Configuration method and system based on BIM digital base

Abstract

The invention relates to the technical field of data processing, provides a configuration method and system based on a BIM digital base, and greatly reduces manual intervention and reduces uncertainty and error risks caused by manual operation through automatic analysis and component feature extraction of a BIM model and configuration element matching and control mechanism derivation. It is ensured that the whole configuration process is more efficient, accurate and stable; a control mechanism is deduced based on a topological relation, and a visual state is dynamically updated according to real-time external data, so that response can be made by closely following the working condition change of the equipment, the smoothness of cooperative work between the equipment is guaranteed, and the overall operation efficiency is improved; an interaction channel among BIM model data, equipment real-time operation data and a control instruction is opened, organic integration of building information, equipment operation information and control logic is realized, and a solid data and technical foundation is laid for advanced applications such as subsequent intelligent operation and maintenance, digital twinning and the like.

Description

Technical Field

[0001] This application relates to the technical field of data processing, and particularly to a configuration method and system based on a BIM digital base. Background Art

[0002] With the continuous development of Building Information Modeling (BIM) technology and industrial automation configuration technology, both have achieved remarkable results in their respective fields. The BIM technology uses a three-dimensional digital model as a carrier to integrate data such as geometric information and semantic attributes of the entire life cycle of a building or equipment, and can achieve accurate modeling and visual expression of physical objects, and is widely used in fields such as building design, construction management, and facility operation and maintenance; the configuration technology associates sensor data with virtual components through a graphical interface to achieve real-time monitoring and control of industrial equipment, and plays an important role in industrial automation scenarios such as power, chemical industry, and manufacturing.

[0003] The existing technology still has deficiencies. On the one hand, traditional BIM applications focus on static model display, lack in-depth interaction with external data, and are difficult to dynamically reflect changes in equipment operation status, and cannot meet the requirements of real-time and interactivity for intelligent equipment operation and maintenance; on the other hand, industrial configuration technology mostly relies on manual configuration, and it is necessary to manually draw components and write control logic. When facing complex equipment systems, not only is the efficiency low, but it is also difficult to accurately match the spatial structure and topological relationship of the equipment, resulting in the disconnection between the control logic and the actual working conditions.

[0004] In addition, in the existing technology, BIM and configuration technology are independent of each other, and cannot realize the collaborative linkage of building information, equipment operation data, and control instructions, and cannot provide a comprehensive and unified data and control basis for intelligent operation and maintenance. Most of them do not solve how to realize the integration of the BIM model and configuration technology to complete model parsing, configuration component matching, and visual display. Summary of the Invention

[0005] Aiming at the deficiencies of the existing technology, this application provides a configuration method and system based on a BIM digital base.

[0006] In a first aspect, this application provides a configuration method based on a BIM digital base. The method includes: parsing the BIM model of the target equipment and extracting component attributes including the geometric information, semantic attributes, and topological relationship of the components, identifying and marking configurable components, and simultaneously generating component features including component identifiers, spatial coordinates, and component attributes; Based on the domain knowledge base and component features, match the configuration components associated with the configurable components, determine the initial layout position of the configuration components in the BIM model according to the spatial coordinates, and deduce the control mechanism of the configuration components based on the topological relationship. At the same time, establish a data channel with external data, determine the binding relationship between the display attributes of the configuration components and the visualization state of the components and the external data, and determine the state mapping relationship; Monitor the real-time external data of the target device, update the display attributes of the configuration components according to the binding relationship, and dynamically drive the change of the visualization state of the components based on the state mapping relationship. At the same time, respond to the user's operation on the configuration components and send control instructions to the target device, and synchronously update the visualization state of the configuration components and the components based on the execution results of the target device.

[0007] As an alternative implementation, the generation strategy of the component features includes: Identify and mark the configurable components from the BIM model according to the preset judgment rules and component attributes, and at the same time analyze the spatial position relationship between the components through the spatial topology analysis algorithm to form spatial coordinates; Encode the geometric information and semantic attributes of the components through the cryptographic hash algorithm to generate component identifiers; Integrate the component identifiers, spatial coordinates and component attributes to generate component features.

[0008] As an alternative implementation, the extraction sub-strategy of the component attributes includes: Analyze the structured data of the BIM model of the target device, divide the geometric contours of the components in the BIM model through the point cloud segmentation algorithm, and extract the geometric information of the components; Perform semantic analysis on the attribute text of the BIM model through natural language processing technology, extract the semantic attributes of the components, and associate and match the semantic attributes of the components with the domain knowledge base; Construct a component topology structure based on the graph theory algorithm, use the components as nodes and the connection relationships between the components as edges, analyze the component topology structure to determine the topological relationship of the components, and form component attributes.

[0009] As an alternative implementation, the derivation strategy of the control mechanism of the configuration components includes: Based on the component topology structure, identify the connection paths and hierarchical relationships between the components through the path search algorithm, and retrieve the basic control mechanism based on the domain knowledge base. At the same time, mark the nodes of the component topology structure as the trigger points of the basic control mechanism; Extract the spatial constraint conditions according to the initial layout position of the configuration components in the BIM model, and transform the spatial constraint conditions into additional rules of the basic control mechanism to form a constraint control mechanism; Take the constraint control mechanism as the initial policy of reinforcement learning, use the operating efficiency and energy consumption of the target device as the reward function of reinforcement learning, and combine external data for multiple rounds of iterative training to derive the control mechanism of the configuration components.

[0010] As an alternative implementation, the matching sub-policy of the configuration component includes: Match the semantic attributes of the configurable component with the semantic labels of the configuration components in the domain knowledge base, and calculate the semantic similarity score between the configurable component and the configuration component to screen out candidate configuration components; Construct the state space of reinforcement learning by combining the topological relationship and geometric information of the components, and use the adaptability of the candidate configuration components as the reward function of reinforcement learning; Through multiple rounds of iterative training of reinforcement learning, dynamically optimize the candidate configuration components and match the configuration components associated with the configurable component.

[0011] As an alternative implementation, the determination sub-policy of the binding relationship includes: Conduct multi-dimensional analysis on the external data and extract the data features of the external data through principal component analysis. At the same time, analyze the display attributes of the configuration components and the parameter requirements of the visualization state of the components to form attribute parameters; Take the data features of the external data as the horizontal dimension and the attribute parameters as the vertical dimension to construct a feature mapping matrix, and calculate the matching degree between the data features of the external data and the attribute parameters through the cosine similarity algorithm to generate the initial binding relationship; Take the data transmission efficiency and visualization accuracy as the fitness function, and iteratively optimize the initial binding relationship through the genetic algorithm. Through selection, crossover and mutation operations, determine the binding relationship between the external data and the attribute parameters.

