Space layout optimization method for intelligent power distribution room based on BIM technology

Abstract

The application relates to the technical field of computer-aided design, in particular to an intelligent power distribution room space layout optimization method based on BIM technology, which comprises the following steps: generating an initial device connection weight topology graph according to the device nodes of a power distribution room electrical single-line diagram, the first connection weight value of a bus connection line and the second connection weight value of a cable connection line; extracting the device mass value and the outer shape boundary size frame of a power distribution room three-dimensional entity model, and obtaining an initial module cutting data set containing a subset based on a Laplacian matrix. The application can improve the space utilization by the following steps: performing full permutation combination on the candidate placement positions, mapping the three-dimensional space normal vector of the local butt flange and the center point coordinates of the local butt flange into global pose parameters, and screening a target global placement combination scheme meeting a preset assembly tolerance extreme value according to the three-dimensional distance difference and the included angle value, and then completing unified checking before outputting the intelligent power distribution room three-dimensional assembly coordinate set.

Description

Technical Field

[0001] This invention relates to the field of computer-aided design technology, and in particular to a method for optimizing the spatial layout of intelligent power distribution rooms based on BIM technology. Background Technology

[0002] Computer-aided design technology refers to the technology of using computers to digitally express, model, analyze, and optimize the structure, space, dimensions, connection relationships, functional parameters, and assembly relationships of engineering objects.

[0003] Current computer-aided design (CAD) technologies typically focus on the digital representation, modeling, analysis, and optimization of the structure, space, dimensions, connections, functional parameters, and assembly relationships of engineering objects. In practice, design objects often exist separately in the form of drawings, 3D models, and attribute tables. Electrical connections are primarily used to express the logical connections between equipment, while 3D models are mainly used to represent the external outline and spatial location. There is a lack of continuous linkage between equipment mass distribution, hoisting center of gravity deviation, group boundaries, and docking postures. This means that while the spatial layout results formed during the design phase may meet geometric placement requirements, it is difficult to simultaneously guarantee the hoisting stability of prefabricated modules and the accuracy of on-site assembly. For example, when arranging based solely on equipment dimensions, even if heavy equipment is concentrated on one side of the virtual base of a standard prefabricated module, significant center of gravity shifts may still occur without external collisions, requiring additional counterweights or adjustments to the hoisting points during the hoisting process. Therefore, improvements are needed. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by proposing a smart power distribution room spatial layout optimization method based on BIM technology.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a method for optimizing the spatial layout of an intelligent power distribution room based on BIM technology, comprising the following steps:

[0006] An initial equipment connection weight topology diagram is generated based on the first connection weight value of the equipment nodes and busbar connection lines and the second connection weight value of the cable connection lines in the electrical single-line diagram of the power distribution room.

[0007] Extract equipment mass values ​​and external boundary dimensions from the 3D solid model of the power distribution room, and obtain an initial module segmentation dataset containing subsets based on the Laplacian matrix;

[0008] The relative three-dimensional centroid coordinates of the outer boundary dimension frame are obtained by local three-dimensional orthogonal nesting within the virtual base of the standard prefabricated cabin.

[0009] The overall center of gravity coordinate point is calculated by combining the equipment mass value with the relative three-dimensional center of mass coordinate point, and the hoisting center of gravity deviation distance value is obtained by comparing the overall center of gravity coordinate point with the geometric center coordinate point.

[0010] For the target subset whose hoisting center of gravity deviation distance exceeds the preset hoisting safety threshold, the connection weight values ​​of key equipment nodes and adjacent nodes are reduced based on the torque contribution, and an updated equipment connection weight topology map is generated.

[0011] Obtain the updated module cutting dataset, re-execute local 3D orthogonal nesting and calculate the hoisting center of gravity deviation distance until it meets the preset hoisting safety threshold, and output the balanced prefabricated cabin equipment grouping;

[0012] Combine the layout position parameters and the outer boundary size frame to generate the group outer boundary size frame; extract the coordinate items of the cuboid cavity from the three-dimensional solid model of the power distribution room by removing the structural columns and the area occupied by the passage, and generate a linked list of usable three-dimensional blank areas;

[0013] Candidate regions are extracted by matching the clustered bounding boundary size box with the available 3D blank area linked list, generating a dataset of candidate locations for prefabricated cabins.

[0014] Preferably, the method further includes:

[0015] Generate a global placement combination scheme from candidate placement position combinations; based on the candidate position coordinates, map the local flange 3D space normal vector and the local flange center point coordinates to the global flange 3D space normal vector and the global flange center point coordinates;

[0016] Calculate the three-dimensional distance difference and included angle values ​​to construct the module docking pose error matrix; lock the target global placement combination scheme that meets the preset assembly tolerance extreme value, extract the three-dimensional placement parameters, and generate the three-dimensional assembly coordinate set of the intelligent power distribution room.

[0017] Preferably, the step of obtaining the initial device connection weight topology graph is as follows:

[0018] Read the electrical single-line diagram of the power distribution room, analyze the equipment element identifier, equipment number field and port connection position of the electrical single-line diagram line by line, extract all equipment nodes with independent access ports, distinguish bus connection lines and cable connection lines according to the layer attributes, line type attributes and port bridging relationships of the connection lines, assign a first connection weight value to the bus connection lines and a second connection weight value to the cable connection lines to form the equipment node connection weight identifier;

[0019] Based on the device node connection weight identifier, all device nodes are converted into topology vertices one by one, and bus connection lines carrying the first connection weight value and cable connection lines carrying the second connection weight value are converted into topology edges one by one. The starting device node, ending device node and connection weight value corresponding to each topology edge are recorded, and invalid topology edges that do not match the starting device node and ending device node at the same time are removed to generate an initial device connection weight topology graph.

[0020] Preferably, the initial module's steps for obtaining the segmented dataset are as follows:

[0021] From the 3D solid model of the power distribution room, match the corresponding 3D equipment entities one by one according to the equipment number field, extract the equipment mass value, length boundary value, width boundary value and height boundary value corresponding to each 3D equipment entity, form the outer shape boundary size box corresponding to each equipment node, construct the corresponding Laplacian matrix according to the connection weight value of the topological vertex in the initial equipment connection weight topology graph, perform eigenvalue decomposition on the Laplacian matrix, divide the equipment node affiliation relationship according to the eigenvalue decomposition result, and obtain the initial module segmentation dataset.

[0022] Preferably, the step of obtaining the hoisting center of gravity deviation distance value is as follows:

[0023] Read the device node number corresponding to each subset in the initial module cutting dataset one by one, retrieve the outer boundary dimension box corresponding to each device node according to the device node number, extract the length boundary value, width boundary value and height boundary value in the outer boundary dimension box, arrange the length boundary value along the horizontal axis of the standard prefabricated cabin virtual base, arrange the width boundary value along the vertical axis of the standard prefabricated cabin virtual base, arrange the height boundary value along the vertical axis of the standard prefabricated cabin virtual base, adjust the horizontal axis position or vertical axis position for device nodes with overlapping boundaries, until the overlap volume between any two outer boundary dimension boxes in the same subset is zero, and obtain the relative three-dimensional centroid coordinate point of each device node in the standard prefabricated cabin virtual base;

[0024] Based on the relative three-dimensional centroid coordinates of each equipment node within the virtual base of the standard prefabricated cabin, the equipment mass values ​​corresponding to each equipment node are retrieved according to the equipment node number. The equipment mass values ​​of each equipment node within the same subset are associated with the corresponding relative three-dimensional centroid coordinates. The horizontal coordinate components, vertical coordinate components, and vertical coordinate components are weighted and accumulated separately. Then, the weighted accumulation result is normalized according to the accumulated result of all equipment mass values ​​within the same subset to obtain the comprehensive centroid coordinates of each subset.

[0025] Read the length, width, and height boundary values ​​of the standard prefabricated module virtual base, and determine the lateral, longitudinal, and vertical center coordinates of the standard prefabricated module virtual base respectively, forming the geometric center coordinate point of the standard prefabricated module virtual base. Calculate the lateral, longitudinal, and vertical coordinate differences between the comprehensive center of gravity coordinate points of each subset and the geometric center coordinate point of the standard prefabricated module virtual base, respectively. Determine the three-dimensional spatial distance based on the lateral, longitudinal, and vertical coordinate differences, and obtain the hoisting center of gravity deviation distance value.

[0026] Preferably, the step of obtaining the updated device connection weight topology map is as follows:

[0027] The hoisting center of gravity deviation distance value corresponding to each subset is read item by item. The hoisting center of gravity deviation distance value is compared with the preset hoisting safety threshold value item by item according to the subset number. The subset with the hoisting center of gravity deviation distance value greater than the preset hoisting safety threshold value is marked as the subset to be fed back. The subset with the hoisting center of gravity deviation distance value less than or equal to the preset hoisting safety threshold value is marked as the stable subset. The subset number, equipment node number and hoisting center of gravity deviation distance value corresponding to the subset to be fed back are retained to obtain the target subset.

