Tree structured conversion and analysis method based on building information model electromechanical system

Abstract

The invention relates to a tree structured conversion and analysis method based on an electromechanical system of a building information model, and the method comprises the steps: building a system topological map according to the characteristics of the electromechanical system; determining the basic flow direction of the system as the development direction of the tree structure; identifying an annular pipe network structure according to a system basic flow direction, packaging the annular pipe network structure into annular nodes, and replacing related nodes in a system topological map, so that the system topological map has a tree-shaped basic characteristic; converting a system topology map into a system tree of a tree structure from a system starting point along a system basic flow direction; and performing feature analysis through the established system tree according to different requirements. While automatic transcription is realized, visualization and interaction characteristics are supplemented, and the efficiency and quality of building electromechanical professional design are improved. The converted electromechanical system data is stored and transmitted in a standardized data structure form, so that it is possible to quickly convert a design result into standard data.

Description

Technical Field

[0001] The invention relates to information technology of building electromechanical system, and in particular to a tree structure conversion and analysis method of electromechanical system based on building information model. Background Art

[0002] With the digital transformation of the architectural design industry, it is an urgent need for the industry to use digital methods to automate and upgrade the system design work of traditional electromechanical professionals. The building electromechanical system designed using the Building Information Model (BIM) is good at expressing the position and connection status of components, but it does not effectively utilize and analyze the large amount of potential information it contains. The design of building electromechanical systems is a systematic work, and the position and connection status of components are not enough to reflect the overall picture of the electromechanical system. This makes it difficult for designers to conduct in-depth and accurate systematic analysis of electromechanical systems in building information models, and it is quite inefficient to manually view the details of system components in the model one by one and draw systematic conclusions.

[0003] The root of the above problems lies in the fact that the understanding of the electromechanical system is not limited to the layout of components or connection relationships, but must also be analyzed and deconstructed from multiple levels such as functional requirements, performance requirements, process specifications, operation requirements, and integration requirements. This requires not only a data structure that focuses on expressing the topological relationship of the electromechanical system, but also a general method that can transcribe the electromechanical system into this data structure. Moreover, for common building electromechanical systems, such as ventilation systems, smoke exhaust systems, sprinkler systems, multi-connected systems, etc., their medium flow must be directed and can often be summarized as a tree structure. If a method can be realized to automatically transcribe the electromechanical system and summarize it into a directed tree structure, supplemented by visualization and interactivity, it will be able to provide data and tool foundations for building electromechanical system designers to efficiently perform system design, calculation, review and other tasks, and improve the efficiency and quality of building electromechanical professional design.

[0004] However, there are still many difficulties in implementing this method of tree-structuring the electromechanical system in the existing building information model (BIM), including path identification in the process of directing the undirected static connection relationship of the electromechanical system in the building information model (BIM), local connection loss caused by modeling accuracy, complex and irregular management of system component parameter information, and fuzzy characteristics of complex system risers. The biggest difficulty lies in the conflict between the tree structure and the loop problem: many electromechanical systems involve complex topological relationships of electromechanical system loops and multiple branches in actual operation, and the tree structure itself is acyclic, which leads to difficulties in directly converting the existing building information model (BIM) electromechanical system into a tree structure. Existing system analysis tools cannot automatically identify or process these complex topological features, which easily leads to logical disconnection during the design and analysis process. Summary of the invention

[0005] To address the above problems, a tree-structured conversion and analysis method based on the electromechanical system of the building information model was proposed. The topological relationship of the building electromechanical system was transcribed using structured data, and the complex electromechanical pipe network system was summarized and processed into tree-structured data, which was conducive to data standardization and subsequent analysis.

[0006] The technical solution of the present invention is: a tree structure conversion method and analysis method based on a mechanical and electrical system of a building information model, which specifically includes the following steps:

[0007] Step 1: Create a system topology map. Take a component in the electromechanical system as the starting point to obtain the system topology map of the electromechanical system. The implementation steps are as follows:

[0008] Step 1.1: Create a system element class for expressing various unit components in the electromechanical system. The system element class stores the element name, unique identification ID, system ID and name, spatial location, room information, and internal connector information. The system element class is subdivided into five subdivision types: pipeline type, pipe fitting type, pipeline accessory type, power equipment type, and terminal type according to the characteristics of the specific instance.

[0009] According to the internal connector information in the system element class, the next closest system element class to which the individual instance of the system element class is connected can be queried;

[0010] Step 1.2: Prepare the system map class and create two system map classes, namely map vertex class and map edge class;

[0011] The map vertex class is used to express the system element class whose connection quantity status is non-two-way; a map vertex class stores two list information, namely the adjacent map vertex class list and the map edge class list;

[0012] The map edge class is used to express the edge between two map vertex classes, which is an ordered list of system element classes;

[0013] Both the map vertex class and the map edge class can characterize their total geometric length according to the pipeline type system element class they represent;

[0014] Step 1.3: Extract the system topology map class and create a system topology map class instance, which stores all map vertex classes and map edge classes;

[0015] Starting from any unit component in the electromechanical system of a building information model, create the first system element class and use it as the first map vertex class to start extracting the entire system map structure; traverse and query the next element connected to the system element class, create an equal number of map edge classes according to its connection quantity status, and decide whether to create a new map vertex class according to the element or add it to its corresponding map edge class according to the connection quantity status of the next element, and mark all decided system element classes as processed to avoid repeated addition; recursively perform the above method until all connected system element classes in the system are marked as processed; all newly created map vertex classes and map edge classes are finally added to the system topology map class created in this step, thus completing the system topology map extraction step and obtaining a system topology map class that fully expresses the connection information of the electromechanical system;

[0016] Step 2: Formulate the basic flow direction. According to any non-starting component in the electromechanical system, determine the basic flow direction of the system as the development direction of the tree structure.

[0017] Step 3: Ring node encapsulation: Identify the ring network structure according to the basic flow direction of the system and encapsulate it as a ring node, replace the relevant vertices in the system topology map in step 1, so that the system topology map has the basic characteristics of tree-like structure; the implementation steps are as follows:

[0018] Step 3.1: Preparation of the system path class; create a system path class for expressing the path in the system topology map, which is stored in an ordered manner as a list of map vertex classes, describing any path in the system map structure with ordered adjacent vertices as stations. The vertices corresponding to the starting point and the end point of the path are the starting point and the end point of the system path class;

[0019] The system path class also stores characteristic attributes used to analyze the electromechanical system, including the starting and ending points on the path, as well as the number, size, length, and spatial position of all system elements on the path;

[0020] Step 3.2: Determine the path search weight attribute; select a certain characteristic attribute as the weight value according to the characteristic attribute of the electromechanical system itself or customized, and the characteristic attribute is also used as the weight value standard for subsequent loops and recursive execution of breadth-first and depth-first algorithms;

[0021] Step 3.3: Determine the temporary main path step; starting from the set system starting point, execute the breadth-first search algorithm, and determine the temporary main path in combination with the basic flow direction;

[0022] Step 3.4: Identify the existence of a ring structure; use a cyclic stepping method to perform a depth-first search algorithm on the temporary main path to identify the ring structure in the electromechanical system pipe network;

[0023] Step 3.5: Determine the main path and bypass path inside the ring structure; perform breadth-first and depth-first algorithms respectively at the entrances and exits represented by the temporary main path at the boundary of the ring structure to determine the main path and bypass path inside the ring structure;

[0024] Step 3.6: Ring node replacement step: encapsulate all components and path information in the ring structure as a ring node to replace the relevant nodes in the topology structure map;

[0025] Step 3.7: loop and recursion step; based on the system topology map partially replaced with ring nodes, repeat steps 3.3 to 3.6 until all ring structures in the system complete topology map are replaced with ring nodes;

[0026] Step 4: Tree structure conversion: starting from the set system starting point and following the system basic flow, the system topology map after executing the ring node encapsulation is converted into a tree structured system tree;

[0027] Step 5: System tree feature analysis, use different weight values ​​to perform system tree feature analysis, and use visualization to express the system tree structure for specific subsequent applications after the structured conversion of the building information model electromechanical system tree.

[0028] Furthermore, the system element class in step 1.1 is subdivided into five subdivision types, namely pipeline type, pipe fitting type, pipeline accessory type, power equipment type and terminal type, according to the characteristics of the specific instance;

[0029] In addition to storing element information, the system element class of pipeline type also stores the geometric length of the pipeline, and is classified into vertical risers and non-vertical risers according to their geometric characteristics;

[0030] In addition to storing element information, all system element classes of pipe fitting types also store the connection quantity status. According to the actual connection quantity of its internal connectors, the system element class is determined to be single-way, two-way, three-way, ..., n-way, expressing the connection quantity status characterized by quantity.