[0012] As an alternative implementation, the update strategy for the visualization state of the configuration components and components includes: Monitor and respond to user operations on the configuration components, convert the user operations on the configuration components into control instructions based on the domain knowledge base, and transmit the control instructions to the target device; Receive the execution result of the target device, verify the execution effectiveness of the expected result and the execution result of the control instruction, and modify the display attributes of the configuration components according to the execution effectiveness and the binding relationship; Update the geometric shape and appearance state of the component based on the execution effectiveness and the mapping action, and trigger the state update of the associated component according to the topological relationship of the component.

[0013] As an alternative implementation, the update sub-policy of the display attributes of the configuration components includes: Monitor the real-time external data of the target device, perform multi-scale analysis on the real-time external data to extract spatio-temporal correlation features, calculate the matching degree between the spatio-temporal correlation features and the feature templates in the binding relationship through cosine similarity, and screen out update actions; Dynamically adjust the display attributes of the configuration components according to the update actions, and judge the type of the display attributes to determine the rendering parameters of the display attributes. The types of the display attributes include numerical attributes, trend attributes, and abnormal attributes; Construct the state space of reinforcement learning according to the spatio-temporal correlation features and the type of the display attributes, use the combination of the rendering parameters of the display attributes as the action space of reinforcement learning, and comprehensively determine the reward function of reinforcement learning according to the information transfer efficiency, visual comfort, and interaction convenience, and train the reinforcement learning to optimize the visual mapping effect of the display attributes.

[0014] As an optional implementation manner, the driving sub-strategy for the visualization state change of the component includes: Find the mapping actions associated with the component from the state mapping relationship according to the spatio-temporal correlation features, and convert the external data into the visualization parameters of the component. The visualization parameters include geometric form, appearance state, and rendering effect; Determine the influence range of the visualization state change of the component based on the connection relationship between the components, and modify the geometric form and appearance state of the component according to the mapping actions; Dynamically adjust the rendering effect of the component according to the importance of the visualization state, and respond to the user's operation on the configuration component to feedback the visualization parameters of the component.

[0015] In a second aspect, the present application provides a configuration system based on a BIM digital base. The system includes: a feature extraction module, a configuration configuration module, and a mapping feedback module.

[0016] The feature extraction module is used to parse the BIM model of the target device and extract the component attributes including the geometric information, semantic attributes, and topological relationships of the components, identify and mark the configurable components, and generate component features including component identifiers, spatial coordinates, and component attributes.

[0017] The configuration configuration module is used to match the configuration components associated with the configurable components based on the domain knowledge base and the component features, determine the initial layout position of the configuration components in the BIM model according to the spatial coordinates, deduce the control mechanism of the configuration components based on the topological relationships, establish a data channel with the external data, configure the display attributes of the configuration components and the binding relationship between the visualization state of the components and the external data, and determine the state mapping relationship.

[0018] The mapping feedback module is used to monitor the real-time external data of the target device, update the display properties of the configuration elements according to the binding relationship, and dynamically drive the visual state changes of the components based on the state mapping relationship. At the same time, it responds to the user's operation on the configuration elements and sends control instructions to the target device, and synchronously updates the visual state of the configuration elements and components based on the execution results of the target device.

[0019] Compared with the prior art, the beneficial effects of the present application are: through the automated parsing of BIM models and extraction of component features, as well as configuration element matching and control mechanism derivation, human intervention is greatly reduced and the uncertainty and error risks caused by human operation are reduced, ensuring that the entire configuration process is more efficient, accurate and stable; the control mechanism is derived based on the topological relationship, and the visualization status is dynamically updated according to real-time external data, so that it can closely follow the changes in equipment operating conditions and respond, ensuring the smoothness of collaborative work between equipment and improving overall operating efficiency; real-time and intuitive visualization status feedback, as well as a convenient user operation response mechanism, enable users to quickly and clearly grasp the equipment operating status, and efficiently control and manage the equipment, significantly reducing the difficulty and complexity of operation and maintenance operations; it opens up the interaction channel between BIM model data, real-time equipment operation data and control instructions, and realizes the organic integration of building information, equipment operation information and control logic, laying a solid data and technical foundation for subsequent advanced applications such as smart operation and maintenance and digital twins.

[0020] Through point cloud segmentation algorithm, natural language processing technology and graph theory algorithm, the geometric information, semantic attributes and topological relationships of components are extracted respectively. Compared with a single data extraction method, it can obtain complete information of components in an all-round and in-depth manner, ensure the comprehensiveness and accuracy of data, and provide reliable data support for subsequent work; identify and mark configurable components, accurately screen out components that really need to be configured, avoid invalid data processing and calculation of irrelevant components, greatly improve overall operating efficiency, and reduce resource waste; generate component identification and integrate various types of information into component features, build a standardized and structured data format, and provide a unified and standardized data interface for the follow-up, ensure the consistency and traceability of data in the interaction process of each link, and effectively reduce the possibility of errors in data interaction.

[0021] Perform configuration component matching based on the domain knowledge base and component features, change the limitations of traditional semantic - only matching, evaluate the adaptability of configuration components and components from multiple dimensions, significantly improve the accuracy and reliability of matching, ensure that the selected configuration components can perfectly fit the component functions, and avoid the problem that the functions cannot be realized normally due to improper selection of configuration components; Derive the control mechanism based on spatial coordinates and topological relationships, break through the bottleneck of low efficiency and difficulty in adapting to complex working conditions in traditional manual writing of control rules. The automatically derived control mechanism can better conform to the actual operation logic of the equipment, not only improve the efficiency of generating the control mechanism, but also enhance the applicability and effectiveness of the control mechanism in actual operation.

