plurality of span divisions of the main frame that can be adopted for the volume of the building can be generated, a sufficient pattern and amount of teacher data can be generated.

In addition, the simulation unit serves as a guideline for earthquake resistance, a guideline for workability, a guideline for material cost, and a guideline for building balance for each of the generated frame shapes. The evaluation value may be calculated. By using the four evaluation values as objective functions, it becomes easier for the user who is the designer to evaluate the quality of the frame shape, and the accuracy of machine learning can be further improved.

In addition, the simulation unit calculates the stress level generated in the cross section of the main frame using the standard load stored in the storage device in advance in association with the application for each frame shape, and determines the predetermined allowable stress level. The Pareto solution set may be obtained by using a frame shape that does not exceed the frame shape. In this way, among the frame shapes that are automatically generated, those that satisfy the constraint conditions required for the building can be adopted.

The evaluation value that serves as a guideline for seismic resistance is the maximum interlayer deformation angle with respect to a predetermined seismic load, and the evaluation value that serves as a guideline for workability is the number of pieces of the main frame, which is a guideline for material cost. Is the weight of the steel frame, and the evaluation value that serves as a guide for the balance of the building may be an evaluation value that represents the rigidity, the eccentricity, or the equality of stress with respect to the load. Specifically, such an evaluation value can be used as an objective function.

Further, the simulation unit may obtain the Pareto solution set using a frame shape in which the maximum interlayer deformation angle does not exceed a predetermined reference. Even in this way, among the frame shapes that are automatically generated, those that satisfy the constraint conditions required for the building can be adopted.

In addition, the initial condition acquisition unit further acquires the use of the building, and the simulation unit reads out the standard load stored in the storage device in advance in association with the acquired use, and uses the read out load to seismic load. The maximum interlayer deformation angle with respect to the above may be calculated. By doing so, it becomes possible to set appropriate parameters and perform a simulation by a simple input process.

The contents described in the means for solving the problems can be combined as much as possible without departing from the problems and technical ideas of the present invention. Further, the content of the means for solving the problem can be provided as a device such as a computer or a system including a plurality of devices, a method executed by the computer, or a program executed by the computer. The program can also be run on the network. Further, a recording medium for holding the program may be provided.

According to the present invention, it is possible to support the creation of design data used as teacher data in machine learning.

FIG. 1 is a block diagram showing an example of a frame data generation device according to the present invention. FIG. 2 is a processing flow diagram showing an example of the frame data generation processing according to the present embodiment. FIG. 3 is a diagram showing an example of a screen for allowing the user to input initial conditions. FIG. 4 is a diagram showing an example of a load load that is predetermined in association with the application. FIG. 5 is a diagram showing an example of the generated frame shape. FIG. 6 is a processing flow diagram showing an example of simulation processing. FIG. 7 is a diagram showing an example of the output evaluation value. FIG. 8 is a diagram showing an example of the Pareto solution set three-dimensionally.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.

<Device configuration>
FIG. 1 is a block diagram showing an example of a frame data generation device according to the present invention. The frame data generation device 1 is a general computer, and generates a plurality of variations of the frame shape of a building within a range that satisfies a given initial condition. Further, in the frame data generation device 1, the number of steel frames (evaluation value as a guideline for material cost), the number of pieces of members (evaluation value as a guideline for workability), and the interlayer deformation angle (reference value for seismic performance) are used according to the frame shape. The Pareto solution for multiple evaluation values such as (evaluation value), rigidity, eccentricity, and stress uniformity with respect to load (evaluation value as a guideline for building balance) is obtained. By presenting and evaluating the information obtained in this way to a user who is an experienced designer, it is possible to use it as teacher data for machine learning a preferable frame shape. That is, according to the frame data generation device 1, it becomes possible to sufficiently accumulate design data that can be used as teacher data. By using at least three of the above evaluation values, it is possible to examine based on a sufficient number of viewpoints based on empirical rules, and it becomes easier for the user who is the designer to evaluate the quality of the frame shape. At the same time, the accuracy of machine learning can be improved.

