CN116150839B - Steel structure factory building component arrangement design method and system based on field knowledge enhancement - Google Patents
Steel structure factory building component arrangement design method and system based on field knowledge enhancement Download PDFInfo
- Publication number
- CN116150839B CN116150839B CN202211627394.4A CN202211627394A CN116150839B CN 116150839 B CN116150839 B CN 116150839B CN 202211627394 A CN202211627394 A CN 202211627394A CN 116150839 B CN116150839 B CN 116150839B
- Authority
- CN
- China
- Prior art keywords
- plant
- design
- roof
- arrangement
- plane
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/13—Architectural design, e.g. computer-aided architectural design [CAAD] related to design of buildings, bridges, landscapes, production plants or roads
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/80—Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Geometry (AREA)
- Evolutionary Computation (AREA)
- Computer Hardware Design (AREA)
- Software Systems (AREA)
- Medical Informatics (AREA)
- Multimedia (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Optimization (AREA)
- Structural Engineering (AREA)
- Pure & Applied Mathematics (AREA)
- Civil Engineering (AREA)
- Mathematical Analysis (AREA)
- Architecture (AREA)
- Computational Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- General Health & Medical Sciences (AREA)
- Instructional Devices (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a method and a system for arranging and designing steel structure plant components based on field knowledge enhancement, which are characterized in that target plant separation characteristic data and plane design treaty characteristics are input into a plant plane component arranging and generating model with the field knowledge enhancement to obtain plant plane structure component arranging and cross section characteristic data, and the plant plane structure component arranging and cross section design data are obtained based on the mapping relation between a steel structure component size diagram and the component arranging and characteristic data; inputting the layout and section characteristic data of the plant plane structural members and the roof design treaty characteristic data into a pre-trained plant roof member layout generation model to obtain the layout and section characteristic data of the plant roof structural members, and obtaining the layout and section design data of the plant roof structural members based on the mapping relation between the steel structural member size diagram and the member layout characteristic data; and integrating the data to obtain a preliminary design result and a model. The invention solves the problem of low efficiency of the manually designed steel structure factory building.
Description
Technical Field
The invention relates to the technical field of artificial intelligence building design, in particular to a method and a system for arranging and designing steel structure factory building components based on field knowledge enhancement.
Background
In the scheme design stage of the steel structure factory building, in order to ensure the safety and economy of the structural design scheme of the steel structure factory building, quick and reasonable positioning and size design of structural members of the steel structure factory building are required on the basis of the structural arrangement scheme of the steel structure factory building and under the constraint of structural design conditions of the steel structure factory building.
However, the positioning and dimension design scheme of the traditional steel structure factory building structural member mainly depends on personal experience of structural engineers, and the manual design mode is time-consuming and labor-consuming, and has low efficiency in the design process and difficult inheritance of the existing design experience. At present, a method for positioning and designing the size of a structural member of an intelligent steel structure factory building is not available.
Disclosure of Invention
The invention provides a method and a system for arranging and designing steel structure factory building components based on field knowledge enhancement, which are used for solving the problems that the time and the labor are consumed for manually designing a steel structure factory building, the efficiency of the design process is low, the existing design experience is difficult to inherit, and the design scheme often breaks against the basic common sense of structural engineers when the intelligent steel structure factory building is designed, so that the high-efficiency, accurate positioning and dimensional design of the steel structure factory building structural components are realized.
The invention provides a field knowledge enhancement-based steel structure factory building component arrangement design method, which comprises the following steps:
acquiring a conceptual design drawing and a design condition text of a steel structure factory building, extracting factory building separation characteristic data from the conceptual design drawing, and extracting plane design condition characteristic data and roof design condition characteristic data from the design condition text;
fusing the plant separation characteristic data and the plane design treaty characteristic data into plant plane layout characteristic tensors, inputting the plant plane layout characteristic tensors into a plant plane component layout generating model with enhanced field knowledge which is trained in advance to obtain layout and section characteristic data of plant plane structure components, and extracting layout and section design data of the plant plane structure components from the layout and section characteristic data of the plant plane structure components;
fusing the layout and section characteristic data of the plant plane structural members with the roof design condition characteristic data to form a plant roof layout characteristic tensor, inputting the plant roof layout characteristic tensor into a plant roof member layout generating model with enhanced field knowledge which is trained in advance, obtaining plant roof member layout and section characteristic data, and extracting the layout and section design data of the plant roof structural members from the plant roof member layout characteristic data;
And matching and integrating the layout and section design data of the plant plane structural members and the layout and section design data of the plant roof structural members to obtain a preliminary design result of the steel frame structural design of the steel structure plant.
According to the steel structure plant member arrangement design method based on field knowledge enhancement, the plant separation characteristic data and the plane design treaty characteristic data are fused into a plant plane arrangement characteristic tensor, and the method specifically comprises the following steps:
the plant separation characteristic data constrains the arrangement range of key structural members of a plant plane;
the plane design treaty characteristic data constrains the structural form of a factory building, anti-seismic design information, the control stress ratio of main components, crane load and mechanical property information of plane key components;
characterizing the plant separation characteristic data as a second-order matrix to obtain a plant separation characteristic matrix, wherein the positions in the plant separation characteristic matrix, where key structural members of a plant plane can be arranged, are marked as 1, and the rest positions are marked as 0;
characterizing the planar design treaty feature data as zero-order scalar, and copying the zero-order scalar as a homomorphic second-order matrix with the factory building separation feature matrix to obtain a planar design condition feature matrix;
And respectively carrying out normalization processing on the plant separation characteristic matrix and the plane design condition characteristic matrix, and stacking normalization processing results of the plant separation characteristic matrix and the plane design condition characteristic matrix to obtain a plant plane layout characteristic tensor.
According to the field knowledge enhancement-based steel structure plant member arrangement design method provided by the invention, the plant planar structural member arrangement and section design data are extracted from the plant planar structural member arrangement and section characteristic data, and the method specifically comprises the following steps:
the arrangement and section characteristic data of the plant plane structural members represent the positioning and section selection matrix of the target steel structure plant plane structural members;
wherein, the non-zero element position in the plane structural member positioning and selecting matrix refers to the existence of a member, the value on the non-zero element position is filled with the undetermined code, and different codes represent steel structural members with different sizes;
and decoding the arrangement and section characteristic data of the plant plane structural members according to the preset mapping relation between the steel structure member size diagram and the non-zero elements in the plane structural member positioning and selecting matrix to obtain the arrangement and section design data of the plant plane structural members, thereby obtaining the design sizes corresponding to various types of members in the target plant plane.