[0022] Monitor real - time external data and update the display attributes of configuration components in a timely manner according to the binding relationship, and dynamically drive the visual state changes of components based on the state mapping relationship, changing the defect of lag in traditional static monitoring information. It can reflect the changes in the equipment operation state in real - time and dynamically, enabling users to master the latest situation of the equipment in the first time and providing strong support for timely decision - making; By responding to user operations, verifying the execution results of control instructions, and synchronously updating the visual states, a complete closed - loop control system is constructed. Compared with the traditional one - way control process, this closed - loop system can timely detect and correct the deviations in the control process, ensure the accurate execution of control instructions, and effectively improve the stability and reliability of operation. Brief Description of the Drawings

[0023] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for description in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings. Among them: Figure 1 It is the method flow chart of a configuration method based on the BIM digital base provided by the embodiment of the present application; Figure 2 It is the matching sub - strategy diagram of configuration components of a configuration method based on the BIM digital base provided by the embodiment of the present application; Figure 3 It is the determination sub - strategy diagram of the binding relationship of a configuration method based on the BIM digital base provided by the embodiment of the present application. Detailed Embodiments

[0024] To make the objectives, technical solutions, and advantages of the embodiments of the present application more obvious and understandable, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the specification. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments.

[0025] Example 1 As shown in Figure 1 the figure, the embodiment of the present application provides a method flow chart of a configuration method based on a BIM digital base. The method includes: S1. Analyze the BIM model of the target device and extract component attributes including geometric information, semantic attributes, and topological relationships of components, identify and mark configurable components, and at the same time generate component features including component identifiers, spatial coordinates, and component attributes.

[0026] The sub-strategies for extracting component attributes include: Analyze the structured data of the BIM model of the target device, divide the geometric contours of the components in the BIM model through a point cloud segmentation algorithm, and extract the geometric information of the components; Perform semantic analysis on the attribute text of the BIM model through natural language processing technology, extract the semantic attributes of the components, and associate and match the semantic attributes of the components with the domain knowledge base; Construct a component topology structure based on graph theory algorithms, use components as nodes and the connection relationships between components as edges, analyze the component topology structure to determine the topological relationships of components, and form component attributes.

[0027] The geometric information of the components in the BIM model is the basis for realizing visual configuration. Accurate geometric contours and dimensional parameters can ensure the authenticity and accuracy of the subsequent layout of configuration components and visual effects; extract the BIM model of the target device, convert the unstructured data in the BIM model into structured data, and import the BIM model of the target device into the parsing engine. Process the BIM model through a point cloud segmentation algorithm. The point cloud segmentation algorithm first converts the three-dimensional data in the BIM model into point cloud data, analyzes the density and distribution characteristics of the point cloud, and divides the point cloud belonging to the same component together through a regional growth algorithm to complete the division of the geometric contour. Then, extract the dimensional parameters, shape features, and spatial positioning information of the components from the divided geometric contours. The dimensional parameters of the components include length, width, and height, the shape features include cylinders and cuboids, and the spatial positioning information includes coordinates and orientations.

[0028] Thus, the automatic and accurate extraction of the geometric information of the components in the BIM model is realized. Compared with manual measurement or simple model parsing, the efficiency and accuracy of information extraction are greatly improved, human errors are reduced, and the extracted geometric information provides key data for the generation of component attributes. Among them, the spatial positioning information is directly used to generate spatial coordinates, and the dimensions and shape features, as important components of component attributes, also provide basic data support for subsequent identification of configurable components and determination of layout positions according to component attributes.

[0029] The semantic attributes of components contain important information such as the functions and materials of components. These information are the key to understanding the characteristics and uses of components, and help to perform accurate function matching and control mechanism derivation during the configuration process. By analyzing the attribute text of the BIM model through natural language processing techniques, through operations such as word segmentation, part-of-speech tagging, and named entity recognition, keywords in the attribute text are extracted to obtain the semantic attributes of components. Then, the extracted semantic attributes are associated and matched with the domain knowledge base, which pre-stores the standard descriptions and association relationships of the semantic attributes of various components. Through keyword matching and semantic similarity calculation, the semantic description of components is improved. Preferably, information such as the category to which the valve component belongs and the applicable working conditions are supplemented for the valve component. Deeply excavate the semantic information of components from the attribute text of the BIM model to make the description of components more comprehensive and accurate, overcome the problem that it is difficult to understand the functional characteristics of components only relying on geometric information, and provide rich semantic support for subsequent intelligent configuration.

[0030] The topological relationship between components determines the way they are connected and influenced functionally. Clarifying the topological relationship helps to deduce the control mechanism of configuration components and achieve the coordinated operation and linkage control of target devices. Based on graph theory algorithms, components are regarded as nodes and the connection relationships between components are regarded as edges to construct a component topological structure. The connection relationships between components include pipeline connections and circuit connections. Then, by analyzing the component topological structure, the connection paths and hierarchical relationships between components are identified through path search algorithms, and then the topological relationship of components is determined. Preferably, through depth-first search, the series topological structure of the pipeline connecting the pump and the valve and the roles and influence relationships of each component in it are judged.

[0031] Constructing the topological relationship based on graph theory can clearly and systematically describe the complex connection relationships between components, provide intuitive and structured data support for the derivation of control mechanisms. The determined topological relationship, as the core content of component attributes, is used for the derivation of the control mechanism of configuration components after the generation of component characteristics, and also affects the linkage relationship of associated components when the visualization state changes.

[0032] The generation strategies of component characteristics include: According to the preset judgment rules and component attributes, identifiable and markable configurable components are identified from the BIM model, and at the same time, the spatial position relationships between components are analyzed through spatial topology analysis algorithms to form spatial coordinates; Encode the geometric information and semantic attributes of components through an encrypted hash algorithm to generate component identifiers; Integrate the component identifier, spatial coordinates, and component attributes to generate component characteristics.

[0033] Not all components require configuration operations. Identifying configurable components can reduce unnecessary calculations and processing, improve configuration efficiency, and spatial coordinates are the key to determining the layout positions of configuration elements, directly affecting the visualization effect and user operation experience. According to the preset judgment rules, that is, whether the component involves parameter adjustment and whether it is a key control component, etc., and then combined with the extracted component attributes, the components that can be configured are screened out from the BIM model and marked as configurable components. At the same time, the spatial position relationship between components is analyzed through the spatial topology analysis algorithm, and the spatial coordinates are formed by calculating the coordinate values and orientation parameters of the components in the three-dimensional space. Thus, the accurate screening of configurable components is realized, avoiding resource waste, improving operation efficiency, and the accurate spatial coordinates lay the foundation for the reasonable layout of subsequent configuration elements in the BIM model, making the visualization interface more in line with the spatial structure of the actual equipment, reducing the complexity of layout planning, and facilitating users to intuitively understand and operate.