[0028] The equipment mass value, relative three-dimensional centroid coordinate point, and geometric center coordinate point of the standard prefabricated cabin virtual base are read one by one for each equipment node in the target subset. The lateral offset, longitudinal offset, and vertical offset of each equipment node are calculated respectively. The equipment mass value is associated with the lateral offset, longitudinal offset, and vertical offset respectively. A lateral moment component, a longitudinal moment component, and a vertical moment component are formed for each equipment node. The results of the combined results of the moment components in the three directions are sorted to obtain the moment contribution of each equipment node relative to the geometric center coordinate point of the standard prefabricated cabin virtual base.

[0029] The critical device nodes that cause imbalance are selected according to the ranking of torque contribution. All adjacent nodes corresponding to the critical device nodes in the initial device connection weight topology graph are retrieved. The connection weight values ​​between the critical device nodes and their adjacent nodes are read. The connection weight values ​​are then reduced in a targeted manner according to the ranking of torque contribution. The reduced connection weight values ​​are written back to the corresponding topology edges. The connection weight values ​​between the unselected device nodes are retained, and an updated device connection weight topology graph is generated.

[0030] Preferably, the step of obtaining the prefabricated cabin equipment cluster is as follows:

[0031] The device node number, topological edge connection relationship, and connection weight value of the updated device connection weight topology graph are read one by one. A matrix row and column correspondence relationship is established according to the device node number. The connection weight value of each device node to other device nodes is filled into the corresponding matrix position. The total connection weight values ​​associated with each device node are accumulated and filled into the corresponding diagonal position to form the Laplacian matrix corresponding to the updated device connection weight topology graph. The matrix elements in the Laplacian matrix are decomposed into eigenvalues. The device node affiliation identifier is read according to the order of eigenvalues ​​to obtain the updated module segmentation dataset.

[0032] Read the device node number corresponding to each subset in the updated module cutting dataset one by one, retrieve the outer boundary dimension frame and equipment mass value according to the device node number, map the outer boundary dimension frame to the virtual base of the standard prefabricated cabin, check the occupied boundary of the outer boundary dimension frame according to the horizontal axis, vertical axis and vertical axis respectively, adjust the layout position of the outer boundary dimension frames with overlapping volume one by one until the overlapping volume between any two outer boundary dimension frames in the same subset is equal to zero, read the relative three-dimensional centroid coordinate point of each device node, calculate the comprehensive centroid coordinate point of each subset in combination with the equipment mass value, accumulate the three-dimensional coordinate difference between the comprehensive centroid coordinate point and the geometric center coordinate point of the standard prefabricated cabin virtual base to obtain the hoisting centroid deviation distance value;

[0033] The hoisting center of gravity deviation value corresponding to each subset is read item by item. The hoisting center of gravity deviation value is compared with the preset hoisting safety threshold according to the subset number. If the hoisting center of gravity deviation value corresponding to all subsets is less than or equal to the preset hoisting safety threshold, the device node affiliation relationship corresponding to each subset in the current updated module segmentation dataset is determined as an output cluster. If the hoisting center of gravity deviation value corresponding to any subset is greater than the preset hoisting safety threshold, the current iteration number, the change amplitude of the connection weight value, and the number of device node affiliation changes are read. It is determined whether the current iteration number has reached the preset iteration limit, whether the change amplitude of the connection weight value is less than the preset change amplitude threshold, and whether the number of device node affiliation changes is equal to zero. If any of the judgment results are satisfied, the device node affiliation relationship corresponding to each subset in the current updated module segmentation dataset is determined as an output cluster, and the balanced prefabricated cabin equipment cluster is output.

[0034] Preferably, the steps for obtaining the prefabricated module candidate location dataset are as follows:

[0035] Read the equipment node number inside each of the balanced prefabricated cabin equipment groups one by one, retrieve the corresponding outer boundary dimension frame and layout position parameters according to the equipment node number, read the horizontal minimum boundary, horizontal maximum boundary, vertical minimum boundary, vertical maximum boundary, vertical minimum boundary and vertical maximum boundary of the outer boundary dimension frame respectively, and calculate the occupied boundary of each outer boundary dimension frame in the balanced prefabricated cabin equipment group in combination with the layout position parameters. Take the outer extreme value of all occupied boundaries in the same balanced prefabricated cabin equipment group to generate the group outer boundary dimension frame of each balanced prefabricated cabin equipment group.

[0036] Scan the 3D solid model of the power distribution room, read the wall boundaries, floor boundaries, structural column boundaries, and passageway occupancy boundaries layer by layer, remove the structural column boundaries and passageway occupancy boundaries from the internal space of the 3D solid model of the power distribution room, divide the remaining space into cuboids according to the horizontal continuous range, vertical continuous range, and vertical continuous range, record the horizontal start and end coordinates, vertical start and end coordinates, and vertical start and end coordinates of each cuboid cavity coordinate item, remove cuboid cavity coordinate items with zero dimension in any direction, and generate a linked list of usable 3D blank areas;

[0037] The length, width, and height values ​​of the grouped outer boundary size frame are read one by one. The length, width, and height values ​​of each cuboid cavity coordinate item in the available 3D blank area list are read one by one. It is determined whether the length, width, and height values ​​of the cuboid cavity coordinate items are greater than or equal to the length, width, and height values ​​of the grouped outer boundary size frame, respectively. The cuboid cavity coordinate items that meet the size matching are marked as candidate regions. Then, the horizontal coordinate range, vertical coordinate range, and vertical coordinate range between candidate regions are overlapped and verified. Candidate regions with zero overlapping volume are retained as placement items, generating a prefabricated cabin candidate position dataset.

[0038] Preferably, the steps for obtaining the three-dimensional assembly coordinate set of the intelligent power distribution room are as follows:

[0039] Based on the prefabricated cabin candidate location dataset, the candidate placement location number, candidate location coordinates, candidate area length, candidate area width, and candidate area height values ​​corresponding to each balanced prefabricated cabin equipment group are read one by one. A placement location selection sequence is established according to the group number of the balanced prefabricated cabin equipment group. The candidate placement locations of each balanced prefabricated cabin equipment group are written into the placement location selection sequence item by item. The candidate location coordinates between different balanced prefabricated cabin equipment groups are combined and enumerated. The combination results where the same candidate location coordinates are repeatedly occupied are eliminated, and several global placement combination schemes are generated.

[0040] Read the group number of adjacent balanced prefabricated cabin equipment in each of the global placement combination schemes one by one, retrieve the three-dimensional space normal vector of the local docking flange, the coordinates of the center point of the local docking flange and the coordinates of the candidate position of the adjacent balanced prefabricated cabin equipment group, perform coordinate offset conversion on the coordinates of the center point of the local docking flange according to the candidate position coordinates, and perform direction conversion on the three-dimensional space normal vector of the local docking flange according to the candidate placement posture to obtain the three-dimensional space normal vector of the global docking flange and the coordinates of the center point of the global docking flange.

[0041] The three-dimensional distance difference between the center point coordinates of the global docking flanges of adjacent prefabricated equipment groups is calculated item by item. The angle between the three-dimensional spatial normal vectors of the global docking flanges of adjacent prefabricated equipment groups is calculated item by item. The three-dimensional distance difference and angle are written into the corresponding matrix positions according to the global placement combination scheme number. The module docking posture error matrix corresponding to the global placement combination scheme is constructed. The absolute difference between the angle value and 180 degrees in the module docking posture error matrix is ​​judged item by item. The three-dimensional distance difference is judged item by item. The target global placement combination scheme that meets the judgment conditions is locked. The group number, candidate position coordinates, candidate placement posture and the center point coordinates of the global docking flange in the target global placement combination scheme are extracted and integrated to form three-dimensional placement parameters, and a three-dimensional assembly coordinate set of the intelligent power distribution room is generated.