According to the method for arranging and designing the steel structure plant components based on field knowledge enhancement, which is provided by the invention, the arrangement and section characteristic data of the plant planar structural components and the roof design treaty characteristic data are fused into a plant roof arrangement characteristic tensor, and the method specifically comprises the following steps:
the arrangement and section characteristic data of the plant plane structural members represent the target steel structure plant plane structural member positioning and section selection matrix, and plane positioning and section selection information is provided for plant roof member arrangement;
the roof design treaty characteristic data constrains the structural form of a factory building, earthquake-resistant design information, roof gradient, roof geometric information, roof load and mechanical property information of roof key components;
the arrangement and section feature data of the plant plane structure members are second-order matrixes, the roof design condition feature data are characterized as zero-order scalar, and the zero-order scalar is copied as homomorphic second-order matrixes with the arrangement and section feature data of the plant plane structure members, so that a roof design condition feature matrix is obtained;
and respectively carrying out normalization processing on the arrangement and section characteristic data of the plant plane structural members and the roof design condition characteristic matrix, and stacking the normalization processing results of the arrangement and section characteristic data of the plant plane structural members and the roof design condition characteristic matrix to obtain a plant roof arrangement characteristic tensor.
According to the field knowledge enhancement-based steel structure plant member arrangement design method provided by the invention, the arrangement and section design data of plant roof structural members are extracted from the plant roof member arrangement and section characteristic data, and the method specifically comprises the following steps:
the plant roof component arrangement and section characteristic data represent the target steel structure plant roof structural component positioning and section selection matrix;
wherein, the non-zero element position in the roof structural member positioning and selecting matrix refers to the existence of a member, the value on the non-zero element position is filled with undetermined codes, and different codes represent steel structural members with different sizes;
and decoding the layout and section characteristic data of the plant roof components according to the mapping relation between the preset steel structure component size diagram and the non-zero elements in the roof structure component positioning and selecting matrix to obtain layout and section design data of the plant roof structure components, thereby obtaining the design sizes corresponding to various types of components in the target plant roof.
According to the method for designing the layout of the steel structure plant components based on the field knowledge enhancement, the construction process of the field knowledge enhancement plant plane component layout generation model and the field knowledge enhancement plant roof component layout generation model comprises the following steps:
The plant plane component arrangement generating model with the enhanced domain knowledge and the plant roof component arrangement generating model with the enhanced domain knowledge are composed of a generating module and an evaluating module with the enhanced domain knowledge.
The generating module is obtained by connecting a convolution neural network and a deconvolution neural network, the convolution neural network is used for extracting high-dimensional characteristics in a plant plane arrangement characteristic tensor and a plant roof arrangement characteristic tensor, and the deconvolution neural network is used for upsampling the high-dimensional characteristics output by the convolution neural network to obtain arrangement and section characteristic data of plant plane structure members and arrangement and section characteristic data of plant roof structure members;
the evaluation module for enhancing the domain knowledge is used for evaluating the generating module to obtain the arrangement and section characteristic data of the plant plane structural members and the arrangement and section characteristic data of the plant roof members based on the comprehensive design loss, wherein the comprehensive design loss is formed by weighting the domain design knowledge loss corresponding to various types of members by image similarity loss, the domain design knowledge corresponding to the various types of members comprises: plane domain design knowledge and roof domain design knowledge;
The image similarity loss in the comprehensive design loss is determined based on an image loss function, the image similarity loss is determined based on differences between the image loss function and plant plane member arrangement ideal feature data and plant plane structure member arrangement and section feature data for the field knowledge enhanced plant plane member arrangement generating model, and the image similarity loss is determined based on differences between the image loss function and plant plane member arrangement ideal feature data and plant plane member arrangement and section feature data for the field knowledge enhanced plant plane member arrangement generating model, wherein the plant plane member arrangement ideal feature data is obtained based on an ideal member size diagram of a steel structure plant structure;
constructing an optimizer for optimizing parameters of the neural network structure to be trained by taking the minimum comprehensive design loss as a target in a training stage;
and training and evaluating the neural network structure to be trained by utilizing a steel structure factory building structure drawing sample, an evaluation module and an optimizer to obtain the factory building plane member arrangement generating model with enhanced domain knowledge and the factory building roof member arrangement generating model with enhanced domain knowledge.
According to the method for arranging and designing the steel structure factory building components based on the field knowledge enhancement, the field design knowledge corresponding to various types of components comprises the following steps: plane domain design knowledge and roof domain design knowledge;
the various components comprise: a planar member and a roofing member; the planar member includes: rigid support among steel frame columns, wind-resistant columns and columns; the roofing component includes: steel frame beam, roofing purlin and roofing flexible support.
For planar domain design knowledge, the domain knowledge design criteria for the steel frame column include: the end span column spacing arrangement should be no greater than the intermediate span column spacing arrangement; the middle cross-column spacing arrangement should be as uniform as possible;
for roofing component design knowledge, the domain knowledge design criteria for roofing purlins include: the purlines are uniformly arranged.
The invention also provides a steel structure factory building component arrangement design system based on field knowledge enhancement, which comprises:
the data acquisition module is used for acquiring a conceptual design drawing and a design condition text of the steel structure factory building, extracting factory building separation characteristic data from the conceptual design drawing, and extracting plane design condition characteristic data and roof design condition characteristic data from the design condition text;
The factory building planar structural member design module is used for fusing the factory building separation characteristic data and the planar design treaty characteristic data into factory building planar arrangement characteristic tensors, inputting the factory building planar arrangement characteristic tensors into a factory building planar member arrangement generating model with enhanced field knowledge which is trained in advance, obtaining arrangement and section characteristic data of factory building planar structural members, and extracting arrangement and section design data of the factory building planar structural members from the arrangement and section characteristic data of the factory building planar structural members;
the factory building roof structural member design module is used for fusing the layout and section characteristic data of the factory building planar structural members with the roof design condition characteristic data to form factory building roof layout characteristic tensors, inputting the factory building roof layout characteristic tensors into a factory building roof member layout generating model with enhanced field knowledge which is trained in advance to obtain factory building roof member layout and section characteristic data, and extracting layout and section design data of the factory building roof structural members from the roof member layout characteristic data;
and the integration module is used for matching and integrating the layout and section design data of the plant plane structural members and the layout and section design data of the plant roof structural members to obtain a preliminary design result of the steel frame structural design of the steel structure plant.
