Bridge model conflict coordination method and system based on BIM (Building Information Modeling) and electronic equipment

Abstract

The invention provides a BIM-based bridge model conflict coordination method and system and electronic equipment, and the method comprises the steps: carrying out the multi-party collaborative design of a bridge model through a BIM platform, and carrying out the collision conflict detection of the designed bridge model; the collision conflict data are exported, and each piece of collision conflict data comprises a collision component ID, a collision component attribute and a collision component coordinate; performing feature extraction on each piece of collision conflict data, and determining a corresponding input feature vector; and inputting the input feature vector into the collision conflict coordination model to obtain an output result of the collision conflict coordination model, namely a conflict coordination scheme. According to the scheme, the technology of an existing BIM platform can be fully utilized to detect collision conflicts, the collision conflict data can be utilized, feature extraction can be performed, input feature vectors can be formed, the collision conflict coordination model of external training can be utilized to generate the conflict coordination scheme, dependence on artificial experience is effectively reduced, and the intelligent collision conflict coordination scheme is provided.

Description

Technical Field

[0001] The present application relates to the field of bridge model design, and more specifically, to a BIM-based bridge model conflict coordination method, system, and electronic equipment. Background Art

[0002] Bridge model design often utilizes the collaborative design capabilities of a BIM platform, allowing different teams and designers to design and modify designs on a unified design model. However, during the model design process, various design coordination issues are inevitable due to the collaboration between different designers and teams. BIM platforms provide collision detection capabilities to detect collision issues (such as collisions between components) in bridge model designs. Mainstream tools (such as Navisworks and BIM 360) can efficiently detect geometric collisions, but solutions rely on manual experience (for example, for a certain collision, the designer needs to determine whether to open a hole or bypass it). This is time-consuming and limited by the ability to communicate and coordinate with multiple parties. It is unable to automatically generate optimization suggestions, requiring designers to handle each one individually. Summary of the Invention

[0003] The purpose of the embodiments of the present application is to provide a BIM-based bridge model conflict coordination method, system and electronic equipment, which provide an intelligent collision conflict coordination solution by deploying a collision conflict coordination model externally.

[0004] In order to achieve the above objectives, the embodiments of the present application are implemented in the following manner: In the first aspect, an embodiment of the present application provides a BIM-based bridge model conflict coordination method, including: using a BIM platform to conduct multi-party collaborative design of a bridge model, and performing collision conflict detection on the designed bridge model; exporting collision conflict data, wherein each collision conflict data includes a collision component ID, collision component attributes, and collision component coordinates; performing feature extraction on each collision conflict data to determine a corresponding input feature vector; inputting the input feature vector into a collision conflict coordination model to obtain an output result of the collision conflict coordination model, wherein the output result reveals a conflict coordination solution corresponding to the collision conflict data.

[0005] In combination with the first aspect, in a first possible implementation method of the first aspect, feature extraction is performed on each collision conflict data to determine a corresponding input feature vector, including: for each collision conflict data: parsing the collision conflict data to determine the collision component ID, collision component attributes, and collision component coordinates in the collision conflict data, wherein the collision component ID reveals the unique number of the collision component that causes the collision conflict, the collision component attributes reveal the component type, geometric dimensions, adjustability, and component priority of the collision component, and the collision component coordinates reveal the coordinate position of the collision component, including the center of mass coordinates, direction vector, and boundary information; feature extraction is performed on the collision component ID, collision component attributes, and collision component coordinates to form an input feature vector corresponding to the collision conflict data.

[0006] In combination with the first possible implementation method of the first aspect, in the second possible implementation method of the first aspect, when the number n of collision components in a single collision conflict data is 2, feature extraction is performed on the collision component ID, collision component attributes, and collision component coordinates to form an input feature vector corresponding to the collision conflict data, including: for each collision component: converting the collision component ID into a fixed-length feature vector v1; performing one-hot encoding on the component type, adjustability, and component priority in the collision component attributes to obtain a feature vector v2; standardizing the geometric dimensions in the collision component attributes to obtain a feature vector v3; extracting the center of mass coordinates, direction vector, and boundary information in the collision component coordinates to obtain a feature vector v4; splicing the feature vector v1, feature vector v2, feature vector v3, and feature vector v4 of each collision component to obtain a comprehensive feature vector for each collision component; splicing the comprehensive feature vectors of two collision components to obtain the input feature vector corresponding to the collision conflict data.