[0031] Furthermore, the implementation of the basic flow direction formulation method in step 2 is achieved by implementing a depth-first search algorithm in the system topology map class. The specific steps are as follows:

[0032] Step 2.1: Determine the system starting point; manually determine. If the manually determined system starting point is not a map vertex class, then find the map edge class to which it belongs and use the closest map vertex class endpoint in the map edge class as the system starting point;

[0033] Step 2.2: Determine the calculation direction point; select any non-starting element in the system topology map. To ensure correctness, this selection is completely done manually. Same as the method of determining the system starting point, if the manually determined system starting point is not a map vertex class, then find the map edge class to which it belongs and use the closest map vertex class endpoint in the map edge class as the calculation direction point.

[0034] Step 2.3: Find the shortest path; starting from the system starting point, use the breadth-first search algorithm to calculate the shortest path from the system starting point to the calculation direction point. The path consists of a set of ordered map vertex lists, and the first vertex in this set of vertex lists is the system starting point;

[0035] Step 2.4: Determine the nearest directional vertex and the nearest system element class; according to the obtained shortest path, the second vertex in the map vertex class list is the nearest directional vertex; a unique system map edge class can be determined by the system starting point and the nearest directional vertex, and the second object in the internal system element list in this system map edge class is the system element class closest to the system starting point. The connection direction between the system starting point and the system element class is the physical space direction of the system basic flow.

[0036] Furthermore, the steps 3.4 and 3.5: create a ring node class for encapsulating the ring structure in the system topology map, which inherits from the system vertex class, rewrites the properties of its adjacent vertex list and adjacent edge list, and is used to express the import and export of the entire ring structure; the internal structure of the ring node is similar to the system path class, and an ordered map vertex class list is used for storage and expression, and new internal main path and bypass path properties are added to express the complete system detail information within the ring structure.

[0037] Further, the specific implementation steps of step 4 are as follows:

[0038] Step 4.1: Preparation of the phylogenetic tree class, creating a phylogenetic tree class based on a tree structure for expressing a phylogenetic topology map, which stores the phylogenetic path expressed by the current phylogenetic tree itself, the branch code of the current phylogenetic tree, the depth of the current tree, a list of any number of phylogenetic trees following the current phylogenetic tree, and the parent tree object of the current phylogenetic tree;

[0039] The system tree also stores characteristic attributes used to analyze the electromechanical system, including the number, size, length, and spatial position of all system elements on the current path, as well as characteristic attributes of subsequent system trees, and obtains statistical values ​​for specific characteristic attributes of the system tree through arithmetic methods of summation and averaging;

[0040] The system tree nodes are derived from the non-ring common vertices and encapsulated ring nodes in the system topology map encapsulated by the ring nodes;

[0041] Step 4.2: The creation step of the system tree class is based on the system topology map. From the set starting point, along the system basic flow, the system tree class instance is recursively constructed; each path is used as the current path of the current system tree class instance, and the new system tree class instance corresponding to the sub-path is added to the sub-system tree collection of the parent system tree class instance; the above steps are repeated until the entire system topology map is traversed and finally converted into a tree-structured system tree.

[0042] Furthermore, in step 5, the system tree feature analysis: the system is distributed according to the floor, the system on each floor is connected through the riser, and the riser path in the system tree is automatically searched to analyze the electromechanical system; the search steps are as follows:

[0043] Step 5.1.1: Confirm the riser starting point; starting from a certain point in the system tree, find the first riser closest to the starting point along the development direction of the system tree, and use the system tree where the riser is located as the current system tree to start the riser search; Step 5.1.2: Calculate the current path vertical rate of the current system tree;

[0044] Step 5.1.3: Calculate the current path vertical rate of the next level subtree of the current phylogenetic tree;

[0045] Step 5.1.4: Determine whether it is in the main riser path; calculate whether the verticality of the current path of the system tree is greater than the verticality of any subtree at the next level; if it is affirmative, use the subtree with the maximum verticality as the new current system tree;

[0046] Step 5.1.5: Extract the riser path; recursively perform steps 5.1.1 to 5.1.4. If a negative situation occurs, end the riser search, then record the phylogenetic tree of all recursive pathways, extract their paths and merge them into a complete riser path.

[0047] Further, in step 5, the system tree feature analysis: in the building mechanical and electrical system, the designer pays attention to two types of system floor information, one is the floor where the terminal is located, and the other is the floor through which the pipeline passes; the floor definition method of the system tree is as follows:

[0048] Step 5.2.1: Obtain the system riser path by automatic riser search or manual designation method. It can be seen that the subsystem trees branched from each node in the riser path and having overlapping paths with the riser path are all floor branch system trees developed on the riser path;

[0049] Step 5.2.2: Obtain the floor information of the branch system tree of each floor by counting the spatial information of the terminal and pipeline, and take the floor with the largest proportion of components as the characteristic floor of the branch system tree of the floor;

[0050] Step 5.2.3: Sort the branch system trees of each floor by floor elevation to obtain a floor branch system tree that matches the number of floors in the building, and conduct further electromechanical system characteristic analysis based on the floor information.

[0051] Furthermore, in step 5, the system tree feature analysis is as follows: in the design of the building mechanical and electrical system, the flow statistics of the system are obtained by reversely deducing the design demand flow at the end along the system flow direction.

[0052] Furthermore, the system tree feature analysis in step 5: implementation of the system tree hydraulic calculation method; implementation of the system tree hydraulic calculation method based on the completion of the system tree flow statistics; precondition parameters need to be configured for the implementation of hydraulic calculation, including the resistance coefficient along the way, the local resistance coefficient of various types of pipe fittings, terminals, and equipment; the precondition parameters are created through other databases or custom methods, and applied to the hydraulic calculation process of the system tree during calculation; since the tree-structured electromechanical system topology map already has significant tree structure characteristics, the complex ring pipe network is node-abstracted, and the hydraulic characteristics of each branch of the system tree can be obtained through simple data calculation when the precondition conditions are complete.

[0053] Furthermore, in step 5, the system tree feature analysis: implementation of the system tree material statistics method; material statistics work is implemented on any system tree, by formulating material statistics rules for different types of components, rules for merging and counting terminal equipment according to characteristic parameters, classification and piece-counting rules for water pipe fittings, and calculation rules for duct material usage for duct fittings. Through the specification of the above rules, material statistics are performed on the current system tree and its branches at each level, each floor branch, and each room branch, so that the granularity of material statistics is refined as the needs of the business scenario, and accurate statistical results are obtained in real time.

[0054] The beneficial effect of the present invention is that the present invention is based on the tree structure conversion and analysis method of the electromechanical system of the building information model, and completes the analysis and conversion of the electromechanical system from the building information model to the tree structure by creating relevant data structures and design matching methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 A step diagram of the tree structure conversion and analysis method of the electromechanical system based on the building information model of the present invention;

[0056] Figure 2 A diagram of steps for creating a system topology map in the method of the present invention;

[0057] Figure 3 The present invention is a tree structure conversion and analysis system module diagram based on the mechanical and electrical system of the building information model;

[0058] Figure 4This is a schematic diagram of an air duct elbow pipe fitting according to an embodiment of the present invention;

[0059] Figure 5 A data structure diagram of the topological map class of the system of the present invention;

[0060] Figure 6 A schematic diagram of the topological relationship of the acyclic system vertices of the present invention;

[0061] Figure 7 Analyze the riser path schematic diagram for the present invention;

[0062] Figure 8 It is a schematic diagram of the basic flow direction designation method of the present invention;

[0063] Fig. 9 A display diagram showing the concepts involved in implementing the depth-first search algorithm of the present invention on a topological map of an electromechanical system;

[0064] Fig.10 It is a schematic diagram of the ring node packaging method of the present invention;

[0065] Fig.11 This is a diagram showing an embodiment of a riser path in a system path class of the present invention;

[0066] Fig.12 This is a prototype morphological diagram of the ring node of the present invention;

[0067] Fig.13 A graphical representation of a system topology map of the concepts provided by the method for determining a temporary primary path of the present invention;

[0068] Fig.14 , 15 , 16, 17 are the present invention Fig.13 The embodiment is a diagram of the process of searching for the entrance of the ring structure;

[0069] Fig.18 , 19 The present invention shows a process diagram of reverse step search and finding the ring interface exit on the temporary main path;

[0070] Fig. 20 The embodiment diagram of collecting the inner vertices of the ring structure for the present invention;

[0071] Fig.21 The present invention shows that based on the Fig.13 The system topology map shown is a state diagram after encapsulation and replacement of ring nodes;

[0072] Fig. 22 This is a step diagram of the system topology map tree structure conversion method of the present invention;

[0073] Fig.23The system topology map tree containing ring nodes of the present invention is structured to become a system tree and a state diagram with numbering completed;

[0074] Fig.24 This is a schematic diagram of the hierarchical expression based on tree structure numbering and paths of the present invention;

[0075] Fig.25 A diagram of analysis steps that can be executed by the phylogenetic tree feature analysis module of the present invention;

[0076] Fig.26 is a sample diagram of the ventilation system of the present invention;

[0077] Fig. 27 For the present invention Fig.26 System analysis result diagram after sample conversion;

[0078] Fig.28 A diagram showing the execution steps of the custom pruning and grafting module of the present invention;

[0079] Fig.29 A schematic diagram of two fire sprinkler systems of the present invention that are not connected to each other;

[0080] Fig.30 A complete fire sprinkler system diagram with virtual connections formed for the present invention. DETAILED DESCRIPTION

[0081] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.