[0042] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0043] This invention transforms electrical connections from simple graphical representations into calculable topological relationships by reading equipment nodes, busbar connection lines, and cable connection lines from the electrical single-line diagram of a power distribution room and assigning them different connection weight values. The strength of electrical connections between equipment can then be used in subsequent grouping decisions. By extracting equipment mass values ​​and external boundary dimensions, and combining this with Laplace matrix decomposition to obtain an initial module segmentation dataset, equipment grouping no longer depends solely on spatial adjacency but is simultaneously influenced by the tightness of electrical connections, equipment physical dimensions, and equipment mass distribution. By mapping external boundary dimensions onto a standard prefabricated cabin virtual base for local three-dimensional orthogonal arrangement and calculating the comprehensive centroid coordinates based on relative three-dimensional centroid coordinates and equipment mass values, the deviation distance of the hoisting centroid can be directly quantified during the equipment grouping stage, avoiding the risk of off-center loading in subsequent hoisting stages. By analyzing the torque contribution of target subsets exceeding the preset hoisting safety threshold and selectively reducing the connection weight values ​​between key equipment nodes and adjacent nodes, the connection... The weights can be adjusted based on spatial center of gravity feedback to achieve a synergistic balance between electrical connection constraints and hoisting safety constraints. By re-cutting, re-layouting, and recalculating the hoisting center of gravity deviation distance, and outputting balanced prefabricated cabin equipment clusters when the preset hoisting safety threshold or convergence iteration constraint conditions are met, the stability and feasibility of the clustering results can be improved. By fusing the layout position parameters and the outer boundary dimension frame to generate the cluster outer boundary dimension frame, and extracting the cuboid cavity coordinate items after removing the structural columns and passageway occupied areas from the 3D solid model of the power distribution room, the candidate position selection can avoid actual unoccupied space. By performing a full permutation and combination of candidate placement positions, and mapping the 3D spatial normal vector of the local docking flange and the coordinates of the center point of the local docking flange to global pose parameters, and then selecting the target global placement combination scheme that meets the preset assembly tolerance extreme value based on the 3D distance difference and included angle value, the unified verification of space occupation, hoisting center of gravity, docking posture and assembly tolerance can be completed before outputting the 3D assembly coordinate set of the intelligent power distribution room, thereby improving space utilization. Attached Figure Description

[0044] Figure 1 A schematic diagram showing the distance of the hoisting center of gravity deviation;

[0045] Figure 2 This is a schematic diagram of the pose error matrix for module docking. Detailed Implementation

[0046] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0047] Please see Figure 1-2This invention provides a technical solution: a method for optimizing the spatial layout of an intelligent power distribution room based on BIM technology, comprising the following steps:

[0048] An initial equipment connection weight topology map is generated based on the first connection weight values ​​of the equipment nodes and busbar connection lines and the second connection weight values ​​of the cable connection lines in the electrical single-line diagram of the power distribution room; the equipment mass values ​​and outer boundary dimensions of the 3D solid model of the power distribution room are extracted, and an initial module cutting dataset containing subsets is obtained based on the Laplacian matrix;

[0049] The relative three-dimensional centroid coordinates are obtained by locally orthogonally laying out the outer boundary dimension frame within the virtual base of the standard prefabricated cabin; the comprehensive centroid coordinates are calculated by combining the equipment mass value with the relative three-dimensional centroid coordinates, and the hoisting centroid deviation distance is obtained by comparing the comprehensive centroid coordinates with the geometric center coordinates.

[0050] For the target subset whose hoisting center of gravity deviation distance exceeds the preset hoisting safety threshold, the connection weight values ​​of key equipment nodes and adjacent nodes are reduced based on the torque contribution, and an updated equipment connection weight topology map is generated.

[0051] Obtain the updated module cutting dataset, re-execute local 3D orthogonal nesting and calculate the hoisting center of gravity deviation distance until it meets the preset hoisting safety threshold, and output the balanced prefabricated cabin equipment grouping;

[0052] Combine the layout position parameters and the outer boundary size box to generate the group outer boundary size box; extract the coordinate items of the cuboid cavity from the 3D solid model of the power distribution room by removing the structural columns and the passage occupied area, and generate a list of available 3D blank areas; match the group outer boundary size box with the list of available 3D blank areas to extract candidate areas and generate a dataset of candidate locations for prefabricated cabins.

[0053] Generate a global placement combination scheme from candidate placement position combinations; map the local flange 3D space normal vector and local flange center point coordinates to the global flange 3D space normal vector and global flange center point coordinates based on the candidate position coordinates; calculate the 3D distance difference and included angle values ​​to construct the module docking pose error matrix; lock the target global placement combination scheme that meets the preset assembly tolerance extreme value, extract the 3D placement parameters, and generate the intelligent power distribution room 3D assembly coordinate set.

[0054] The steps for obtaining the initial device connection weight topology graph are as follows:

[0055] Read the electrical single-line diagram of the power distribution room, analyze the equipment element identifier, equipment number field and port connection position of the electrical single-line diagram line by line, extract all equipment nodes with independent access ports, distinguish bus connection lines and cable connection lines according to the layer attributes, line type attributes and port bridging relationships of the connection lines, assign a first connection weight value to the bus connection lines and a second connection weight value to the cable connection lines to form the equipment node connection weight identifier;

[0056] Based on the device node connection weight identifier, all device nodes are converted into topology vertices one by one, and bus connection lines carrying the first connection weight value and cable connection lines carrying the second connection weight value are converted into topology edges one by one. The starting device node, ending device node and connection weight value corresponding to each topology edge are recorded. Invalid topology edges that do not match the starting device node and ending device node at the same time are removed, and an initial device connection weight topology graph is generated.

[0057] Specifically, the process involves reading the electrical single-line diagram of the power distribution room, identifying and traversing all graphic elements in the diagram, and extracting the graphic element attribute data for each element. This includes equipment element identifiers, such as the national standard symbols for circuit breakers and transformers, the equipment number field of the associated text object, and the port connection location information connected by leads. All equipment elements with independent electrical access ports are filtered out and compiled into an initial list containing all equipment nodes. Then, all connection line elements in the diagram are traversed. Based on preset layer naming rules (e.g., lines with the layer name "BUSBAR" are identified as busbar connection lines, and lines with the layer name "CABLE" are identified as cable connection lines), line type attributes (e.g., solid lines represent busbars, dashed lines represent cables), and whether the line endpoints cross different equipment port locations, a double verification is performed to distinguish between busbar connection lines and cable connection lines. Different weight values ​​are assigned to these two types of connection lines. The weight value setting here is based on engineering experience; considering the rigid physical characteristics of busbar connections and the strong constraints on equipment layout, their weight should be much greater than that of flexible cable connections. For example, a first connection weight value is set. A value of 10.0 represents a busbar connection, setting the second connection weight value. A value of 1.0 represents a cable connection. This 10:1 ratio is used in the subsequent community partitioning algorithm to ensure that devices with strong electrical connections are preferentially assigned to the same module. Each device node and its associated connection lines are bound to the corresponding weight value to form a device node connection weight identifier.

[0058] Based on the device node connection weight identifier, the topology graph construction process is initiated. Each device node in the list of all device nodes extracted in the previous step is instantiated as an independent topology vertex in the graph data structure, and its device number is used as a unique identifier. Then, all connection relationships recorded in the device node connection weight identifier are traversed, and each relationship carrying the first connection weight value is processed. Busbar connection lines and carrying second connection weight values The cable connection lines are converted into a weighted undirected topological edge connecting the corresponding two topological vertices. During the conversion process, detailed attributes are recorded for each generated topological edge, including the starting device node number, the ending device node number, and the corresponding connection weight value. or A temporary set of topological edges is constructed. After the transformation of all connecting lines is completed, an integrity check is performed on this temporary set of topological edges. Specifically, the starting and ending device nodes of each topological edge are checked one by one to confirm that both nodes exist in the created set of topological vertices. If it is found that any endpoint of a topological edge cannot find a corresponding device node in the vertex set, for example, due to errors in the original CAD drawing causing the line to be suspended or connected to non-device primitives, then the topological edge is considered invalid data and removed from the set. After this purification step, all the remaining valid topological vertices and topological edges together constitute the initial device connection weight topological graph.

[0059] The steps to obtain the initial module split dataset are as follows:

[0060] From the 3D solid model of the power distribution room, match the corresponding 3D equipment entities one by one according to the equipment number field, extract the equipment mass value, length boundary value, width boundary value and height boundary value of each 3D equipment entity to form the outline boundary size box of each equipment node, construct the corresponding Laplacian matrix according to the connection weight value of the topological vertex in the initial equipment connection weight topology diagram, perform eigenvalue decomposition on the Laplacian matrix, divide the equipment node affiliation relationship according to the eigenvalue decomposition result, and obtain the initial module segmentation dataset.

