A method for arranging the living area of a construction site based on a knowledge graph and intelligent algorithms

Abstract

The present invention discloses a method for arranging the living area of a construction site based on a knowledge graph and intelligent algorithms, which relates to the technical field of model application. By establishing a knowledge graph and using intelligent algorithms for automatic optimization design, the present invention can efficiently and accurately optimize the design according to user requirements, improve the design quality and save manual intervention. Moreover, the method of the present invention realizes the full-process automation from design requirements to the automatic generation of BIM models, without manual intervention, reduces the cumbersome steps and potential errors in the design process. And the present invention has better customized design, can optimize the design according to specific space types and area requirements, can achieve a more personalized design that meets the actual needs, improves the rationality and practicability of the design. And through the comprehensive evaluation criteria to guide the design optimization, it is ensured that the design result not only meets the space requirements, but also can achieve the optimal balance of comprehensive factors such as environment and function.

Description

Technical Field

[0001] The present invention relates to the technical field of model applications, and specifically to a method for arranging the living area of a construction site based on a knowledge graph and intelligent algorithms. Background Art

[0002] The layout of a construction site includes the layout of the construction area and the layout of the living area.

[0003] Among them, the layout of the living area needs to take into account requirements such as the functional zoning and traffic flow of the living area.

[0004] Currently, the design process mainly relies on designers to manually adjust according to design experience and design cases, resulting in a waste of a large amount of time and being unable to quickly generate an optimized layout plan.

[0005] At the same time, in terms of the layout of a construction site, there are currently many methods for the layout of the construction area, and fewer methods for the layout of the living area.

[0006] Therefore, a new solution needs to be proposed for the above problems. Summary of the Invention

[0007] The purpose of the present invention is to provide a method for arranging the living area of a construction site based on a knowledge graph and intelligent algorithms to solve the technical problems raised in the background art.

[0008] To achieve the above purpose, the present invention provides the following technical solution: A method for arranging the living area of a construction site based on a knowledge graph and intelligent algorithms, at least including the following steps:

[0009] S1: Construction and extraction of the knowledge graph for the layout of the living area of a construction site. Through the established knowledge graph of the user's design requirements, the specific space types and their areas required are obtained respectively, and at the same time, the comprehensive evaluation criteria for the design are obtained;

[0010] S2: Establish an intelligent algorithm for the layout of the living area of a construction site. Use the space types and their areas as the design conditions for the optimized design, and the comprehensive evaluation criteria as the goal of the optimized design, and use the intelligent algorithm for automatic optimized design;

[0011] S3: Automatically obtain the BIM model of the living area of the construction site through Dynamo software in Revit software based on the optimized design result.

[0012] Further, the construction and extraction of the knowledge graph for the layout of the living area of the construction site at least includes data collection and processing, construction of the knowledge graph, and extraction of design conditions and design goals.

[0013] Further, the data collection and processing at least includes the following steps:

[0014] First, sort out the current relevant specifications for the layout of the living area on the construction site and the previous design documents;

[0015] For the collected relevant specifications and design documents in text form, perform text standardization;

[0016] After standardizing the text, remove the content that has nothing to do with the layout of the living area on the construction site;

[0017] Finally, split the continuous text into paragraphs.

[0018] Furthermore, the construction of the knowledge graph includes at least the following steps:

[0019] First, according to the term frequency - inverse design specification frequency, perform entity recognition for the layout of the living area on the construction site, and then evaluate the importance of words in the design specifications, as shown in formulas (1)-(3):

[0020] (1)

[0021] (2)

[0022] (3)

[0023] Among them, TF is the term frequency, IDF is the inverse design specification frequency, is the number of times the word t appears in the design specification d, is the sum of the number of times all words appear in the design specification d, N is the total number of design specifications, and df(t) is the number of design specifications containing the word t;

[0024] Then, perform dependency analysis according to formula (4), and determine the semantic relationship between the verb and the relevant entity in the sentence according to formula (5);

[0025] (4)

[0026] (5)

[0027] Among them, entity represents the entity, relation represents the relationship, Count represents the statistical number, S is the subject, V is the verb, O is the object, F is the feature function, and w is the weight vector;

[0028] Finally, convert the entity into a node in the graph, and then establish a relationship edge between the corresponding entities according to the recognized relationship.