[0007] In combination with the first possible implementation method of the first aspect, in the third possible implementation method of the first aspect, when the number n of collision components in a single collision conflict data exceeds 2, feature extraction is performed on the collision component ID, collision component attributes, and collision component coordinates to form an input feature vector corresponding to the collision conflict data, including: for each collision component: converting the collision component ID into a fixed-length feature vector u1; performing unique-hot encoding on the component type, adjustability, and component priority in the collision component attributes to obtain a feature vector u2; standardizing the geometric dimensions in the collision component attributes to obtain a feature vector u3; extracting the collision component coordinates to obtain a fixed-length feature vector u1; performing unique-hot encoding on the component type, adjustability, and component priority in the collision component attributes to obtain a feature vector u2; standardizing the geometric dimensions in the collision component attributes to obtain a feature vector u3; extracting the collision component coordinates to obtain a fixed-length feature vector u1; The center of mass coordinates, direction vector and boundary information in the mark are used to obtain the feature vector u4; the feature vector u1, feature vector u2, feature vector u3 and feature vector u4 of each collision component are spliced to obtain the comprehensive feature vector of each collision component; the collision components are combined in pairs to form C(n,r)=n! / [r!(nr)!] combinations, where C(n,r) represents the number of combinations, n is the number of collision components in a single collision conflict data, and r is the number of collision components in each combination; the comprehensive feature vectors of the corresponding two collision components are spliced according to the combination to obtain C(n,r) groups of input feature vectors corresponding to the collision conflict data.

[0008] In combination with the first aspect, in a fourth possible implementation of the first aspect, the collision conflict coordination model adopts a decision tree model and introduces a component rule library, which is obtained after training with training data, wherein the rule library contains multiple set rules.

[0009] On the second aspect, an embodiment of the present application provides a BIM-based bridge model conflict coordination system, including: a collision detection unit, used to use the BIM platform to perform multi-party collaborative design of the bridge model, and perform collision conflict detection on the designed bridge model; a data export unit, used to export collision conflict data, wherein each collision conflict data includes a collision component ID, collision component attributes, and collision component coordinates; a feature extraction unit, used to perform feature extraction on each collision conflict data, and determine the corresponding input feature vector; a conflict coordination unit, used to input the input feature vector into the collision conflict coordination model to obtain the output result of the collision conflict coordination model, wherein the output result reveals the conflict coordination solution corresponding to the collision conflict data.

[0010] In combination with the second aspect, in the first possible implementation method of the second aspect, the feature extraction unit is specifically used to: for each collision conflict data: parse the collision conflict data, determine the collision component ID, collision component attributes, and collision component coordinates in the collision conflict data, wherein the collision component ID reveals the unique number of the collision component that causes the collision conflict, the collision component attributes reveal the component type, geometric dimensions, adjustability, and component priority of the collision component, and the collision component coordinates reveal the coordinate position of the collision component, including the center of mass coordinates, direction vector, and boundary information; perform feature extraction on the collision component ID, collision component attributes, and collision component coordinates to form an input feature vector corresponding to the collision conflict data.

[0011] In combination with the first possible implementation method of the second aspect, in the second possible implementation method of the second aspect, when the number n of collision components in a single collision conflict data is 2, the feature extraction unit is specifically used to: for each collision component: convert the collision component ID into a fixed-length feature vector v1; perform one-hot encoding on the component type, adjustability, and component priority in the collision component attributes to obtain a feature vector v2; standardize the geometric dimensions in the collision component attributes to obtain a feature vector v3; extract the center of mass coordinates, direction vector and boundary information in the collision component coordinates to obtain a feature vector v4; splice the feature vector v1, feature vector v2, feature vector v3 and feature vector v4 of each collision component to obtain a comprehensive feature vector for each collision component; splice the comprehensive feature vectors of two collision components to obtain the input feature vector corresponding to the collision conflict data.