[0034] Generating a unique identifier for each component can accurately distinguish and track different components in the complex BIM model and data interaction process, ensuring the accuracy and consistency of data processing and operations. Through the cryptographic hash algorithm, key features in the geometric information and semantic attributes of the component are selected for combination, that is, the key dimension parameters and the unique name identifier are combined, and the combined data is input into the cryptographic hash algorithm to generate a fixed-length and unique code as the component identifier. Due to the irreversibility and uniqueness of the cryptographic hash algorithm, the uniqueness and security of each component identifier can be guaranteed. Thus, a reliable unique identity identifier is provided for the component. During the data transmission, storage, and processing processes, the component can be quickly located and identified through the component identifier, avoiding confusion caused by repeated names or similar information, improving the accuracy and efficiency of data management. The component identifier, as the core element of the component characteristics, is used to associate the component with information such as configuration elements and external data in subsequent steps. Preferably, during the determination of the binding relationship, the corresponding relationship between the visualization state of the component and the external data is accurately established through the component identifier to ensure the accuracy of data binding.

[0035] Integrate the component identifier, spatial coordinates, and component attributes into component features to form a unified data structure, facilitating the subsequent steps to call and process component information and improving the data usage efficiency; establish a structured data format, and store and integrate the generated component identifier, spatial coordinates, and complete component attributes according to the preset data format and hierarchical relationship to form complete component features; thereby construct standardized and structured component features, making component information clearer and more orderly, facilitating users to quickly read and process, reducing the difficulty of data interaction and analysis. The integrated component features provide a comprehensive and standardized data foundation for the subsequent steps. Based on the component features, it is possible to efficiently match configuration components, deduce control mechanisms, and establish data binding relationships, etc., promoting the smooth progress of the entire configuration process.

[0036] S2. Based on the domain knowledge base and component features, match the configuration components associated with the configurable components, determine the initial layout position of the configuration components in the BIM model according to the spatial coordinates, deduce the control mechanism of the configuration components based on the topological relationship, and at the same time establish a data channel with external data, determine the display attributes of the configuration components and the binding relationship between the visualization state of the components and the external data, and determine the state mapping relationship.

[0037] As Figure 2 shown, the matching sub-strategies of the configuration components include: Match the semantic attributes of the configurable components with the semantic labels of the configuration components in the domain knowledge base, and calculate the semantic similarity score between the configurable components and the configuration components to screen out candidate configuration components; Construct the state space of reinforcement learning by combining the topological relationship and geometric information of the components, and use the adaptability of the candidate configuration components as the reward function of reinforcement learning; Through multiple rounds of iterative training of reinforcement learning, dynamically optimize the candidate configuration components and match the configuration components associated with the configurable components.

[0038] The functions of the components and the configuration components in the BIM model need to be accurately corresponding. By semantic label matching, the screening range can be quickly narrowed, avoiding blind matching, and improving the matching efficiency and accuracy of the configuration components; compare the extracted semantic attributes of the configurable components with the semantic labels of the configuration components in the domain knowledge base, calculate the semantic similarity score between the semantic labels of the components and each configuration component through the text similarity algorithm, and set the similarity threshold. Screen the configuration components with a semantic similarity score greater than the similarity threshold as candidate configuration components; realize automatic screening based on semantic attributes, reduce manual intervention, quickly locate the suitable candidate configuration components, provide a basis for subsequent optimization matching, reduce the matching error rate, and the screened candidate configuration components provide data objects for subsequent reinforcement learning optimization, limit the optimization range, and improve the overall matching efficiency.

[0039] Single semantic matching is difficult to comprehensively evaluate the adaptability of configuration components. It is necessary to construct a dynamic evaluation system by combining the topological relationship and geometric information of components to ensure that the matching results meet the actual operation requirements. Combine the topological relationship and geometric information of components to construct the state space of reinforcement learning, where the geometric information serves as the spatial constraint condition, that is, the limitation of the installation position size, and define the action space as selecting different candidate configuration components, and the reward function as the adaptability score of the candidate configuration components in the current state space, including whether it conforms to the topological connection relationship and whether it adapts to the geometric space. Construct a dynamic evaluation system through the fusion of multi-dimensional information, making the matching process more in line with the actual working conditions, improving the matching accuracy between configuration components and components, avoiding operation conflicts caused by static matching. The constructed state space, action space, and reward function provide a framework for the iterative training of reinforcement learning, enabling the dynamic optimization of the matching scheme of configuration components by simulating different operations.

[0040] There will be various potential problems with candidate configuration components in actual operation. Through multiple rounds of iterative training of reinforcement learning, the matching strategy can be dynamically adjusted to obtain the optimal matching result. The agent of reinforcement learning randomly selects one from the candidate configuration components as the initial action, and calculates the reward value according to the current state space, that is, getting positive points if the selected valve is correctly topologically connected to the pipeline, and negative points if the size does not match. By continuously performing actions, observing the environmental feedback, and updating the strategy, after multiple rounds of iteration, the selection strategy is gradually optimized to make the selected configuration component have the highest adaptability score under the reward function, and judge whether the iteration termination condition is met to determine the final configuration component associated with the configurable component. Thus, the adaptive optimization of the configuration component matching is realized, which can cope with the complex and changeable working condition requirements, improve the reliability and practicality of the matching scheme, ensure that the configuration component plays the best efficiency in actual operation, and the determined associated configuration component provides an object for the derivation of the subsequent control mechanism. Its matching result affects the pertinence and effectiveness of the control mechanism, and at the same time provides basic data for the determination of the binding relationship and the state mapping relationship, clarifying the main body of data interaction and visualization.

[0041] The derivation strategy of the control mechanism of configuration components includes: Based on the component topological structure, identify the connection paths and hierarchical relationships between components through the path search algorithm, and retrieve the basic control mechanism based on the domain knowledge base, and at the same time mark the nodes of the component topological structure as the trigger points of the basic control mechanism; Extract the spatial constraint conditions according to the initial layout position of the configuration component in the BIM model, and transform the spatial constraint conditions into additional rules of the basic control mechanism to form a constraint control mechanism; Take the constraint control mechanism as the initial strategy of reinforcement learning, use the operating efficiency and energy consumption of the target device as the reward function of reinforcement learning, and conduct multiple rounds of iterative training in combination with external data to derive the control mechanism of the configuration component.