[0029] Furthermore, the extraction of the design conditions and design goals includes at least the following steps:

[0030] According to the input design requirements, use the graph database query language (Cypher for Neo4j) to query each spatial type and its corresponding area, and use them as design conditions;

[0031] Meanwhile, use the graph database to query the relevant specification requirements, and then take weighted values as the final optimization objective, as shown in Equation (6):

[0032] Score (6)

[0033] Among them, w represents different weights, indicating the score of the nth objective.

[0034] Furthermore, the intelligent algorithm for the layout of the living area of the construction site is a two-stage dynamic intelligent algorithm;

[0035] First, use the dynamic genetic algorithm for global optimization, and then use the hill climbing algorithm for local optimization to finally obtain the optimal solution;

[0036] The dynamic genetic algorithm includes population dynamic adjustment, dynamic crossover probability, and dynamic mutation probability;

[0037] The population dynamic adjustment is as shown in Equation (7):

[0038] (7)

[0039] Among them, is the population individual of the t-th generation, is the diversity control parameter, is the set of newly generated individuals;

[0040] The dynamic crossover probability and dynamic mutation probability are as shown in Equations (8) and (9):

[0041] (8)

[0042] (9)

[0043] Among them, and are the crossover and mutation probabilities of the t-th generation, and are the initial crossover and mutation probabilities, is the measure of the degree of environmental change, that is, the number of iterations; and are the adjustment parameters set by the user.

[0044] Furthermore, the application of the intelligent algorithm for the layout of the living area of the construction site at least includes the following steps:

[0045] First, a mathematical model needs to be established for the layout of the living area on the construction site;

[0046] Then, the variables are parameterized;

[0047] Finally, an intelligent algorithm for the layout of the living area on the construction site is used for calculation.

[0048] Furthermore, according to the indexing of the design conditions and design objectives by the knowledge graph, the following mathematical modeling process is determined for the mathematical modeling:

[0049] S2.1: Location constraints of building facilities. All building facilities need to be within the living area and at the same time within the range that the tower crane cannot reach, as shown in Equation (10):

[0050] (10)

[0051] Among them, represents the location of the i-th building facility, represents the living area, represents the range that the tower crane can reach;

[0052] S2.2: The regulation of the dormitories in the living area needs to be of a specified type because the commonly used dormitory types of different construction units are usually several fixed types, as shown in Equation (11):

[0053] (11)

[0054] Among them, is the type of the j-th dormitory, and T is the set of allowed dormitory types;

[0055] S2.3: The capacity of the dormitories and the office area needs to meet the requirements of the design specifications;

[0056] S2.4: The adjacent constraints of past design cases need to be met between each building;

[0057] S2.5: The flow line of the entire living area on the construction site needs to meet the fire prevention requirements, that is, the fire truck can reach smoothly and carry out fire extinguishing operations, as shown in Equation (12):

[0058] (12)

[0059] Among them, x represents the location of any building facility in the living area, P(x) represents the path from the fire truck entrance to location x, and Set lives represents the set of all building facilities in the living area;

[0060] Equation (12) indicates that for each building facility location in the living area , there exists a reachable path P(x).

[0061] Furthermore, the parameterization at least includes parameterizing the terrain and parameterizing the buildings. Parameterizing the terrain is to obtain the locations of each building functional area;

[0062] The parameterization of the terrain at least includes the following steps:

[0063] First, the area within the building red line is rasterized at a certain step size to obtain grid points ;

[0064] Meanwhile, all grid points are numbered, that is, each position i can correspond to a code ;

[0065] The parameterization of the buildings at least includes the following steps:

[0066] First, the minimum rectangular bounding rectangle is constructed according to the building plane of different functional areas, and the center of the building area plane is used as the anchor point of the building plane. The position of this anchor point is a parameter;

[0067] Meanwhile, the angle by which the plane rotates around the anchor point is used as another parameter;

[0068] The step size of the rotation angle is taken as 45 degrees;

[0069] Finally, for building area i, positioning is achieved through two parameters, that is .

[0070] Compared with the prior art, the beneficial effects of the present invention are:

[0071] 1. By establishing a knowledge graph and using intelligent algorithms for automatic optimization design, the present invention can efficiently and accurately perform design optimization according to user requirements, improve design quality and save manual intervention. Moreover, the method of the present invention realizes the full-process automation from design requirements to the automatic generation of BIM models, without manual intervention, reducing the cumbersome steps and potential errors in the design process.

[0072] 2. The present invention has better customized design, can perform optimization design according to specific space types and area requirements, can achieve more personalized designs that meet actual needs, improve the rationality and practicality of the design, and through the comprehensive evaluation criteria to guide the design optimization, ensure that the design results not only meet the space requirements, but also can achieve the optimal balance of comprehensive factors such as environment and function.