[0012] In combination with the first possible implementation method of the second aspect, in the third possible implementation method of the second aspect, when the number n of collision components in a single collision conflict data exceeds 2, the feature extraction unit is specifically used to: for each collision component: convert the collision component ID into a fixed-length feature vector u1; perform one-hot encoding on the component type, adjustability, and component priority in the collision component attributes to obtain the feature vector u2; standardize the geometric dimensions in the collision component attributes to obtain the feature vector u3; extract the center of mass coordinates, direction vector and boundary information in the collision component coordinates to obtain the feature vector u4; splice the feature vector u1, feature vector u2, feature vector u3 and feature vector u4 of each collision component to obtain a comprehensive feature vector of each collision component; combine the collision components in pairs to form C(n,r)=n! / [r!(nr)!] combinations, where C(n,r) represents the number of combinations, n is the number of collision components in a single collision conflict data, and r is the number of collision components in each combination; splice the comprehensive feature vectors of the corresponding two collision components according to the combination to obtain C(n,r) groups of input feature vectors corresponding to the collision conflict data.

[0013] In a third aspect, an embodiment of the present application provides an electronic device comprising a memory and a processor, wherein the memory is used to store information including program instructions, and the processor is used to control the execution of the program instructions, and when the program instructions are loaded and executed by the processor, the steps of the BIM-based bridge model conflict coordination method described in the first aspect or any one of the possible implementation methods of the first aspect are implemented.

[0014] Beneficial effects: This solution uses the BIM platform to conduct multi-party collaborative design of bridge models and perform collision conflict detection. This allows the use of existing BIM platform technology to detect collision conflicts. After collision conflicts are detected, the collision conflict data can be exported (including collision component ID, collision component attributes, collision component coordinates, or API calls can be used to obtain relevant collision conflict data). Feature extraction is performed on each collision conflict data to determine the corresponding input feature vector, which is then input into the collision conflict coordination model to obtain the output result of the collision conflict coordination model (i.e., the conflict coordination solution corresponding to the collision conflict data). This approach can fully utilize the technology of the existing BIM platform to detect collision conflicts. By using collision conflict data, feature extraction can be performed to form input feature vectors, and the externally trained collision conflict coordination model can be used to generate conflict coordination solutions, effectively reducing reliance on manual experience and providing an intelligent collision conflict coordination solution.

[0015] During the feature extraction process, by parsing the collision data, the key features of various elements in the collision conflict data are extracted, and the feature extraction of information such as collision component ID, collision component attributes (component type, geometric dimensions, adjustability, component priority), and collision component coordinates (center of mass coordinates, direction vector, and boundary information) is achieved. Simple collision conflicts when the number of collision components n in a single collision conflict data is 2 and complex collision conflicts when the number of collision components n exceeds 2 are considered separately. Complex collision conflicts are then decomposed into multiple groups of simple collision conflicts. Complex multi-component conflicts are decomposed into multiple groups of binary problems, which are processed pair by pair and then comprehensively optimized to improve adaptability to complex scenarios and help maintain the consistency of the input feature vector. The model training is based on a decision tree model and introduces a component rule library (containing multiple set rules). Training is performed using training data. This type of model can integrate human experience and industry regulations, making the conflict coordination solution provided by the model highly effective and feasible, with good application results. Moreover, based on the collision component ID, the component that needs to be adjusted can be determined from the conflict coordination plan, and the designer can be further determined based on the collision component ID, which greatly reduces the communication inconvenience in multi-party collaborative design and significantly improves the efficiency of design plan adjustment in multi-party collaborative design bridge models.

[0016] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments of the present application. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without creative work.

[0018] Figure 1 A flowchart of a BIM-based bridge model conflict coordination method provided in an embodiment of the present application.

[0019] Figure 2 This is a structural block diagram of an electronic device.

[0020] Figure 3 A schematic diagram for exporting collision conflict data.

[0021] Figure 4 This is the overall block diagram of the BIM-based bridge model conflict coordination system.

[0022] Icons: 10-electronic equipment; 11-memory; 12-communication module; 13-bus; 14-processor; 20-BIM-based bridge model conflict coordination system; 21-collision detection unit; 22-data export unit; 23-feature extraction unit; 24-conflict coordination unit. DETAILED DESCRIPTION

[0023] The technical solutions in the embodiments of the present application will be described below in conjunction with the drawings in the embodiments of the present application.

[0024] See also Figure 1 , Figure 1 A flowchart of a BIM-based bridge model conflict coordination method provided in an embodiment of the present application.

[0025] In this embodiment, the BIM-based bridge model conflict coordination method includes steps S10, S20, S30, and S40, and is applied to electronic devices such as Figure 2 As shown, the electronic device 10 can be a server (such as a cloud server, a server cluster, etc.) or a smart terminal (such as a personal computer).