[0061] Specifically, from the 3D solid model of the power distribution room, all 3D equipment entities in the model are traversed through the API interface or IFC file parser. Using the equipment node number obtained in the previous step as a query index, an attribute-based search is performed in the 3D model to match each 3D equipment entity. After a successful match, its predefined physical parameters are extracted from the entity's attribute set, including the equipment mass value and the three dimensions of its spatial boundary frame, namely the length boundary value, the width boundary value, and the height boundary value. This physical information is associated and stored with the corresponding equipment node. Based on the initial equipment connection weight topology graph, a Laplace matrix for community discovery is constructed. The matrix is ​​constructed as follows: ,in, Given a weighted adjacency matrix, if the node and nodes If there are connections between them, then the matrix elements The value is equal to the connection weight between them ( or If there are no connections, the value is 0. The degree matrix is ​​a diagonal matrix, and its diagonal elements are... For nodes The sum of the weights of all connected edges, after construction, is used to construct the Laplace matrix. Eigenvalue decomposition is performed to calculate all eigenvalues ​​and their corresponding eigenvectors. Based on spectral clustering theory, the device nodes are partitioned using the eigenvector (Fiedler vector) corresponding to the second smallest eigenvalue (Fiedler value). Specifically, all component values ​​in the Fiedler vector are grouped according to their sign, thus initially dividing all device nodes into two subsets. If more than two subsets are needed, the first... The eigenvectors corresponding to the smallest eigenvalues ​​form an eigenvector matrix, and the K-means clustering algorithm is applied to the row vectors of this matrix for clustering. The number of modules is predetermined, and the affiliation of each device node is determined based on the clustering results to obtain the initial module segmentation dataset.

[0062] The steps to obtain the hoisting center of gravity deviation distance value are as follows:

[0063] Read the device node number corresponding to each subset in the initial module cutting dataset one by one, retrieve the outer boundary dimension box corresponding to each device node according to the device node number, extract the length boundary value, width boundary value and height boundary value in the outer boundary dimension box, arrange the length boundary value along the horizontal axis of the standard prefabricated cabin virtual base, arrange the width boundary value along the vertical axis of the standard prefabricated cabin virtual base, arrange the height boundary value along the vertical axis of the standard prefabricated cabin virtual base, adjust the horizontal axis position or vertical axis position for device nodes with overlapping boundaries, until the overlap volume between any two outer boundary dimension boxes in the same subset is zero, and obtain the relative three-dimensional centroid coordinate point of each device node in the standard prefabricated cabin virtual base;

[0064] Based on the relative three-dimensional centroid coordinates of each equipment node within the virtual base of the standard prefabricated cabin, the equipment mass values ​​corresponding to each equipment node are retrieved according to the equipment node number. The equipment mass values ​​of each equipment node within the same subset are associated with the corresponding relative three-dimensional centroid coordinates. The horizontal coordinate components, vertical coordinate components, and vertical coordinate components are weighted and accumulated separately. Then, the weighted accumulation result is normalized according to the accumulated result of all equipment mass values ​​within the same subset to obtain the comprehensive centroid coordinates of each subset.

[0065] Read the length, width, and height boundary values ​​of the standard prefabricated module virtual base, and determine the lateral, longitudinal, and vertical center coordinates of the standard prefabricated module virtual base to form the geometric center coordinate point of the standard prefabricated module virtual base. Calculate the lateral, longitudinal, and vertical coordinate differences between the comprehensive center of gravity coordinate points of each subset and the geometric center coordinate point of the standard prefabricated module virtual base. Determine the three-dimensional spatial distance based on the lateral, longitudinal, and vertical coordinate differences to obtain the hoisting center of gravity deviation distance value.

[0066] Specifically, the device node numbers in each subset of the initial module's segmented dataset are read one by one. For each subset, its internal 3D nesting process is initiated. Based on the device node numbers within the subset, the bounding box data corresponding to each node, including length, width, and height values, is retrieved from the previously acquired equipment physical information database. Nesting is then performed within a pre-defined standard prefabricated cabin virtual base in 3D space. This virtual base has fixed length, width, and height dimensions, for example, 12m x 3m x During the layout, the orthogonal placement principle is followed, meaning that the length of the equipment's outer boundary dimension frame is uniformly arranged along the horizontal X-axis of the virtual base, the width along the vertical Y-axis, and the height along the vertical Z-axis. The layout algorithm adopts a heuristic placement strategy. For example, starting from a corner of the virtual base, the equipment is placed one by one. When it is detected that the outer boundary dimension frames of newly placed equipment and already placed equipment overlap in three-dimensional space, the horizontal axis (X-axis) is first attempted to be translated and adjusted. If the horizontal space is insufficient, the vertical axis (Y-axis) is adjusted until a position without any overlap is found. This process is repeated until all equipment in the subset is placed in the virtual base without overlap. After the layout is completed, the coordinates of the geometric center point of the outer boundary dimension frame of each equipment in the coordinate system of the virtual base are determined. This coordinate point is used as the relative three-dimensional centroid coordinate point of this equipment in the prefabricated module.

[0067] Based on the relative three-dimensional centroid coordinates of each equipment node within the virtual base of the standard prefabricated cabin, a precise calculation of the comprehensive centroid is performed for each subset of the initial module segmentation dataset. First, based on the equipment node number within the subset, the equipment mass value corresponding to each device is retrieved from the equipment physical information database. Then, the mass value of each device within the same subset is used as the weight of its relative three-dimensional centroid coordinates, and a weighted average is calculated for the coordinate components of all devices. The specific calculation process is as follows: For a subset containing... A subset of devices, whose composite centroid coordinates This is achieved by performing mass-weighted accumulation on the three coordinate components: horizontal, vertical, and longitudinal. First, the sum of the products of the mass and coordinate of all devices on each coordinate axis is calculated; that is, the mass-weighted accumulation value of the horizontal coordinate component is... Vertical Vertical ,in It is the first The quality of each piece of equipment. It is its relative three-dimensional centroid coordinate point, and then the total mass of all devices within this subset is calculated. Finally, the weighted cumulative mass result is normalized using the total mass, that is, the weighted cumulative mass value in each direction is divided by the total mass to obtain the comprehensive centroid coordinates of each subset.

[0068] Read the dimensional parameters of the standard prefabricated cabin virtual base itself, that is, its total length value. Total width value and total height value Based on these dimensional parameters, the geometric center coordinates of the virtual base are calculated, and the coordinates of this point are: This point represents the physical center of the prefabricated module under an ideal uniform mass distribution. The comprehensive center-of-gravity coordinates of each subset calculated in the previous step are compared with the geometric center coordinates of this standard prefabricated module's virtual base to quantify the degree of center-of-gravity offset during hoisting. For each subset, the lateral coordinate difference, longitudinal coordinate difference, and vertical coordinate difference are calculated separately. Specifically, if the comprehensive center-of-gravity coordinates of the subset are... The difference in horizontal coordinates is The difference in vertical coordinates is The vertical coordinate difference is Finally, based on the coordinate differences in these three directions, the three-dimensional straight-line distance between the comprehensive center of gravity and the geometric center is determined using the spatial two-point distance calculation method. This distance is the hoisting center of gravity deviation value of this subset (prefabricated module).

[0069] The steps to obtain the updated device connection weight topology map are as follows:

[0070] Read the hoisting center of gravity deviation distance value corresponding to each subset item by item. According to the subset number, compare the hoisting center of gravity deviation distance value with the preset hoisting safety threshold item by item. The subset with the hoisting center of gravity deviation distance value greater than the preset hoisting safety threshold is marked as the subset to be fed back. The subset with the hoisting center of gravity deviation distance value less than or equal to the preset hoisting safety threshold is marked as the stable subset. The subset number, equipment node number and hoisting center of gravity deviation distance value corresponding to the subset to be fed back are retained to obtain the target subset.

[0071] The equipment mass value, relative three-dimensional centroid coordinates, and geometric center coordinates of the standard prefabricated cabin virtual base are read one by one for each equipment node in the target subset. The lateral offset, longitudinal offset, and vertical offset of each equipment node are calculated respectively. The equipment mass value is associated with the lateral offset, longitudinal offset, and vertical offset respectively. A lateral moment component, a longitudinal moment component, and a vertical moment component are formed for each equipment node. The results of the combined results of the moment components in the three directions are sorted to obtain the moment contribution of each equipment node relative to the geometric center coordinates of the standard prefabricated cabin virtual base.

[0072] The critical equipment nodes that cause imbalance are selected according to the ranking of torque contribution. All adjacent nodes corresponding to the critical equipment nodes in the initial equipment connection weight topology graph are retrieved. The connection weight values ​​between the critical equipment nodes and their adjacent nodes are read. The connection weight values ​​are then reduced in a targeted manner according to the ranking of torque contribution. The reduced connection weight values ​​are written back to the corresponding topology edges. The connection weight values ​​between the unselected equipment nodes are retained, and an updated equipment connection weight topology graph is generated.

[0073] Specifically, the previously calculated hoisting center of gravity deviation distance values ​​corresponding to each subset are read item by item, and these values ​​are compared with a preset hoisting safety threshold. This hoisting safety threshold is set based on national hoisting safety regulations and the prefabricated module structural strength analysis report. For example, for a prefabricated module of standard weight and size, the tilt angle and force uniformity requirements of its hoisting slings dictate that the center of gravity deviation cannot exceed 5% of the length of its bottom short side. For example, if the width of the prefabricated module is 3000mm, the hoisting safety threshold can be set as follows: During numerical comparison, if the deviation distance of the hoisting center of gravity of a subset is greater than 150mm, the subset is judged as unstable and marked as "subset to be fed back" in the system. Conversely, if its deviation distance is less than or equal to 150mm, it is marked as "stable subset". After judging all subsets, all subsets marked as "subset to be fed back" are filtered out, and the core data related to them are retained, including the unique subset number, the list of all equipment node numbers contained in the subset, and the specific hoisting center of gravity deviation distance that caused it to be marked. After integrating this information, the target subset for subsequent optimization and adjustment is obtained.