[0082] As attached Figure 1 The tree structure conversion and analysis method steps based on the mechanical and electrical system of the building information model are shown in the figure. The specific steps of implementation are as follows:

[0083] Step 1: System topology map creation step: Based on a certain component in the electromechanical system as a starting point, obtain the system topology map of the electromechanical system; Figure 2 The implementation steps are as follows:

[0084] Step 1.1: System element class preparation step; create a system element class for expressing various unit components in the electromechanical system. The system element class stores the element name, unique identification ID, system ID and name, spatial location, room information, and internal connector information. The system element class can be subdivided into five subdivision types: pipeline type, pipe fitting type, pipeline accessory type, power equipment type, and terminal type according to the characteristics of the specific instance;

[0085] In addition to storing element information, the pipeline type system element class also stores the geometric length of the pipeline, and can be classified into vertical risers and non-vertical risers according to their geometric characteristics;

[0086] All system element classes of pipe fitting types store not only element information but also connection quantity status. According to the actual connection quantity of internal connectors, the system element class can be determined as single-way, two-way, three-way, four-way and other connection quantity status characterized by quantity.

[0087] According to the internal connector information in the system element class, the next closest system element class to which the individual instance of the system element class is connected can be queried.

[0088] Step 1.2: System map class preparation step; create two system map classes, namely map vertex class and map edge class;

[0089] The map vertex class is used to express the system element class whose connection quantity status is non-two-way; a map vertex class stores two list information, namely the adjacent map vertex class list and the map edge class list;

[0090] The map edge class is used to express the edge between two map vertex classes, which is an ordered list of system element classes;

[0091] Both the map vertex class and the map edge class can characterize their total geometric length according to the pipeline type system element class they represent;

[0092] Step 1.3: System topology map class extraction step: Create a system topology map class instance, which stores all map vertex classes and map edge classes; starting from any unit component in the electromechanical system of a building information model, create the first system element class, and use it as the first map vertex class to start extracting the entire system map structure; traverse and query the next element connected to the system element class, create an equal number of map edge classes according to its connection quantity status, and at the same time, decide whether to create a new map vertex class according to the element or add it to the corresponding map edge class according to the connection quantity status of the next element, and also mark all decided system element classes as processed to avoid repeated addition; recursively perform the above method until all connected system element classes in the system are marked as processed; all newly created map vertex classes and map edge classes are finally added to the system topology map class created in this step, thus completing the system topology map extraction step, and obtaining a system topology map class that fully expresses the connection information of the electromechanical system;

[0093] Step 1.4: Optional, vertexization step of the vertical state transition points of the pipeline network; this step is a supplement to the system map structure extraction step. Considering that identifying and separating the riser part in the electromechanical pipeline network is of practical significance for the subsequent system feature analysis, the system element class with the feature of two-way type but vertical and non-vertical pipeline types before and after is also created as a map vertex class.

[0094] Step 2: Basic flow direction formulation step: According to any non-starting component in the electromechanical system, determine the basic flow direction of the system as the development direction of the tree structure. In the system tree to be created, the branch point of its tree structure is the node of the tree. Generally, the map vertex class with multi-pass characteristics can be used as a system node, but for the ring structure that conflicts with the tree structure, it is expressed by encapsulating it as a ring node.

[0095] Step 3: Ring node encapsulation step: Identify the ring network structure according to the basic flow direction of the system and encapsulate it as a ring node, replace the relevant vertices in the system topology map, so that the system topology map has the basic characteristics of being tree-like; the implementation steps are as follows:

[0096] Step 3.1: Preparation of the system path class; create a system path class for expressing the path in the system topology map, which is stored in an ordered manner as a list of map vertex classes, describing any path in the system map structure with ordered adjacent vertices as stations. The vertices corresponding to the starting point and end point of the path are the starting point and end point of the system path class; the map edge class can only express the path between two adjacent vertices, and the introduction of the system path class can express any path that spans more than two vertices; the system path class also stores characteristic attributes for analyzing electromechanical systems, including the starting point and end point on the path, as well as the number, size, length, and spatial position of all system elements on the path;

[0097] Step 3.2: Determine the path search weight attribute; select a certain characteristic attribute as the weight value according to the characteristic attributes of the electromechanical system itself or customized. This characteristic attribute also serves as the weight value standard for subsequent loops and recursive execution of breadth-first and depth-first algorithms.

[0098] Step 3.3: Determine the temporary main path step; starting from the set system starting point, execute the breadth-first search algorithm, and determine the temporary main path in combination with the basic flow direction;

[0099] Step 3.4: Identify the existence of a ring structure; use a cyclic stepping method to perform a depth-first search algorithm on the temporary main path to identify the ring structure in the electromechanical system pipe network;

[0100] Step 3.5: Determine the main path and bypass path inside the ring structure; perform breadth-first and depth-first algorithms respectively on the entrances and exits represented by the temporary main path at the boundary of the ring structure to determine the main path and bypass path inside the ring structure; the main path is the main path section with the entrance and exit determined on the temporary main path, and the bypass path is the non-main path with the same entrance and exit;

[0101] Step 3.6: Ring node replacement step: encapsulate all components and path information in the ring structure as a ring node to replace the relevant nodes in the topology structure map;

[0102] Step 3.7: loop and recursive step; based on the system topology map partially replaced with ring nodes, repeat steps 3.3 to 3.6 until all ring structures in the system's complete topology map are replaced with ring nodes.

[0103] Step 4: Tree structure conversion step: Starting from the set system starting point and following the system basic flow, the system topology map after executing the ring node encapsulation is converted into a tree structured system tree; the implementation steps are as follows:

[0104] Step 4.1: Preparation steps for the phylogenetic tree class; create a phylogenetic tree class based on a tree structure for expressing a system topology map, which stores the system path expressed by the current phylogenetic tree itself, the current phylogenetic tree branch code, the depth of the current tree, a list of any number of phylogenetic trees following the current phylogenetic tree (a sub-tree set), and the parent tree object of the current phylogenetic tree (also a phylogenetic tree class); the phylogenetic tree also stores characteristic attributes for analyzing electromechanical systems, including the number, size, length, and spatial position of all system elements on the current path, as well as the characteristic attributes of subsequent phylogenetic trees, and obtains statistical values ​​for specific characteristic attributes of the phylogenetic tree through arithmetic methods of summation and averaging; the phylogenetic tree nodes come from the ordinary vertices in the non-ring of the system topology map encapsulated by the ring nodes and the encapsulated ring nodes.

[0105] Step 4.2: Creation step of the system tree class; Based on the system topology map, recursively construct the system tree class instance from the set starting point along the system base flow. Nodes are classified according to the number of connections: nodes with a connection number of 1 are terminal nodes, nodes with a connection number of 2 are path nodes, and nodes with a connection number greater than or equal to 3 are branch nodes. During the traversal process, the visited nodes are marked to avoid duplication. Each path is used as the current path of the current system tree class instance, and the new system tree class instance corresponding to the sub-path is added to the sub-system tree collection of the parent system tree class instance. Repeat the above steps until the entire system topology map is traversed and finally converted into a tree-structured system tree.

[0106] Step 5: System tree feature analysis step; use different weight values ​​to perform system tree feature analysis, and use visualization to express the system tree structure for specific subsequent applications after the structured conversion of the building information model electromechanical system tree.