The invention also provides electronic equipment, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor realizes the steel structure factory building component arrangement design method based on the domain knowledge enhancement when executing the program.
The present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements a method of domain knowledge based enhanced steel structure plant member layout design as described in any of the above.
According to the method and the system for arranging and designing the steel structure plant components based on the field knowledge enhancement, the arrangement and design data of the plane structural components, the arrangement and the section design data of the plant roof structural components and the arrangement and the section design data of the plant roof structural components are matched and integrated by constructing the field knowledge enhancement plant plane component arrangement generation model and the field knowledge enhancement plant roof component arrangement generation model, so that the preliminary design result and the model of the steel frame structural design of the steel structure plant are obtained. The intelligent design of the quick and reliable steel structure factory building is realized, and the stability of the design result is higher because the intelligent design does not depend on manual experience.
Drawings
In order to more clearly illustrate the invention or the technical solutions of the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described, and it is obvious that the drawings in the description below are some embodiments of the invention, and other drawings can be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow diagram of a method for designing layout of structural plant components of a steel structure based on domain knowledge enhancement;
FIG. 2 is a schematic diagram of a domain knowledge enhanced neural network planar design model architecture provided by the present invention;
FIG. 3 is a schematic diagram of a domain knowledge enhanced neural network roof design model architecture provided by the present invention;
FIG. 4 is a schematic diagram of a system module connection for designing the layout of steel structure plant components based on domain knowledge enhancement;
fig. 5 is a schematic structural diagram of an electronic device provided by the present invention.
Reference numerals:
110: a data acquisition module; 120: plant plane structural member design modules; 130: building roof structural member design modules; 140: an integration module;
510: a processor; 520: a communication interface; 530: a memory; 540: a communication bus.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The following describes a method for designing arrangement of steel structure factory building components based on field knowledge enhancement with reference to fig. 1, which comprises the following steps:
s100, acquiring a conceptual design drawing and a design condition text of a steel structure factory building, extracting factory building separation characteristic data from the conceptual design drawing, and extracting plane design condition characteristic data and roof design condition characteristic data from the design condition text;
wherein the plant separation characteristic data constrains the arrangement range of plant plane key structural members (such as columns, supports and the like). It will be appreciated that the planar key structural members can only be arranged within the limit of the arrangement;
The plane design treaty characteristic data constrains the structural form of the factory building, the earthquake-resistant design information, the control stress ratio of main components, the crane load and the mechanical property information of plane key components; the summary of the planar features employed in this example is shown in table 1.
TABLE 1
The calculation formula of the earthquake influence coefficient alpha is as follows:
in the above, alpha max For maximum value of seismic influence coefficient, T g The characteristic period(s) of the field is represented by T, and the characteristic period(s) of the field is represented by a basic self-vibration period of the structure;
the calculation formula of the structural basic self-oscillation period T is as follows:
in the above, h s The structure height, b is the structure width.
As shown below, table 2 illustrates the maximum value of the seismic impact coefficient α max Table 3 illustrates the manner in which the site characteristic period(s) is determined.
TABLE 2
TABLE 3 Table 3
The roof design treaty characteristic data constrains the structural form of the factory building, the earthquake-proof design information, the roof gradient, the roof geometric information, the roof load and the mechanical property information of the roof key components. The summary of the planar features employed in this example is shown in table 4.
TABLE 4 Table 4
S200, fusing the plant separation characteristic data and the plane design treaty characteristic data into plant plane layout characteristic tensors, inputting the plant plane layout characteristic tensors into a plant plane component layout generating model with enhanced field knowledge, which is trained in advance, obtaining layout and section characteristic data of plant plane structural components, and extracting layout and section design data of the plant plane structural components from the layout and section characteristic data of the plant plane structural components;
Wherein, in order to obtain the plant floor plan characteristic tensor. Firstly, characterizing the plant separation characteristic data as a second-order matrix to obtain a plant separation characteristic matrix; the position in the plant separation characteristic matrix, where the plant plane key structural members can be arranged, is marked as 1, and the rest positions are marked as 0; then, the planar design treaty feature data is required to be characterized as a zero-order scalar, and the zero-order scalar is copied as a homomorphic second-order matrix of the plant separation feature matrix, so that a planar design condition feature matrix is obtained; on the basis, the plant separation characteristic matrix and the plane design condition characteristic matrix are respectively normalized, and the normalization processing results of the plant separation characteristic matrix and the plane design condition characteristic matrix are stacked to obtain a plant plane layout characteristic tensor.
Wherein the model architecture of one embodiment of the domain knowledge enhanced plant planar member arrangement generation model is shown in fig. 2. The plant plane member arrangement generating model with enhanced domain knowledge consists of a generating module and an evaluating module with enhanced domain knowledge.
The generating module is obtained by connecting a convolution neural network and a deconvolution neural network; the convolutional neural network is used for extracting high-dimensional characteristics in the plant plane layout characteristic tensor. The embodiment performs abstraction and feature extraction on the input feature tensor through the convolutional neural network, so as to determine high-dimensional features related to the positions and the sizes of the key components of the steel structure plane, for example, the high-dimensional features of 1×1×4000 are extracted from the input feature tensor of 512×512×5; the deconvolution neural network is used for upsampling the high-dimensional characteristics output by the convolution neural network to obtain the arrangement and section characteristic data of the plant plane structural members; for example, a 512×512×1 design feature tensor is generated based on the aforementioned 1×1×4000 high-dimensional features.
The evaluation module with enhanced domain knowledge is used for evaluating the generating module based on comprehensive design loss to obtain the arrangement of the plant plane structural members and the advantages and disadvantages of the section characteristic data; the comprehensive design loss is formed by weighting the image similarity loss and the field design knowledge loss corresponding to each type of components of the steel structure factory building plane;
wherein the image similarity loss in the integrated design loss is determined based on an image loss function. Generating a model for the plant plane component arrangement with enhanced domain knowledge, wherein the image similarity loss is determined based on the difference between an image loss function and ideal characteristic data of the plant plane component arrangement and the arrangement and section characteristic data of the plant plane structural component; the ideal characteristic data of the plant plane component arrangement is obtained based on an ideal component size diagram of a steel structure plant structure;
constructing an optimizer for optimizing parameters of the neural network structure to be trained by taking the minimum comprehensive design loss as a target in a training stage; the comprehensive design loss in the dimensional design model is calculated by the following formula:
Loss final ＝λ pix ×Loss pix,weighted +λ k,1 ×Loss k,1 +λ k,2 ×Loss k,2
in the above, loss pix Loss of image similarity, loss of Loss k,1 And Loss of k,2 The specific calculation method is described below for the loss of domain knowledge corresponding to each type of components of the steel structure factory building plane; lambda (lambda) pix ，λ k,1 And lambda is k,2 The weight of the image similarity loss and the domain knowledge of the plane members of the various steel structure factory buildings are respectively; the weight coefficients used in this example are shown in table 5.