[0074] The relevant data for each device node in the target subset is read one by one, including its device mass value and its relative three-dimensional centroid coordinates within the prefabricated cabin. Simultaneously, the geometric center coordinates of the standard prefabricated cabin's virtual base are retrieved. For each device within the target subset, its "contribution" to the overall centroid offset of the subset is calculated individually. First, the offsets of the device's relative three-dimensional centroid coordinates to the prefabricated cabin's geometric center coordinates in the horizontal, vertical, and longitudinal directions are calculated—that is, the differences between the three coordinate components. Then, the device's mass value is multiplied by these three offsets to obtain the device's mass value. The torque components generated on the three axes of horizontal, vertical, and longitudinal are comprehensively reflected in the magnitude of the equipment's impact on the overall imbalance in three-dimensional space. The absolute values ​​of these three torque components are summed or the magnitude of their vector sum is calculated to obtain a comprehensive torque contribution value. The torque contribution values ​​of all equipment nodes in the subset are sorted in descending order. After sorting, the equipment at the top of the list is the "culprit" that contributes the most to the center of gravity deviation problem. The torque contribution of each equipment node relative to the geometric center coordinate point of the standard prefabricated cabin virtual base is obtained.

[0075] Based on the ranking of torque contribution, the top-ranked devices are selected from highest to lowest as the critical device nodes causing the imbalance. The number of selected devices can be a fixed value, such as the top three devices in terms of contribution, or devices whose total contribution accounts for 80% of the overall contribution. After determining the critical device nodes, all adjacent nodes directly connected to these critical device nodes are retrieved from the initial device connection weight topology diagram, and the connection weight values ​​between the critical device nodes and each adjacent node are read. These connection weight values ​​are then targeted for reduction, with the reduction amount related to the torque contribution. For example, for the critical device node with the highest torque contribution, its connection weight values ​​are reduced by a factor of 100. The connection weights between adjacent nodes are reduced by 50%, the next highest contribution is reduced by 30%, and so on. The purpose of this targeted processing is to reduce the tendency for "overweight" or "off-center" devices to be grouped with other devices in the next community partitioning, and encourage them to be partitioned into other subsets to balance the center of gravity. Specifically, the calculated new weight values ​​are written back to the corresponding topological edge attributes in the topology graph. For edges in the topology graph that are not connected to any key device nodes, their connection weight values ​​remain unchanged. After adjusting the relevant weights of key device nodes in all target subsets, an updated device connection weight topology graph is generated.

[0076] The steps for obtaining the prefabricated cabin equipment clusters are as follows:

[0077] Read the device node number, topological edge connection relationship and connection weight value of the updated device connection weight topology graph one by one. Establish the matrix row and column correspondence according to the device node number. Fill the connection weight value of each device node to other device nodes into the corresponding matrix position. Accumulate all the connection weight values ​​associated with each device node and fill them into the corresponding diagonal position to form the Laplacian matrix corresponding to the updated device connection weight topology graph. Perform eigenvalue decomposition on the matrix elements in the Laplacian matrix. Read the device node affiliation identifier according to the eigenvalue arrangement order to obtain the updated module segmentation dataset.

[0078] Read the device node number corresponding to each subset in the updated module cutting dataset one by one, retrieve the outer boundary dimension frame and equipment mass value according to the device node number, map the outer boundary dimension frame to the virtual base of the standard prefabricated cabin, check the occupied boundary of the outer boundary dimension frame according to the horizontal axis, vertical axis and vertical axis respectively, adjust the layout position of the outer boundary dimension frames with overlapping volume one by one until the overlapping volume between any two outer boundary dimension frames in the same subset is equal to zero, read the relative three-dimensional centroid coordinate point of each device node, calculate the comprehensive centroid coordinate point of each subset in combination with the equipment mass value, accumulate the three-dimensional coordinate difference between the comprehensive centroid coordinate point and the geometric center coordinate point of the standard prefabricated cabin virtual base to obtain the hoisting centroid deviation distance value;

[0079] Read the hoisting center of gravity deviation distance value corresponding to each subset item by item. Compare the hoisting center of gravity deviation distance value with the preset hoisting safety threshold according to the subset number. If the hoisting center of gravity deviation distance value corresponding to all subsets is less than or equal to the preset hoisting safety threshold, then the device node affiliation relationship corresponding to each subset in the currently updated module segmentation dataset is determined as an output cluster. If the hoisting center of gravity deviation distance value corresponding to any subset is greater than the preset hoisting safety threshold, then read the current iteration number, the change amplitude of the connection weight value, and the number of changes in device node affiliation. Determine whether the current iteration number has reached the preset iteration limit, whether the change amplitude of the connection weight value is less than the preset change amplitude threshold, and whether the number of changes in device node affiliation is equal to zero. If any of the judgment results are satisfied, the device node affiliation relationship corresponding to each subset in the currently updated module segmentation dataset is determined as an output cluster, and the balanced prefabricated cabin equipment cluster is output.

[0080] Specifically, the updated device connection weight topology graph is read one by one, including device node numbers, topology edge connections, and updated connection weight values. Based on this information, the Laplace matrix is ​​reconstructed. First, an N×N zero matrix is ​​created based on the total number of device nodes N, and a correspondence is established between device node numbers and matrix row and column indices. Then, each edge in the topology graph is traversed, connecting the nodes... and nodes The connection weight values ​​are filled into the matrix. and The system constructs a weighted adjacency matrix based on the location of each node. Simultaneously, it calculates the sum of all connection weights associated with each node and inserts this sum into the corresponding diagonal of the degree matrix. Subtracting the adjacency matrix from the degree matrix yields the updated Laplacian matrix corresponding to the updated device connection weight topology. This updated Laplacian matrix is ​​then subjected to eigenvalue decomposition, and, as in the initial partitioning, spectral clustering or K-means clustering analysis is performed based on its eigenvectors. According to the device node affiliation identifiers output by the clustering analysis, all device nodes are reassigned to their respective subsets, resulting in the updated module segmentation dataset.

[0081] The process involves sequentially reading the device node numbers corresponding to each subset in the updated module segmentation dataset, and repeating the 3D layout and center of gravity verification process for each updated subset. First, the device node numbers are used to retrieve their external boundary dimensions and mass values. These external boundary dimensions are then mapped onto the internal space of the standard prefabricated cabin virtual base. During the layout process, the orthogonality principle of the horizontal, vertical, and longitudinal axes is strictly followed, and the occupancy of each device's external boundary dimensions is checked in real time. If any newly placed device is found to have overlapping volume with existing devices, it is immediately corrected by adjusting the horizontal or vertical alignment. The layout position is adjusted until the overlap volume between any two outer boundary dimension frames within the subset is exactly zero. After completing the non-overlapping layout, the updated relative three-dimensional centroid coordinates of each device are recorded. Combined with the mass value of each device, the updated comprehensive centroid coordinates of each subset are calculated again using the mass-weighted average method. Finally, the three-dimensional spatial distance between this new comprehensive centroid coordinate and the geometric center coordinate of the standard prefabricated cabin virtual base is calculated. That is, the sum of the squares of the differences between the two points on the three coordinate axes is taken as the square root to obtain the latest hoisting centroid deviation distance value of each subset in the current iteration.

[0082] The system reads the hoisting center of gravity deviation value calculated for each subset after the new iteration and compares these values ​​with a preset hoisting safety threshold (e.g., 150mm). If, in this check, the hoisting center of gravity deviation value for all subsets is less than or equal to the threshold, it indicates that the current equipment grouping scheme has met the requirements for hoisting safety, and the iteration process has successfully converged. At this point, the affiliation relationship of each subset's equipment node recorded in the updated module segmentation dataset is finally determined as an output grouping scheme, and the iteration ends. If the check finds that at least one subset still has a hoisting center of gravity deviation value greater than the preset hoisting safety threshold, it is necessary to determine whether the iteration termination condition is met. The body judgment includes: first, checking whether the current iteration count has reached the preset iteration limit, for example, set to 50 times, to prevent infinite loops; second, calculating the overall change of the connection weight values ​​in this iteration and determining whether it is less than a very small preset change threshold, for example, 0.01, indicating that the weight adjustment has tended to stabilize; finally, comparing the module segmentation dataset of this round with that of the previous round and determining whether the number of changes in the affiliation of device nodes is equal to zero, indicating that the community structure has no longer changed. When any of the above termination conditions are met, even if the center of gravity deviation problem is not completely resolved, the iteration is forcibly terminated, and the latest device node affiliation relationship is output as the optimal solution under the current constraints, forming a balanced prefabricated cabin equipment cluster.