[0107] The present invention also provides a tree structure conversion and analysis system based on a building information model electromechanical system, which is used to implement a tree structure conversion and analysis method based on a building information model electromechanical system, as shown in the attached Figure 3 The system module diagram shown includes:

[0108] The system topology map creation module is used to execute step 1, including step 1.1, step 1.2, and step 1.3, through the system starting point of the system topology map, to implement the creation of a system topology map describing a certain building mechanical and electrical system, and to execute step 2 at the same time to record the basic flow direction of the system;

[0109] Ring node encapsulation module; used to execute step 3, including steps 3.1 to 3.7, to implement the encapsulation replacement of the ring structure in the built system topology map, so that the system topology map is represented externally as a tree-like structure;

[0110] System tree conversion module; used to execute step 4, including step 4.1 and step 4.2, to convert the system topology map into a system tree;

[0111] A system tree feature analysis module; used to execute step 5, implement feature analysis of the system tree according to different weight values, including: filtering the trunk vertical riser path of the system tree, analyzing the internal pipeline length of the system tree, and analyzing the usage of components of different sizes of the system tree; implement visual expression of the system tree, including: using a tree view to express the hierarchical structure of the system tree, and using points and lines to display the graphical structure of the system tree through a node coordinate sorting algorithm;

[0112] System tree custom pruning and grafting module; used to implement leaf node pruning (deletion) and grafting (splicing) of new subtrees based on the converted system tree; create a new system tree by executing steps 1 to 4 again, and add it to a leaf node of the established system tree to achieve custom pruning and grafting of the system tree.

[0113] In order to solve the problem of understanding the electromechanical system in the building information model, the present invention realizes a data structure that focuses on expressing the topological relationship of the electromechanical system by deconstructing the topological relationship and logical process of the pipe network, and at the same time realizes a method of converting an undirected graph structure analysis into a tree structure, and implements it through a specific device.

[0114] According to an embodiment of the present invention, a tree structured conversion method based on a mechanical and electrical system of a building information model is provided, which is specifically implemented by a tree structured conversion and analysis device based on a mechanical and electrical system of a building information model provided by the present invention, based on 5 main steps and 4 modules, as shown in the attached Figure 1 and attached Figure 3 shown.

[0115] For step 1, for example: the system element class of pipeline type stores the geometric length and size of the pipeline, and can be classified into vertical risers and non-vertical risers according to its geometric characteristics; the judgment logic of the vertical riser is based on the relative relationship between the starting point and the end point of the line representing the direction of the pipeline. If the distance between its projection points on the horizontal plane is within a given error range, it can be judged as a vertical riser.

[0116] For example, for the two types of system element classes, power equipment type and terminal type, in a specific type of electromechanical system, they have clearer specific type definitions, such as air outlets, upright sprinklers, fire smoke exhaust fans, etc., and the family names of system element classes in the building information model may be non-standard. In order to allow these two types of system element classes to have standardized and clear type definitions, in this embodiment, a mapping table is prepared to allow users to manually specify and semantic analysis to make judgments to clarify their specific type definitions.

[0117] Internal connector information is the only way to obtain the adjacency relationship of the system element class and is important information in the system element class. Therefore, it is also necessary to build a dedicated internal connector class to express this concept. The parent object of the internal connector is the system element class to which it belongs. Internal connectors are divided into two states: used connectors and unused connectors. Each used connector has a next connector attribute. By querying its parent object, you can get the next system element class connected to the current system element class.

[0118] Each used connector has a unique system ID. During the structured conversion process, only objects with the same system ID as the connector need to be processed, avoiding situations such as entering an electrical system pipeline object when the conversion target is a duct system.

[0119] Internal connectors also have spatial position and direction information, but their direction may be opposite to the actual flow direction of the system. The actual flow direction of the system can be obtained through tree-structured conversion, and the directions of internal connectors of all system objects are sorted out. This allows the actual flow direction of each object in the electromechanical system to be visually and accurately expressed in the building information model.

[0120] As attached Figure 4The figure shows a duct elbow fitting. The duct elbow fitting has two connectors 1 and 2 at both ends, and the two connectors have default directions.

[0121] Step 1.2: System map class preparation step; create two system map classes, namely map vertex class and map edge class;

[0122] The map vertex class is used to express the system element class whose connection quantity status is non-two-way; a map vertex class stores two list information, namely the adjacent map vertex class list and the map edge class list;

[0123] An instance of a map vertex class can be understood as the concept of a subgraph in graph theory. In graph theory, a subgraph is a graph consisting of some vertices and edges selected from a graph, which retains some structural features of the original graph. By analogy, in the mechanical and electrical system of the building information model, the storage method of the map vertex class can be regarded as a method of building a subsystem. By selecting a specific list of map vertex classes, a data structure describing the subsystem can be formed, and these subsystems can represent branch components in the mechanical and electrical system.

[0124] The map edge class is used to express the edge between two map vertex classes. It is a sequence list composed of system element classes. The number of instances in the system element class sequence list is unlimited, so the number of corresponding relationships is unlimited. The map edge class can express its two endpoints composed of map vertex classes. In addition to the endpoints, the map edge class may not contain any other system element classes. In the electromechanical system, the two endpoints can be understood as the splicing of two pipes with multi-pass features.

[0125] Map edge classes are of practical significance in describing the specific paths of electromechanical systems. For example, map edge classes can express information such as the total geometric length, riser ratio, and nominal size based on the pipeline type system element class they contain.

[0126] Step 1.3: System topology map class extraction step: Create a system topology map class instance, which stores all map vertex classes and map edge classes; starting from any unit component in the electromechanical system of a building information model, create the first system element class, and use it as the first map vertex class to start extracting the entire system map structure; traverse and query the next element connected to the system element class, create an equal number of map edge classes according to its connection quantity status, and at the same time, decide whether to create a new map vertex class according to the element or add it to the corresponding map edge class according to the connection quantity status of the next element, and also mark all decided system element classes as processed to avoid repeated addition; recursively perform the above method until all connected system element classes in the system are marked as processed; all newly created map vertex classes and map edge classes are finally added to the system topology map class created in this step, thus completing the system topology map extraction step, and obtaining a system topology map class that fully expresses the connection information of the electromechanical system; Figure 5 It expresses the data structure of the system topology map class, including the system ID, the base point for generation, the map vertex class set, and the map edge class set; Figure 6 It expresses an acyclic system vertex topology relationship, in which the map edge connection between two map vertices V0 and V1 is composed of one system element class, and the map edge connection between two map vertices V3 and V4 is composed of three system element classes. Figure 5 The data structure can clearly show the system topology relationship.

[0127] Step 1.4: Optional, vertexization step of the vertical state transition points of the pipeline network; this step is a supplement to the system map structure extraction step. Considering that identifying and separating the riser part in the electromechanical pipeline network is of practical significance for the subsequent system feature analysis, the system element class with the feature of two-way type but vertical and non-vertical pipeline types before and after is also created as a map node class.

[0128] The practical significance of implementing the vertexization step of the vertical state transition point of the pipe network is that when analyzing the electromechanical system, it is often necessary to split the cross pipe and the riser part. By implementing the method of nodeization of the elbow between the cross pipe and the riser, the cross pipe and the riser are separated by a two-way node, so that they can be split into two map edge classes with significant cross pipe or riser geometric features for separate processing.

[0129] Attached Figure 7 It expresses that in order to analyze the riser path, the system elements such as the riser end elbow are noded.

[0130] According to an embodiment of the present invention, a basic flow direction designation method is provided, as shown in the attached Figure 8 shown.

[0131] The implementation of the basic flow direction specification method is achieved by implementing a depth-first search algorithm in the system topology map class, which can be divided into the following four steps. Fig. 9 To understand,

[0132] Step 2.1: Determine the system starting point; this is usually determined manually. If the manually determined system starting point is not a map vertex class, then find the map edge class to which it belongs and use the closest map vertex class endpoint in the map edge class as the system starting point (V1).

[0133] Step 2.2: Determine the calculation direction point; select any non-starting element in the system topology map ( Fig. 9 In the user-specified direction element, this selection is completely manually selected to ensure correctness. Similar to the determination of the system starting point, if the manually determined system starting point is not a map vertex class, the map edge class to which it belongs is searched, and the nearest map vertex class endpoint (V9) in the map edge class is used as the calculation direction point.

[0134] Step 2.3: Find the shortest path; starting from the system starting point, use the breadth-first search algorithm to calculate the shortest path from the system starting point to the calculation direction point. The path consists of a set of ordered map vertex class lists (V1, V2, V3, V4, V5, V7, V9), and the first vertex in this set of vertex lists is the system starting point.

[0135] Step 2.4: Determine the nearest directional vertex and the nearest system element class; according to the shortest path obtained, the second vertex in the map vertex class list is the nearest directional vertex (V2). A unique system map edge class can be determined by the system starting point and the nearest directional vertex. The second object in the internal system element list of this system map edge class is the system element class closest to the system starting point ( Fig. 9 The connection direction between the system starting point and the system element class is the physical space direction of the system basic flow direction.