TABLE 5
The end span column spacing arrangement of the domain knowledge design rule of the steel frame column is not more than the corresponding loss calculation formula of the middle span column spacing arrangement, and the method is specifically as follows:
Loss k,1 ＝Loss z1,l +Loss z1,r
dis lcs,i ＝ED(col side,i ,col side,i+1 )
dis lcm,i ＝ED(col mid,i ,col mid,i+1 )
col side,i ＝Col_location(Pred plane )
col mid,i ＝Col_location(Pred plane )
in the Loss k,1 Representing planar domain design knowledge, the domain knowledge design criteria of the steel frame columns, namely, the end span column spacing arrangement should not be larger than the Loss function of the middle span column spacing arrangement, loss z1,l Loss function representing that "end span column spacing arrangement should be no greater than intermediate span column spacing arrangement" for left steel frame column perpendicular to span direction, loss z1,r The "end span column pitch arrangement" representing the right steel frame column perpendicular to the span direction should be no greater than the loss function of the intermediate span column pitch arrangement "; middle dis lcs,i Representing the steel frame column pitch, dis, of the ith side span of the left steel frame column perpendicular to the span direction lcm,i Representation ofThe steel frame column distance of the middle span of the left steel frame column vertical to the span direction, max represents a maximum function, and n represents that the left steel frame column vertical to the span direction has n side spans; where ED represents the Euclidean distance function, col side,i Representing coordinates corresponding to an ith steel frame column of the side span; chinese col mid,i Representing the coordinate corresponding to the ith steel frame column in the middle span; pred in the formula plane Representing the arrangement and section characteristic data of the plant plane structural members, and Col_location represents the function of positioning steel frame columns in the arrangement and section characteristic data of the plant plane structural members.
The corresponding loss calculation formula of the domain knowledge design rule 'the middle cross column distance arrangement should be as uniform as possible' of the steel frame column is as follows:
Loss k,2 ＝Loss z2,l +Loss z2,r
in the Loss k,2 Representing planar domain design knowledge, the domain knowledge design criteria of the steel frame columns are Loss functions of which the intermediate column spacing arrangement should be as uniform as possible z2,l Loss function representing "middle column pitch arrangement should be as uniform as possible" for left steel frame column perpendicular to span direction, loss z2,r A loss function representing that the "middle column pitch arrangement of the right steel frame column perpendicular to the span direction should be as uniform as possible"; the present embodiment provides a Loss of z2,l In the formula, mean represents a mean function, m represents the number of middle columns, max represents a maximum value, min represents a minimum value, dis lcm,i The acquisition method of (a) is as described above.
And training and evaluating the neural network structure to be trained by utilizing a steel structure factory building structure drawing sample, and obtaining the factory building plane member arrangement generating model with the enhanced domain knowledge.
S300, fusing the layout and section characteristic data of the plant plane structural members with the roof design condition characteristic data to form a plant roof layout characteristic tensor, inputting the plant roof layout characteristic tensor into a plant roof member layout generation model with enhanced field knowledge which is trained in advance to obtain plant roof member layout and section characteristic data, and extracting the layout and section design data of the plant roof structural members from the plant roof member layout characteristic data;
wherein, in order to obtain the plant roof layout characteristic tensor, the layout and section characteristic data of plant plane structural members and the roof design treaty characteristic data are required to be utilized. The layout and section characteristic data of the plant plane structural members show the target steel structure plant plane structural member positioning and section selection matrix, and plane positioning and section selection information can be provided for layout of plant roof members (such as purlines, braces) and the like. The roof design treaty characteristic data constrains the structural form of a factory building, earthquake-resistant design information, roof gradient, roof geometric information, roof load and mechanical property information of roof key components; the arrangement and section feature data of the plant plane structure members are second-order matrixes, the roof design condition feature data are characterized as zero-order scalar, and the zero-order scalar is copied as homomorphic second-order matrixes with the arrangement and section feature data of the plant plane structure members, so that a roof design condition feature matrix is obtained; and respectively carrying out normalization processing on the arrangement and section characteristic data of the plant plane structural members and the roof design condition characteristic matrix, and stacking the normalization processing results of the arrangement and section characteristic data of the plant plane structural members and the roof design condition characteristic matrix to obtain a plant roof arrangement characteristic tensor.
Wherein the model architecture of one embodiment of the domain knowledge enhanced plant roofing component arrangement generative model is shown in fig. 3. The plant roof component arrangement generating model with enhanced domain knowledge consists of a generating module and an evaluating module with enhanced domain knowledge.
The generating module is obtained by connecting a convolution neural network and a deconvolution neural network; the convolutional neural network is used for extracting high-dimensional characteristics in the plant roof layout characteristic tensor. In the embodiment, the input characteristic tensor is abstracted and extracted through the convolutional neural network, so that high-dimensional characteristics related to the positions and the sizes of key components of the steel structure roof are determined, for example, the high-dimensional characteristics of 1 multiplied by 4000 are extracted from the input characteristic tensor of 512 multiplied by 5; the deconvolution neural network is used for upsampling the high-dimensional characteristics output by the convolution neural network to obtain plant roof component arrangement and section characteristic data; for example, a 512×512×1 design feature tensor is generated based on the aforementioned 1×1×4000 high-dimensional features.