[0083] The steps for obtaining the candidate location dataset for prefabricated modules are as follows:

[0084] Read the equipment node number within each balanced prefabricated cabin equipment group one by one, retrieve the corresponding outer boundary dimension frame and layout position parameters according to the equipment node number, read the horizontal minimum boundary, horizontal maximum boundary, vertical minimum boundary, vertical maximum boundary, vertical minimum boundary, and vertical maximum boundary of the outer boundary dimension frame respectively, and calculate the occupied boundary of each outer boundary dimension frame within the balanced prefabricated cabin equipment group in combination with the layout position parameters. Take the outer extreme value of all occupied boundaries within the same balanced prefabricated cabin equipment group to generate the group outer boundary dimension frame of each balanced prefabricated cabin equipment group.

[0085] Scan the 3D solid model of the power distribution room, read the wall boundaries, floor boundaries, structural column boundaries, and passageway occupancy boundaries layer by layer, remove the structural column boundaries and passageway occupancy boundaries from the internal space of the 3D solid model of the power distribution room, divide the remaining space into cuboids according to the horizontal continuous range, vertical continuous range, and vertical continuous range, record the horizontal start and end coordinates, vertical start and end coordinates, and vertical start and end coordinates of each cuboid cavity coordinate item, remove cuboid cavity coordinate items with zero dimension in any direction, and generate a linked list of usable 3D blank areas;

[0086] The length, width, and height values ​​of the grouped outer boundary size box are read one by one. The length, width, and height values ​​of each cuboid cavity coordinate item in the available 3D blank area list are read one by one. It is determined whether the length, width, and height values ​​of the cuboid cavity coordinate items are greater than or equal to the length, width, and height values ​​of the grouped outer boundary size box, respectively. The cuboid cavity coordinate items that meet the size matching are marked as candidate regions. Then, the horizontal coordinate range, vertical coordinate range, and vertical coordinate range of the candidate regions are overlapped and verified. The candidate regions with zero overlapping volume are retained as placement items, and a prefabricated cabin candidate position dataset is generated.

[0087] Specifically, each balanced prefabricated cabin equipment group is read one by one, and the geometric envelope calculation is performed on the equipment layout within the group. First, all equipment node numbers contained in the group are traversed. Based on the number, the outer boundary size frame and its layout position parameters of each equipment after layout within the group are retrieved from the database. For each outer boundary size frame, its six boundary values ​​in the group coordinate system are read: horizontal minimum boundary, horizontal maximum boundary, vertical minimum boundary, vertical maximum boundary, vertical minimum boundary, and vertical maximum boundary. These boundary values ​​are the absolute space occupancy boundaries calculated by combining the equipment's own size and its layout position parameters within the group. Then, the occupancy boundaries of all equipment in the same balanced prefabricated cabin equipment group are compared to find the minimum value among all horizontal minimum boundaries, the maximum value among all horizontal maximum boundaries, and the corresponding extreme values ​​in the vertical and longitudinal directions. These six extreme values ​​together define a three-dimensional cuboid space that can completely enclose all equipment in the group. This cuboid is the group outer envelope size frame of the balanced prefabricated cabin equipment group.

[0088] The established 3D solid model of the power distribution room is scanned, and a space availability analysis program is initiated. The program scans layer by layer, starting from the ground floor and working upwards, reading and parsing the structural information of each floor. This includes the geometric data of wall boundaries, floor boundaries, and structural column boundaries, as well as the boundaries of passageway occupancy areas pre-defined according to design specifications or fire safety requirements, such as main inspection passages and equipment operation and maintenance passages with a width of no less than 1.2 meters. After obtaining the complete spatial outline of the power distribution room, a spatial Boolean operation is performed to "subtract" all identified structural column entities and virtual entities of passageway occupancy areas from the internal space volume of the power distribution room, resulting in one or more... For the remaining available space of regular shapes, a three-dimensional space segmentation algorithm, such as octree segmentation or scan-line-based cuboid partitioning algorithm, is applied to these remaining available spaces. The cuboid fitting segmentation is maximized according to the horizontal, vertical and vertical continuous range of the space, decomposing the irregular space into a series of non-overlapping cuboid cavities. For each segmented cuboid cavity, its horizontal start and end coordinates, vertical start and end coordinates and vertical start and end coordinates in the global coordinate system are recorded to form cuboid cavity coordinate items. Finally, all coordinate items are traversed, and those tiny cavities with a size of zero or less than the minimum installation requirement of the equipment in any direction are eliminated to generate a linked list of available three-dimensional blank areas.

[0089] The process involves sequentially reading the outer boundary dimension frames of each previously generated prefabricated balanced cabin equipment group, extracting their length, width, and height values. Simultaneously, iterating through the available 3D blank area linked list for each cuboid cavity coordinate item and reading its corresponding length, width, and height values, performing a size matching filtering process. Specifically, the length, width, and height values ​​of the group's outer boundary dimension frame are compared with the length, width, and height values ​​of the cuboid cavity. It's important to note that since the groups can be rotated, the two cases of interchanged length and width of the group's outer boundary dimension frame need to be considered. The goal is to ensure that the length, width, and height values ​​of the cuboid cavity are simultaneously greater than or equal to the length, width, and height values ​​of the group's outer boundary dimension frame in any rotational direction. The length, width, and height values ​​are used to determine if the coordinates of the cuboid cavity meet the size matching condition and it is marked as a candidate region for that group. After this screening is completed for all groups, a list of candidate regions for each group is obtained. These candidate regions are then subjected to a final validity check. Since a group can only be placed in one candidate region, and the candidate regions of different groups may overlap in space, it is necessary to verify the overlap of the horizontal, vertical, and longitudinal coordinate ranges between the candidate regions. Candidate region combinations with non-zero overlapping volumes are eliminated. The final set of candidate regions that can be placed without overlap for all groups is used as the final placement item, generating a prefabricated cabin candidate location dataset.

[0090] The steps for obtaining the 3D assembly coordinate set of the intelligent power distribution room are as follows:

[0091] Based on the prefabricated cabin candidate location dataset, the candidate placement location number, candidate location coordinates, candidate area length, candidate area width, and candidate area height values ​​corresponding to each balanced prefabricated cabin equipment cluster are read one by one. A placement location selection sequence is established according to the cluster number of the balanced prefabricated cabin equipment cluster. The candidate placement locations of each balanced prefabricated cabin equipment cluster are written into the placement location selection sequence item by item. The candidate location coordinates between different balanced prefabricated cabin equipment clusters are combined and enumerated. The combination results where the same candidate location coordinates are repeatedly occupied are eliminated, and several global placement combination schemes are generated.

[0092] Read the group number of adjacent balanced prefabricated cabin equipment in each global placement combination scheme one by one, retrieve the three-dimensional space normal vector of the local docking flange, the coordinates of the center point of the local docking flange and the coordinates of the candidate position of the adjacent balanced prefabricated cabin equipment group, perform coordinate offset conversion on the coordinates of the center point of the local docking flange according to the candidate position coordinates, and perform direction conversion on the three-dimensional space normal vector of the local docking flange according to the candidate placement posture to obtain the three-dimensional space normal vector of the global docking flange and the coordinates of the center point of the global docking flange.

[0093] The three-dimensional distance difference between the center point coordinates of the global docking flanges of adjacent prefabricated equipment groups is calculated item by item. The angle between the three-dimensional space normal vectors of the global docking flanges of adjacent prefabricated equipment groups is calculated item by item. The three-dimensional distance difference and angle values ​​are written into the corresponding matrix positions according to the global placement combination scheme number. The module docking posture error matrix corresponding to the global placement combination scheme is constructed. The absolute difference between the angle value in the module docking posture error matrix and 180 degrees is judged item by item. The three-dimensional distance difference is judged item by item. The target global placement combination scheme that meets the judgment conditions is locked. The group number, candidate position coordinates, candidate placement posture and global docking flange center point coordinates in the target global placement combination scheme are extracted and integrated to form three-dimensional placement parameters, generating a three-dimensional assembly coordinate set for the intelligent power distribution room.

[0094] Specifically, based on the prefabricated cabin candidate location dataset, all possible global layout schemes are systematically generated. First, each balanced prefabricated cabin device is grouped, and the corresponding candidate placement location numbers, coordinates, and candidate area size information are read. According to the unique numbers of the groups, such as group 1, group 2, and group 3, a selection sequence of placement locations is established. Then, through looping and recursion, a candidate location is selected for the first group in the sequence, then for the second group, and so on, until a placement location is selected for all groups, which constitutes a combination. During this process, the validity of each generated combination is checked, that is, it is determined whether multiple groups are assigned to the same candidate location coordinates. If such duplicate occupation exists, the combination is invalid and is eliminated. By enumerating all permutations of all candidate locations for all groups and eliminating all invalid combinations, a series of feasible global placement combination schemes are finally obtained.