[0136] According to an embodiment of the present invention, a ring node encapsulation method is provided, the purpose of which is to realize the operation of identifying and removing rings in the system topology map of the electromechanical system, and replace the ring structure in the system topology map with the encapsulated ring node, so as to prepare for the next step of tree structuring. Fig.10 As shown, the ring node encapsulation method includes the following steps:

[0137] Step 3.1: Preparation of the system path class; create a system path class for expressing the path in the system topology map, which stores a list of map vertex classes in an ordered manner, describing any path in the system map structure with ordered adjacent vertices as stations. The vertices corresponding to the starting point and end point of the path are the starting point and end point of the system path class; the map edge class can only express the path between two adjacent vertices, and the introduction of the system path class can express any path that spans more than two vertices;

[0138] The system path class also stores characteristic attributes used to analyze the electromechanical system, including the starting point and end point on the path, as well as the number, size, length, spatial position and other information of all system element classes on the path;

[0139] The system path class has the characteristics of being split and connectable. The split operation is implemented based on the existing key system vertex classes on its path, and the connection operation requires that the first and last vertices of the two system path classes to be connected coincide.

[0140] Attached Fig.11 The figure shows a schematic diagram of a standpipe path in a system path class. The ordered list of map vertex classes with travel directions is Vertex V2, Vertex V3, and Vertex V5. In addition to the vertex attributes themselves, it also includes information on all system element classes on the path.

[0141] The ring node class identifies and determines the structure; a ring node class is created to encapsulate the ring structure in the system topology map. It inherits from the system vertex class and rewrites the properties of its adjacent vertex list and adjacent edge list to express the import and export of the entire ring structure (ring entrance V2, ring exit V5). The structure inside the ring is similar to the path class, which is also an ordered map vertex class list. The properties of the internal main path and bypass path are added to express the complete system details in the ring structure.

[0142] Attached Fig.12 The figure shows a prototype form of a ring node, in which the conceptual expression of the internal main path (the ordered list of map vertex classes is V2, V3, V4, V5) and the bypass path can be seen.

[0143] Step 3.3: Determine the temporary main path step; starting from the system starting point, select a certain weight value according to the system characteristics, perform a breadth-first search algorithm on all terminal vertices, and filter out a temporary main path by combining the basic flow direction and the weight value;

[0144] This step processes one terminal vertex each time, and performs steps 3.4 to 3.6 according to the temporary main path obtained for the terminal vertex. The next terminal vertex is cycled based on the system topology map after the ring node is replaced in step 3.6. This cycle process is specifically described by step 3.7.

[0145] The purpose of determining the temporary main path is that the algorithm for identifying the ring structure needs to be based on an imaginary main path. By default, the physical quantity of length is selected as the weight value for determining the temporary main path, so that the imaginary main path is closer to the actual main path of the electromechanical system. The step of determining the temporary main path is carried out according to the following sub-steps:

[0146] Step 3.3.1: Filter out all terminal vertices in the system topology map; the characteristic of terminal vertices is a node with only one valid connector, that is, all map vertex classes with the number of objects in the adjacent vertex list being 1.

[0147] Step 3.3.2: Obtain the nearest direction vertex based on the basic flow direction of the system determined in the previous step 2 (step 2.4); use the nearest direction vertex as the vertex of the necessary path to perform a breadth-first search algorithm on all terminal vertices relative to the starting point to obtain the shortest route to the terminal vertex, and use the system path class to express the shortest routes to all terminal vertices and form a list.

[0148] Step 3.3.3: Filter out the temporary main path according to the preset weight; sort all the system path lists obtained in step 3.3.2 in descending order by length. Mark the end vertex of the temporary main path as a processed vertex, and determine the loop order of step 3.7 according to the end vertex of each path object in this system path list as the loop processing order.

[0149] Attached Fig.13 Shown is a graphical representation of the system topology map of the concepts set out in steps 3.3.1 to 3.3.3.

[0150] Attached Fig.13 The list of system paths obtained by the breadth-first search algorithm between the starting point V1 and the end vertex is shown below in order from the longest to the shortest total path length.

[0151] System path 1: V1-V2-V3-V4-V5-V6-V7-V8-V9-VT3;

[0152] System path 2: V1-V2-V3-V4-V5-V6-V7-VT2;

[0153] System path 3: V1-V2-V3-V4-V5-V6-V7-VT4;

[0154] System path 4: V1-V2-V3-V10-V11-VT1;

[0155] System path 5: V1-V2-V13-V14-V15-VT5;

[0156] Among them, system path 1 is the longest, so it is used as the first temporary main path to first perform ring structure identification.

[0157] Step 3.4: Identify the existence of a ring structure; use a cyclic stepping method to perform a depth-first search algorithm on the temporary main path to identify the ring structure in the electromechanical system pipe network;

[0158] The criterion for judging whether a loop structure exists between any two vertices is that there are more than one possible route between the two vertices. With the determination of the temporary main path, the step of identifying the existence of a loop structure can be decomposed into the following sub-steps for implementation.

[0159] Step 3.4.1: Determine whether there is a ring structure; based on the starting point and end point of the temporary main path, perform a depth-first search algorithm to query the number of feasible routes connecting the two points. If the number exceeds one, it can be determined that a ring structure exists on the temporary main path, and subsequent ring structure processing steps are required.

[0160] Attached Fig.14 It means that by running the depth-first search algorithm between V1 and VT3, two feasible routes are obtained, and it can be determined that there is a ring structure between V1 and VT3.

[0161] Step 3.4.2: Find the entrance of the ring structure; Fig.14 Taking the temporary main path shown in the figure as an example, the depth-first search algorithm (Step 1) is executed step by step with V1 as the starting point and VT3 as the end point. At the same time, the starting point of the previous step is used as the shielding point. The subsequent search conditions are obviously V2-VT3 (shielding V1, Step 2), V3-VT3 (shielding V2, V1, Step 3), Vn-VT3. As long as the number of possible paths returned each time is consistent with the number of possible paths of V1-VT3, the search can continue to move forward until V4-VT3 (shielding V3, V2, V1, Step 4). At this time, the number of possible paths decreases (1), and the search ends at this time. V3 is recorded as the starting point of the ring structure on this path, and it can be known that the shielded point V2 is an external adjacent vertex of the entrance of the ring structure. The whole process is shown in the attached figure. Fig.14 , 15 , 16, and 17.

[0162] Step 3.4.3: Find the exit of the ring structure; Fig.13 To Attachment Fig.17Taking the temporary main path shown as an example, in order to find the exit of the ring structure, the search order that needs to be executed is opposite to the method of finding the entrance of the ring structure, that is, starting from VT3 and ending at the ring entrance V3, and performing the depth-first search algorithm backward to search, it can be found that the number of feasible paths is 2. The subsequent search conditions are obviously V9-V3 (shielding VT3), V8-V3 (shielding VT3, V9)...Vn+1-V3, and the starting point of the previous step is used as the shielding point. As long as the number of feasible paths returned remains unchanged each time the search continues, until V4-V3, the number of feasible paths drops to only 1, then the search ends, and the starting point V5 of the previous reverse step is recorded as the starting point of the ring structure on this path. It can be seen that V6 is an external adjacent vertex of the exit of the ring structure. Attached Fig.18 , 19 The process of reverse step search and finding the ring interface exit on the temporary main path is demonstrated.

[0163] Step 3.4.4: Collect the internal vertices of the ring structure; Fig. 20 As shown in the figure, for the discovered ring structure entrance V3 and exit V5, the depth-first search algorithm is used to search all possible paths with the two as endpoints. For the returned system path list, all system vertex classes on the path are extracted and deduplicated, thus completing the collection of all vertices inside the ring structure. Fig. 20 In this step, the following vertex sets will be collected, including: V3, V4, V5, V10, V11, and V12.

[0164] Step 3.4.5: Identify the external vertices of the ring structure; use all the collected internal vertices of the ring structure as comparison objects, traverse the adjacent vertex list of each internal vertex, and if there is an adjacent vertex that does not belong to the internal vertex of the ring structure, the vertex can be determined to be an external vertex. Obviously, the entrance and exit of the known ring structure must be external vertices. Fig. 20 As shown, the available external vertex is V11, and it contains an external adjacent point VT1.

[0165] Step 3.5: Determine the main path and bypass path inside the ring structure; perform breadth-first and depth-first algorithms at the entrances and exits represented by the temporary main path at the boundary of the ring structure to determine the main path and bypass path inside the ring structure;

[0166] For electromechanical systems with tree-like structural characteristics, the main path is usually used to express its main functional segment, and the bypass path often appears in the system's auxiliary functions, such as maintenance, distribution, etc. Therefore, for ring nodes, identifying its internal main path and bypass path is of practical significance for understanding the function of the electromechanical system.

[0167] The steps of determining the main path and bypass path inside the ring structure are performed according to the following sub-steps:

[0168] Step 3.5.1: Determine the main path inside the ring structure; Generally, in electromechanical systems, the main path that realizes the system function is characterized by the internal main path of the ring structure, and it is often expressed as the shortest path. Therefore, under the default configuration, the length can be used as the weight, and the shortest path calculated by the breadth-first search algorithm at the entrance and exit of the ring structure can be used as the internal main path. Taking into account exceptional scenarios, the electromechanical system designer can specify the necessary nodes of the main path, and use the depth-first search algorithm to obtain the possible paths of all necessary nodes, and then sort the results by the length of the path to assist in determining the internal main path. Appendix Fig. 20 The V3, V4, and V5 nodes in the figure describe the internal main path in a ring structure.