The field knowledge enhanced evaluation module is used for evaluating the generating module based on comprehensive design loss to obtain the advantages and disadvantages of the arrangement of the plant roof components and the section characteristic data; the comprehensive design loss is formed by weighting the image similarity loss and the field design knowledge loss corresponding to each type of components of the roof of the steel structure factory building;
Wherein the image similarity loss in the integrated design loss is determined based on an image loss function. Generating a model for the plant roof component arrangement with enhanced domain knowledge, wherein the image similarity loss is determined based on the difference between the image loss function and ideal characteristic data of the plant roof component arrangement and section characteristic data; the ideal characteristic data of the plant roof component arrangement are obtained based on an ideal component size diagram of a steel structure plant structure;
constructing an optimizer for optimizing parameters of the neural network structure to be trained by taking the minimum comprehensive design loss as a target in a training stage; the comprehensive design loss in the dimensional design model is calculated by the following formula:
Loss final ＝λ pix ×Loss pix,weighted +λ k,3 ×Loss k,3
in the above, loss pix Loss of image similarity, loss of Loss k,3 All kinds of roof of factory building with steel structureThe domain knowledge loss corresponding to the model component is calculated by a specific calculation method as described below; lambda (lambda) pix ，λ k,3 The weight of the image similarity loss and the domain knowledge of the roof components of various steel structure plants are respectively calculated; the weight coefficients used in this example are shown in table 6.
TABLE 6
The loss calculation formula corresponding to the field knowledge design rule of the roofing purlines, namely purline arrangement should be uniform as much as possible, is specifically as follows:
dis purline,i ＝ED(pur i ,pur i+1 )
pur i ＝Pur_location(Pred roof )
In the Loss k,3 Knowledge of roofing component design, loss function, dis, of roofing purlin domain knowledge design criteria "purlin arrangement should be as uniform as possible purline,i Representing the distance between the ith purlines of the row of roof purlines parallel to the span direction; where ED represents a Euclidean distance function, pur i Representing the coordinates corresponding to the ith purline of the roof; pred in roof Representing the plant roof component arrangement and section characteristic data, and Pur_location represents a function of locating purlines in the plant roof component arrangement and section characteristic data.
And training and evaluating the neural network structure to be trained by utilizing a steel structure factory building structure drawing sample, and obtaining the factory building roof component arrangement generating model with the enhanced domain knowledge.
S400, matching and integrating the layout and section design data of the plant plane structural members and the layout and section design data of the plant roof structural members to obtain a preliminary design result of the steel frame structural design of the steel structure plant.
After the preliminary design model of the steel structure factory building to be designed is obtained, computer-aided deep design can be further performed based on interaction of engineers and structural design software (such as PKPM, YJK, ETABS and the like), and auxiliary drawing of a construction drawing is completed.
In the invention, the structural layout and design conditions of the steel structure factory building are collected first. Thereby extracting the arrangement position and the size of an ideal component from the structural layout of the steel structure factory building, and constructing a training data set, a verification data set and a test data set for training the domain knowledge-enhanced neural network plane (roof) design model;
in the training stage, the training data set is utilized to supervise and learn the neural network plane (roof) design model with enhanced domain knowledge, so that the training data set is used for mastering the design experience, the design specification and the domain common knowledge of engineers; in the training stage, performing super-parameter tuning by using the verification data set to select a model with the best performance on the verification set for evaluation;
in the evaluation stage, performance evaluation is carried out on the neural network plane (roof) design model with enhanced domain knowledge obtained in the training stage by utilizing the test data set, and a network architecture with qualified evaluation can be put into application;
in the application stage, firstly, plant separation characteristic data and plane (roof) design treatise characteristics are obtained according to a conceptual design diagram and design conditions of a steel structure plant to be designed. Then, inputting target plant separation characteristic data and plane design treaty characteristics into a plant plane component layout generating model with enhanced field knowledge which is trained in advance, and obtaining layout and section characteristic data of plant plane structural components; inputting the layout and section characteristic data of the plant plane structural members and the roof design treaty characteristic data into a plant roof member layout generation model which is trained in advance, so as to obtain plant roof member layout and section characteristic data;
And in the integration stage, integrating the arrangement and section characteristic data of the plant plane structural members, the arrangement and section characteristic data of the plant roof members and the mapping relation between the pre-stored steel structural member dimension diagram and the arrangement characteristic data of the members to obtain the preliminary design result and the model of the steel frame structural design of the steel structural plant.
Referring to fig. 4, the invention also discloses a steel structure factory building component arrangement design system based on field knowledge enhancement, the system comprises:
the data acquisition module is used for acquiring a conceptual design drawing and a design condition text of the steel structure factory building, extracting factory building separation characteristic data from the conceptual design drawing, and extracting plane design condition characteristic data and roof design condition characteristic data from the design condition text;
the factory building planar structural member design module is used for fusing the factory building separation characteristic data and the planar design treaty characteristic data into factory building planar arrangement characteristic tensors, inputting the factory building planar arrangement characteristic tensors into a factory building planar member arrangement generating model with enhanced field knowledge which is trained in advance, obtaining arrangement and section characteristic data of factory building planar structural members, and extracting arrangement and section design data of the factory building planar structural members from the arrangement and section characteristic data of the factory building planar structural members;
The factory building roof structural member design module is used for fusing the layout and section characteristic data of the factory building planar structural members with the roof design condition characteristic data to form factory building roof layout characteristic tensors, inputting the factory building roof layout characteristic tensors into a factory building roof member layout generating model with enhanced field knowledge which is trained in advance to obtain factory building roof member layout and section characteristic data, and extracting layout and section design data of the factory building roof structural members from the roof member layout characteristic data;
and the integration module is used for matching and integrating the layout and section design data of the plant plane structural members and the layout and section design data of the plant roof structural members to obtain a preliminary design result of the steel frame structural design of the steel structure plant.
According to the field knowledge enhancement-based steel structure plant member arrangement design system, the field knowledge enhancement plant plane member arrangement generation model and the field knowledge enhancement plant roof member arrangement generation model are constructed, the arrangement design data of the plane structural members, the arrangement and section design data of the plant roof structural members are obtained, the arrangement and section design data of the plant plane structural members, the arrangement and section design data of the plant roof structural members are matched and integrated, and the steel frame structural design primary design result and model of the steel structure plant are obtained. The intelligent design of the quick and reliable steel structure factory building is realized, and the stability of the design result is higher because the intelligent design does not depend on manual experience.