[0026] Exemplarily, the electronic device 10 may include: a communication module 12 connected to the outside world via a network, one or more processors 14 for executing program instructions, a bus 13, and various forms of memory 11, such as a disk, ROM, or RAM, or any combination thereof. The memory 11, communication module 12, and processor 14 may be connected via bus 13. The memory 11 stores programs, which the processor 14 can access and execute from the memory 11. This allows the BIM-based bridge model conflict coordination method to be implemented by executing these programs.

[0027] First, the electronic device may execute step S10.

[0028] Step S10: Utilize the BIM platform to conduct multi-party collaborative design of the bridge model, and perform collision detection on the designed bridge model.

[0029] In this embodiment, a BIM platform can be used to perform multi-party collaborative design of a bridge model and perform collision detection on the designed bridge model. Examples of BIM platforms include the Autodesk series, which implements multi-party collaborative design and collision detection through Revit + BIM360 / Autodesk Construction Cloud, and the Bentley Systems series, which implements multi-party collaborative design and collision detection through MicroStation + ProjectWise + Bentley Navigator.

[0030] After the collision is detected, the electronic device may execute step S20.

[0031] Step S20: exporting collision conflict data, wherein each collision conflict data includes a collision component ID, collision component attributes, and collision component coordinates.

[0032] In this embodiment, the electronic device can export collision conflict data (such as Figure 3 Specifically, you can use the data export function of the BIM platform or the open API call function to obtain relevant collision and conflict data.

[0033] Exemplarily, after exporting the collision conflict data, for each collision conflict data: the electronic device can parse the collision conflict data to determine the collision component ID, collision component attributes, and collision component coordinates in the collision conflict data, the collision component ID reveals the unique number of the collision component that caused the collision conflict, the collision component attributes reveal the component type, geometric dimensions, adjustability, and component priority of the collision component, and the collision component coordinates reveal the coordinate position of the collision component, including the center of mass coordinates, direction vector, and boundary information (the boundary information can be represented by a geometric bounding box, such as the minimum circumscribed cuboid of the collision component, which can quickly determine the spatial occupancy range of the collision component).

[0034] Afterwards, the electronic device may execute step S30.

[0035] Step S30: extracting features from each collision data to determine a corresponding input feature vector.

[0036] In this embodiment, the electronic device may perform feature extraction on the collision component ID, collision component attributes, and collision component coordinates to form an input feature vector corresponding to the collision conflict data.

[0037] For example, for a simple collision conflict scenario, that is, when the number n of collision components in a single collision conflict data is 2, for each collision component: The electronic device can convert the collision component ID into a fixed-length feature vector v1. In this embodiment, taking a collision component ID with a digital number of a specific length (e.g., 10 digits) as an example, the digital number of the collision component ID can be directly used to obtain an 8-dimensional feature vector v1.

[0038] Furthermore, the electronic device can perform one-hot encoding on the collision component attributes, including component type, adjustability, and component priority, to obtain a feature vector v2. For example, the component type is encoded using an 8-bit one-hot encoding; adjustability is encoded using a 1-bit one-hot encoding, where 1 indicates adjustable and 0 indicates unadjustable; and component priority is encoded using a 1-bit one-hot encoding, where the value ranges from 1 to 5 (or 0 to 9). The larger the value, the higher the priority level. When adjustments are necessary, lower-level collision components are prioritized (of course, higher-level components can also be prioritized, depending on the priority allocation of each component). This is then concatenated according to component type, adjustability, and component priority, resulting in a 10-dimensional feature vector v2.

[0039] Furthermore, the electronic device can standardize the geometric dimensions of the collision component attributes to obtain a feature vector v3. The geometric dimensions here include information such as the length, width, height, diameter, angle, area (or volume) of the collision component. The geometric dimensions of the collision component are standardized according to the maximum and minimum values of the component type of the collision component. The standardization process is not described in detail here. After standardization, the length, width, height, diameter, angle (angle does not need to be standardized), and area (or volume) are spliced in sequence, with each value occupying one dimension, and the splicing is a 6-dimensional feature vector v3.