[0169] Step 3.5.2: Determine the bypass path inside the ring structure; In the electromechanical system, in addition to the internal main path, the bypass path of a local ring structure has practical application significance and is generally used to implement auxiliary functions, such as balancing, measurement, maintenance and other application scenarios. Under the condition that the internal main path has been determined, the bypass path can be defined as a possible path that connects the inlet and outlet and does not overlap with the internal main path.

[0170] The specific way to implement this step is: take the middle vertex of the internal main path as the exclusion point, that is, exclude the remaining vertices of the entry and exit points to avoid repeated path finding, perform the depth-first search algorithm for the entry and exit points to calculate, and obtain the possible path list and sort it in ascending order by length weight. Based on the vertex objects other than the entry and exit points, remove the overlapping paths in the possible path list. The overlapping paths that appear in this step indicate that there is a ring nested structure inside the ring structure. It is necessary to nest and recursively execute steps 3.6 and 3.5.3 in the steps to complete the encapsulation and node replacement of all ring structures of the ring nodes.

[0171] Step 3.5.3: Recursively identify nested rings inside the ring structure; if an internal nested ring is identified in the ring structure, execute steps 3.4 to 3.6 with the nested ring as the temporary system, and exclude the adjacent node relationship outside the nested ring during execution, and lock the path search to execute inside the nested ring.

[0172] Step 3.6: Ring node replacement step: encapsulate all components and path information in the ring structure into a prepared ring node class instance, and replace the relevant nodes in the topology map with the instance;

[0173] Step 3.7: After the topology map is refreshed, execute steps 3.3 to 3.6 repeatedly until all temporary paths are traversed and all ring nodes in the electromechanical system are encapsulated and replaced. At this time, the topology map has the basic conditions for subsequent tree structure processing. Fig.21 It shows that based on the Fig.13 The system topology map shown is the state after the ring nodes are encapsulated and replaced.

[0174] According to step 4 of the embodiment of the present invention, a system topology map tree structure conversion method is provided, such as Fig. 22 shown.

[0175] Step 4.1: Preparation of the phylogenetic tree class; create a phylogenetic tree class based on a tree structure for expressing the phylogenetic map structure, which stores the phylogenetic path expressed by the current phylogenetic tree itself, the branch code of the current phylogenetic tree, the depth of the current tree, a list of any number of phylogenetic trees following the current phylogenetic tree (a sub-phylogenetic tree set), and the parent tree object of the current phylogenetic tree (also a phylogenetic tree class); the phylogenetic tree also stores characteristic attributes for analyzing electromechanical systems, including the number, size, length, and spatial position of all system elements on the current path, as well as the characteristic attributes of subsequent phylogenetic trees, and obtains statistical values ​​for specific characteristic attributes of the phylogenetic tree through arithmetic methods such as summation and averaging;

[0176] Step 4.2: the step of creating a system tree class; create a system tree class instance, take the system topology map after the ring node encapsulation in step 3 as the data source basis, start from the system starting point along the system basic flow direction, identify and create a system path class, and use the identified system path as the current path of the current system tree class instance; start from the end point of the current system path, search and create all the subsequent system path classes of its adjacent non-current paths, create a new system tree class instance to store the subsequent system path, and add the new system tree class instance to the list (sub-system tree set) composed of any number of system trees subsequent to the current system tree of the upper-level system tree class instance, recursively repeat the above steps until the entire system topology map is searched, and the system tree with a tree structure after the system topology map is converted is output;

[0177] Optional step 4.3: Phylogenetic tree numbering step; set a default numbering rule to use "-" as a connector to number each level of phylogenetic tree. The root tree is the source of the phylogenetic tree, and the numbering starts with the number 1. The nth tree used in the subsequent word number set is numbered according to the "1-n" rule. And so on, for example: the number 1-2-3 means the third subsequent subtree of the second subtree in the root tree. According to this rule, each subtree of a complete phylogenetic tree has a unique number.

[0178] Attached Fig.23It indicates that a system topology map tree with ring nodes is structured to become a system tree and completed with numbering. Fig.24 It is a hierarchical expression based on tree structure numbers and paths. It can be seen that the system tree after vertex analysis can summarize and express the common complex routes in electromechanical systems, and record their topological relationships. The encapsulation of the ring structure allows the local bypass paths or bypass routes in the electromechanical system to be integrated in the system tree, and its main functional segments are represented externally.

[0179] According to step 5 of the embodiment of the present invention, a system tree feature analysis method is provided, which is implemented by using a system tree feature analysis module in a tree structured conversion and analysis device based on a building information model electromechanical system provided by the present invention.

[0180] The system tree feature analysis step is used to perform system tree feature analysis, use different weights to analyze features, and use visualization to express the system tree structure, which is used to carry out specific subsequent applications after the mechanical and electrical system tree structure conversion of the building information model. Specifically, the system tree feature analysis step can be implemented through the following method embodiments. Fig.25 As shown, the analysis steps that can be performed by the phylogenetic tree feature analysis module include the following methods:

[0181] Method 5.1: Implementation of the automatic search method for risers in the system tree; In the mechanical and electrical systems of buildings, the systems are often distributed according to floors, and the systems on each floor are connected through risers. The riser path can be implemented by manually specifying: manually specify the riser top component, and search the system tree for the shortest path from the starting point to the top of the riser. In addition to manually specifying the riser path, automatically searching the riser path in the system tree has practical significance for analyzing the mechanical and electrical system, which can be decomposed into the following steps:

[0182] Step 5.1.1: Confirm the riser starting point; starting from a certain point in the system tree, find the first riser closest to the starting point along the development direction of the system tree, and use the system tree where the riser is located as the current system tree to start the riser search.

[0183] Step 5.1.2: Calculate the current path vertical rate of the current phylogenetic tree;

[0184] Step 5.1.3: Calculate the current path vertical rate of the next level subtree of the current phylogenetic tree;

[0185] Step 5.1.4: Determine whether it is in the main riser path; calculate whether the verticality of the current path of the system tree is greater than the verticality of any subtree at the next level. If it is positive, the subtree with the maximum verticality is used as the new current system tree.

[0186] Step 5.1.5: Extract the riser path; recursively perform steps 5.1.1 to 5.1.4. If a negative situation occurs, end the riser search, then record the phylogenetic tree of all recursive pathways, extract their paths and merge them into a complete riser path.

[0187] Method 5.2: Implementation of the system tree floor definition method; In the building mechanical and electrical system, the designer pays attention to two types of system floor information. One is the floor where the terminal is located, and the other is the floor through which the pipeline passes. The floor definition method of defining the system tree includes the following two steps:

[0188] Step 5.2.1: Obtain the system riser path through automatic riser search or manual designation method. It can be seen that the subsystem trees branching from each node in the riser path that have no overlapping paths with the riser path are all floor branch system trees developed on the riser path.

[0189] Step 5.2.2: The floor information of the branch system tree of each floor is obtained by counting the spatial information of the terminals and pipelines. Generally, the floor with the largest proportion of components is the characteristic floor of the branch system tree of that floor.

[0190] Step 5.2.3: Sort the branch system trees of each floor by floor elevation to obtain a floor branch system tree that matches the number of floors in the building, and conduct further electromechanical system characteristic analysis based on the floor information.

[0191] Attached Fig.26 A sample of a ventilation system is shown, which contains two fans connected in parallel to form a ring structure, and the whole system is divided into two layers. The ends of the system are air outlets and rain shutters. In addition to branch pipes such as tees and crosses, the air duct is also equipped with air duct accessories such as check valves and regulating valves. By implementing the tree structure conversion method of the electromechanical system described in the present invention to obtain a system tree, and implementing the system tree riser automatic search method in the system tree feature analysis method, a system tree in the shape of the attached figure can be obtained. Fig. 27 system analysis results.

[0192] Method 5.3: Implementation of the system tree flow identification method; The tree-structured processing of the system topology map essentially completes the flow arrangement process of the building mechanical and electrical system. For the current path of any system tree, its starting point to the end point represents the flow of the system. The flow information is ultimately reflected through the direction attribute of the virtual connector, so the actual flow direction of the connectors of all system components in the system tree can be determined based on the development direction of the tree structure. Finally, by recording the coordinates and directions of the connectors, the flow display arrows of all mechanical and electrical system components can be generated in three-dimensional space, helping mechanical and electrical designers to master comprehensive system flow information.