The frame data generation device 1 according to the present invention is a general computer, and includes a communication I / F (Interface) 11, a storage device 12, an input / output I / F (Interface) 13, a processor 14, and a bus 15. To be equipped with.

The communication I / F 11 is, for example, a wired network card or a wireless communication module, and transmits / receives information to / from another computer via a network (not shown).

The storage device 12 includes a main storage device such as a RAM (Random Access Memory) and a ROM (Read Only Memory), and an auxiliary storage device (secondary storage) such as an HDD (Hard-disk Drive), an SSD (Solid State Drive), and a flash memory. Device). The main memory caches programs and data read by the processor and secures a work area for the processor. The auxiliary storage device stores the program executed by the processor and the design data represented according to the standard such as BIM (Building Information Modeling).

The input / output I / F 13 is connected to, for example, an input device such as a keyboard, a mouse, or a touch panel, or an output device such as a printer. It accepts user operations and outputs information to the user via the input / output I / F13.

The processor 14 is an arithmetic processing unit such as a CPU (Central Processing Unit), and executes each process according to the present embodiment by executing a program. Specifically, the processor 14 functions as an initial condition registration unit 141, a frame shape generation unit 142, and a simulation unit 143.

The initial condition registration unit 141 accepts a user's operation via the input / output I / F 13 and acquires initial conditions such as the outer shape of the building, the number of floors, and the usage. The outer shape is specified by, for example, the width, depth and height of a rectangular parallelepiped building. The outer shape of the building may be specified by a complicated shape other than a rectangular parallelepiped. The number of floors represents the number of floors of a building. The floor height of each floor may be calculated from the height of the building and the number of floors, or the floor height may be further acquired for each floor by having the user input the floor height. Further, the use is for a building, and is used as a load condition in the present embodiment. In addition, the usage may be acquired in association with each floor. The load per unit area is stored in advance in association with the intended use, and the load is used in the structural calculation.

The frame shape generation unit 142 automatically changes parameters such as span division, beam arrangement, brace arrangement, and cross section of columns and beams within the range of the acquired initial conditions, and generates a plurality of frame shapes. In this embodiment, it is described as designing an S structure, but it may be applied to an RC building to change the number and arrangement of reinforcing bars in addition to the cross sections of columns and beams. ..

The simulation unit 143 calculates the structure of the generated frame shape and solves the multi-objective optimization problem using at least one of the maximum interlayer deformation angle, steel frame weight, number of pieces, rigidity, eccentricity, uniformity, etc. as the objective function. Therefore, an approximate Pareto optimal solution is obtained for each generated frame shape. The plurality of data obtained in this way can be used as teacher data in machine learning.

The above components are connected via the bus 15. The frame data generation device 1 does not have to include some components such as communication I / F11 and input / output I / F13. Further, the frame data generation device 1 may be provided on a so-called cloud and output the result of the processing according to the present embodiment to another computer via a network (not shown) such as the Internet.

<Framework data generation process>
FIG. 2 is a processing flow diagram showing an example of the frame data generation processing according to the present embodiment. The initial condition registration unit 141 of the frame data generation device 1 receives input of initial conditions from the user via the input / output I / F 13 (FIG. 2: S1). In this step, the initial condition registration unit 141 acquires information and usage (load) that define the outer shape of the building. Information that defines the outline of a building includes width, depth, height, and number of floors. Applications include types such as offices, apartments, hotels, logistics / warehouses, etc. The storage device 12 can store the load in advance in association with the type of application and use it for structural calculation. In addition, the information that defines the outer shape of the building may include the roof shape such as a flat roof, a single-flow roof, and a gable roof, the height of the roof, the slope, and the presence or absence of hollow pillars.

FIG. 3 is a diagram showing an example of a screen for allowing the user to input initial conditions. In the example of FIG. 3, the width, depth, and height of a rectangular parallelepiped building, an input field for the number of floors, and an input field for the purpose of the building are included. It should be noted that a fixed load such as a roof, a reference floor, a parapet, a wall, and a fireproof coating may be further input for each floor.