[0040] Furthermore, the electronic device can extract the centroid coordinates, direction vector, and boundary information from the collision component coordinates to obtain a feature vector v4. For example, the centroid coordinates are three-dimensional coordinates (the default coordinate values are corrected to the three-dimensional space of the bridge model); the direction vector is a three-dimensional vector (which can be converted into a unit vector) that describes the main orientation of the collision component in the three-dimensional space of the bridge model; the boundary information is information about an axis-aligned bounding box (a rectangular bounding box is used as an example in this embodiment), which contains a six-dimensional coordinate information description, specifically the minimum corner coordinates (3-dimensional) and the maximum corner coordinates (3-dimensional). Finally, the electronic device can splice the centroid coordinates, direction vector, and boundary information to obtain a 12-dimensional feature vector v4.

[0041] Afterwards, the electronic device can splice the feature vector v1, feature vector v2, feature vector v3 and feature vector v4 of each collision component to obtain a comprehensive feature vector v of each collision component. The comprehensive feature vector v is a feature vector of 8+10+6+12=36 dimensions.

[0042] Since the number of collision components n in a simple collision conflict scenario is 2, the comprehensive feature vectors of the two collision components can be spliced to obtain the input feature vector corresponding to the collision conflict data, that is, a 72-dimensional feature vector.

[0043] For complex collision scenarios, that is, when the number of collision components n in a single collision data exceeds 2, for each collision component: The electronic device can convert the collision component ID into a fixed-length feature vector u1 (8 dimensions), perform unique-hot encoding on the component type, adjustability, and component priority in the collision component attributes to obtain a feature vector u2 (10 dimensions), standardize the geometric dimensions in the collision component attributes to obtain a feature vector u3 (6 dimensions), extract the center of mass coordinates, direction vector, and boundary information in the collision component coordinates to obtain a feature vector u4 (12 dimensions), and concatenate the feature vectors u1, u2, u3, and u4 of each collision component to obtain a comprehensive feature vector u (36 dimensions) for each collision component. The specific extraction process can be found in the previous article and will not be repeated here.

[0044] After extracting the comprehensive feature vector u for each collision component in the current collision data, the electronic device needs to break down the collision situation of the complex collision scenario into a combination of multiple simple collision scenarios. Specifically, the electronic device can combine the collision components in pairs to form C(n,r)=n! / [r!(nr)!] combinations, where C(n,r) represents the number of combinations, n is the number of collision components in a single collision data set, and r is the number of collision components in each combination. The comprehensive feature vectors of the two corresponding collision components can then be concatenated according to the combination to obtain C(n,r) sets of input feature vectors corresponding to the collision data.

[0045] Considering that complex collision conflict scenarios are decomposed into a combination of multiple simple collision scenarios, it is also necessary to establish the association relationship between these simple collision scenario combinations for verification (that is, using the new information of each collision component, importing it into the BIM platform for collision detection verification, and determining that there is no collision conflict between the components in the coordination plan), and ensuring that the output coordination plan is feasible, so as to avoid the problem of collision conflict remaining when a single group resolves the collision conflict and then combines them.

[0046] In this embodiment, after extracting the input feature vector corresponding to each collision conflict data, the electronic device may execute step S40.

[0047] Step S40: inputting the input feature vector into the collision conflict coordination model to obtain an output result of the collision conflict coordination model, wherein the output result reveals a conflict coordination solution corresponding to the collision conflict data.

[0048] In this embodiment, the collision conflict coordination model uses a decision tree model as the basis to construct a rule library (including multiple rules: for example, no holes are opened in load-bearing components, low-priority components are bypassed or opened, the position of prefabricated components cannot be adjusted, etc.). In this way, human experience and industry standards can be integrated into the model through the rule library to improve the reliability of the model.

[0049] To obtain the training dataset, we need to export historical reported clash data from the BIM platform and obtain the corresponding conflict coordination solutions (as labels). Using the feature extraction process described above, we process each historical report's clash data and use its corresponding conflict coordination solution as a label to construct a training dataset for training the decision tree model. The training process of the decision tree model is not detailed here. Once training is complete, the clash coordination model is obtained.

[0050] Based on this, the electronic device can input the input feature vector into the collision conflict coordination model to obtain the output of the collision conflict coordination model, that is, the conflict coordination solution corresponding to the collision conflict data. For example, the conflict coordination solution corresponding to a certain collision conflict data may be: collision component A detours; or the conflict coordination solution corresponding to another collision conflict data may be: collision component B opens a hole.