[0073] 3. Based on the optimized design results, the present invention realizes the automatic generation of the BIM model for the living area of the construction site through Revit and Dynamo, improving the efficiency of the connection between design and construction. The technologies of intelligent optimized design and automatic generation of BIM models can greatly improve the design efficiency, reduce human errors at the same time, and ensure the design accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] In order to illustrate the technical solutions of the embodiments of the present invention more clearly, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0075] Figure 1 It is a schematic diagram of the method architecture of the present invention;

[0076] Figure 2 It is a parametric schematic diagram of the present invention;

[0077] Figure 3 It is a schematic diagram of the two-stage dynamic intelligent algorithm of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0078] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention.

[0079] The input of the present invention is the design requirements of the living and office area, and the final output is the BIM model of the living area of the construction site.

[0080] Please refer to Figure 1 , a method for arranging the living area of the construction site based on a knowledge graph and an intelligent algorithm, at least including the following steps:

[0081] S1: Construction of a knowledge graph for the layout of the living area of the construction site and extraction. The design requirements of the user pass through the established knowledge graph to obtain the specific spatial types and their areas required respectively, and at the same time obtain the comprehensive evaluation criteria for the design;

[0082] S2: Establish an intelligent algorithm for the layout of the living area of the construction site. Use the spatial types and their areas as the design conditions for the optimized design, and the comprehensive evaluation criteria as the goal of the optimized design, and use the intelligent algorithm for automatic optimized design;

[0083] S3: Automatically obtain the BIM model of the living area of the construction site based on the optimized design results through the Dynamo software in the Revit software.

[0084] The construction and extraction of the knowledge graph for the layout of the living area at the construction site include at least data collection and processing, knowledge graph construction, and extraction of design conditions and design objectives.

[0085] The collection and processing of data include at least the following steps:

[0086] First, sort out the current relevant specifications for the layout of the living area at the construction site and previous design documents;

[0087] For the collected relevant specifications and design documents in text form, perform text normalization, such as standardizing term expressions, for example, standardizing expressions like "living area design" and "living area site layout" to "layout of the living area at the construction site";

[0088] After normalizing the text, remove the content unrelated to the layout of the living area at the construction site;

[0089] Finally, split the continuous text into paragraphs.

[0090] The construction of the knowledge graph includes at least the following steps:

[0091] First, based on the term frequency-inverse design specification frequency (TF-IDF), perform entity recognition for the layout of the living area at the construction site, and then evaluate the importance of words such as "office location" in the design specifications, as shown in formulas (1)-(3):

[0092] (1)

[0093] (2)

[0094] (3)

[0095] Among them, TF is the term frequency, IDF is the inverse design specification frequency, is the number of times the word t appears in the design specification d, is the sum of the number of times all words appear in the design specification d, N is the total number of design specifications, and df(t) is the number of design specifications containing the word t;

[0096] Then, perform dependency relationship analysis according to formula (4) and determine the semantic relationship between the verb and the relevant entity in the sentence according to formula (5);

[0097] (4)

[0098] (5)

[0099] Among them, entity represents an entity, relation represents a relationship, Count represents the statistical number, S is the subject, V is the verb, O is the object, F is the feature function, and w is the weight vector;

[0100] Finally, the entity is converted into a node in the graph, and then a relationship edge is established between the corresponding entities according to the identified relationship.

[0101] The extraction of design conditions and design objectives includes at least the following steps:

[0102] According to the input design requirements, use the graph database query language (Cypher for Neo4j) to query each spatial type and its corresponding area, and use it as the design condition;

[0103] At the same time, use the graph database to query the relevant specification requirements, and then take weighting as the final optimization objective, as shown in Equation (6):

[0104] Score (6)

[0105] Among them, w represents different weights, indicating the score of the nth objective.

[0106] Refer to Figure 3 , the intelligent algorithm for the layout of the living area of the construction site is a two-stage dynamic intelligent algorithm;

[0107] First, use the dynamic genetic algorithm for global optimization, and then use the hill climbing algorithm for local optimization to finally obtain the optimal solution;

[0108] The dynamic genetic algorithm includes population dynamic adjustment, dynamic crossover probability, and dynamic mutation probability;

[0109] The population dynamic adjustment is as shown in Equation (7):

[0110] (7)

[0111] Among them, is the population individual of the t-th generation, is the diversity control parameter, is the newly generated individual set;

[0112] The dynamic crossover probability and dynamic mutation probability are as shown in Equations (8) and (9):

[0113] (8)

[0114] (9)

[0115] Among them, and is the crossover and mutation probability for the t-th generation, and is the initial crossover and mutation probability, is a measure of the degree of environmental change, i.e., the number of iterations; and are adjustment parameters set by the user.