[0042] The topological relationship of components implies functional connection logic. Based on this, the basic control mechanism is retrieved, and a preliminary control framework can be quickly established, providing a basis for the derivation of complex control mechanisms. Based on the component topological structure, the connection paths and hierarchical relationships between components are identified through a path search algorithm. Using the topological structure as an index, the corresponding basic control mechanism is retrieved from the domain knowledge base. Preferably, for the series topological structure of a pipeline connecting a pump and a valve, the retrieved basic control mechanism is that the valve opens after the pump starts, and the key nodes in the topological structure are marked as the trigger points of the basic control mechanism, where the key nodes include the pump and the valve. Automatically retrieving the basic control mechanism based on the topological relationship reduces the workload of manual design of control logic, improves the derivation efficiency of the control mechanism, and ensures that the basic control mechanism conforms to the operating principle of the actual equipment. The obtained basic control mechanism and trigger points provide an original framework for subsequent addition of spatial constraint conditions and formation of a constrained control mechanism, gradually improving the control mechanism.

[0043] There are spatial limitations in the actual installation of components. Relying solely on the basic control mechanism cannot meet the actual operation requirements. It is necessary to add spatial constraint rules in combination with the initial layout positions of configuration components to ensure the feasibility of the control mechanism. According to the initial layout positions of configuration components in the BIM model, spatial constraint conditions are extracted, that is, the spacing requirements between components and the limitations of the pipe bending radius. These spatial constraint conditions are transformed into additional conditions for the basic control mechanism. Preferably, on the basis of the basic control mechanism that the valve opens after the pump starts, the added constraint condition is that when the distance between the valve and the pump is less than the safety threshold, the valve delays opening, thus forming a constrained control mechanism. This enables the control mechanism to fully consider actual spatial layout factors, avoids the inability to execute the control mechanism due to spatial conflicts, improves the practicality and reliability of the control mechanism, and the formed constrained control mechanism serves as the initial strategy for reinforcement learning, providing a more practical framework for subsequent dynamic optimization in combination with external data.

[0044] During actual operation, the equipment conditions are constantly changing, and a fixed constrained control mechanism is difficult to adapt to. Through reinforcement learning combined with external data for iterative optimization, dynamic adaptive adjustment of the control mechanism can be achieved. The constrained control mechanism is used as the initial strategy for reinforcement learning, and the operating efficiency and energy consumption of the target equipment are used as the reward function of reinforcement learning. The operating efficiency of the target equipment includes the flow rate per unit time, and the energy consumption includes power consumption. Then, the monitored external data is obtained in real time. The agent of reinforcement learning selects control actions according to the current conditions, that is, adjusts the valve opening and changes the pump speed, calculates the action score according to the reward function, that is, a positive score is obtained for efficiency improvement, while a negative score is obtained for increased energy consumption. The strategy is continuously updated through the Q-learning algorithm. After multiple rounds of training, the control mechanism of configuration components adapted to different conditions is derived.

[0045] Thus, the dynamic optimization of the control mechanism is achieved, enabling the target device to maintain efficient and energy-saving operation under different working conditions, improving the intelligent level and operation stability. The determined control mechanism provides a logical basis for sending control instructions, ensuring that the control instructions meet the operation requirements of the target device. At the same time, it also affects the determination of the state mapping relationship, synchronizing the visual state changes of components with the control mechanism.

[0046] As Figure 3 shown, the sub-strategies for determining the binding relationship include: Conduct multi-dimensional analysis on external data and extract the data characteristics of external data through principal component analysis. At the same time, analyze the display attributes of configuration components and the parameter requirements of the visual state of components to form attribute parameters. Taking the data characteristics of external data as the horizontal dimension and the attribute parameters as the vertical dimension, construct a feature mapping matrix, calculate the matching degree between the data characteristics of external data and the attribute parameters through the cosine similarity algorithm, and generate an initial binding relationship. Taking data transmission efficiency and visualization accuracy as fitness functions, iteratively optimize the initial binding relationship through genetic algorithms, and determine the binding relationship between external data and attribute parameters through selection, crossover, and mutation operations.

[0047] The external data needs to accurately correspond to the display attributes of configuration components and the visual state of components. Extracting the features and parameters of both sides is the premise for achieving accurate binding. Conduct multi-dimensional analysis on the monitored external data, extract the data characteristics of external data through principal component analysis. The data characteristics include temperature change trends and pressure fluctuation peaks. At the same time, analyze the display attributes of configuration components and the parameter requirements of the visual state of components to form attribute parameters. The display attributes of configuration components include color, size, and transparency. The visual state of components includes geometric deformation and animation effects. The attribute parameters include the RGB value range of color and the size scaling ratio. Preferably, take the mean and variance of temperature data as data characteristics and the RGB interval corresponding to color changes as attribute parameters. Extract key information from a large amount of external data, clarify the required parameters for visualization, provide a clear data basis for constructing the binding relationship, and avoid data redundancy and invalid binding.

[0048] Establish a preliminary binding relationship between external data and attribute parameters, providing a starting point for subsequent optimization; taking the data characteristics of the extracted external data as the horizontal dimension and the attribute parameters as the vertical dimension, construct a feature mapping matrix, calculate the matching degree between the data characteristics of the external data and the attribute parameters through the cosine similarity algorithm, generate an initial binding relationship according to the size of the matching degree. If the data characteristics of the temperature data have a high matching degree with the color attribute parameters, then bind the temperature value to a specific color range, and the initial relationship formed is that the temperature increase causes the color to turn red; thus quickly establishing the association relationship between external data and attribute parameters, providing a basic rule for visual display, being able to initially display the visual effect corresponding to data changes. The initial binding relationship is used as the object of genetic algorithm optimization and is improved through iterative operations to achieve a better mapping effect of data visualization.

[0049] There are inaccuracies or inefficiencies in the initial binding relationship. By simulating the natural evolution process through the genetic algorithm, perform iterative optimization on it to improve data transmission efficiency and visualization accuracy; taking data transmission efficiency and visualization accuracy as fitness functions, perform genetic algorithm operations on the initial binding relationship. Among them, data transmission efficiency includes reducing the amount of data transmitted, and visualization accuracy includes accurately reflecting data changes. Through selection, crossover, and mutation operations, after multiple rounds of iteration, determine the final binding relationship between external data and attribute parameters. Among them, selection is to retain the binding relationship with high fitness, the crossover operation is to exchange part of the rules of different binding relationships, and the mutation operation is to randomly modify the binding relationship. Preferably, through the crossover operation, partially fuse the binding rule between temperature and color and the binding relationship between pressure and size, and then adjust the detailed parameters through mutation.