[0051] After obtaining the conflict coordination plan corresponding to the collision conflict data, the designer corresponding to the collision component can be determined based on the collision component adjusted in the conflict coordination plan, so that the conflict coordination plan can be distributed to the corresponding designers for efficient collision conflict resolution.

[0052] See also Figure 4 Based on the same inventive concept, an embodiment of the present application provides a BIM-based bridge model conflict coordination system 20, including: a collision detection unit 21, used to use the BIM platform to perform multi-party collaborative design of the bridge model, and perform collision conflict detection on the designed bridge model; a data export unit 22, used to export collision conflict data, wherein each collision conflict data includes a collision component ID, collision component attributes, and collision component coordinates; a feature extraction unit 23, used to extract features from each collision conflict data and determine the corresponding input feature vector; a conflict coordination unit 24, used to input the input feature vector into the collision conflict coordination model to obtain the output result of the collision conflict coordination model, wherein the output result reveals the conflict coordination solution corresponding to the collision conflict data.

[0053] In this embodiment, the feature extraction unit 23 is specifically used to: for each collision conflict data: parse the collision conflict data, determine the collision component ID, collision component attributes, and collision component coordinates in the collision conflict data, wherein the collision component ID reveals the unique number of the collision component that causes the collision conflict, the collision component attributes reveal the component type, geometric dimensions, adjustability, and component priority of the collision component, and the collision component coordinates reveal the coordinate position of the collision component, including the center of mass coordinates, direction vector, and boundary information; perform feature extraction on the collision component ID, collision component attributes, and collision component coordinates to form an input feature vector corresponding to the collision conflict data.

[0054] In this embodiment, when the number n of collision components in a single collision conflict data is 2, the feature extraction unit 23 is specifically used to: for each collision component: convert the collision component ID into a fixed-length feature vector v1; perform one-hot encoding on the component type, adjustability, and component priority in the collision component attributes to obtain a feature vector v2; standardize the geometric dimensions in the collision component attributes to obtain a feature vector v3; extract the center of mass coordinates, direction vector and boundary information in the collision component coordinates to obtain a feature vector v4; splice the feature vector v1, feature vector v2, feature vector v3 and feature vector v4 of each collision component to obtain a comprehensive feature vector of each collision component; splice the comprehensive feature vectors of two collision components to obtain the input feature vector corresponding to the collision conflict data.

[0055] In this embodiment, when the number n of collision components in a single collision conflict data exceeds 2, the feature extraction unit 23 is specifically used to: for each collision component: convert the collision component ID into a fixed-length feature vector u1; perform one-hot encoding on the component type, adjustability, and component priority in the collision component attributes to obtain a feature vector u2; standardize the geometric dimensions in the collision component attributes to obtain a feature vector u3; extract the center of mass coordinates, direction vector and boundary information in the collision component coordinates to obtain a feature vector u4; splice the feature vector u1, feature vector u2, feature vector u3 and feature vector u4 of each collision component to obtain a comprehensive feature vector of each collision component; combine the collision components in pairs to form C(n,r)=n! / [r!(nr)!] combinations, where C(n,r) represents the number of combinations, n is the number of collision components in a single collision conflict data, and r is the number of collision components in each combination; splice the comprehensive feature vectors of the corresponding two collision components according to the combination to obtain C(n,r) groups of input feature vectors corresponding to the collision conflict data.

[0056] In summary, the embodiments of the present application provide a BIM-based bridge model conflict coordination method, system, and electronic device, which utilizes the BIM platform to perform multi-party collaborative design of bridge models and perform collision conflict detection. In this way, the technology of the existing BIM platform can be utilized to detect collision conflicts, and then the collision conflict data (including collision component ID, collision component attributes, collision component coordinates, or calling an API to obtain relevant collision conflict data) can be exported; feature extraction is performed on each collision conflict data to determine the corresponding input feature vector, and then the input feature vector is input into the collision conflict coordination model to obtain the output result of the collision conflict coordination model (i.e., the conflict coordination solution corresponding to the collision conflict data). In this way, the technology of the existing BIM platform can be fully utilized to detect collision conflicts, and the collision conflict data can be used to perform feature extraction to form an input feature vector, and the externally trained collision conflict coordination model can be used to generate a conflict coordination solution, effectively reducing the dependence on manual experience and providing an intelligent collision conflict coordination solution.