[0116] The application of the intelligent algorithm for the layout of the living area of the construction site includes at least the following steps:

[0117] First, a mathematical model is established for the problem of the layout of the living area of the construction site;

[0118] Then, the variables are parameterized;

[0119] Finally, the intelligent algorithm for the layout of the living area of the construction site is used for calculation.

[0120] Based on the indexing of the design conditions and design objectives in the knowledge graph, the following mathematical modeling process is determined for the mathematical modeling:

[0121] S2.1: Location constraints of building facilities. All building facilities (such as workers' accommodation areas, meeting rooms, etc.) need to be within the living area and at the same time within the range that the tower crane cannot reach, as shown in Equation (10):

[0122] (10)

[0123] Among them, represents the location of the i-th building facility, represents the living area, represents the range that the tower crane can reach;

[0124] S2.2: The regulation of the dormitories in the living area needs to be of a specified type because the common dormitory types of different construction units are usually several fixed ones, as shown in Equation (11):

[0125] (11)

[0126] Among them, is the type of the j-th dormitory, and T is the set of allowed dormitory types;

[0127] S2.3: The capacity of the dormitories and the office area needs to meet the requirements of the design specifications;

[0128] S2.4: Adjacent constraints between buildings need to be met according to past design cases. For example, the cafeteria should be as close as possible to the dormitory area, the toilet should be as far as possible from the kitchen, and the meeting room should be as close as possible to the entrance of the gate;

[0129] S2.5: The flow line of the living area of the entire construction site needs to meet the fire prevention requirements, that is, the fire truck can reach smoothly and carry out fire extinguishing operations, as shown in Equation (12):

[0130] (12)

[0131] Where x represents the location of any building facility in the living area, P(x) represents the path from the fire truck entrance to location x, and Set lives represents the set of all building facilities in the living area;

[0132] Equation (12) indicates that for each building facility location in the living area , there exists a reachable path P(x).

[0133] Parametrization includes at least parametrizing the terrain and parametrizing the buildings. Parametrizing the terrain is to obtain the location of each building functional area;

[0134] Parametrizing the terrain includes at least the following steps:

[0135] First, rasterize the area within the building red line at a certain step size to obtain grid points ; as Figure 2 shown;

[0136] At the same time, number all grid points, that is, each location i can correspond to a code ;

[0137] Parametrizing the buildings includes at least the following steps:

[0138] First, construct the minimum rectangular bounding rectangle according to the building plan of different functional areas, as Figure 2 shown. Take the center of the building area plane as the anchor point of the building plane, and the position of this anchor point is a parameter;

[0139] At the same time, take the angle by which the plane rotates around the anchor point as another parameter;

[0140] The step size of the rotation angle is taken as 45 degrees;

[0141] Finally, for building area i, positioning is achieved through two parameters, that is .

[0142] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above-described exemplary embodiments, and the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention. Therefore, in any aspect, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be embraced within the present invention. Any reference signs in the claims should not be construed as limiting the claims involved.