FIG. 4 is a diagram showing an example of a load load that is predetermined in association with the application. The table shown in FIG. 4 includes the attributes of use and load. In addition, a standard load capacity is set according to the purpose of the building such as an office, a housing complex, a hotel, a distribution / warehouse, etc., and can be used for structural calculation. The load may be set separately for slabs / beams, frames (for long-term columns / beams), earthquakes (for earthquakes of columns / beams), and the like.

Further, the frame shape generation unit 142 determines a predetermined parameter and generates a frame shape (FIG. 2: S2). In this step, as predetermined parameters, span division, beam division, cross section of columns and beams, arrangement of braces, etc. are randomly determined, for example, within a predetermined shape and quantity. In the span division, for example, the number of spans is randomly changed by equal division. The beam split is arranged between the beams in a shape such as a one-way type (long side), a one-way type (short side), or a fish bone type. The cross sections of the columns and beams are randomly selected by storing the list of steel materials used as the materials for the columns and beams in the storage device 12 in advance. Further, columns and beams continuous in a predetermined direction may be grouped based on the above span, and steel materials having the same cross section may be assigned to the columns and beams of the same group.

FIG. 5 is a diagram showing an example of the generated frame shape. In the example of FIG. 5, it is divided into 4 spans in the X direction and 2 spans in the Y direction, and columns and girders are formed of steel frames. In S2 (FIG. 2), the number of spans can be set randomly, not limited to the example shown in FIG. In addition, for example, it may be set at random within the range of the allowable span length. Further, for example, a part of the span such as a common area may be fixed, and only the span of another part such as an office space may be randomly determined. Further, the frame shape as shown in FIG. 5 may be generated by a so-called BIM (Building Information Modeling) model.

Then, the simulation unit 143 performs a predetermined simulation on the generated frame shape, and the steel frame weight (material cost), the number of pieces of the member (workability), the interlayer deformation angle (seismic resistance), the rigidity, and the eccentricity. , Equality and other evaluation values are calculated (Fig. 2: S3). The process of this step will be described with reference to FIG.

FIG. 6 is a processing flow diagram showing an example of simulation processing. The simulation unit 143 performs stress analysis based on the generated frame shape (FIG. 6: S11). In this step, the degree of stress applied to columns and girders is calculated. That is, the simulation unit 143 calculates the degree of normal stress, the degree of shear stress, and the degree of bending stress applied to members such as columns and girders. Existing methods can be used to calculate the stress level.

Further, the simulation unit 143 determines whether the stress degree satisfies a predetermined constraint condition (FIG. 6: S12). In this step, for example, it is determined whether the stress degree is equal to or less than the allowable stress degree (verification ratio 1.0 or less). When it is determined that the stress degree does not satisfy the constraint condition (S12: NO), the process returns to the process of S2 (FIG. 2) via the connector “A”. Even if it is determined that the stress degree does not satisfy the constraint condition, the process may proceed to S13 and an operation may be performed to lower the evaluation value described later as a penalty.

When it is determined that the stress degree satisfies the constraint condition (S12: YES), the simulation unit 143 determines the maximum interlayer deformation angle with respect to a predetermined seismic load based on the frame shape generated in S2 (FIG. 2). Obtain (FIG. 6: S13). First, the interlayer deformation angle r of one floor is a value obtained by dividing the horizontal displacement of the floor at the time of an earthquake by the floor height, and is obtained by the following equation (1).
r = δ / h ・ ・ ・ (1)
Note that δ is the inter-story displacement and h is the floor height. The maximum interlayer deformation angle is obtained as the largest value among the interlayer deformation angles of each floor.

Then, the simulation unit 143 determines whether the maximum interlayer deformation angle satisfies a predetermined constraint condition (FIG. 6: S14). In this step, it is determined whether the maximum interlayer deformation angle is equal to or less than a predetermined reference. The predetermined standard may be a statutory standard (for example, 1/200). When it is determined that the maximum interlayer deformation angle does not satisfy the constraint condition (S14: NO), the process returns to the process of S2 (FIG. 2) via the connector “A”. Even if it is determined that the maximum interlayer deformation angle does not satisfy the constraint condition, the transition to S15 may be performed and an operation such as lowering the evaluation value described later as a penalty may be performed.