Fig. 5 illustrates a physical schematic diagram of an electronic device, as shown in fig. 5, which may include: processor 510, communication interface (Communications Interface) 520, memory 530, and communication bus 540, wherein processor 510, communication interface 520, memory 530 complete communication with each other through communication bus 540. Processor 510 may invoke logic instructions in memory 530 to perform a method of domain knowledge based enhanced steel structure plant component placement design, the method comprising: acquiring a conceptual design drawing and a design condition text of a steel structure factory building, extracting factory building separation characteristic data from the conceptual design drawing, and extracting plane design condition characteristic data and roof design condition characteristic data from the design condition text;
fusing the plant separation characteristic data and the plane design treaty characteristic data into plant plane layout characteristic tensors, inputting the plant plane layout characteristic tensors into a plant plane component layout generating model with enhanced field knowledge which is trained in advance to obtain layout and section characteristic data of plant plane structure components, and extracting layout and section design data of the plant plane structure components from the layout and section characteristic data of the plant plane structure components;
Fusing the layout and section characteristic data of the plant plane structural members with the roof design condition characteristic data to form a plant roof layout characteristic tensor, inputting the plant roof layout characteristic tensor into a plant roof member layout generating model with enhanced field knowledge which is trained in advance, obtaining plant roof member layout and section characteristic data, and extracting the layout and section design data of the plant roof structural members from the plant roof member layout characteristic data;
and matching and integrating the layout and section design data of the plant plane structural members and the layout and section design data of the plant roof structural members to obtain a preliminary design result and a model of the steel frame structural design of the steel structure plant.
Further, the logic instructions in the memory 530 described above may be implemented in the form of software functional units and may be stored in a computer-readable storage medium when sold or used as a stand-alone product. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
In another aspect, the present invention also provides a computer program product, the computer program product including a computer program, the computer program being storable on a non-transitory computer readable storage medium, the computer program, when executed by a processor, being capable of executing a method of designing a steel structure plant member arrangement based on domain knowledge enhancement provided by the above methods, the method comprising: acquiring a conceptual design drawing and a design condition text of a steel structure factory building, extracting factory building separation characteristic data from the conceptual design drawing, and extracting plane design condition characteristic data and roof design condition characteristic data from the design condition text;
fusing the plant separation characteristic data and the plane design treaty characteristic data into plant plane layout characteristic tensors, inputting the plant plane layout characteristic tensors into a plant plane component layout generating model with enhanced field knowledge which is trained in advance to obtain layout and section characteristic data of plant plane structure components, and extracting layout and section design data of the plant plane structure components from the layout and section characteristic data of the plant plane structure components;
Fusing the layout and section characteristic data of the plant plane structural members with the roof design condition characteristic data to form a plant roof layout characteristic tensor, inputting the plant roof layout characteristic tensor into a plant roof member layout generating model with enhanced field knowledge which is trained in advance, obtaining plant roof member layout and section characteristic data, and extracting the layout and section design data of the plant roof structural members from the plant roof member layout characteristic data;
and matching and integrating the layout and section design data of the plant plane structural members and the layout and section design data of the plant roof structural members to obtain a preliminary design result and a model of the steel frame structural design of the steel structure plant.
In yet another aspect, the present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, is implemented to perform a method of field knowledge based enhanced steel structure plant component layout design provided by the above methods, the method comprising: acquiring a conceptual design drawing and a design condition text of a steel structure factory building, extracting factory building separation characteristic data from the conceptual design drawing, and extracting plane design condition characteristic data and roof design condition characteristic data from the design condition text;
Fusing the plant separation characteristic data and the plane design treaty characteristic data into plant plane layout characteristic tensors, inputting the plant plane layout characteristic tensors into a plant plane component layout generating model with enhanced field knowledge which is trained in advance to obtain layout and section characteristic data of plant plane structure components, and extracting layout and section design data of the plant plane structure components from the layout and section characteristic data of the plant plane structure components;
fusing the layout and section characteristic data of the plant plane structural members with the roof design condition characteristic data to form a plant roof layout characteristic tensor, inputting the plant roof layout characteristic tensor into a plant roof member layout generating model with enhanced field knowledge which is trained in advance, obtaining plant roof member layout and section characteristic data, and extracting the layout and section design data of the plant roof structural members from the plant roof member layout characteristic data;
and matching and integrating the layout and section design data of the plant plane structural members and the layout and section design data of the plant roof structural members to obtain a preliminary design result and a model of the steel frame structural design of the steel structure plant.
The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course may be implemented by means of hardware. Based on this understanding, the foregoing technical solution may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the respective embodiments or some parts of the embodiments.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (10)
1. The method for arranging and designing the steel structure factory building components based on field knowledge enhancement is characterized by comprising the following steps of:
acquiring a conceptual design drawing and a design condition text of a steel structure factory building, extracting factory building separation characteristic data from the conceptual design drawing, and extracting plane design condition characteristic data and roof design condition characteristic data from the design condition text;
fusing the plant separation characteristic data and the plane design treaty characteristic data into plant plane layout characteristic tensors, inputting the plant plane layout characteristic tensors into a plant plane component layout generating model with enhanced field knowledge which is trained in advance to obtain layout and section characteristic data of plant plane structure components, and extracting layout and section design data of the plant plane structure components from the layout and section characteristic data of the plant plane structure components;
Fusing the layout and section characteristic data of the plant plane structural members with the roof design condition characteristic data to form a plant roof layout characteristic tensor, inputting the plant roof layout characteristic tensor into a plant roof member layout generating model with enhanced field knowledge which is trained in advance, obtaining plant roof member layout and section characteristic data, and extracting the layout and section design data of the plant roof structural members from the plant roof member layout characteristic data;
and matching and integrating the layout and section design data of the plant plane structural members and the layout and section design data of the plant roof structural members to obtain a preliminary design result of the steel frame structural design of the steel structure plant.
2. The steel structure plant member arrangement design method based on domain knowledge enhancement according to claim 1, wherein the fusing of the plant separation feature data and the planar design treaty feature data into a plant planar arrangement feature tensor specifically comprises:
the plant separation characteristic data constrains the arrangement range of key structural members of a plant plane;
the plane design treaty characteristic data constrains the structural form of a factory building, anti-seismic design information, the control stress ratio of main components, crane load and mechanical property information of plane key components;
Characterizing the plant separation characteristic data as a second-order matrix to obtain a plant separation characteristic matrix, wherein the positions in the plant separation characteristic matrix, where key structural members of a plant plane can be arranged, are marked as 1, and the rest positions are marked as 0;
characterizing the planar design treaty feature data as zero-order scalar, and copying the zero-order scalar as a homomorphic second-order matrix with the factory building separation feature matrix to obtain a planar design condition feature matrix;
and respectively carrying out normalization processing on the plant separation characteristic matrix and the plane design condition characteristic matrix, and stacking normalization processing results of the plant separation characteristic matrix and the plane design condition characteristic matrix to obtain a plant plane layout characteristic tensor.