[0095] Each global placement combination scheme is read one by one, and the coordinates of the docking interfaces of the spatially adjacent balanced prefabricated cabin equipment groups in the scheme are converted. First, the local docking flange information for inter-module connection is retrieved from the BIM model database for the two adjacent groups. This includes the coordinates of the local docking flange center point in the coordinate system of their respective modules, and the local docking flange three-dimensional space normal vector perpendicular to the flange plane. Then, according to the candidate position coordinates and candidate placement postures (e.g., whether to rotate 90 degrees) specified for these two groups by the current global placement combination scheme, the local coordinates are transformed globally. Specifically, the coordinates of the local docking flange center point are added to the candidate position coordinates of its module (i.e., the global origin offset) to obtain the coordinates of the global docking flange center point. At the same time, according to the candidate placement posture, the corresponding rotation matrix transformation is applied to the local docking flange three-dimensional space normal vector to obtain the global docking flange three-dimensional space normal vector.

[0096] Calculate the three-dimensional spatial distance difference between the center points of the global docking flanges of each pair of adjacent prefabricated pod equipment groups within the same global placement assembly scheme. Also calculate the spatial angle between the three-dimensional spatial normal vectors of their corresponding global docking flanges. Fill the calculated three-dimensional distance difference and angle values ​​into a pre-constructed matrix according to the global placement assembly scheme number and the pairing relationship of adjacent groups, forming the module docking pose error matrix corresponding to that scheme. Evaluate the values ​​in this matrix to select schemes that meet the precision assembly requirements. The evaluation criteria include two preset assembly tolerance extremes: one is the angle tolerance, and the other is the distance tolerance. The angle tolerance is set based on the flange... The sealing and fixing requirements for the docking are as follows: for example, the absolute difference between the angle between the normal vector and the ideal reverse parallel state (180 degrees) must be less than or equal to 0.5 degrees. The distance tolerance is based on the operability of on-site installation and the allowable gap of the connectors, for example, it is set to be less than or equal to 5mm. The error matrix of all global placement combination schemes is traversed to lock the target global placement combination schemes that simultaneously meet the angle and distance tolerance judgment conditions for the docking errors between all adjacent modules. Finally, the subgroup number, selected candidate position coordinates, candidate placement posture and global docking flange center point coordinates used for precise positioning are extracted from these target schemes. These parameters are integrated to generate a three-dimensional assembly coordinate set for the intelligent power distribution room.

[0097] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for optimizing the spatial layout of intelligent power distribution rooms based on BIM technology, characterized in that, Includes the following steps: An initial equipment connection weight topology diagram is generated based on the first connection weight value of the equipment nodes and busbar connection lines and the second connection weight value of the cable connection lines in the electrical single-line diagram of the power distribution room. Extract equipment mass values ​​and external boundary dimensions from the 3D solid model of the power distribution room, and obtain an initial module segmentation dataset containing subsets based on the Laplacian matrix; The relative three-dimensional centroid coordinates of the outer boundary dimension frame are obtained by local three-dimensional orthogonal nesting within the virtual base of the standard prefabricated cabin. The overall center of gravity coordinate point is calculated by combining the equipment mass value with the relative three-dimensional center of mass coordinate point, and the hoisting center of gravity deviation distance value is obtained by comparing the overall center of gravity coordinate point with the geometric center coordinate point. For the target subset whose hoisting center of gravity deviation distance exceeds the preset hoisting safety threshold, the connection weight values ​​of key equipment nodes and adjacent nodes are reduced based on the torque contribution, and an updated equipment connection weight topology map is generated. Obtain the updated module cutting dataset, re-execute local 3D orthogonal nesting and calculate the hoisting center of gravity deviation distance until it meets the preset hoisting safety threshold, and output the balanced prefabricated cabin equipment grouping; Generate the grouping outer boundary dimension frame by combining the layout position parameters and the outer boundary dimension frame; The coordinates of the cuboid cavity are extracted from the 3D solid model of the power distribution room by removing the structural columns and passageway occupied areas, and a linked list of usable 3D blank areas is generated. Candidate regions are extracted by matching the clustered bounding boundary size box with the available 3D blank area linked list, generating a dataset of candidate locations for prefabricated cabins.

2. The method for optimizing the spatial layout of an intelligent power distribution room based on BIM technology according to claim 1, characterized in that, The method further includes: Generate a global placement combination scheme from candidate placement position combinations; based on the candidate position coordinates, map the local flange 3D space normal vector and the local flange center point coordinates to the global flange 3D space normal vector and the global flange center point coordinates; Calculate the three-dimensional distance difference and included angle values ​​to construct the module docking pose error matrix; lock the target global placement combination scheme that meets the preset assembly tolerance extreme value, extract the three-dimensional placement parameters, and generate the three-dimensional assembly coordinate set of the intelligent power distribution room.

3. The method for optimizing the spatial layout of an intelligent power distribution room based on BIM technology according to claim 1, characterized in that, The steps for obtaining the initial device connection weight topology graph are as follows: Read the electrical single-line diagram of the power distribution room, analyze the equipment element identifier, equipment number field and port connection position of the electrical single-line diagram line by line, extract all equipment nodes with independent access ports, distinguish bus connection lines and cable connection lines according to the layer attributes, line type attributes and port bridging relationships of the connection lines, assign a first connection weight value to the bus connection lines and a second connection weight value to the cable connection lines to form the equipment node connection weight identifier; Based on the device node connection weight identifier, all device nodes are converted into topology vertices one by one, and bus connection lines carrying the first connection weight value and cable connection lines carrying the second connection weight value are converted into topology edges one by one. The starting device node, ending device node and connection weight value corresponding to each topology edge are recorded, and invalid topology edges that do not match the starting device node and ending device node at the same time are removed to generate an initial device connection weight topology graph.

4. The method for optimizing the spatial layout of an intelligent power distribution room based on BIM technology according to claim 1, characterized in that, The steps for obtaining the initial module-segmented dataset are as follows: From the 3D solid model of the power distribution room, match the corresponding 3D equipment entities one by one according to the equipment number field, extract the equipment mass value, length boundary value, width boundary value and height boundary value corresponding to each 3D equipment entity, form the outer shape boundary size box corresponding to each equipment node, construct the corresponding Laplacian matrix according to the connection weight value of the topological vertex in the initial equipment connection weight topology graph, perform eigenvalue decomposition on the Laplacian matrix, divide the equipment node affiliation relationship according to the eigenvalue decomposition result, and obtain the initial module segmentation dataset.

5. The method for optimizing the spatial layout of an intelligent power distribution room based on BIM technology according to claim 1, characterized in that, The steps for obtaining the hoisting center of gravity deviation distance value are as follows: Read the device node number corresponding to each subset in the initial module cutting dataset one by one, retrieve the outer boundary dimension box corresponding to each device node according to the device node number, extract the length boundary value, width boundary value and height boundary value in the outer boundary dimension box, arrange the length boundary value along the horizontal axis of the standard prefabricated cabin virtual base, arrange the width boundary value along the vertical axis of the standard prefabricated cabin virtual base, arrange the height boundary value along the vertical axis of the standard prefabricated cabin virtual base, adjust the horizontal axis position or vertical axis position for device nodes with overlapping boundaries, until the overlap volume between any two outer boundary dimension boxes in the same subset is zero, and obtain the relative three-dimensional centroid coordinate point of each device node in the standard prefabricated cabin virtual base; Based on the relative three-dimensional centroid coordinates of each equipment node within the virtual base of the standard prefabricated cabin, the equipment mass values ​​corresponding to each equipment node are retrieved according to the equipment node number. The equipment mass values ​​of each equipment node within the same subset are associated with the corresponding relative three-dimensional centroid coordinates. The horizontal coordinate components, vertical coordinate components, and vertical coordinate components are weighted and accumulated separately. Then, the weighted accumulation result is normalized according to the accumulated result of all equipment mass values ​​within the same subset to obtain the comprehensive centroid coordinates of each subset. Read the length, width, and height boundary values ​​of the standard prefabricated module virtual base, and determine the lateral, longitudinal, and vertical center coordinates of the standard prefabricated module virtual base respectively, forming the geometric center coordinate point of the standard prefabricated module virtual base. Calculate the lateral, longitudinal, and vertical coordinate differences between the comprehensive center of gravity coordinate points of each subset and the geometric center coordinate point of the standard prefabricated module virtual base, respectively. Determine the three-dimensional spatial distance based on the lateral, longitudinal, and vertical coordinate differences, and obtain the hoisting center of gravity deviation distance value.