On the other hand, when it is determined that the maximum interlayer deformation angle satisfies the constraint condition (S14: YES), the simulation unit 143 calculates the steel frame weight, the number of pieces, and the like (FIG. 6: S15). The steel frame weight is the quantity (weight) of the members used in the building. The number of pieces is the number of members used in the building. In this step, the number of members and the number of members are integrated based on the generated frame shape, and the weight of the steel frame and the number of pieces are obtained. Further, the rigidity may be calculated for each floor in the X direction and the Y direction, and the minimum value in a plurality of floors may be obtained for each of the X direction and the Y direction. Further, the eccentricity rate indicating the magnitude of the difference between the center of gravity and the center of gravity may be calculated in the X and Y directions of each floor, and the maximum value in a plurality of floors may be obtained in each of the X and Y directions. .. Further, the value of uniformity representing the distribution of the load applied to the girder may be obtained. As the value of uniformity, the standard deviation of the stress applied to the girder may be averaged for each floor in each of the X direction and the Y direction, and the average value in a plurality of floors may be obtained for each of the X direction and the Y direction. .. For the stress, for example, the larger of the long-term loads applied to both ends of the girder is used. Further, the stress applied to the girder may be obtained for each sign. That is, girders with the same span, such as short girders, standard girders, and long girders, are given the same reference numerals. Then, the stress applied to each girder is obtained, and these standard deviations are obtained for each direction, each floor, and each girder with the same reference numeral. Then, the average of the standard deviations is calculated for each floor, and the average value of the entire building with a plurality of averages is calculated in the X direction and the Y direction, and used as the equality value.

Then, the simulation unit 143 outputs the calculated evaluation value in association with the frame shape (FIG. 6: S16). In this step, the interlayer deformation angle, the steel frame weight, the number of pieces, the rigidity, the eccentricity, and the uniformity are stored in the storage device 12 as evaluation values in association with the frame shape generated in S2 (FIG. 2).

FIG. 7 is a diagram showing an example of the output evaluation value. In S16 (FIG. 6), information such as an evaluation value is stored in a table having attributes as shown in FIG. 7. The table shown in FIG. 7 shows the load, the number of short spans in the X (width) direction, the number of short spans in the X direction, the number of short spans in the Y (depth) direction, the number of short spans in the Y direction, the number of standard spans in the X direction, and X. Standard span in the direction, standard number of spans in the Y direction, standard span in the Y direction, number of long spans in the X direction, long span in the X direction, number of long spans in the Y direction, long span in the Y direction, floor height, beam splitting , Steel weight, span deformation angle, number of pieces, rigidity (minimum value in X direction), rigidity (minimum value in Y direction), eccentricity (maximum value in X direction), eccentricity (maximum value in Y direction) , Equality (maximum value in the X direction), Equality (maximum value in the Y direction). In the load capacity field, the load capacity according to the application is registered. In the fields related to short span, standard span, and long span, a number is registered for each span classified into each size. In the floor height field, a value obtained by dividing the height of the building by the number of floors or the floor height input by the user is registered. The value of the steel frame weight calculated in S15 (FIG. 6) is registered in the steel frame weight field. The value of the maximum interlayer deformation angle calculated in S13 (FIG. 6) is registered in the interlayer deformation angle field. The value of the number of pieces calculated in S15 (FIG. 6) is registered in the field of the number of pieces. Each of the records registered in the table shown in FIG. 7 may be further associated with the frame shape as shown in FIG. 5, for example. In the rigidity field, the minimum values of the rigidity in the X direction and the Y direction calculated in S15 (FIG. 6) are registered. In the field of eccentricity, the maximum value of the eccentricity calculated in S15 (FIG. 6) in the X direction and the Y direction is registered. In the field of uniformity, the average value in the X direction and the Y direction calculated in S15 (FIG. 6) is registered.

After S16 (FIG. 6), the process returns to FIG. 2, and the frame shape generation unit 142 determines whether to end the generation of the frame shape (FIG. 2: S4). In this step, when the processes of S2 and S3 are repeated and, for example, a predetermined number of frame shapes are generated, the frame shape generation unit 142 determines that the process is completed. If it is determined that the process is not completed (S4: NO), the process returns to S2 and the process is repeated.