[0057] During the feature extraction process, by parsing the collision data, the key features of various elements in the collision conflict data are extracted, and the feature extraction of information such as collision component ID, collision component attributes (component type, geometric dimensions, adjustability, component priority), and collision component coordinates (center of mass coordinates, direction vector, and boundary information) is achieved. Simple collision conflicts when the number of collision components n in a single collision conflict data is 2 and complex collision conflicts when the number of collision components n exceeds 2 are considered separately. Complex collision conflicts are then decomposed into multiple groups of simple collision conflicts. Complex multi-component conflicts are decomposed into multiple groups of binary problems, which are processed pair by pair and then comprehensively optimized to improve adaptability to complex scenarios and help maintain the consistency of the input feature vector. The model training is based on a decision tree model and introduces a component rule library (containing multiple set rules). Training is performed using training data. This type of model can integrate human experience and industry regulations, making the conflict coordination solution provided by the model highly effective and feasible, with good application results. Moreover, based on the collision component ID, the component that needs to be adjusted can be determined from the conflict coordination plan, and the designer can be further determined based on the collision component ID, which greatly reduces the communication inconvenience in multi-party collaborative design and significantly improves the efficiency of design plan adjustment in multi-party collaborative design bridge models.

[0058] In this document, relational terms such as first and second, etc. are used merely to distinguish one entity or operation from another entity or operation, but do not necessarily require or imply any actual relationship or order between these entities or operations.

[0059] The above description is merely an embodiment of the present application and is not intended to limit the scope of protection of the present application. For those skilled in the art, various modifications and variations of the present application are possible. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A bridge model conflict coordination method based on BIM, characterized by: include: Use the BIM platform to conduct multi-party collaborative design of bridge models and perform collision detection on the designed bridge models; Exporting collision conflict data, wherein each collision conflict data includes a collision component ID, a collision component attribute, and a collision component coordinate; Perform feature extraction on each collision data to determine the corresponding input feature vector; The input feature vector is input into the collision conflict coordination model to obtain an output result of the collision conflict coordination model, wherein the output result reveals a conflict coordination solution corresponding to the collision conflict data.

2. The BIM-based bridge model conflict coordination method according to claim 1, characterized in that: Perform feature extraction on each collision data to determine the corresponding input feature vector, including: For each collision conflict data: Parsing the collision conflict data to determine the collision component ID, collision component attributes, and collision component coordinates in the collision conflict data, wherein the collision component ID reveals the unique number of the collision component that caused the collision conflict, the collision component attributes reveal the component type, geometric dimensions, adjustability, and component priority of the collision component, and the collision component coordinates reveal the coordinate position of the collision component, including the center of mass coordinates, direction vector, and boundary information; Feature extraction is performed on the collision component ID, collision component attributes, and collision component coordinates to form an input feature vector corresponding to the collision conflict data.

3. The BIM-based bridge model conflict coordination method according to claim 2, characterized in that: When the number n of collision components in a single collision conflict data is 2, feature extraction is performed on the collision component ID, collision component attributes, and collision component coordinates to form an input feature vector corresponding to the collision conflict data, including: For each collision component: Convert the collision component ID into a fixed-length feature vector v1; Perform one-hot encoding on the component type, adjustability, and component priority in the collision component attributes to obtain the feature vector v2; Normalize the geometric dimensions of the collision component attributes to obtain the feature vector v3; Extract the center of mass coordinates, direction vector and boundary information from the collision component coordinates to obtain the feature vector v4; Splicing the eigenvector v1, eigenvector v2, eigenvector v3 and eigenvector v4 of each collision component to obtain a comprehensive eigenvector of each collision component; The comprehensive feature vectors of the two colliding components are spliced together to obtain the input feature vector corresponding to the collision conflict data.