[0193] Method 5.4: Implementation of the system tree flow statistics method; In the design of building mechanical and electrical systems, the flow of common systems can generally be obtained by reversely deducing the design demand flow at the end along the system flow direction.

[0194] In other specific types of systems, the flow accumulation rules also need to be determined according to the different design conditions of the specific system. For example, the fire smoke exhaust system allocates smoke exhaust volume according to the design principles of smoke and fire zones; the automatic sprinkler fire extinguishing system is determined according to the nozzle flow demand of the service area under the most unfavorable working condition; the fire elevator vestibule mechanical pressurized air supply system opens the air supply port on a specific floor according to the fire floor, and its flow needs to be determined according to the most unfavorable working condition to maintain the positive pressure of the vestibule. These systems with specific rule control can implement the flow statistics of the system tree by configuring the professional design characteristics of each system tree to formulate flow statistics rules.

[0195] Method 5.5: Implementation of the system tree hydraulic calculation method; Based on the completion of the system tree flow statistics method, the system tree hydraulic calculation method can be implemented. The implementation of hydraulic calculations requires the configuration of precondition parameters, including the resistance coefficient along the way, the local resistance coefficient of various types of pipes, terminals, and equipment. These precondition parameters can be created through other databases or custom methods, and applied to the hydraulic calculation process of the system tree during calculation. Since the tree-structured electromechanical system topology map already has significant tree structure characteristics, some complex ring pipe networks are node-abstracted. Under the condition of complete precondition conditions, the hydraulic characteristics of each branch of the system tree can be obtained through simple data calculations, which significantly improves the efficiency and accuracy of hydraulic calculations.

[0196] Method 5.6: Implementation of the system tree material statistics method; Material statistics can be implemented on any system tree by formulating material statistics rules for different types of components, such as thickness rules and insulation layer setting rules for air ducts; rules for merging and counting terminal equipment according to characteristic parameters; classification and piece counting rules for water pipe fittings, and calculation rules for air duct material usage for air duct fittings, etc. By specifying the above rules, material statistics can be performed on the current system tree and its branches at each level, each floor branch, and each room branch, so that the granularity of material statistics can be refined according to the needs of business scenarios, and accurate statistical results can be obtained in real time.

[0197] Method 5.7: Implementation of the logical relationship analysis method of system tree components; In the business scenario of electromechanical system design, the logical relationship analysis of some specific component conditions has practical significance. For example, in the automatic sprinkler fire extinguishing system, whether an automatic exhaust valve is installed at the top of the riser, and whether a hydraulic indicator is installed at the starting end of the system; in the fire smoke exhaust system, whether a 280℃ fire damper is installed at the starting end of the floor branch, and whether a normally closed smoke exhaust valve is installed at the entrance of the room branch; in some electromechanical systems that reflect the process of technological processing, analyze whether two components have passed through a specific type of device in a specific order to determine whether the system setting meets the process requirements of the process of technological processing. The above requirements can be completed by extracting and analyzing the system tree on demand, and the implementation steps of the logical relationship analysis of the system tree components can be completed by setting the system path and logical rule conditions.

[0198] The present invention provides a tree-structured conversion and analysis device system tree based on a building information model electromechanical system, as shown in the attached Fig.28 As shown, the custom pruning and grafting module can be implemented through the following embodiments.

[0199] Attach Fig.29 Taking the two fire sprinkler systems shown as examples, which are not connected to each other, in the building electromechanical model, due to potential modeling problems, the two groups of systems that should be the same system may not be physically connected in the model. In the design scenario, it is necessary to regard these two groups of systems as being in the same topology for system analysis. The present invention provides a method for implementing a logical connection between these two physically separated systems. Conversely, for system branches with redundant connections, the separation operation can also be implemented using the method of this module. The connection operation is implemented by the following steps:

[0200] Step 6.1: Perform transformation to obtain a phylogenetic tree; perform tree structure transformation on the two separated phylogenetic trees respectively to obtain two independent phylogenetic trees, which are divided into a parent tree and a new subtree. The goal is to graft the new subtree into a branch of the parent tree.

[0201] Step 6.2: Determine the connection point; select a branch in the parent tree and the next level subtree set of the branch;.

[0202] Step 6.3: Execute the connection; add the new subtree to this collection and set the parent object of the new subtree to the branch in the selected parent tree.

[0203] Step 6.4: Arrange the relationship; re-execute the tree structure analysis process on the grafted parent tree, renumber each node to reflect the impact of the newly inserted subtree on the branch numbering of each level of the tree; according to the business scenario, re-implement the riser analysis and floor analysis methods to obtain new valid results. At this point, the process of grafting the new subtree system to the parent tree is completed.

[0204] Follow steps 6.1 to 6.4 to Fig.29 After the fire sprinkler system shown in the figure is implemented with the grafting method, the attached Fig.30 The complete fire sprinkler system with virtual connection is shown in the figure, on which the subsequent system design and analysis work can be carried out.

[0205] The present invention provides a tree-structured conversion and analysis method for electromechanical systems based on building information models, which realizes automated transcription while being supplemented by visualization, interactivity and other features to improve the efficiency and quality of building electromechanical professional design. Based on the electromechanical system after tree-structured conversion, professional design tasks such as generating two-dimensional planar system diagrams, system pipe network hydraulic calculations, system service object analysis, system material statistics, and system design verification can also be performed. At the same time, the converted electromechanical system data is saved and transmitted in a standardized data structure form, which makes it possible to quickly convert design results into standard data. Performing analysis on a large amount of converted electromechanical system data through intelligent algorithm means such as machine learning and large model training helps to achieve artificial intelligence's system-level understanding of building electromechanical systems.

[0206] The above-mentioned embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A tree structure conversion method and analysis method based on building information model electromechanical system, characterized in that: The specific steps include: Step 1: Create a system topology map. Take a component in the electromechanical system as the starting point to obtain the system topology map of the electromechanical system. The implementation steps are as follows: Step 1.1: Create a system element class for expressing various unit components in the electromechanical system. The system element class stores the element name, unique identification ID, system ID and name, spatial location, room information, and internal connector information. The system element class is subdivided into five subdivision types: pipeline type, pipe fitting type, pipeline accessory type, power equipment type, and terminal type according to the characteristics of the specific instance. According to the internal connector information in the system element class, the next closest system element class to which the individual instance of the system element class is connected can be queried; Step 1.2: Prepare the system map class and create two system map classes, namely map vertex class and map edge class; The map vertex class is used to express the system element class whose connection quantity status is non-two-way; a map vertex class stores two list information, namely the adjacent map vertex class list and the map edge class list; The map edge class is used to express the edge between two map vertex classes, which is an ordered list of system element classes; Both the map vertex class and the map edge class can characterize their total geometric length according to the pipeline type system element class they represent; Step 1.3: Extract the system topology map class and create a system topology map class instance, which stores all map vertex classes and map edge classes; Starting from any unit component in the electromechanical system of a building information model, create the first system element class and use it as the first map vertex class to start extracting the entire system map structure; traverse and query the next element connected to the system element class, create an equal number of map edge classes according to its connection quantity status, and decide whether to create a new map vertex class according to the element or add it to its corresponding map edge class according to the connection quantity status of the next element, and mark all decided system element classes as processed to avoid repeated addition; recursively perform the above method until all connected system element classes in the system are marked as processed; all newly created map vertex classes and map edge classes are finally added to the system topology map class created in this step, thus completing the system topology map extraction step and obtaining a system topology map class that fully expresses the connection information of the electromechanical system; Step 2: Formulate the basic flow direction. According to any non-starting component in the electromechanical system, determine the basic flow direction of the system as the development direction of the tree structure. Step 3: Ring node encapsulation: identify the ring network structure according to the basic flow direction of the system and encapsulate it into a ring node, replace the relevant vertices in the system topology map in step 1, so that the system topology map has the basic characteristics of tree-like structure; The implementation steps are as follows: Step 3.1: Preparation of the system path class; create a system path class for expressing the path in the system topology map, which is stored in an ordered manner as a list of map vertex classes, describing any path in the system map structure with ordered adjacent vertices as stations. The vertices corresponding to the starting point and the end point of the path are the starting point and the end point of the system path class; The system path class also stores characteristic attributes used to analyze the electromechanical system, including the starting and ending points on the path, as well as the number, size, length, and spatial position of all system elements on the path; Step 3.2: Determine the path search weight attribute; select a certain characteristic attribute as the weight value according to the characteristic attribute of the electromechanical system itself or customized, and the characteristic attribute is also used as the weight value standard for subsequent loops and recursive execution of breadth-first and depth-first algorithms; Step 3.3: Determine the temporary main path step; starting from the set system starting point, execute the breadth-first search algorithm, and determine the temporary main path in combination with the basic flow direction; Step 3.4: Identify the existence of a ring structure; use a cyclic stepping method to perform a depth-first search algorithm on the temporary main path to identify the ring structure in the electromechanical system pipe network; Step 3.5: Determine the main path and bypass path inside the ring structure; perform breadth-first and depth-first algorithms respectively at the entrances and exits represented by the temporary main path at the boundary of the ring structure to determine the main path and bypass path inside the ring structure; Step 3.6: Ring node replacement step: encapsulate all components and path information in the ring structure as a ring node to replace the relevant nodes in the topology structure map; Step 3.7: loop and recursion step; based on the system topology map partially replaced with ring nodes, repeat steps 3.3 to 3.6 until all ring structures in the system complete topology map are replaced with ring nodes; Step 4: Tree structure conversion: starting from the set system starting point and following the system basic flow, the system topology map after executing the ring node encapsulation is converted into a tree structured system tree; Step 5: System tree feature analysis, use different weight values ​​to perform system tree feature analysis, and use visualization to express the system tree structure for specific subsequent applications after the structured conversion of the building information model electromechanical system tree.