[0050] The optimized binding relationship can more efficiently and accurately implement the mapping between external data and attribute parameters, improve data processing and display capabilities, provide a clearer and more accurate visual interface for users, and the determined binding relationship is used to update the display attributes of configuration components in subsequent steps to ensure that the display effect reflects the changes in external data in real time and accurately, and at the same time provides a data association basis for determining the state mapping relationship.

[0051] External data and component attributes are combined through feature splicing to form a fused feature vector. Based on the fused feature vector, a dynamic state mapping relationship is constructed using a fuzzy logic algorithm. Fuzzy language variables and membership functions are predefined, and an initial rule base is set according to historical data. At the same time, the stability of device operation and user operation feedback are used as reward signals for reinforcement learning to continuously optimize the state mapping relationship. Through continuous optimization, the agent selects mapping actions according to the changes in current external data and the operating conditions of components, and obtains reward feedback based on the execution results, where the execution results include whether the state of the target device has improved and whether the user confirms. Then, the value function of the state and action is updated through the Q-learning algorithm to dynamically adjust the state mapping relationship to make it more in line with actual requirements.

[0052] S3. Monitor the real-time external data of the target device, update the display attributes of the configuration components according to the binding relationship, and dynamically drive the visualization state change of the components based on the state mapping relationship. At the same time, respond to the user's operation on the configuration components and send control instructions to the target device, and synchronously update the visualization states of the configuration components and the components based on the execution results of the target device.

[0053] The update sub-strategy of the display attributes of the configuration components includes: Monitor the real-time external data of the target device, perform multi-scale analysis on the real-time external data to extract spatio-temporal correlation features, calculate the matching degree between the spatio-temporal correlation features and the feature templates in the binding relationship through cosine similarity, and filter out update actions; Dynamically adjust the display attributes of the configuration components according to the update actions, and determine the rendering parameters of the display attributes by judging the type of the display attributes. The types of display attributes include numerical attributes, trend attributes, and abnormal attributes; Construct the state space of reinforcement learning according to the spatio-temporal correlation features and the types of display attributes, use the combination of rendering parameters of the display attributes as the action space of reinforcement learning, and comprehensively determine the reward function of reinforcement learning according to information transfer efficiency, visual comfort, and interaction convenience, and train the reinforcement learning to optimize the visual mapping effect of the display attributes.

[0054] Real-time external data has temporal and spatial correlations. By extracting spatio-temporal features and matching update actions, dynamic and precise updates of the display attributes of configuration components can be achieved. Through wavelet transform, multi-scale analysis of real-time external data is carried out to extract features in different time windows and spatial dimensions, including pressure fluctuation frequency and temperature gradient changes. The spatio-temporal correlation features extracted are used to calculate the cosine similarity with the feature templates preset in the binding relationship to obtain the matching degree, and a matching degree threshold is configured. When the matching degree is greater than the matching degree threshold, the corresponding update actions are screened out. Through spatio-temporal feature matching, rapid response to external data changes is realized, making the update of the display attributes of configuration components more timely and targeted, avoiding ineffective updates. The screened update actions provide a basis for subsequent adjustment of display attributes and directly determine the adjustment direction and amplitude of display attributes.

[0055] Different types of display attributes require different rendering methods. By clarifying the attribute type and determining the rendering parameters, the visualization effect can be optimized. According to the update actions, the display attributes of configuration components are dynamically adjusted, where the display attributes include color, size, and transparency. Then, the type of the display attribute is judged. If it is a numerical attribute, rendering parameters such as digital font, color, and precision are determined. If it is a trend attribute, rendering parameters such as line chart style and time window are determined. If it is an abnormal attribute, rendering parameters such as warning icons and blinking frequency are determined. Among them, the numerical attributes include temperature values, the trend attributes include pressure change trends, and the abnormal attributes include equipment failures. Adopting a differentiated rendering strategy for different types of display attributes improves the information transmission efficiency, enabling users to more intuitively obtain key data and abnormal states. The determined rendering parameters provide an initial configuration for reinforcement learning optimization and affect the training direction of subsequent visual mapping effects.

[0056] The initial rendering parameters are not optimal. Through dynamic optimization by reinforcement learning, the balance among information transmission efficiency, visual comfort, and interaction convenience can be achieved. A reinforcement learning model is constructed, with the spatio-temporal correlation features and the type of display attributes as the state space, and the combination of rendering parameters of display attributes as the action space. A reward function is designed based on information transmission efficiency, visual comfort, and interaction convenience. Among them, information transmission efficiency includes the speed at which users recognize data, visual comfort includes color contrast and layout rationality, and interaction convenience includes operation response speed. Through iterative training using the Q-learning algorithm, the combination of rendering parameters of display attributes is continuously adjusted to maximize the value of the reward function. Thus, the adaptive optimization of visual mapping of display attributes is realized, enhancing the human-computer interaction experience, reducing the user's cognitive burden, improving the monitoring efficiency. The optimized visual mapping effect provides a unified display standard for the visualization state changes of components, ensuring the coordination of their visual performances.

[0057] The driving sub-strategies for the visualization state changes of components include: According to the spatiotemporal correlation characteristics, the mapping action associated with the component is found from the state mapping relationship, and the external data is converted into the visualization parameters of the component. The visualization parameters include geometric form, appearance state and rendering effect; Determine the impact range of component visualization state changes based on the connection relationship between components, and modify the component's geometry and appearance state according to the mapping action; The rendering effect of the component is dynamically adjusted according to the importance of the visualization status, and the visualization parameters of the component are fed back in response to the user's operation on the configuration element.

[0058] To convert external data into visualization parameters, it is necessary to establish the correspondence between data and visual representation through state mapping relationships to achieve accurate driving of state changes; according to the spatiotemporal correlation characteristics, the mapping actions associated with the component are found from the state mapping relationship, and the external data is converted into the visualization parameters of the component. The visualization parameters include geometric shape, appearance shape and rendering effect; thereby achieving standardized conversion from data to visual, ensuring that the state changes of the component are consistent with the actual working conditions, and improving the authenticity and credibility of the visualization.