4. The BIM-based bridge model conflict coordination method according to claim 2, characterized in that: When the number of collision components n in a single collision conflict data exceeds 2, feature extraction is performed on the collision component ID, collision component attributes, and collision component coordinates to form an input feature vector corresponding to the collision conflict data, including: For each collision component: Convert the collision component ID into a fixed-length feature vector u1; Perform one-hot encoding on the component type, adjustability, and component priority in the collision component attributes to obtain the feature vector u2; Standardize the geometric dimensions of the collision component attributes to obtain the characteristic vector u3; Extract the center of mass coordinates, direction vector and boundary information from the collision component coordinates to obtain the feature vector u4; Splicing the eigenvector u1, eigenvector u2, eigenvector u3 and eigenvector u4 of each collision component to obtain a comprehensive eigenvector of each collision component; Combine the collision components in pairs to form C(n,r)=n! / [r!(nr)!] combinations, where C(n,r) represents the number of combinations, n is the number of collision components in a single collision data, and r is the number of collision components in each combination; By combining and splicing the comprehensive feature vectors of the corresponding two collision components, a C(n, r) group of input feature vectors corresponding to the collision conflict data is obtained.

5. The BIM-based bridge model conflict coordination method according to claim 1, characterized in that: The collision conflict coordination model adopts a decision tree model and introduces a component rule base, which is obtained after training with training data. The rule base contains multiple set rules.

6. A BIM-based bridge model conflict coordination system, characterized by: include: A collision detection unit is used to conduct multi-party collaborative design of bridge models using the BIM platform and perform collision detection on the designed bridge models; A data exporting unit, configured to export collision conflict data, wherein each collision conflict data includes a collision component ID, collision component attributes, and collision component coordinates; A feature extraction unit is used to extract features from each collision data and determine the corresponding input feature vector; The conflict coordination unit is used to input the input feature vector into the collision conflict coordination model to obtain an output result of the collision conflict coordination model, wherein the output result reveals a conflict coordination solution corresponding to the collision conflict data.

7. The BIM-based bridge model conflict coordination system according to claim 6, characterized in that: Feature extraction unit, specifically used for: For each collision conflict data: parse the collision conflict data, determine the collision component ID, collision component attributes, and collision component coordinates in the collision conflict data, wherein the collision component ID reveals the unique number of the collision component that causes the collision conflict, the collision component attributes reveal the component type, geometric dimensions, adjustability, and component priority of the collision component, and the collision component coordinates reveal the coordinate position of the collision component, including the center of mass coordinates, direction vector, and boundary information; perform feature extraction on the collision component ID, collision component attributes, and collision component coordinates to form an input feature vector corresponding to the collision conflict data.

8. The BIM-based bridge model conflict coordination method according to claim 7, characterized in that: When the number n of collision components in a single collision conflict data is 2, the feature extraction unit is specifically configured to: For each collision component: convert the collision component ID into a fixed-length feature vector v1; perform one-hot encoding on the component type, adjustability, and component priority in the collision component attributes to obtain the feature vector v2; standardize the geometric dimensions in the collision component attributes to obtain the feature vector v3; extract the center of mass coordinates, direction vector, and boundary information in the collision component coordinates to obtain the feature vector v4; splice the feature vector v1, feature vector v2, feature vector v3, and feature vector v4 of each collision component to obtain the comprehensive feature vector of each collision component; splice the comprehensive feature vectors of two collision components to obtain the input feature vector corresponding to the collision conflict data.

9. The BIM-based bridge model conflict coordination method according to claim 7, characterized in that: When the number n of collision components in a single collision conflict data exceeds 2, the feature extraction unit is specifically configured to: For each collision component: convert the collision component ID into a fixed-length feature vector u1; perform one-hot encoding on the component type, adjustability, and component priority in the collision component attributes to obtain the feature vector u2; standardize the geometric dimensions in the collision component attributes to obtain the feature vector u3; extract the center of mass coordinates, direction vector, and boundary information in the collision component coordinates to obtain the feature vector u4; concatenate the feature vector u1, feature vector u2, feature vector u3, and feature vector u4 of each collision component to obtain a comprehensive feature vector for each collision component; combine the collision components in pairs to form C(n,r)=n! / [r!(nr)!] combinations, where C(n,r) represents the number of combinations, n is the number of collision components in a single collision conflict data, and r is the number of collision components in each combination; concatenate the comprehensive feature vectors of the corresponding two collision components according to the combination to obtain C(n,r) groups of input feature vectors corresponding to the collision conflict data.

10. An electronic device comprising a memory and a processor, wherein the memory is used to store information including program instructions, and the processor is used to control the execution of the program instructions, characterized in that: When the program instructions are loaded and executed by the processor, the steps of the BIM-based bridge model conflict coordination method according to any one of claims 1 to 5 are implemented.

Images