2. The tree structure conversion method and analysis method based on the mechanical and electrical system of the building information model according to claim 1 is characterized in that: In step 1.1, the system element class is subdivided into five subdivided types, namely, pipeline type, pipe fitting type, pipeline accessory type, power equipment type and terminal type, according to the characteristics of the specific instance; the system element class of the pipeline type stores not only the element information but also the geometric length of the pipeline, and is classified into two types, vertical riser and non-vertical riser, according to its geometric characteristics; In addition to storing element information, all system element classes of pipe fitting types also store the connection quantity status. According to the actual connection quantity of its internal connectors, the system element class is determined to be single-way, two-way, three-way, ..., n-way, expressing the connection quantity status characterized by quantity.

3. The tree structure conversion method and analysis method based on the mechanical and electrical system of the building information model according to claim 2 is characterized in that: The implementation of the basic flow direction formulation method in step 2 is achieved by implementing a depth-first search algorithm in the system topology map class. The specific steps are as follows: Step 2.1: Determine the system starting point; manually determine. If the manually determined system starting point is not a map vertex class, then find the map edge class to which it belongs and use the closest map vertex class endpoint in the map edge class as the system starting point; Step 2.2: Determine the calculation direction point; select any non-starting element in the system topology map. To ensure correctness, this selection is completely done manually. Same as the method of determining the system starting point, if the manually determined system starting point is not a map vertex class, then find the map edge class to which it belongs and use the closest map vertex class endpoint in the map edge class as the calculation direction point. Step 2.3: Find the shortest path; starting from the system starting point, use the breadth-first search algorithm to calculate the shortest path from the system starting point to the calculation direction point. The path consists of a set of ordered map vertex lists, and the first vertex in this set of vertex lists is the system starting point; Step 2.4: Determine the nearest directional vertex and the nearest system element class; according to the obtained shortest path, the second vertex in the map vertex class list is the nearest directional vertex; a unique system map edge class can be determined by the system starting point and the nearest directional vertex, and the second object in the internal system element list in this system map edge class is the system element class closest to the system starting point. The connection direction between the system starting point and the system element class is the physical space direction of the system basic flow.

4. The tree structure conversion method and analysis method based on the mechanical and electrical system of the building information model according to claim 3 is characterized in that: The steps 3.4 and 3.5: create a ring node class for encapsulating the ring structure in the system topology map, which inherits from the system vertex class, rewrites the properties of its adjacent vertex list and adjacent edge list, and is used to express the import and export of the entire ring structure; the internal structure of the ring node is similar to the system path class, and an ordered map vertex class list is used for storage and expression, and new internal main path and bypass path properties are added to express the complete system detail information within the ring structure.

5. The tree structure conversion method and analysis method based on the mechanical and electrical system of the building information model according to claim 4 is characterized in that: The specific implementation steps of step 4 are as follows: Step 4.1: Preparation of the phylogenetic tree class, creating a phylogenetic tree class based on a tree structure for expressing a phylogenetic topology map, which stores the phylogenetic path expressed by the current phylogenetic tree itself, the branch code of the current phylogenetic tree, the depth of the current tree, a list of any number of phylogenetic trees following the current phylogenetic tree, and the parent tree object of the current phylogenetic tree; The system tree also stores characteristic attributes used to analyze the electromechanical system, including the number, size, length, and spatial position of all system elements on the current path, as well as characteristic attributes of subsequent system trees, and obtains statistical values ​​for specific characteristic attributes of the system tree through arithmetic methods of summation and averaging; The system tree nodes are derived from the non-ring common vertices and encapsulated ring nodes in the system topology map encapsulated by the ring nodes; Step 4.2: The creation step of the system tree class is based on the system topology map. From the set starting point, along the system basic flow, the system tree class instance is recursively constructed; each path is used as the current path of the current system tree class instance, and the new system tree class instance corresponding to the sub-path is added to the sub-system tree collection of the parent system tree class instance; the above steps are repeated until the entire system topology map is traversed and finally converted into a tree-structured system tree.

6. The tree structure conversion method and analysis method based on the mechanical and electrical system of the building information model according to claim 5 is characterized in that: The system tree feature analysis in step 5: the system is distributed according to the floor, the system on each floor is connected through the riser, and the riser path in the system tree is automatically searched to analyze the electromechanical system; the search steps are as follows: Step 5.1.1: Confirm the riser starting point; starting from a certain point in the system tree, find the first riser closest to the starting point along the development direction of the system tree, and use the system tree where the riser is located as the current system tree to start the riser search; Step 5.1.2: Calculate the current path vertical rate of the current system tree; Step 5.1.3: Calculate the current path vertical rate of the next level subtree of the current phylogenetic tree; Step 5.1.4: Determine whether it is in the main riser path; calculate whether the verticality of the current path of the system tree is greater than the verticality of any subtree at the next level; if it is affirmative, use the subtree with the maximum verticality as the new current system tree; Step 5.1.5: Extract the riser path; Recursively repeat steps 5.1.1 to 5.1.

4. If a negative situation occurs, the riser search is terminated, and the phylogenetic trees of all recursive pathways are recorded, and their paths are extracted and merged into a complete riser path.

7. The tree structure conversion method and analysis method based on the mechanical and electrical system of the building information model according to claim 5 is characterized in that: System tree feature analysis in step 5: In the building mechanical and electrical system, the designer pays attention to two types of system floor information, one is the floor where the terminal is located, and the other is the floor through which the pipeline passes; the steps of the system tree floor definition method are as follows: Step 5.2.1: Obtain the system riser path by automatic riser search or manual designation method. It can be seen that the subsystem trees branched from each node in the riser path and having overlapping paths with the riser path are all floor branch system trees developed on the riser path; Step 5.2.2: Obtain the floor information of the branch system tree of each floor by counting the spatial information of the terminal and pipeline, and take the floor with the largest proportion of components as the characteristic floor of the branch system tree of the floor; Step 5.2.3: Sort the branch system trees of each floor by floor elevation to obtain a floor branch system tree that matches the number of floors in the building, and conduct further electromechanical system characteristic analysis based on the floor information.

8. The tree structure conversion method and analysis method based on the mechanical and electrical system of the building information model according to claim 5 is characterized in that: The system tree feature analysis in step 5: In the design of building mechanical and electrical systems, the flow statistics of the system are obtained by reversely deducing the design demand flow at the end along the system flow direction.

9. The tree structure conversion method and analysis method based on the mechanical and electrical system of the building information model according to claim 8 is characterized in that: The system tree feature analysis in step 5: implementation of the system tree hydraulic calculation method; implementation of the system tree hydraulic calculation method based on the completion of the system tree flow statistics; precondition parameters need to be configured for the implementation of hydraulic calculation, including the resistance coefficient along the way, the local resistance coefficient of various types of pipe fittings, terminals, and equipment; the precondition parameters are created through other databases or custom methods, and applied to the hydraulic calculation process of the system tree during calculation; since the tree-structured electromechanical system topology map already has significant tree structure characteristics, the complex ring pipe network is node-abstracted, and the hydraulic characteristics of each branch of the system tree can be obtained through simple data calculation when the precondition conditions are complete.

10. The tree structure conversion method and analysis method based on the mechanical and electrical system of the building information model according to claim 5, characterized in that: The system tree feature analysis in step 5: implementation of the system tree material statistics method; material statistics work is implemented on any system tree, by formulating material statistics rules for different types of components, rules for merging and counting terminal equipment according to characteristic parameters, classification and piece-counting rules for water pipe fittings, and calculation rules for duct material usage for duct fittings. Through the specification of the above rules, material statistics are performed on the current system tree and its branches at each level, each floor branch, and each room branch, so that the granularity of material statistics is refined as the needs of the business scenario, and accurate statistical results are obtained in real time.

Images