なお、Ｃは鉄骨重量、Ｒ_ｋはｋ階の層間変形角、Ｐはピース数、γ_ｊは制約条件ｊに対するペナルティ関数である。さらに、上記の目的関数のいずれかに代えて、又は上記の３つの目的関数に加えて、剛性率（Ｘ方向の最小値）、剛性率（Ｙ方向の最小値）、偏心率（Ｘ方向の最大値）、偏心率（Ｙ方向の最大値）、均等性（Ｘ方向の平均値）、均等性（Ｙ方向の平均値）のいずれかを目的関数としてもよい。パレート解集合の生成法は、例えばＳＰＥＡ２のような多目的遺伝的アルゴリズム等、既存の手法を用いることができる。すなわち、シミュレーション部１４３は、例えばＳ１６（図６）において図７のテーブルに登録された複数のレコードについて、支配しているレコードの数を求める。そして、支配されているレコードの適合度が、当該レコードを支配しているレコードを支配しているレコードの数を合計した値となるように、適合度を求める。したがって、レコードが非劣解の場合、適合度はゼロになる。このような処理を用いてアーカイブ個体群から探索個体群を選択し、探索を行う。また、シミュレーション部１４３は、得られたパレート解集合を、所定の出力テーブルに出力する。出力テーブルは、図７に示したテーブルと同様の属性を含むものであってもよい。 On the other hand, when it is determined that the processing is completed (S4: YES), the simulation unit 143 outputs the Pareto solution set (FIG. 2: S5). In this step, the simulation unit 143 uses the plurality of frame shapes generated in S2 (FIG. 2) and the evaluation values calculated in S3 (FIG. 2) as objective functions for the interlayer deformation angle, the steel frame weight, and the number of pieces. It is also possible to solve the multi-objective optimization problem. For example, the multi-objective optimization problem is given by the following equation (1).

C is the weight of the steel frame, R _k is the interlayer deformation angle of the k-th order, P is the number of pieces, and γ _j is a penalty function for the constraint condition j. Further, instead of any of the above objective functions, or in addition to the above three objective functions, the rigidity (minimum value in the X direction), the rigidity (minimum value in the Y direction), and the eccentricity (in the X direction). Any of (maximum value), eccentricity (maximum value in the Y direction), uniformity (average value in the X direction), and uniformity (average value in the Y direction) may be used as the objective function. As a method for generating a Pareto solution set, an existing method such as a multipurpose genetic algorithm such as SPEA2 can be used. That is, the simulation unit 143 obtains the number of dominant records for the plurality of records registered in the table of FIG. 7 in, for example, S16 (FIG. 6). Then, the goodness of fit is obtained so that the goodness of fit of the controlled record is the total value of the number of records controlling the record that controls the record. Therefore, if the record is non-inferior, the goodness of fit is zero. A search population is selected from the archive population using such processing, and the search is performed. Further, the simulation unit 143 outputs the obtained Pareto solution set to a predetermined output table. The output table may include the same attributes as the table shown in FIG.

FIG. 8 is a diagram showing an example of the Pareto solution set three-dimensionally. In the example of FIG. 8, each axis of the interlayer deformation angle (Drift Angle), the steel frame weight (Weight), and the number of pieces (Piece) is included, and each of the Pareto solutions is plotted. In S5 (FIG. 2), such a graph may be output and visualized.

This completes the frame data generation process according to this embodiment.

<Effect>
As described above, according to the frame data generation process, an approximate Pareto optimal solution can also be obtained. However, even in the Pareto optimal solution, there are frame shapes that are preferable for experienced designers and frame shapes that are not. For example, if the designer assigns superiority or inferiority to a plurality of Pareto optimal solutions obtained and uses this as teacher data, it can be used for machine learning. Further, according to the present embodiment, it is possible to generate a plurality of frame shapes so as to cover various span divisions that can be adopted for the volume of the building. Therefore, a sufficient pattern and amount of teacher data can be generated.