[0059] There is a topological association between components. The state change of one component will affect other components. The scope of influence is determined and updated in conjunction, so that the operating status can be fully presented. According to the component topological relationship, the breadth-first search algorithm is used, starting from the current component, to traverse all directly or indirectly connected components to determine the scope of influence of the state change. The geometric shape and appearance state of the components within the range are modified according to the mapping action. The geometric shape includes the pipe diameter, and the appearance state includes the color and transparency. Preferably, when the valve is closed, in addition to modifying its own state, the state of the downstream pipeline is also marked as fluid stagnation, and the color is adjusted to gray. This achieves the coordinated update of the status of the associated components, reflects the overall operating status of the components, avoids information deviation caused by isolated display, and improves the integrity of visualization.

[0060] Components in different states have different importances. Dynamically adjusting the rendering effect can highlight key information and respond to user operation feedback to enhance the interactive experience. According to the importance of the component visualization state, the rendering parameters are dynamically adjusted, where the importance of the fault state is greater than that of the normal state. The rendering parameters include brightness, transparency and hierarchy. Highlight, magnify or animate components in important states, reduce transparency or simplify the display of components in secondary states, and monitor user operations on configuration elements, such as clicks and drags, to provide real-time feedback on changes in component visualization parameters, including displaying detailed data and adjusting the viewing angle. By sorting the importance of visualization states, users are guided to focus on key information, and interactive feedback enhances user participation and improves system usability. The adjusted rendering effect directly affects the user's perception of the component state, and thus affects the user's operation decision and the generation of control instructions.

[0061] The update strategy for the visualization states of configuration components and building blocks includes: Monitor and respond to user operations on configuration components, convert the user operations on configuration components into control instructions based on the domain knowledge base, and transmit the control instructions to the target device; Receive the execution results of the target device, verify the execution effectiveness of the expected results and execution results of the control instructions, and modify the display attributes of the configuration components according to the execution effectiveness and binding relationship; Update the geometric form and appearance state of the building blocks based on the execution effectiveness and mapping actions, and trigger the state update of associated building blocks according to the topological relationship of the building blocks.

[0062] To convert user operations into executable control instructions, semantic parsing and logical conversion need to be combined with the domain knowledge base; monitor the operation events of users on configuration components, convert the user operations on configuration components into control instructions based on the domain knowledge base, supplement instruction parameters according to the topological relationship of building blocks and control mechanisms, including pump speed and start time, and generate complete control instructions; thus realizing the seamless conversion from user intent to device control, reducing the operation threshold, improving control efficiency, avoiding control errors caused by semantic ambiguity, directly sending the generated control instructions to the target device, determining the operating state of the device, and further affecting the verification of subsequent execution results and state updates.

[0063] Verify the execution effectiveness of the control instructions, ensure that the execution feedback is consistent with the user's expectations, and update the display attributes of the configuration components according to the results to maintain information timeliness; receive the execution results returned by the target device, compare the expected results with the actual execution results, calculate the execution deviation rate, that is, the execution deviation rate between the set speed in the control instruction and the actual speed. When the execution deviation rate is within the set threshold range, it is determined that the execution is effective, and then update the display attributes of the configuration components according to the binding relationship, that is, change the button color to green; when the execution deviation rate exceeds the set threshold range, it is determined that the execution fails, and at the same time, display a warning message and mark the abnormality; update the display attributes of the configuration components in real time, enabling users to timely understand the operation effect and enhancing the operation confidence. The results of the execution effectiveness and the updated display attributes provide a basis for the visualization state update of the building blocks, determining the direction and method of the state update.

[0064] Update the component status based on the execution result, trigger the chain reaction of associated components, and fully present the change process of the components; update the geometric shape and appearance status of the components according to the execution effectiveness and mapping actions, and identify the affected associated components through the graph traversal algorithm based on the topological relationship of the components, and recursively trigger their status updates. Preferably, after the pump is started, the fluid status and pressure display of the downstream pipeline are automatically updated; thus, the synchronous update of the status of all components is realized, reflecting the collaborative working relationship between devices, providing a complete view of the component operation, assisting the user to comprehensively master the working conditions, and the updated component status serves as the new initial status, affecting the next round of data monitoring and status change driving process, forming a closed-loop feedback mechanism.

[0065] Embodiment 2 The embodiment of the present application provides a configuration system based on a BIM digital base, and the system includes a feature extraction module, a configuration configuration module, and a mapping feedback module.

[0066] The feature extraction module is used to parse the BIM model of the target device and extract the component attributes including the geometric information, semantic attributes, and topological relationship of the components, identify and mark the configurable components, and simultaneously generate the component features including the component identifier, spatial coordinates, and component attributes.

[0067] The configuration configuration module is used to match the configuration elements associated with the configurable components based on the domain knowledge base and component features, determine the initial layout position of the configuration elements in the BIM model according to the spatial coordinates, deduce the control mechanism of the configuration elements based on the topological relationship, establish a data channel with external data at the same time, configure the display attributes of the configuration elements and the binding relationship between the visualization status of the components and external data, and determine the status mapping relationship.

[0068] The mapping feedback module is used to monitor the real-time external data of the target device, update the display attributes of the configuration elements according to the binding relationship, dynamically drive the change of the visualization status of the components based on the status mapping relationship, respond to the user's operation on the configuration elements and send control instructions to the target device, and synchronously update the visualization status of the configuration elements and the components based on the execution result of the target device.

[0069] Since the principle of solving problems of the system in the embodiment of the present application is similar to the above method in the embodiment of the present application, the implementation of the method can refer to the implementation of the method, and the repeated parts will not be elaborated.

Claims

1. A configuration method based on a BIM digital base, characterized in that Including: Analyze the BIM model of the target device and extract component attributes including the geometric information, semantic attributes, and topological relationships of components, identify and mark configurable components, and simultaneously generate component features including component identifiers, spatial coordinates, and component attributes; Based on the domain knowledge base and component features, match the configuration components associated with the configurable components, determine the initial layout position of the configuration components in the BIM model according to the spatial coordinates, and deduce the control mechanism of the configuration components based on the topological relationships. At the same time, establish a data channel with external data, determine the binding relationship between the display attributes of the configuration components and the visualization state of the components and the external data, and determine the state mapping relationship; Monitor the real-time external data of the target device, update the display attributes of the configuration components according to the binding relationship, and dynamically drive the change of the visualization state of the components based on the state mapping relationship. At the same time, respond to the operation of the user on the configuration components and send control instructions to the target device, and synchronously update the visualization states of the configuration components and the components based on the execution results of the target device.