6. The method for optimizing the spatial layout of an intelligent power distribution room based on BIM technology according to claim 1, characterized in that, The steps for obtaining the updated device connection weight topology map are as follows: The hoisting center of gravity deviation distance value corresponding to each subset is read item by item. The hoisting center of gravity deviation distance value is compared with the preset hoisting safety threshold value item by item according to the subset number. The subset with the hoisting center of gravity deviation distance value greater than the preset hoisting safety threshold value is marked as the subset to be fed back. The subset with the hoisting center of gravity deviation distance value less than or equal to the preset hoisting safety threshold value is marked as the stable subset. The subset number, equipment node number and hoisting center of gravity deviation distance value corresponding to the subset to be fed back are retained to obtain the target subset. The equipment mass value, relative three-dimensional centroid coordinate point, and geometric center coordinate point of the standard prefabricated cabin virtual base are read one by one for each equipment node in the target subset. The lateral offset, longitudinal offset, and vertical offset of each equipment node are calculated respectively. The equipment mass value is associated with the lateral offset, longitudinal offset, and vertical offset respectively. A lateral moment component, a longitudinal moment component, and a vertical moment component are formed for each equipment node. The results of the combined results of the moment components in the three directions are sorted to obtain the moment contribution of each equipment node relative to the geometric center coordinate point of the standard prefabricated cabin virtual base. The critical device nodes that cause imbalance are selected according to the ranking of torque contribution. All adjacent nodes corresponding to the critical device nodes in the initial device connection weight topology graph are retrieved. The connection weight values ​​between the critical device nodes and their adjacent nodes are read. The connection weight values ​​are then reduced in a targeted manner according to the ranking of torque contribution. The reduced connection weight values ​​are written back to the corresponding topology edges. The connection weight values ​​between the unselected device nodes are retained, and an updated device connection weight topology graph is generated.

7. The method for optimizing the spatial layout of an intelligent power distribution room based on BIM technology according to claim 1, characterized in that, The steps for obtaining the balanced prefabricated cabin equipment cluster are as follows: The device node number, topological edge connection relationship, and connection weight value of the updated device connection weight topology graph are read one by one. A matrix row and column correspondence relationship is established according to the device node number. The connection weight value of each device node to other device nodes is filled into the corresponding matrix position. The total connection weight values ​​associated with each device node are accumulated and filled into the corresponding diagonal position to form the Laplacian matrix corresponding to the updated device connection weight topology graph. The matrix elements in the Laplacian matrix are decomposed into eigenvalues. The device node affiliation identifier is read according to the order of eigenvalues ​​to obtain the updated module segmentation dataset. Read the device node number corresponding to each subset in the updated module cutting dataset one by one, retrieve the outer boundary dimension frame and equipment mass value according to the device node number, map the outer boundary dimension frame to the virtual base of the standard prefabricated cabin, check the occupied boundary of the outer boundary dimension frame according to the horizontal axis, vertical axis and vertical axis respectively, adjust the layout position of the outer boundary dimension frames with overlapping volume one by one until the overlapping volume between any two outer boundary dimension frames in the same subset is equal to zero, read the relative three-dimensional centroid coordinate point of each device node, calculate the comprehensive centroid coordinate point of each subset in combination with the equipment mass value, accumulate the three-dimensional coordinate difference between the comprehensive centroid coordinate point and the geometric center coordinate point of the standard prefabricated cabin virtual base to obtain the hoisting centroid deviation distance value; The hoisting center of gravity deviation value corresponding to each subset is read item by item. The hoisting center of gravity deviation value is compared with the preset hoisting safety threshold according to the subset number. If the hoisting center of gravity deviation value corresponding to all subsets is less than or equal to the preset hoisting safety threshold, the device node affiliation relationship corresponding to each subset in the current updated module segmentation dataset is determined as an output cluster. If the hoisting center of gravity deviation value corresponding to any subset is greater than the preset hoisting safety threshold, the current iteration number, the change amplitude of the connection weight value, and the number of device node affiliation changes are read. It is determined whether the current iteration number has reached the preset iteration limit, whether the change amplitude of the connection weight value is less than the preset change amplitude threshold, and whether the number of device node affiliation changes is equal to zero. If any of the judgment results are satisfied, the device node affiliation relationship corresponding to each subset in the current updated module segmentation dataset is determined as an output cluster, and the balanced prefabricated cabin equipment cluster is output.

8. The method for optimizing the spatial layout of an intelligent power distribution room based on BIM technology according to claim 1, characterized in that, The steps for obtaining the candidate location dataset for the prefabricated cabin are as follows: Read the equipment node number inside each of the balanced prefabricated cabin equipment groups one by one, retrieve the corresponding outer boundary dimension frame and layout position parameters according to the equipment node number, read the horizontal minimum boundary, horizontal maximum boundary, vertical minimum boundary, vertical maximum boundary, vertical minimum boundary and vertical maximum boundary of the outer boundary dimension frame respectively, and calculate the occupied boundary of each outer boundary dimension frame in the balanced prefabricated cabin equipment group in combination with the layout position parameters. Take the outer extreme value of all occupied boundaries in the same balanced prefabricated cabin equipment group to generate the group outer boundary dimension frame of each balanced prefabricated cabin equipment group. Scan the 3D solid model of the power distribution room, read the wall boundaries, floor boundaries, structural column boundaries, and passageway occupancy boundaries layer by layer, remove the structural column boundaries and passageway occupancy boundaries from the internal space of the 3D solid model of the power distribution room, divide the remaining space into cuboids according to the horizontal continuous range, vertical continuous range, and vertical continuous range, record the horizontal start and end coordinates, vertical start and end coordinates, and vertical start and end coordinates of each cuboid cavity coordinate item, remove cuboid cavity coordinate items with zero dimension in any direction, and generate a linked list of usable 3D blank areas; The length, width, and height values ​​of the grouped outer boundary size frame are read one by one. The length, width, and height values ​​of each cuboid cavity coordinate item in the available 3D blank area list are read one by one. It is determined whether the length, width, and height values ​​of the cuboid cavity coordinate items are greater than or equal to the length, width, and height values ​​of the grouped outer boundary size frame, respectively. The cuboid cavity coordinate items that meet the size matching are marked as candidate regions. Then, the horizontal coordinate range, vertical coordinate range, and vertical coordinate range between candidate regions are overlapped and verified. Candidate regions with zero overlapping volume are retained as placement items, generating a prefabricated cabin candidate position dataset.

9. The method for optimizing the spatial layout of an intelligent power distribution room based on BIM technology according to claim 1, characterized in that, The steps for obtaining the three-dimensional assembly coordinate set of the intelligent power distribution room are as follows: Based on the prefabricated cabin candidate location dataset, the candidate placement location number, candidate location coordinates, candidate area length, candidate area width, and candidate area height values ​​corresponding to each balanced prefabricated cabin equipment group are read one by one. A placement location selection sequence is established according to the group number of the balanced prefabricated cabin equipment group. The candidate placement locations of each balanced prefabricated cabin equipment group are written into the placement location selection sequence item by item. The candidate location coordinates between different balanced prefabricated cabin equipment groups are combined and enumerated. The combination results where the same candidate location coordinates are repeatedly occupied are eliminated, and several global placement combination schemes are generated. Read the group number of adjacent balanced prefabricated cabin equipment in each of the global placement combination schemes one by one, retrieve the three-dimensional space normal vector of the local docking flange, the coordinates of the center point of the local docking flange and the coordinates of the candidate position of the adjacent balanced prefabricated cabin equipment group, perform coordinate offset conversion on the coordinates of the center point of the local docking flange according to the candidate position coordinates, and perform direction conversion on the three-dimensional space normal vector of the local docking flange according to the candidate placement posture to obtain the three-dimensional space normal vector of the global docking flange and the coordinates of the center point of the global docking flange. The three-dimensional distance difference between the center point coordinates of the global docking flanges of adjacent prefabricated equipment groups is calculated item by item. The angle between the three-dimensional spatial normal vectors of the global docking flanges of adjacent prefabricated equipment groups is calculated item by item. The three-dimensional distance difference and angle are written into the corresponding matrix positions according to the global placement combination scheme number. The module docking posture error matrix corresponding to the global placement combination scheme is constructed. The absolute difference between the angle value and 180 degrees in the module docking posture error matrix is ​​judged item by item. The three-dimensional distance difference is judged item by item. The target global placement combination scheme that meets the judgment conditions is locked. The group number, candidate position coordinates, candidate placement posture and the center point coordinates of the global docking flange in the target global placement combination scheme are extracted and integrated to form three-dimensional placement parameters, and a three-dimensional assembly coordinate set of the intelligent power distribution room is generated.

Images