Claims

1. A method for arranging living areas in construction sites based on knowledge graph and intelligent algorithm, characterized in that: At least the following steps are included: S1: Construction and extraction of knowledge graph for the layout of living areas in construction sites. The user's design requirements are obtained through the established knowledge graph, and the specific space types and areas required are obtained, and the comprehensive evaluation criteria for the design are obtained; The knowledge graph construction includes at least the following steps: First, the entity recognition of the living area layout of the construction site is carried out according to the word frequency-inverse design specification frequency, and then the importance of the words in the design specification is evaluated, as shown in formulas (1)-(3): ， Among them, TF is the term frequency, IDF is the inverse design frequency, Yes word t In the design specification d The number of times it appears in It is a design specification d The sum of the occurrences of all words in , N is the total number of design specifications, Contains words t The number of design specifications; Then, the dependency relationship is analyzed according to formula (4), and the semantic relationship between the verbs and related entities in the sentence is determined according to formula (5); ， in, entity Represents an entity, relation Indicates relationship, Count Indicates the statistical number, S is the subject, V It is a verb. O It is the object. F is the characteristic function, w is the weight vector; Finally, the entity is converted into a node in the graph, and then relationship edges are established between corresponding entities based on the identified relationships; The construction and extraction of the knowledge graph of the living area layout of the construction site at least includes data collection and processing, knowledge graph construction, and extraction of design conditions and design objectives; The extraction of the design conditions and design objectives comprises at least the following steps: Based on the input design requirements, use the graph database query language Cypher for Neo4j to query each space type and its corresponding area, and use it as the design condition; At the same time, the graph database is used to query the relevant specification requirements, and then the weighted method is used as the final optimization goal, as shown in formula (6): ， in, w Indicates different weights, Indicates the score of the nth target; S2: Establish an intelligent algorithm for the layout of the living area of ​​the construction site, take the space type and its area as the design conditions for the optimization design, and the comprehensive evaluation criteria as the goal of the optimization design, and use the intelligent algorithm for automatic optimization design; The intelligent algorithm for arranging the living area of ​​the construction site is a two-stage dynamic intelligent algorithm; First, a dynamic genetic algorithm is used for global optimization, and then a hill climbing algorithm is used for local optimization to finally obtain the optimal solution; The dynamic genetic algorithm includes dynamic population adjustment, dynamic crossover probability and dynamic mutation probability; The population dynamic adjustment is shown in formula (7): ， in, is the population individual of generation t, is the diversity control parameter, is a newly generated set of individuals; The dynamic crossover probability and dynamic mutation probability are shown in equations (8) and (9): ， in, and For the t The crossover and mutation probabilities of each generation, and is the initial crossover and mutation probability, is a measure of the degree of environmental change, i.e., the number of iterations; and Adjustment parameters set by the user; S3: Based on the optimization design results, the BIM model of the living area of ​​the construction site is automatically obtained through the Dynamo software in the Revit software; The application of the intelligent algorithm for the layout of the living area of ​​the construction site includes at least the following steps: First, we need to mathematically model the problem of arranging the living area of ​​the construction site; Then, the variables are parameterized; Finally, an intelligent algorithm for the layout of the living area of ​​the construction site is used for calculation; The mathematical modeling determines the following mathematical modeling process based on the index of the design conditions and design goals by the knowledge graph: S2.1: Building facility location constraints. All building facilities must be located within the living area and within the range that the tower crane cannot reach, as shown in formula (10): ， in, Indicates i The location of the building facilities, Indicates the living area, Indicates the range that the tower crane can reach; S2.2: The regulations for dormitories in living areas need to be of a specified type. This is because different construction units often use a few fixed types of dormitories, as shown in formula (11): ， in, For the j The types of dormitories, T Assemble for the types of dormitories permitted; S2.3: The capacity of dormitories and office areas needs to meet the requirements of the design specifications; S2.4: Each building needs to satisfy the adjacent constraints of previous design cases; S2.5: The streamlines of the living area of ​​the entire construction site need to meet the requirements of fire prevention, that is, the fire trucks can arrive and extinguish the fire smoothly, as shown in formula (12): ， in, x Indicates the location of any building facility in the living area. P ( x ) indicates the distance from the fire truck entrance to the location x The path, Represents the collection of all living quarters building facilities; Formula (12) shows that for each building facility location in the living area , there is a path that can reach P ( x ); The parameterization at least includes parameterization of terrain and parameterization of buildings, and the parameterization of terrain is to obtain the position of each building functional area; The parameterization of the terrain comprises at least the following steps: First, the area within the building red line is divided into Rasterize and obtain grid points At the same time, all grid points are numbered, so that each position can be i Corresponding to a code ; The parameterization of the building comprises at least the following steps: First, the minimum rectangular enclosing rectangle is constructed according to the building planes of different functional areas, and the center of the building area plane is used as the anchor point of the building plane. The position of this anchor point is a parameter. The angle at which the plane is rotated around the anchor point As another parameter; The step size of the rotation angle is 45 degrees; Finally, for the building area i Positioning is achieved through two parameters, namely .

2. According to claim 1, a method for arranging living areas in construction sites based on knowledge graph and intelligent algorithm is characterized by: The data collection and processing includes at least the following steps: First, sort out the current relevant specifications for the layout of living areas on construction sites, as well as previous design documents; Normalize the collected text-based specifications and design documents; When the text is standardized, remove the content that is not related to the layout of the living area of ​​the construction site; Finally, the continuous text is divided into paragraphs.

Images