2. The configuration method based on the BIM digital base as described in claim 1, wherein, The generation strategy of the component features includes: According to the preset judgment rules and component attributes, identify and mark configurable components from the BIM model, and at the same time analyze the spatial position relationship between components through a spatial topology analysis algorithm to form spatial coordinates; Encode the geometric information and semantic attributes of the components through an encrypted hash algorithm to generate component identifiers; Integrate the component identifiers, spatial coordinates, and component attributes to generate component features.

3. A configuration method based on a BIM digital base as described in claim 2, characterized in that, The extraction sub-strategy of the component attributes includes: Analyze the structured data of the BIM model of the target device, divide the geometric contours of the components in the BIM model through a point cloud segmentation algorithm, and extract the geometric information of the components; Perform semantic analysis on the attribute text of the BIM model through natural language processing technology, extract the semantic attributes of the components, and associate and match the semantic attributes of the components with the domain knowledge base; Construct a component topology structure based on graph theory algorithms, use components as nodes and the connection relationships between components as edges, analyze the component topology structure to determine the topological relationships of the components, and form component attributes.

4. The configuration method based on the BIM digital base as claimed in claim 3, wherein, The derivation strategy of the control mechanism of the configuration components includes: Based on the component topology structure, identify the connection paths and hierarchical relationships between components through a path search algorithm, and retrieve the basic control mechanism based on the domain knowledge base. At the same time, mark the nodes of the component topology structure as the trigger points of the basic control mechanism; Extract spatial constraint conditions according to the initial layout position of the configuration components in the BIM model, transform the spatial constraint conditions into additional rules of the basic control mechanism, and form a constraint control mechanism; Use the constraint control mechanism as the initial strategy of reinforcement learning, use the operating efficiency and energy consumption of the target device as the reward function of reinforcement learning, and perform multiple rounds of iterative training in combination with external data to deduce the control mechanism of the configuration components.

5. The configuration method based on the BIM digital base according to claim 4, characterized in that The matching sub-strategy of the configuration components includes: Match the semantic attributes of the configurable components with the semantic labels of the configuration components in the domain knowledge base, and calculate the semantic similarity score between the configurable components and the configuration components to screen out candidate configuration components; Construct the state space of reinforcement learning by combining the topological relationship and geometric information of the component, and use the adaptability of the candidate configuration components as the reward function of reinforcement learning; Through multiple rounds of iterative training of reinforcement learning, dynamically optimize the candidate configuration components and match the configuration components associated with the configurable components.

6. The configuration method based on the BIM digital base according to claim 5, characterized in that The determination sub-strategy of the binding relationship includes: Conduct multi-dimensional analysis on the external data and extract the data features of the external data through principal component analysis. At the same time, analyze the display attributes of the configuration components and the parameter requirements of the visualization state of the component to form attribute parameters; Taking the data features of the external data as the horizontal dimension and the attribute parameters as the vertical dimension, construct a feature mapping matrix, calculate the matching degree between the data features of the external data and the attribute parameters through the cosine similarity algorithm, and generate the initial binding relationship; Taking the data transmission efficiency and visualization accuracy as the fitness function, iteratively optimize the initial binding relationship through the genetic algorithm, and determine the binding relationship between the external data and the attribute parameters through selection, crossover and mutation operations.

7. The configuration method based on the BIM digital base according to claim 6, characterized in that, The update strategy for the visualization state of the configuration components and components includes: Monitor and respond to the user's operations on the configuration components, convert the user's operations on the configuration components into control instructions based on the domain knowledge base, and transmit the control instructions to the target device; Receive the execution result of the target device, verify the execution effectiveness of the expected result and the execution result of the control instruction, and modify the display attributes of the configuration components according to the execution effectiveness and the binding relationship; Update the geometric shape and appearance state of the component based on the execution effectiveness and the mapping action, and trigger the state update of the associated component according to the topological relationship of the component.

8. A configuration method based on a BIM digital base as claimed in claim 7, wherein, The update sub-strategy for the display attributes of the configuration components includes: Monitor the real-time external data of the target device, conduct multi-scale analysis on the real-time external data to extract spatio-temporal correlation features, and calculate the matching degree between the spatio-temporal correlation features and the feature template in the binding relationship through cosine similarity to screen out the update actions; Dynamically adjust the display attributes of the configuration components according to the update actions, and judge the type of the display attributes to determine the rendering parameters of the display attributes. The types of the display attributes include numerical attributes, trend attributes and abnormal attributes; Construct the state space of reinforcement learning according to the spatio-temporal correlation features and the type of the display attributes, use the combination of the rendering parameters of the display attributes as the action space of reinforcement learning, and comprehensively determine the reward function of reinforcement learning according to the information transmission efficiency, visual comfort and interaction convenience, and train the reinforcement learning to optimize the visual mapping effect of the display attributes.

9. A configuration method based on a BIM digital base as claimed in claim 8, wherein, The driving sub-strategy for the change of the visualization state of the component includes: Find the mapping actions associated with the component from the state mapping relationship according to the spatio-temporal correlation features, and convert the external data into the visualization parameters of the component. The visualization parameters include geometric shape, appearance state and rendering effect; Determine the influence range of the change of the visualization state of the component based on the connection relationship between the components, and modify the geometric shape and appearance state of the component according to the mapping action; Dynamically adjust the rendering effect of the component according to the importance of the visualization state, and respond to the user's operations on the configuration components to feedback the visualization parameters of the component.

10. A configuration system based on a BIM digital base for implementing a configuration method based on a BIM digital base according to any one of claims 1-9, characterized in that, Include: Feature extraction module, configuration configuration module and mapping feedback module; The feature extraction module is used to parse the BIM model of the target device and extract component attributes including the geometric information, semantic attributes, and topological relationships of components, identify and mark configurable components, and generate component features including component identifiers, spatial coordinates, and component attributes; The configuration module is used to match configuration components associated with the configurable components based on the domain knowledge base and component features, determine the initial layout positions of the configuration components in the BIM model according to the spatial coordinates, deduce the control mechanism of the configuration components based on the topological relationships, establish a data channel with external data, configure the binding relationship between the display attributes of the configuration components and the visualization state of the components and the external data, and determine the state mapping relationship; The mapping feedback module is used to monitor the real-time external data of the target device, update the display attributes of the configuration components according to the binding relationship, dynamically drive the change of the visualization state of the components based on the state mapping relationship, respond to the operations of the user on the configuration components and send control instructions to the target device, and synchronously update the visualization states of the configuration components and the components based on the execution results of the target device.

Images