3. The field knowledge enhancement based steel structure plant member arrangement design method according to claim 1, wherein the plant planar structural member arrangement and section design data is extracted from the plant planar structural member arrangement and section feature data, specifically comprising:
the arrangement and section characteristic data of the plant plane structural members represent the positioning and section selection matrix of the target steel structure plant plane structural members;
Wherein, the non-zero element position in the plane structural member positioning and selecting matrix refers to the existence of a member, the value on the non-zero element position is filled with the undetermined code, and different codes represent steel structural members with different sizes;
according to the mapping relation between the preset steel structure member size diagram and the non-zero elements in the plane structure member positioning and selecting matrix, the arrangement and section characteristic data of the plant plane structure members are decoded to obtain the arrangement and section design data of the plant plane structure members, and accordingly the design sizes corresponding to various types of members in the target plant plane are obtained.
4. The field knowledge enhancement based steel structure plant member arrangement design method according to claim 1, wherein the plant planar structural member arrangement and section feature data and roof design treaty feature data are fused into a plant roof arrangement feature tensor, specifically comprising:
the arrangement and section characteristic data of the plant plane structural members represent a target steel structure plant plane structural member positioning and section selection matrix, and plane positioning and section selection information is provided for plant roof member arrangement;
the roof design treaty characteristic data constrains the structural form of a factory building, earthquake-resistant design information, roof gradient, roof geometric information, roof load and mechanical property information of roof key components;
The arrangement and section feature data of the plant plane structure members are second-order matrixes, the roof design condition feature data are characterized as zero-order scalar, and the zero-order scalar is copied as homomorphic second-order matrixes with the arrangement and section feature data of the plant plane structure members, so that a roof design condition feature matrix is obtained;
and respectively carrying out normalization processing on the arrangement and section characteristic data of the plant plane structural members and the roof design condition characteristic matrix, and stacking the normalization processing results of the arrangement and section characteristic data of the plant plane structural members and the roof design condition characteristic matrix to obtain a plant roof arrangement characteristic tensor.
5. The field knowledge enhancement based steel structure plant member arrangement design method according to claim 1, wherein the plant roof member arrangement and section design data is extracted from the plant roof member arrangement and section feature data, specifically comprising:
the plant roof component arrangement and section characteristic data represent a target steel structure plant roof structural component positioning and section selection matrix;
wherein, the non-zero element position in the roof structural member positioning and selecting matrix refers to the existence of a member, the value on the non-zero element position is filled with undetermined codes, and different codes represent steel structural members with different sizes;
And decoding the layout and section characteristic data of the plant roof components according to the mapping relation between the preset steel structure component size diagram and the non-zero elements in the roof structure component positioning and selecting matrix to obtain layout and section design data of the plant roof structure components, thereby obtaining the design sizes corresponding to various types of components in the target plant roof.
6. The field knowledge reinforced steel structure plant member arrangement design method according to claim 1, wherein the construction process of the field knowledge reinforced plant plane member arrangement generation model and the field knowledge reinforced plant roof member arrangement generation model comprises the following steps:
the plant plane component arrangement generating model with enhanced domain knowledge and the plant roof component arrangement generating model with enhanced domain knowledge comprise generating modules and evaluating modules with enhanced domain knowledge;
the generating module is obtained by connecting a convolution neural network and a deconvolution neural network, the convolution neural network is used for extracting high-dimensional characteristics in a plant plane arrangement characteristic tensor and a plant roof arrangement characteristic tensor, and the deconvolution neural network is used for upsampling the high-dimensional characteristics output by the convolution neural network to obtain arrangement and section characteristic data of plant plane structure members and arrangement and section characteristic data of plant roof structure members;
The evaluation module for enhancing the domain knowledge is used for evaluating the generating module to obtain the arrangement and section characteristic data of the plant plane structural members and the arrangement and section characteristic data of the plant roof members based on the comprehensive design loss, wherein the comprehensive design loss is formed by weighting the domain design knowledge loss corresponding to various types of members by image similarity loss, the domain design knowledge corresponding to the various types of members comprises: plane domain design knowledge and roof domain design knowledge;
the image similarity loss in the comprehensive design loss is determined based on an image loss function, the image similarity loss is determined based on differences between the image loss function and plant plane member arrangement ideal feature data and plant plane structure member arrangement and section feature data for the field knowledge enhanced plant plane member arrangement generating model, and the image similarity loss is determined based on differences between the image loss function and plant plane member arrangement ideal feature data and plant plane member arrangement and section feature data for the field knowledge enhanced plant plane member arrangement generating model, wherein the plant plane member arrangement ideal feature data is obtained based on an ideal member size diagram of a steel structure plant structure;
Constructing an optimizer for optimizing parameters of the neural network structure to be trained by taking the minimum comprehensive design loss as a target in a training stage;
and training and evaluating the neural network structure to be trained by utilizing the steel structure factory building structure drawing sample, the evaluation module and the optimizer to obtain the factory building plane member arrangement generating model with enhanced domain knowledge and the factory building roof member arrangement generating model with enhanced domain knowledge.
7. The field knowledge enhancement based steel structure plant member arrangement design method according to claim 6, wherein the field design knowledge corresponding to each type of member comprises: plane domain design knowledge and roof domain design knowledge;
the various types of members include: a planar member and a roofing member; the planar member includes: rigid support among steel frame columns, wind-resistant columns and columns; the roofing component includes: steel frame beams, roof purlines and roof flexible supports;
for planar domain design knowledge, the domain knowledge design criteria for the steel frame column include: the end span column spacing arrangement is not greater than the middle span column spacing arrangement; the middle cross-column distances are uniformly distributed;
for roofing component design knowledge, the domain knowledge design criteria for roofing purlins include: the purlines are uniformly arranged.