<Others>
The embodiment is an example, and the present invention is not limited to the above-described configuration. The object of the present invention also includes a computer program that executes the above-described processing and a computer-readable recording medium that records the program. The recording medium on which the program is recorded can perform the above-mentioned processing by causing the computer to execute the program.

A computer-readable recording medium is a recording medium that can be read from a computer by accumulating information such as data and programs by electrical, magnetic, optical, mechanical, or chemical action. Among such recording media, those that can be removed from the computer include flexible disks, magneto-optical disks, optical disks, magnetic tapes, memory cards, and the like. Further, as the recording medium fixed to the computer, there are HDD, SSD (Solid State Drive), ROM and the like.

1: Frame data generator 11: Communication I / F
12: Storage device 13: Input / output I / F
14: Processor 141: Initial condition registration unit 142: Frame current generation unit 143: Simulation unit

Claims (8)

Initial condition acquisition part to acquire the plane shape, height and number of floors of the building,
Within the range of the acquired plane shape and the number of floors, the frame shape generation unit that changes the span division of the main frame including columns and girders to generate a plurality of frame shapes,
For each of the generated frame shapes, at least 3 of the evaluation values that are used as a guideline for earthquake resistance, the evaluation values that are used as a guideline for workability, the evaluation values that are used as a guideline for material costs, and the evaluation values that are used as a guideline for building balance. A simulation unit that calculates one and obtains a Pareto solution set using the calculated evaluation value as an objective function and stores it in a storage device.
A frame data generator equipped with. For each of the generated frame shapes, the simulation unit determines the evaluation value as a guideline for earthquake resistance, the evaluation value as a guideline for workability, the evaluation value as a guideline for material cost, and the balance of the building. The frame data generator according to claim 1, which calculates an evaluation value as a guide. The simulation unit calculates the stress degree generated in the cross section of the main frame using a standard load stored in the storage device in advance in association with the application for each of the frame shapes, and also determines the predetermined allowable stress degree. The frame data generation device according to claim 1 or 2, wherein the Pareto solution set is obtained by using a frame shape that does not exceed the above. The evaluation value as a guideline for seismic resistance is the maximum interlayer deformation angle with respect to a predetermined seismic load.
The evaluation value that serves as a guide for workability is the number of pieces in the main frame.
The evaluation value that serves as a guide for the material cost is the weight of the steel frame.
The frame data generation device according to any one of claims 1 to 3, wherein the evaluation value as a guideline for the balance of the building is an evaluation value indicating the rigidity, the eccentricity, or the equality of stress with respect to the load. The frame data generation device according to claim 4, wherein the simulation unit obtains the Pareto solution set using a frame shape in which the maximum interlayer deformation angle does not exceed a predetermined reference. The initial condition acquisition unit further acquires the use of the building,
The simulation unit reads out a standard load stored in the storage device in advance in association with the acquired application, and uses the read out load to calculate the maximum interlayer deformation angle with respect to the seismic load. The frame data generator according to claim 4 or 5, which is cited. Steps to get the plan shape, height and number of floors of a building,
Within the range of the acquired plan shape and the number of floors, the span division of the main frame including columns and girders is changed to generate a plurality of frame shapes.
For each of the generated frame shapes, at least 3 of the evaluation values that are used as a guideline for earthquake resistance, the evaluation values that are used as a guideline for workability, the evaluation values that are used as a guideline for material costs, and the evaluation values that are used as a guideline for building balance. The step of calculating the Pareto solution set using the calculated evaluation value as the objective function and storing it in the storage device.
A method of generating framed data that is executed by a computer. Steps to get the plan shape, height and number of floors of a building,
Within the range of the acquired plan shape and the number of floors, the span division of the main frame including columns and girders is changed to generate a plurality of frame shapes.
For each of the generated frame shapes, at least 3 of the evaluation values that are used as a guideline for earthquake resistance, the evaluation values that are used as a guideline for workability, the evaluation values that are used as a guideline for material costs, and the evaluation values that are used as a guideline for building balance. The step of calculating the Pareto solution set using the calculated evaluation value as the objective function and storing it in the storage device.
A frame data generation program that causes a computer to execute.

Images