8. A steel structure plant component arrangement design system based on domain knowledge enhancement, the system comprising:
the data acquisition module is used for acquiring a conceptual design drawing and a design condition text of the steel structure factory building, extracting factory building separation characteristic data from the conceptual design drawing, and extracting plane design condition characteristic data and roof design condition characteristic data from the design condition text;
the factory building planar structural member design module is used for fusing the factory building separation characteristic data and the planar design treaty characteristic data into factory building planar arrangement characteristic tensors, inputting the factory building planar arrangement characteristic tensors into a factory building planar member arrangement generating model with enhanced field knowledge which is trained in advance, obtaining arrangement and section characteristic data of factory building planar structural members, and extracting arrangement and section design data of the factory building planar structural members from the arrangement and section characteristic data of the factory building planar structural members;
the factory building roof structural member design module is used for fusing the layout and section characteristic data of the factory building planar structural members with the roof design condition characteristic data to form factory building roof layout characteristic tensors, inputting the factory building roof layout characteristic tensors into a factory building roof member layout generating model with enhanced field knowledge which is trained in advance to obtain factory building roof member layout and section characteristic data, and extracting layout and section design data of the factory building roof structural members from the roof member layout characteristic data;
And the integration module is used for matching and integrating the layout and section design data of the plant plane structural members and the layout and section design data of the plant roof structural members to obtain a preliminary design result of the steel frame structural design of the steel structure plant.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the domain knowledge based enhanced steel structure plant component placement design method of any one of claims 1 to 7 when the program is executed.
10. A non-transitory computer readable storage medium having stored thereon a computer program, wherein the computer program when executed by a processor implements the domain knowledge based enhanced steel structure plant component arrangement design method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211627394.4A CN116150839B (en) | 2022-12-16 | 2022-12-16 | Steel structure factory building component arrangement design method and system based on field knowledge enhancement |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211627394.4A CN116150839B (en) | 2022-12-16 | 2022-12-16 | Steel structure factory building component arrangement design method and system based on field knowledge enhancement |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116150839A CN116150839A (en) | 2023-05-23 |
| CN116150839B true CN116150839B (en) | 2023-08-29 |
Family
ID=86357436
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211627394.4A Active CN116150839B (en) | 2022-12-16 | 2022-12-16 | Steel structure factory building component arrangement design method and system based on field knowledge enhancement |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116150839B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110084293A (en) * | 2019-04-18 | 2019-08-02 | 贝壳技术有限公司 | A kind of determination method and apparatus in complete bright pattern house |
| CN111339592A (en) * | 2020-02-26 | 2020-06-26 | 广东众图互联网工程设计有限公司 | Automatic design method, device and system for components of building structure model and server |
| CN114417464A (en) * | 2021-12-31 | 2022-04-29 | 清华大学 | Multi-modal data-driven building structural member generation method and device |
| CN114880741A (en) * | 2022-04-25 | 2022-08-09 | 清华大学 | Building structure component size design method and device embedded with domain knowledge |
| CN114925416A (en) * | 2022-04-25 | 2022-08-19 | 清华大学 | Building structure generation method and device based on data conversion |
| JP7159513B1 (en) * | 2022-05-09 | 2022-10-24 | スパイダープラス株式会社 | ICON ARRANGEMENT SYSTEM, ICON ARRANGEMENT METHOD AND PROGRAM |
-
2022
- 2022-12-16 CN CN202211627394.4A patent/CN116150839B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110084293A (en) * | 2019-04-18 | 2019-08-02 | 贝壳技术有限公司 | A kind of determination method and apparatus in complete bright pattern house |
| CN111339592A (en) * | 2020-02-26 | 2020-06-26 | 广东众图互联网工程设计有限公司 | Automatic design method, device and system for components of building structure model and server |
| CN114417464A (en) * | 2021-12-31 | 2022-04-29 | 清华大学 | Multi-modal data-driven building structural member generation method and device |
| CN114880741A (en) * | 2022-04-25 | 2022-08-09 | 清华大学 | Building structure component size design method and device embedded with domain knowledge |
| CN114925416A (en) * | 2022-04-25 | 2022-08-19 | 清华大学 | Building structure generation method and device based on data conversion |
| JP7159513B1 (en) * | 2022-05-09 | 2022-10-24 | スパイダープラス株式会社 | ICON ARRANGEMENT SYSTEM, ICON ARRANGEMENT METHOD AND PROGRAM |
Non-Patent Citations (1)
| Title |
|---|
| 钢筋混凝土单层工业厂房可靠性评估知识获取与知识分析;陈瑞金 等;四川建筑科学研究(第01期);1-5 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116150839A (en) | 2023-05-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20210287138A1 (en) | Learning to simulate and design for structural engineering | |
| CN113779675B (en) | Physical-data driven intelligent shear wall building structure design method and device | |
| CN114880741B (en) | Building structure component size design method and device embedded with domain knowledge | |
| CN117076931B (en) | Time sequence data prediction method and system based on conditional diffusion model | |
| CN116680778B (en) | Building structure arrangement generation method and system and construction method of generation model | |
| Dai et al. | Agent-based model of land system: Theory, application and modelling framework | |
| JP2019179538A (en) | Prediction method for predicting construction price of building and prediction method for predicting its construction period | |
| CN114741766A (en) | Building structure design method and device based on joint structure optimization neural network | |
| CN114741758B (en) | Building earthquake-resistant toughness preliminary design method and system based on machine learning | |
| CN116150839B (en) | Steel structure factory building component arrangement design method and system based on field knowledge enhancement | |
| CN110852415A (en) | Vegetation index prediction method, system and equipment based on neural network algorithm | |
| CN114154622A (en) | Algorithm model for traffic operation system flow data acquisition missing completion | |
| CN114219701A (en) | Dunhuang fresco artistic style conversion method, system, computer equipment and storage medium | |
| KR102589525B1 (en) | Method and apparatus for providing earthquake fragility function, learning method for providing earthquake fragility function | |
| CN110852475B (en) | Extreme gradient lifting algorithm-based vegetation index prediction method, system and equipment | |
| CN116484904A (en) | Supervision data processing implementation method based on artificial intelligence deep learning | |
| CN115964599A (en) | Large-scale land utilization/coverage change transfer matrix device and using method | |
| CN115271221A (en) | City expansion prediction method, device, medium and equipment | |
| CN110543724B (en) | Satellite structure performance prediction method for overall design | |
| CN114548585A (en) | Urban building earthquake damage prediction method based on neural network | |
| CN107633010B (en) | Identification method and system for GRC plate image with complex modeling | |
| CN114118343A (en) | Layer normalization processing hardware accelerator and method applied to Transformer neural network | |
| CN110704587A (en) | Text answer searching method and device | |
| CN116305427B (en) | Anti-seismic fortification parameter processing method and device, electronic equipment and storage medium | |
| CN117993048B (en) | Method and system for designing dimension of building structural member based on diffusion model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |