Power system building three-dimensional model data processing method and system applying BIM (Building Information Modeling)

Abstract

The invention relates to the technical field of data processing, and particularly provides an electric power system building three-dimensional model data processing method and system applying a BIM (Building Information Modeling). The embodiment of the invention aims to realize comprehensive digital management of a power system building, improve the design precision and the construction efficiency, reduce the operation and maintenance cost and provide powerful support for intelligent upgrading of the power industry.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a method and system for processing three-dimensional model data of a power system building using BIM. Background Art

[0002] In the field of power system buildings, traditional planning, design, operation and maintenance methods often rely on two-dimensional drawings and manual calculations, which is not only inefficient, but also difficult to fully and accurately capture the complex relationships and potential risks in the system. With the continuous development of technology, BIM (Building Information Modeling) technology has provided new possibilities for the digital management of power systems. However, how to efficiently process and analyze BIM model data to achieve comprehensive monitoring and prediction of power system buildings is still a major technical challenge facing the industry. Especially in the operation and maintenance stage, the performance and status of each component element in the power system building will change accordingly with the change of time and usage status. It is difficult for traditional methods to capture and predict such dynamic changes in a timely and accurate manner, thereby increasing operation and maintenance costs and safety risks. Summary of the invention

[0003] In order to improve the above problems, the present application provides a method and system for processing three-dimensional model data of power system buildings using BIM.

[0004] The present application provides a method for processing data of a three-dimensional model of a power system building using BIM, which is applied to a data processing system. The method includes: Extracting model attributes from the first BIM electric construction model data to obtain a first BIM state attribute vector, and extracting model attributes from the second BIM electric construction model data to obtain a second BIM state attribute vector; Performing operation and maintenance trend identification on the first BIM state attribute vector to obtain trend simulation weights of each first multi-mode three-dimensional component element in the first BIM state attribute vector, and performing operation and maintenance trend identification on the second BIM state attribute vector to obtain trend simulation weights of each second multi-mode three-dimensional component element in the second BIM state attribute vector, wherein the trend simulation weights are used to characterize confidence coefficients for operation and maintenance fault risk prediction of BIM electric construction model data; According to the trend simulation weights of the respective first multimodal three-dimensional component elements, constructing element sampling on the first BIM state attribute vector to obtain x first multimodal three-dimensional component elements; and according to the trend simulation weights of the respective second multimodal three-dimensional component elements, constructing element sampling on the second BIM state attribute vector to obtain x second multimodal three-dimensional component elements; Performing operation and maintenance trend identification on the x first multi-mode three-dimensional component elements and the x second multi-mode three-dimensional component elements, obtaining first model annotation information corresponding to the x first multi-mode three-dimensional component elements and second model annotation information corresponding to the x second multi-mode three-dimensional component elements, wherein the first model annotation information is used to indicate mining u first multi-mode three-dimensional component elements in the x first multi-mode three-dimensional component elements that have a power line layout intersection state with the x second multi-mode three-dimensional component elements, and the second model annotation information is used to enhance v second multi-mode three-dimensional component elements in the x second multi-mode three-dimensional component elements that have a power line layout intersection state with the x first multi-mode three-dimensional component elements; An operation and maintenance failure risk prediction is performed in combination with the u first multi-mode three-dimensional component elements and the v second multi-mode three-dimensional component elements to obtain a first operation and maintenance failure risk view between the first BIM electric power construction model data and the second BIM electric power construction model data.

[0005] Optionally, constructing element sampling on the first BIM state attribute vector according to the trend simulation weights of the respective first multimodal three-dimensional component elements to obtain x first multimodal three-dimensional component elements, and constructing element sampling on the second BIM state attribute vector according to the trend simulation weights of the respective second multimodal three-dimensional component elements to obtain x second multimodal three-dimensional component elements, comprises: Determine x according to preset parameters; According to the trend simulation weights of each first multimodal three-dimensional component element and in accordance with the descending arrangement rule of the trend simulation weights, the first x first multimodal three-dimensional component elements in the first BIM state attribute vector are determined as the x first multimodal three-dimensional component elements; and according to the trend simulation weights of each second multimodal three-dimensional component element and in accordance with the descending arrangement rule of the trend simulation weights, the first x second multimodal three-dimensional component elements in the second BIM state attribute vector are determined as the x second multimodal three-dimensional component elements.

[0006] Optionally, determining x according to preset parameters includes: determining x according to the preset parameters and a characteristic dimension of the first BIM state attribute vector, or determining x according to the preset parameters and a characteristic dimension of the second BIM state attribute vector.

[0007] Optionally, the performing operation and maintenance trend identification on the x first multimodal three-dimensional component elements and the x second multimodal three-dimensional component elements to obtain first model annotation information corresponding to the x first multimodal three-dimensional component elements and second model annotation information corresponding to the x second multimodal three-dimensional component elements includes: Determine first basic model annotation information corresponding to the x first multi-mode three-dimensional component elements, and determine second basic model annotation information corresponding to the x second multi-mode three-dimensional component elements; According to the x first multimodal three-dimensional component elements, the x second multimodal three-dimensional component elements, the first basic model annotation information, and the second basic model annotation information, the x first multimodal three-dimensional component elements are adjusted to obtain x first adjusted three-dimensional component elements; and according to the x first multimodal three-dimensional component elements, the x second multimodal three-dimensional component elements, the first basic model annotation information, and the second basic model annotation information, the x second multimodal three-dimensional component elements are adjusted to obtain x second adjusted three-dimensional component elements; According to the x first adjusted three-dimensional component elements, the first basic model annotation information is adjusted to obtain first adjusted model annotation information, and according to the x second adjusted three-dimensional component elements, the second basic model annotation information is adjusted to obtain second adjusted model annotation information; The first model annotation information is determined according to the first adjustment model annotation information, and the second model annotation information is determined according to the second adjustment model annotation information.

[0008] Optionally, before extracting the model attributes of the first BIM electrical construction model data to obtain the first BIM state attribute vector, and extracting the model attributes of the second BIM electrical construction model data to obtain the second BIM state attribute vector, the method further includes: In the case where the feature dimension of the first BIM electrical construction model data and the feature dimension of the second BIM electrical construction model data do not match, converting the first BIM electrical construction model data and the second BIM electrical construction model data to the same feature dimension; The determining of the first basic model annotation information corresponding to the x first multi-mode three-dimensional component elements, and the determining of the second basic model annotation information corresponding to the x second multi-mode three-dimensional component elements, comprises: Determine the first basic model annotation information according to the feature members augmented during the dimensional conversion of the first BIM electrical construction model data, wherein the model query vector corresponding to the quantized value of the feature members augmented in the x first multi-mode three-dimensional component elements in the first basic model annotation information is a annotation keyword; Based on the feature members augmented during the dimensional conversion of the second BIM electrical construction model data, the second basic model annotation information is determined, in which the model query vector corresponding to the quantized value of the feature members augmented in the x second multi-modal three-dimensional component elements is the annotation keyword.

[0009] Optionally, the adjusting the x first multimodal three-dimensional component elements according to the x first multimodal three-dimensional component elements, the x second multimodal three-dimensional component elements, the first basic model annotation information, and the second basic model annotation information to obtain x first adjusted three-dimensional component elements, and the adjusting the x second multimodal three-dimensional component elements according to the x first multimodal three-dimensional component elements, the x second multimodal three-dimensional component elements, the first basic model annotation information, and the second basic model annotation information to obtain x second adjusted three-dimensional component elements include: Using the first basic model annotation information and the x first multimodal three-dimensional component elements, x first spatial structure relationship vectors are obtained; and using the second basic model annotation information and the x second multimodal three-dimensional component elements, x second spatial structure relationship vectors are obtained; The x first spatial structure relationship vectors are subjected to intra-class feature focusing processing by using the first intra-class feature focusing branch, and the x second spatial structure relationship vectors and the x processing results generated by the first intra-class feature focusing branch are subjected to the first inter-class feature focusing branch to obtain the x first adjusted three-dimensional component elements, and the x second spatial structure relationship vectors are subjected to intra-class feature focusing processing by using the second intra-class feature focusing branch, and the x first spatial structure relationship vectors and the x processing results generated by the second intra-class feature focusing branch are subjected to the second inter-class feature focusing branch to obtain the x second adjusted three-dimensional component elements.

[0010] Optionally, the adjusting the first basic model annotation information according to the x first adjusted three-dimensional component elements to obtain first adjusted model annotation information, and adjusting the second basic model annotation information according to the x second adjusted three-dimensional component elements to obtain second adjusted model annotation information include: Performing operation and maintenance trend identification on the x first adjusted three-dimensional component elements to obtain two discrimination weights corresponding to each of the x first adjusted three-dimensional component elements, and performing operation and maintenance trend identification on the x second adjusted three-dimensional component elements to obtain two discrimination weights corresponding to each of the x second adjusted three-dimensional component elements, the two discrimination weights being used to characterize the discrimination weight of the element vector belonging to the cross state of the power line layout and the discrimination weight of the element vector not belonging to the cross state of the power line layout; Determining x first key-value relationship features corresponding to the x first adjusted three-dimensional component elements based on the two discrimination weights corresponding to the respective first adjusted three-dimensional component elements, and determining x second key-value relationship features corresponding to the x second adjusted three-dimensional component elements based on the two discrimination weights corresponding to the respective second adjusted three-dimensional component elements, wherein the first key-value relationship feature is used to characterize authentication information of an attention index to which a maximum discrimination weight belongs among the two discrimination weights corresponding to the first adjusted three-dimensional component elements, and the second key-value relationship feature is used to characterize authentication information of an attention index to which a maximum discrimination weight belongs among the two discrimination weights corresponding to the second adjusted three-dimensional component elements; Performing discriminant analysis on the x first key-value relationship features to obtain x first discriminant weights, and performing discriminant analysis on the x second key-value relationship features to obtain x second discriminant weights; The first adjustment model annotation information is obtained by using the first basic model annotation information and the x first discrimination weights, and the second adjustment model annotation information is obtained by using the second basic model annotation information and the x second discrimination weights.

[0011] Optionally, the adjusting the first basic model annotation information according to the x first adjusted three-dimensional component elements to obtain first adjusted model annotation information, and adjusting the second basic model annotation information according to the x second adjusted three-dimensional component elements to obtain second adjusted model annotation information include: According to the x first adjusted three-dimensional component elements, the x second adjusted three-dimensional component elements, the first adjustment model annotation information, and the second adjustment model annotation information, the x first adjusted three-dimensional component elements are adjusted to obtain x third adjusted three-dimensional component elements; and according to the x first adjusted three-dimensional component elements, the x second adjusted three-dimensional component elements, the first adjustment model annotation information, and the second adjustment model annotation information, the x second adjusted three-dimensional component elements are adjusted to obtain x fourth adjusted three-dimensional component elements; adjusting the first adjustment model annotation information according to the x third adjusted three-dimensional component elements to obtain third adjustment model annotation information, and adjusting the second adjustment model annotation information according to the x fourth adjusted three-dimensional component elements to obtain fourth adjustment model annotation information; The first model annotation information is determined based on the third adjustment model annotation information, and the second model annotation information is determined based on the fourth adjustment model annotation information, the first model annotation information is the adjustment model annotation information obtained when the number of cyclic adjustments of the first basic model annotation information is not less than the set number of times, and the second model annotation information is the adjustment model annotation information obtained when the number of cyclic adjustments of the second basic model annotation information is not less than the set number of times.

[0012] Optionally, extracting model attributes from the first BIM electrical construction model data to obtain a first BIM state attribute vector, and extracting model attributes from the second BIM electrical construction model data to obtain a second BIM state attribute vector, include: Performing model attribute extraction on the first BIM electrical construction model data to obtain the first BIM state attribute vector and the third BIM state attribute vector, and performing model attribute extraction on the second BIM electrical construction model data to obtain the second BIM state attribute vector and the fourth BIM state attribute vector, wherein the feature dimension of the first BIM state attribute vector is different from the feature dimension of the third BIM state attribute vector, and the feature dimension of the second BIM state attribute vector is different from the feature dimension of the fourth BIM state attribute vector; Wherein, after obtaining the first operation and maintenance failure risk view between the first BIM electric power construction model data and the second BIM electric power construction model data, the method further includes: According to the first operation and maintenance failure risk viewpoint, the characteristic dimension of the first BIM state attribute vector, the characteristic dimension of the third BIM state attribute vector and the third BIM state attribute vector, construct element sampling for the third BIM state attribute vector to obtain u BIM electrical construction model data subsets corresponding to the u first multi-mode three-dimensional component elements, and according to the first operation and maintenance failure risk viewpoint, the characteristic dimension of the second BIM state attribute vector, the characteristic dimension of the fourth BIM state attribute vector and the fourth BIM state attribute vector, construct element sampling for the fourth BIM state attribute vector to obtain v BIM electrical construction model data subsets corresponding to the v second multi-mode three-dimensional component elements; adjusting the u BIM electric power construction model data subsets according to the u BIM electric power construction model data subsets and the v BIM electric power construction model data subsets to obtain u model data adjustment subsets, and adjusting the v BIM electric power construction model data subsets according to the u BIM electric power construction model data subsets and the v BIM electric power construction model data subsets to obtain v model data adjustment subsets; Detecting the key data clusters of the u model data adjustment subsets to obtain u key data clusters, and performing fine-grained optimization processing on the v model data adjustment subsets according to the model fine-grained comparison result between the third BIM electric power construction model data and the fourth BIM electric power construction model data to obtain v target model data subsets; Performing distribution feature mapping processing on the v target model data subsets to obtain v distribution mapping features corresponding to the v target model data subsets; Combine the u key data clusters and the v distribution mapping features to perform state jump analysis, and a second operation and maintenance fault risk perspective between the first BIM electric construction model data and the second BIM electric construction model data.

[0013] Optionally, performing distribution feature mapping processing on the v target model data subsets to obtain v distribution mapping features corresponding to the v target model data subsets includes: Acquire the relative relationship features of each model unit in the first target model data subset among the v target model data subsets; The relative relationship characteristics and influence coefficients of each model unit in the first target model data subset are enhanced to obtain the distribution mapping characteristics of the first target model data subset.

[0014] In some independent embodiments, the method further comprises: Extracting model attributes from the first BIM electric construction model data sample to obtain a first BIM state attribute vector sample, and extracting model attributes from the second BIM electric construction model data sample to obtain a second BIM state attribute vector sample; Using a trend simulation processing algorithm, the operation and maintenance trend of the first BIM state attribute vector sample is identified to obtain the trend simulation weight of each first multi-mode three-dimensional component element in the first BIM state attribute vector sample; and using the trend simulation processing algorithm, the operation and maintenance trend of the second BIM state attribute vector sample is identified to obtain the trend simulation weight of each second multi-mode three-dimensional component element in the second BIM state attribute vector sample, wherein the trend simulation weight is used to characterize the confidence coefficient of the operation and maintenance fault risk prediction of the BIM electric construction model data sample; According to the trend simulation weights of the respective first multimodal three-dimensional component elements, constructing element sampling on the first BIM state attribute vector sample to obtain x first multimodal three-dimensional component elements; and according to the trend simulation weights of the respective second multimodal three-dimensional component elements, constructing element sampling on the second BIM state attribute vector sample to obtain x second multimodal three-dimensional component elements; Using a BIM annotation classification network, operation and maintenance trends of the x first multi-mode three-dimensional component elements and the x second multi-mode three-dimensional component elements are identified to obtain first model annotation information corresponding to the x first multi-mode three-dimensional component elements and second model annotation information corresponding to the x second multi-mode three-dimensional component elements, wherein the first model annotation information is used to indicate mining u first multi-mode three-dimensional component elements in the x first multi-mode three-dimensional component elements that have a power line layout intersection state with the x second multi-mode three-dimensional component elements, and the second model annotation information is used to enhance v second multi-mode three-dimensional component elements in the x second multi-mode three-dimensional component elements that have a power line layout intersection state with the x first multi-mode three-dimensional component elements; The trend simulation processing algorithm is debugged according to the trend simulation weights of the first multimodal three-dimensional component elements and the trend simulation weights of the second multimodal three-dimensional component elements, and the BIM annotation classification network is debugged according to the first model annotation information and the second model annotation information.

[0015] In some independent embodiments, the step of debugging the trend simulation processing algorithm according to the trend simulation weights of the first multimodal three-dimensional component elements and the trend simulation weights of the second multimodal three-dimensional component elements includes: A first algorithm debugging error is constructed based on the difference between the model annotation information used to characterize whether the first multi-mode three-dimensional component element in the first BIM state attribute vector sample has abnormal thermal characteristics and the trend simulation weight of the first multi-mode three-dimensional component element in the first BIM state attribute vector sample; and a second algorithm debugging error is constructed based on the difference between the model annotation information used to characterize whether the second multi-mode three-dimensional component element in the second BIM state attribute vector sample has abnormal thermal characteristics and the trend simulation weight of the second multi-mode three-dimensional component element in the second BIM state attribute vector sample; The trend simulation processing algorithm is debugged according to the first algorithm debugging error and the second algorithm debugging error.

[0016] In some independent embodiments, debugging the BIM annotation classification network according to the first model annotation information and the second model annotation information includes: A third algorithm debugging error is constructed according to the difference between the first model annotation information and the prior model annotation information, and a fourth algorithm debugging error is constructed according to the difference between the second model annotation information and the prior model annotation information; wherein the prior model annotation information is model annotation information for characterizing whether the cross-state of the power line layout of the x first multi-mode three-dimensional component elements and the x second multi-mode three-dimensional component elements has abnormal thermal characteristics; The BIM annotation classification network is debugged according to the third algorithm debugging error and the fourth algorithm debugging error.

[0017] An embodiment of the present application provides a data processing system, comprising at least one processor and a memory; the memory stores computer-executable instructions; the at least one processor executes the computer-executable instructions stored in the memory, so that the at least one processor performs the above method.

[0018] An embodiment of the present application provides a computer-readable storage medium on which a computer program is stored. The computer program implements the above method when running.

[0019] In the embodiment of the present application, through a series of innovative steps such as model attribute extraction, operation and maintenance trend identification, construction element sampling, and operation and maintenance fault risk prediction, the embodiment of the present application aims to achieve comprehensive digital management of power system buildings, improve design accuracy and construction efficiency, while reducing operation and maintenance costs, and provide strong support for the intelligent upgrade of the power industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 A flowchart of a method for processing power system building three-dimensional model data using BIM provided in an embodiment of the present application.

[0021] Figure 2 A schematic diagram of the structure of a data processing system provided in an embodiment of the present application. DETAILED DESCRIPTION

[0022] In order to better understand the above technical scheme, the technical scheme of the present application is described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present application and the specific features in the embodiments are detailed descriptions of the technical scheme of the present application, rather than limitations on the technical scheme of the present application. In the absence of conflict, the embodiments of the present application and the technical features in the embodiments can be combined with each other.

[0023] Figure 1 A method for processing three-dimensional model data of an electric power system building using BIM is shown, which is applied to a data processing system. The method includes S101-S105.

[0024] S101. Extract model attributes from first BIM electric construction model data to obtain a first BIM state attribute vector, and extract model attributes from second BIM electric construction model data to obtain a second BIM state attribute vector.

[0025] In the embodiment of the present application, the first BIM electric construction model data refers to the three-dimensional model data created for the power system building through BIM technology at a specific time point or stage. These data contain various detailed information of the building, such as structure, equipment, pipelines, etc., and are stored and managed in a digital form. And "first" can represent that this is a data version at an earlier time point, which is used for comparative analysis with subsequent data versions.

[0026] Similar to the first BIM electric construction model data, the second BIM electric construction model data is also a three-dimensional model data created for the power system building through BIM technology, but it represents the building status at another time point or stage. The "second" can be understood as being obtained after the first data, reflecting the change or update of the building status.

[0027] Model attribute extraction is used to extract key attribute information from BIM electrical construction model data. These attributes can include the size, location, material, connection method, etc. of the component, which is the basis for subsequent analysis and prediction. The extracted attribute information can be stored in a structured data format to facilitate efficient calculation and analysis.

[0028] The first BIM state attribute vector is a vector composed of multiple numerical features, which is used to fully describe the state of the first BIM electrical construction model data. For example, the state attribute vector of a component may include numerical features such as its length, width, height, position coordinates (x, y, z), material type, etc. These features are combined into a vector that can fully and accurately represent the state of the component in the BIM model. By comparing the state attribute vectors at different time points, the changing trend of the component state can be analyzed.

[0029] For example, there is a power line component, whose first BIM status attribute vector can be as follows: Length: 10m; Diameter: 0.1m; Material type: copper (can be converted into a numerical code, such as 1 for copper); Starting position coordinates: (0, 0, 0); End position coordinates: (10, 0, 0).

[0030] The above vectors can fully describe the status of the power line component in the first BIM electric construction model data.

[0031] Similar to the first BIM state attribute vector, the second BIM state attribute vector is also a vector composed of multiple numerical features, but is used to describe the component status in the second BIM electrical construction model data. These numerical features can correspond to the features in the first BIM state attribute vector for direct comparison and analysis. By comparing the state attribute vectors at two different time points, the changes in the component status can be revealed, and the possible operation and maintenance failure risks can be predicted.

[0032] Taking the same power line component as an example, its second BIM state attribute vector may be as follows (for example, some attributes have changed): Length: 10 meters (unchanged); Diameter: 0.12 m (changed); Material Type: Copper (unchanged); Starting position coordinates: (0, 0, 0.1) (changed); End position coordinates: (10, 0, 0.1) (changed).

[0033] In the field of power system buildings, as the building is used and over time, its state may change. In order to capture these changes and predict possible operation and maintenance failure risks, it is first necessary to extract key information from the BIM power construction model data at two different time points. This process is called model attribute extraction. For the first BIM power construction model data, the component elements in the model can be analyzed one by one. For each component, its key attributes, such as size, position, material, etc., are extracted, and this information is integrated into a numerical feature vector, namely the first BIM state attribute vector. This vector not only fully describes the initial state of the component in the model, but also provides basic data for subsequent trend analysis and risk prediction. Similarly, for the second BIM power construction model data, the same attribute extraction process is performed. By comparing the state attribute vectors at two time points, the changes in the component state can be observed, such as size adjustment, position movement, or material replacement. These changes may indicate potential operation and maintenance risk points, which require further attention and analysis.

[0034] S102. Perform operation and maintenance trend identification on the first BIM status attribute vector to obtain trend simulation weights of each first multi-mode three-dimensional component element in the first BIM status attribute vector, and perform operation and maintenance trend identification on the second BIM status attribute vector to obtain trend simulation weights of each second multi-mode three-dimensional component element in the second BIM status attribute vector, wherein the trend simulation weights are used to characterize the confidence coefficient of operation and maintenance fault risk prediction for BIM electric power construction model data.

[0035] In the embodiment of the present application, operation and maintenance trend identification is used to identify the change trend of each component element in the BIM electric construction model during the operation and maintenance process. By comparing the BIM state attribute vectors at different time points, the change in the state of the component element can be analyzed, and then its future development trend can be predicted. This identification helps to discover potential operation and maintenance problems in a timely manner and provide strong support for fault risk prediction.

[0036] The first multi-mode three-dimensional component elements refer to the three-dimensional component elements in the first BIM electric construction model data. They not only have physical structural characteristics, but also integrate mechanical structural modes and power connection relationship modes. For example, a power line component may be structurally manifested as a cable with a certain length, diameter and material; in the mechanical structural mode, it may have a specific vibration frequency and damping characteristics; in the power connection relationship mode, it carries the task of transmitting current and forms a specific power connection network with other components. The fusion of these multi-modal features enables each three-dimensional component element to have a unique and comprehensive description in the BIM model.

[0037] The trend simulation weight is a numerical indicator used to characterize the importance and change trend of each component element in the BIM power construction model data during the operation and maintenance trend identification process. The size of the weight reflects the significance of the change in the state of the component element and the degree of impact on the overall operation and maintenance status. By assigning a trend simulation weight to each component element, the risk of operation and maintenance failure can be predicted more accurately and corresponding maintenance strategies can be formulated.

[0038] Similar to the first multi-mode three-dimensional component element, the second multi-mode three-dimensional component element refers to the three-dimensional component element in the second BIM electrical construction model data, which also integrates the characteristics of mechanical structure mode and power connection relationship mode. These elements represent the state of the building at different time points, and through comparative analysis, changes in the building state and possible operation and maintenance problems can be revealed.

[0039] The confidence coefficient of the operation and maintenance fault risk prediction of BIM electric construction model data is a numerical indicator used to evaluate the reliability of the prediction results. In the process of operation and maintenance fault risk prediction, due to the uncertainty of the data and the limitations of the model, the prediction results may have certain errors. The confidence coefficient reflects the reliability of the prediction results. The higher the value, the more reliable the prediction results. The trend simulation weight can be used as a form of confidence coefficient to quantify the reliability of the prediction results.

[0040] In the process of power system building operation and maintenance, in-depth analysis of BIM power construction model data is crucial. The core goal of S102 is to assign trend simulation weights to each multi-mode three-dimensional component element in the first BIM state attribute vector and the second BIM state attribute vector through operation and maintenance trend identification. First, the data processing system conducts a detailed analysis of each first multi-mode three-dimensional component element in the first BIM state attribute vector. These elements not only contain basic information such as the physical size and position of the component, but also integrate the complex characteristics of the mechanical structure mode and the power connection relationship mode. By comparing historical data with the current state, the system identifies the change trend of each element, such as the increase or decrease of size, the movement of position, or the change of power connection relationship. Then, the system assigns a trend simulation weight to each first multi-mode three-dimensional component element based on these change trends. This weight is actually a confidence coefficient, which reflects the importance and reliability of the element in the prediction of operation and maintenance fault risk. The higher the weight, the greater the impact of the state change of the element on the overall operation and maintenance status, so more attention needs to be paid in the subsequent fault risk prediction. Similarly, the data processing system will also perform the same operation and maintenance trend identification and weight allocation process on each second multi-mode three-dimensional component element in the second BIM state attribute vector. In this way, two sets of BIM state attribute vectors with trend simulation weights are obtained, providing strong data support for subsequent operation and maintenance failure risk prediction.

[0041] S103. According to the trend simulation weights of the respective first multimodal three-dimensional component elements, construct element sampling is performed on the first BIM state attribute vector to obtain x first multimodal three-dimensional component elements; and according to the trend simulation weights of the respective second multimodal three-dimensional component elements, construct element sampling is performed on the second BIM state attribute vector to obtain x second multimodal three-dimensional component elements.

[0042] In an embodiment of the present application, construction element sampling refers to selecting a portion of representative multi-mode three-dimensional component elements from the BIM state attribute vector for analysis according to specific standards or weights during data processing. This process is intended to reduce the amount of data processing while ensuring that the selected elements can fully reflect the characteristics and trends of the overall data. In the BIM electric construction model data processing, construction element sampling can be based on the trend simulation weights of each component element to ensure the scientificity and effectiveness of the sampling.

[0043] In detail, S103 is a key link in the data processing process, which involves sampling the construction elements of the first BIM state attribute vector and the second BIM state attribute vector according to the trend simulation weights calculated in the previous step. First, the data processing system reviews the trend simulation weights assigned to each first multi-mode three-dimensional component element in S102. These weights reflect the changing trends of each component element during the operation and maintenance process and the degree of its influence on the operation and maintenance status of the entire system. Based on these weights, the system uses a scientific sampling method, such as weighted random sampling, to select x representative first multi-mode three-dimensional component elements from the first BIM state attribute vector. Similarly, for the second BIM state attribute vector, the data processing system will also sample according to the trend simulation weights of each second multi-mode three-dimensional component element. This process is similar to the sampling method of the first BIM state attribute vector, which aims to ensure that the selected x second multi-mode three-dimensional component elements can fully and accurately reflect the overall characteristics and operation and maintenance trends of the second BIM electric construction model data. Through the construction element sampling of S103, the data processing system not only reduces the amount of calculation for subsequent analysis, but also improves the pertinence and accuracy of the analysis. The selected x first multi-mode three-dimensional component elements and x second multi-mode three-dimensional component elements will serve as basic data for subsequent operation and maintenance trend identification and fault risk prediction, providing strong support for the stable operation of the power system.

[0044] S104. Perform operation and maintenance trend identification on the x first multi-mode three-dimensional component elements and the x second multi-mode three-dimensional component elements to obtain first model annotation information corresponding to the x first multi-mode three-dimensional component elements and second model annotation information corresponding to the x second multi-mode three-dimensional component elements, wherein the first model annotation information is used to indicate mining u first multi-mode three-dimensional component elements among the x first multi-mode three-dimensional component elements that have a power line layout intersection state with the x second multi-mode three-dimensional component elements, and the second model annotation information is used to enhance v second multi-mode three-dimensional component elements among the x second multi-mode three-dimensional component elements that have a power line layout intersection state with the x first multi-mode three-dimensional component elements.

[0045] In an embodiment of the present application, the first model annotation information refers to the annotation information generated for each element after the operation and maintenance trend of the x first multi-mode three-dimensional component elements in the first BIM state attribute vector is identified. This information not only reveals the status and change trend of these elements during the operation and maintenance process, but is also particularly used to identify the parts that have a power line layout intersection state with the elements in the second BIM state attribute vector. In short, the first model annotation information helps to discover the intersecting or overlapping power line parts that may exist in the two models.

[0046] Similar to the first model annotation information, the second model annotation information is generated after the operation and maintenance trend identification of the x second multi-mode three-dimensional component elements in the second BIM state attribute vector. Its main function is to enhance the identification of the parts where there are cross-laying power lines with the elements in the first BIM state attribute vector. This information provides an important basis for subsequent fault risk prediction.

[0047] The u first multi-mode three-dimensional component elements with power line layout crossing states refer to the elements in the x sampled elements of the first BIM state attribute vector that are identified by the first model annotation information as having power line layout crossing states with the elements in the second BIM state attribute vector. These elements exist in both models, and their power lines cross or overlap, so they may be key risk points in operation and maintenance.

[0048] Similar to the above u elements, the v second multi-mode three-dimensional component elements with cross-layout of power lines are the parts of the x sampling elements of the second BIM state attribute vector that are identified by the second model annotation information as having cross-layout of power lines with the first BIM state attribute vector elements. These elements are also regarded as potential risk points in operation and maintenance.

[0049] In S104, the data processing system performs an in-depth analysis of the x first multi-mode three-dimensional component elements and x second multi-mode three-dimensional component elements previously sampled from the first BIM state attribute vector and the second BIM state attribute vector. The core of this step is to use advanced operation and maintenance trend identification technology to generate detailed model annotation information for these elements. For example, the system first performs a detailed state and trend analysis on each element, which includes but is not limited to their position, size, material properties, and changes between two different time points. Then, based on these analyses, the system generates first model annotation information and second model annotation information for each element. These annotation information not only reveals the operation and maintenance status and trend of the element itself, but more importantly, they are used to identify the three-dimensional component elements that exist in both BIM models and have a cross state in the power line layout. This is achieved by comparing the annotation information of the two models to find out the commonalities and differences therein. Finally, through S104, the key points that exist in both models and may have operation and maintenance risks can be accurately located, that is, the first multi-mode three-dimensional component elements (u) and the second multi-mode three-dimensional component elements (v) with a cross state in the power line layout. This information provides valuable decision support for subsequent fault risk prediction and optimized maintenance of power systems.

[0050] S105. Perform operation and maintenance fault risk prediction based on the u first multi-mode three-dimensional component elements and the v second multi-mode three-dimensional component elements to obtain a first operation and maintenance fault risk view between the first BIM electric power construction model data and the second BIM electric power construction model data.

[0051] In an embodiment of the present application, operation and maintenance failure risk prediction is a predictive analysis method that aims to assess the failure risks that may be encountered during the operation and maintenance of the power system. This prediction can be based on historical data, real-time monitoring data, and information such as element attributes and operation and maintenance trends in the BIM model. By using advanced algorithms and models, the system can predict possible failures of equipment or components in the future, as well as the impact these failures may have on the overall performance and safety of the system. This prediction helps to detect potential problems in advance, so as to intervene and prevent them in time and ensure the stable operation of the power system.

[0052] The first operation and maintenance failure risk viewpoint refers to the comprehensive judgment or opinion on the operation and maintenance failure risk formed by the data processing system after in-depth analysis and comparison of the first BIM power construction model data and the second BIM power construction model data. This viewpoint is based on the operation and maintenance trend identification, state change analysis, and failure risk prediction of specific component elements in the two models (such as three-dimensional component elements with cross-laying of power lines). It can help power system operators understand current and future failure risk points and provide an important basis for formulating targeted maintenance plans and optimization strategies.

[0053] In S105, the data processing system takes the u first multi-mode three-dimensional component elements and v second multi-mode three-dimensional component elements identified in the previous step as key analysis objects. Since these elements exist in the BIM electric construction model data at two different time points and their power line layouts are in a cross state, they are considered to be important reference points for operation and maintenance failure risk prediction. The system first integrates various types of information of these key elements, including but not limited to their location, size, material properties, historical operation and maintenance records, and simulation weights obtained through operation and maintenance trend identification. Next, using advanced prediction models and algorithms, such as machine learning models or statistical analysis models, the system conducts in-depth failure risk prediction for these elements. During the prediction process, the system considers a variety of factors that may affect the failure risk of elements, such as environmental factors (temperature, humidity, etc.), usage conditions (load conditions, operating time, etc.), and maintenance history. Through a comprehensive analysis of these factors, the system can predict the probability of failure of each element in the future and the possible types of failures. Finally, based on these prediction results, the data processing system generates a first operation and maintenance failure risk view. This view not only includes the failure risk prediction results of each key element, but also includes the response strategies and optimization suggestions for these risks. In this way, power system operators can formulate more accurate and effective maintenance plans based on this view, thereby ensuring the safe and stable operation of the power system.

[0054] In the field of power system construction, the application of BIM (Building Information Modeling) technology is very critical. The data processing system plays a vital role in the in-depth analysis and processing of BIM model data. The following is an exemplary application scenario based on the above S101-S105, which shows how the data processing system can realize the processing and analysis of BIM power construction model data.

[0055] First, the data processing system processes the BIM electric construction model data at two different time points, namely the first BIM electric construction model data and the second BIM electric construction model data. The data processing system extracts the first BIM state attribute vector and the second BIM state attribute vector from the first BIM electric construction model data and the second BIM electric construction model data respectively through the model attribute extraction technology (which can also be understood as model attribute feature mining processing). The first BIM state attribute vector and the second BIM state attribute vector contain the attribute information of each component element in the model, such as size, position, material, etc.

[0056] Next, the data processing system uses AI algorithms to identify the operation and maintenance trends of the extracted BIM status attribute vectors. The purpose of this step is to analyze the changing trends of each multi-mode three-dimensional component element during the operation and maintenance process and assign a trend simulation weight to it. The trend simulation weight can be understood as a confidence coefficient, which is used to characterize the reliability of the operation and maintenance fault risk prediction of the BIM electric construction model data.

[0057] With these trend simulation weights, the data processing system performs construction element sampling. The data processing system samples x elements from the first BIM state attribute vector according to the weights of each first multi-mode three-dimensional component element; similarly, it also samples x elements from the second BIM state attribute vector. This sampling method can ensure that the selected elements are highly representative in the operation and maintenance trend analysis.

[0058] Then, the data processing system further identifies the operation and maintenance trends of the 2x multi-mode three-dimensional component elements (x first multi-mode three-dimensional component elements and x second multi-mode three-dimensional component elements), and annotates them with the first model annotation information and the second model annotation information respectively. The first model annotation information and the second model annotation information are crucial for subsequent analysis because they can help the system accurately identify those three-dimensional component elements that exist in the BIM electric construction model data at two different time points and whose power line layouts are crossed.

[0059] Finally, the data processing system combines these three-dimensional component elements with cross-states to predict the operation and maintenance failure risk. By comparing and analyzing the state changes of these elements in the two models, the data processing system can generate an operation and maintenance failure risk view between the first BIM power construction model data and the second BIM power construction model data. The operation and maintenance failure risk view not only points out the possible failure risk points, but also provides strong data support for the maintenance and optimization of the power system.

[0060] In summary, the data processing system plays a core role in the data processing of the BIM power system building three-dimensional model. Through a series of steps such as extracting model attributes, identifying operation and maintenance trends, sampling analysis, and fault risk prediction, the system can provide scientific and accurate decision-making basis for the stable operation of the power system.

[0061] The embodiment of the present application uses BIM technology to achieve efficient processing and analysis of the three-dimensional model data of the power system building, significantly improving the planning, design and operation and maintenance capabilities of the power system. First, by extracting model attributes from the first BIM power construction model data and the second BIM power construction model data, the first BIM state attribute vector and the second BIM state attribute vector are obtained respectively, thereby fully capturing the key information in the model. Furthermore, through operation and maintenance trend identification, trend simulation weights are assigned to the multi-mode three-dimensional component elements in these state attribute vectors. This innovative step not only quantifies the changing trend of the components during the operation and maintenance process, but also provides a scientific confidence coefficient for subsequent fault risk prediction.

[0062] Another highlight of the embodiment of the present application is that it samples the construction elements of the BIM state attribute vector according to the trend simulation weight, and accurately selects representative component elements for further analysis. By conducting in-depth operation and maintenance trend identification on these sampled elements and marking components with cross-layout of power lines, the embodiment of the present application successfully reveals the potential operation and maintenance risk points in the power system.

[0063] Finally, combining these key component elements, the embodiment of the present application predicts the risk of operation and maintenance failures, and obtains the first view of operation and maintenance failure risks, providing strong data support for the stable operation of the power system. This innovative method not only improves the accuracy of design and the efficiency of construction, but also effectively reduces the operation and maintenance costs by realizing the comprehensive digital management of power system buildings, injecting new vitality into the sustainable development of the power industry.

[0064] In some optional embodiments, the first BIM state attribute vector is sampled for constructing elements according to the trend simulation weights of the respective first multimodal three-dimensional component elements to obtain x first multimodal three-dimensional component elements, and the second BIM state attribute vector is sampled for constructing elements according to the trend simulation weights of the respective second multimodal three-dimensional component elements to obtain x second multimodal three-dimensional component elements, including: determining x according to preset parameters; determining the first x first multimodal three-dimensional component elements in the first BIM state attribute vector as the x first multimodal three-dimensional component elements according to a descending order rule of trend simulation weights according to the trend simulation weights of the respective first multimodal three-dimensional component elements, and determining the first x second multimodal three-dimensional component elements in the second BIM state attribute vector as the x second multimodal three-dimensional component elements according to a descending order rule of trend simulation weights according to the trend simulation weights of the respective second multimodal three-dimensional component elements.

[0065] In this embodiment, the data processing system will perform a series of precise operations to sample and select key three-dimensional component elements from the first BIM electrical construction model data and the second BIM electrical construction model data for analysis. This process is based on the first BIM state attribute vector and the second BIM state attribute vector obtained by model attribute extraction, and the trend simulation weight assigned to each component element through operation and maintenance trend identification.

[0066] First, the data processing system determines the number of samples x according to preset parameters. The preset parameters can be set according to historical experience, system performance or user needs to ensure the representativeness of the samples and the processing efficiency.

[0067] Next, the data processing system sorts the first multi-mode three-dimensional component elements according to their trend simulation weights. The sorting rule is to sort in descending order of trend simulation weights, that is, the element with the highest weight is at the front. The purpose of this is to ensure that during the sampling process, those component elements that have the greatest impact on the operation and maintenance trend and are the most representative can be selected first.

[0068] After arranging in descending order, the data processing system selects the first x first multi-mode three-dimensional component elements from the first BIM state attribute vector. These elements are considered to be the most important part of the first BIM electric construction model data because they have a higher weight in the identification of operation and maintenance trends, and they play a key role in the subsequent fault risk prediction.

[0069] Similarly, the data processing system also performs the same operation on the second multi-mode three-dimensional component element in the second BIM state attribute vector. The system arranges the trend simulation weights of each element in descending order and selects the first x elements as the sampling result.

[0070] In this way, the data processing system can ensure that the component elements extracted from the BIM electric construction model data at two different time points are the most obvious and representative parts of the operation and maintenance trends in their respective models. This not only improves the accuracy of subsequent analysis, but also provides a more reliable data basis for the prediction of operation and maintenance failure risks. In other words, the data processing system realizes efficient sampling of key component elements in the BIM electric construction model data. The use of trend simulation weights for descending sorting and sampling selection ensures that the selected elements have high importance and representativeness in the operation and maintenance analysis. This method not only optimizes the data processing process, but also improves the accuracy and efficiency of the prediction of operation and maintenance failure risks. By reducing the amount of unnecessary data analysis, the system can locate potential fault risk points more quickly, thereby providing more timely and effective support for the stable operation of the power system.

[0071] In a further technical solution of the above embodiment, determining x based on preset parameters includes: determining x based on the preset parameters and the characteristic dimension of the first BIM state attribute vector, or determining x based on the preset parameters and the characteristic dimension of the second BIM state attribute vector.

[0072] In detail, in this technical solution, the data processing system first refers to preset parameters, which may include a series of rules, thresholds or constants required in the algorithm for data processing. These parameters can be set by experts according to actual needs, aiming to guide the system on how to analyze and process data more accurately.

[0073] In the process of determining x, the data processing system considers two BIM (Building Information Modeling) state attribute vectors: the first BIM state attribute vector and the second BIM state attribute vector. The BIM state attribute vector refers to a vector that describes the various attributes and states of the building model, which may include data on the building structure, materials, systems, etc. Feature dimensions refer to specific elements or features in these attribute vectors, such as the length, width, height, material type, etc. of the building.

[0074] The data processing system determines the value of x based on the preset parameters and the characteristic dimensions of the BIM state attribute vector. For example, the system may use an algorithm or formula to weight, calculate or otherwise process these parameters and dimensions to ultimately derive a specific x value. This x value may be a key indicator for further data analysis or an important parameter for a process in the control system.

[0075] For example, in the scenario of construction project management, the data processing system may need to determine the overall budget (i.e., the value of x) of a project based on characteristic dimensions such as the building volume, the type of materials used, and the complexity of the design, as well as preset parameters such as the construction cost and duration. This budget will serve as an important basis for subsequent project management and resource allocation.

[0076] In actual operation, the data processing system may first obtain the first BIM state attribute vector, which may contain the basic physical information and design parameters of the building. The system processes and analyzes these characteristic dimensions according to preset parameters, such as cost estimation rules, material prices, etc., to preliminarily determine the value of x. On the other hand, if the system needs to consider more dynamic factors, such as construction progress, environmental factors, etc., it may further introduce a second BIM state attribute vector, which may contain more real-time data and changing factors. By comprehensively analyzing these two BIM state attribute vectors and combining them with preset parameters, the data processing system can more accurately determine the value of x.

[0077] In this way, the data processing system can flexibly determine the key parameter x according to different needs and scenarios, thereby providing accurate data support for subsequent decision-making and optimization. This not only improves the accuracy and flexibility of data processing, but also helps professionals in related fields better understand and control complex systems or processes, ultimately achieving efficient use of resources and optimized project management.

[0078] Under some preferred design ideas, the operation and maintenance trend identification of the x first multimodal three-dimensional component elements and the x second multimodal three-dimensional component elements is performed to obtain the first model annotation information corresponding to the x first multimodal three-dimensional component elements and the second model annotation information corresponding to the x second multimodal three-dimensional component elements, including: determining the first basic model annotation information corresponding to the x first multimodal three-dimensional component elements, and determining the second basic model annotation information corresponding to the x second multimodal three-dimensional component elements; adjusting the x first multimodal three-dimensional component elements according to the x first multimodal three-dimensional component elements, the x second multimodal three-dimensional component elements, the first basic model annotation information and the second basic model annotation information to obtain x first adjusted The invention relates to a three-dimensional component element, and according to the x first multi-mode three-dimensional component elements, the x second multi-mode three-dimensional component elements, the first basic model annotation information and the second basic model annotation information, the x second multi-mode three-dimensional component elements are adjusted to obtain x second adjusted three-dimensional component elements; according to the x first adjusted three-dimensional component elements, the first basic model annotation information is adjusted to obtain first adjusted model annotation information, and according to the x second adjusted three-dimensional component elements, the second basic model annotation information is adjusted to obtain second adjusted model annotation information; the first model annotation information is determined according to the first adjusted model annotation information, and the second model annotation information is determined according to the second adjusted model annotation information.

[0079] Based on this design idea, the data processing system identifies the operation and maintenance trends of x first multi-mode three-dimensional component elements and x second multi-mode three-dimensional component elements. This process involves multiple steps, aiming to accurately obtain the model annotation information of these three-dimensional component elements.

[0080] First, the data processing system determines the first basic model annotation information corresponding to the x first multi-mode three-dimensional component elements, and the second basic model annotation information corresponding to the x second multi-mode three-dimensional component elements. These basic model annotation information may include basic information such as the type, size, and material of the component, which is the basis for subsequent analysis and adjustment.

[0081] Next, the data processing system adjusts the x first multi-mode three-dimensional component elements and the x second multi-mode three-dimensional component elements based on these three-dimensional component elements and the basic model annotation information. This adjustment process may involve fine-tuning the position, direction, size and other attributes of the components to ensure that they are more consistent with the actual building model. After the adjustment, the system obtains x first adjusted three-dimensional component elements.

[0082] Then, the data processing system makes corresponding adjustments to the previously determined basic model annotation information based on these adjusted three-dimensional component elements. The purpose of this step is to ensure the accuracy and completeness of the model annotation information so that it can more realistically reflect the actual situation of the three-dimensional component elements. After this step, the system obtains the first adjusted model annotation information and the second adjusted model annotation information.

[0083] Finally, the data processing system determines the final first model annotation information and second model annotation information based on these adjusted model annotation information. This information will serve as an important output for operation and maintenance trend identification and provide strong data support for subsequent building operation and maintenance management.

[0084] In this way, the data processing system can more accurately identify the operation and maintenance trends of multi-mode three-dimensional component elements and provide accurate model annotation information. This not only helps to improve the efficiency and accuracy of building operation and maintenance, but also provides strong technical support for the digital transformation of the construction industry. At the same time, this technical solution also demonstrates the powerful ability and flexibility of the data processing system in processing complex three-dimensional data, and provides a useful reference for technological innovation and application expansion in related fields.

[0085] Under some other preferred design ideas, before extracting model attributes from the first BIM electric power construction model data to obtain a first BIM state attribute vector, and extracting model attributes from the second BIM electric power construction model data to obtain a second BIM state attribute vector, the method also includes: when the feature dimension of the first BIM electric power construction model data and the feature dimension of the second BIM electric power construction model data do not match, converting the first BIM electric power construction model data and the second BIM electric power construction model data to the same feature dimension. Then the determining of the first basic model annotation information corresponding to the x first multimodal three-dimensional component elements, and the determining of the second basic model annotation information corresponding to the x second multimodal three-dimensional component elements, include: determining the first basic model annotation information based on the feature members augmented during the dimensional conversion process of the first BIM electric construction model data, wherein the model query vector corresponding to the quantized value of the feature members augmented in the x first multimodal three-dimensional component elements in the first basic model annotation information is the annotation keyword; determining the second basic model annotation information based on the feature members augmented during the dimensional conversion process of the second BIM electric construction model data, wherein the model query vector corresponding to the quantized value of the feature members augmented in the x second multimodal three-dimensional component elements in the second basic model annotation information is the annotation keyword.

[0086] In the application scenario of processing the three-dimensional model data of the power system building, the data processing system faces the challenge of processing BIM (Building Information Modeling) electric construction model data from different sources with different feature dimensions. The feature dimensions of these data may not match due to differences in modeling software, versions or standards, so dimension conversion is required to ensure the consistency and comparability of the data. Based on this embodiment, before extracting the model properties of the first BIM electric construction model data and the second BIM electric construction model data, the data processing system will first check whether the feature dimensions of the two match. If the feature dimensions are found to be mismatched, the system takes measures to convert the two BIM electric construction model data to the same feature dimensions. This conversion process may involve adding, deleting or modifying certain feature members to ensure the structural consistency of the two models.

[0087] After the dimensional conversion is completed, the data processing system continues its workflow to determine the basic model annotation information corresponding to the x first multi-mode three-dimensional component elements and the x second multi-mode three-dimensional component elements. In this process, the system determines these basic model annotation information based on the feature members augmented by the BIM electric construction model data during the dimensional conversion process. For example, for the first BIM electric construction model data, the system determines the first basic model annotation information based on the feature members added during the conversion process. Among these annotation information, the model query vectors corresponding to the quantized values ​​of the augmented feature members will be regarded as annotation keywords, which are crucial for the subsequent identification and adjustment of three-dimensional component elements.

[0088] Similarly, for the second BIM electrical construction model data, the data processing system will also determine the second basic model annotation information based on the feature members augmented during the dimensional conversion process, and use the model query vector corresponding to the quantized value of the augmented feature member as the annotation keyword.

[0089] Through this method, the data processing system can ensure that BIM electric construction model data from different sources and dimensions can be compared and analyzed on the same feature dimension after processing. This not only improves the accuracy and efficiency of data processing, but also provides a solid foundation for subsequent operation and maintenance trend identification, model adjustment and other steps.

[0090] In this way, by introducing the dimension conversion mechanism, the problem of mismatch in the feature dimensions of BIM power construction model data is effectively solved, allowing the data processing system to more flexibly respond to data from different sources. At the same time, by augmenting the feature members to determine the basic model annotation information, the system can more accurately identify and annotate the three-dimensional component elements, thereby providing more powerful data support for the building operation and maintenance management of the power system. This not only improves the level of intelligent data processing of the three-dimensional model of the power system building, but also lays a solid foundation for the digital transformation and upgrading of related industries.

[0091] In other optional embodiments, the adjusting the x first multimodal three-dimensional component elements according to the x first multimodal three-dimensional component elements, the x second multimodal three-dimensional component elements, the first basic model annotation information and the second basic model annotation information to obtain x first adjusted three-dimensional component elements, and adjusting the x second multimodal three-dimensional component elements according to the x first multimodal three-dimensional component elements, the x second multimodal three-dimensional component elements, the first basic model annotation information and the second basic model annotation information to obtain x second adjusted three-dimensional component elements, includes: using the first basic model annotation information and the x first multimodal three-dimensional component elements to obtain x first spatial structure relationship vectors, and using the second basic model annotation information The x second multimodal three-dimensional component elements are annotated with the basic model, and the x second multimodal three-dimensional component elements are obtained by combining the annotation information of the basic model with the x second multimodal three-dimensional component elements; the x first spatial structural relationship vectors are subjected to intra-class feature focusing processing by using the first intra-class feature focusing branch, and the x second spatial structural relationship vectors and the x processing results generated by the first intra-class feature focusing branch are subjected to the first inter-class feature focusing branch to obtain the x first adjusted three-dimensional component elements, and the x second spatial structural relationship vectors are subjected to intra-class feature focusing processing by using the second intra-class feature focusing branch, and the x first spatial structural relationship vectors and the x processing results generated by the second intra-class feature focusing branch are subjected to the second inter-class feature focusing branch to obtain the x second adjusted three-dimensional component elements.

[0092] In an embodiment of the present application, the data processing system needs to process and analyze a large number of three-dimensional component elements to ensure the accuracy and efficiency of the building model. In this process, adjusting the three-dimensional component elements is a key step, which can help the system better understand and optimize the spatial structure of the building model. In detail, the data processing system adjusts these three-dimensional component elements based on x first multi-mode three-dimensional component elements, x second multi-mode three-dimensional component elements, first basic model annotation information, and second basic model annotation information. This process can be divided into several key steps.

[0093] First, the data processing system uses the first basic model annotation information and x first multi-mode three-dimensional component elements to generate x first spatial structure relationship vectors. These vectors describe the spatial relationship and attributes between component elements and are the basis for subsequent analysis. Similarly, the system also uses the second basic model annotation information and x second multi-mode three-dimensional component elements to generate x second spatial structure relationship vectors.

[0094] Next, the system uses the first intra-class feature focusing branch to perform intra-class feature focusing on the x first spatial structure relationship vectors. The purpose of this step is to highlight the key features in the vectors so that the system can more easily identify and adjust the three-dimensional component elements. At the same time, the system also uses the first inter-class feature focusing branch to combine the processing results generated by the second spatial structure relationship vector and the first intra-class feature focusing branch to obtain the x first adjusted three-dimensional component elements.

[0095] Similarly, the data processing system will also use the second intra-class feature focusing branch to perform intra-class feature focusing on the x second spatial structure relationship vectors, and use the second inter-class feature focusing branch to combine the first class spatial structure relationship vectors and the processing results generated by the second intra-class feature focusing branch to obtain x second adjusted three-dimensional component elements.

[0096] These adjusted 3D component elements not only take into account the properties of the component itself, but also integrate the information of other related components, so they are more accurate and comprehensive. Through this adjustment method, the data processing system can better understand and optimize the spatial structure of the building model, providing strong support for subsequent building operation and maintenance management.

[0097] In this way, by introducing the concepts of spatial structural relationship vectors and feature focused branches, the accuracy and efficiency of the adjustment of three-dimensional component elements are improved. It not only helps to improve the intelligent level of data processing of three-dimensional models of power system buildings, but also provides a useful reference and reference for the digital transformation and upgrading of related industries. Through this method, the data processing system can more accurately identify and adjust three-dimensional component elements, thereby providing more reliable and efficient data support for the building operation and maintenance management of the power system.

[0098] In some optional technical solutions, the first basic model annotation information is adjusted according to the x first adjusted three-dimensional component elements to obtain the first adjusted model annotation information, and the second basic model annotation information is adjusted according to the x second adjusted three-dimensional component elements to obtain the second adjusted model annotation information, including: operating and maintenance trend identification of the x first adjusted three-dimensional component elements to obtain two discrimination weights corresponding to each of the x first adjusted three-dimensional component elements, and operating and maintenance trend identification of the x second adjusted three-dimensional component elements to obtain two discrimination weights corresponding to each of the x second adjusted three-dimensional component elements, the two discrimination weights being used to characterize the discrimination weights of the element vectors belonging to the cross state of the power line layout and the discrimination weights of the element vectors not belonging to the cross state of the power line layout; determining the x first adjusted three-dimensional component elements according to the two discrimination weights corresponding to each of the first adjusted three-dimensional component elements. x first key-value relationship features corresponding to the first adjusted three-dimensional component element, and x second key-value relationship features corresponding to the x second adjusted three-dimensional component elements are determined based on the two discrimination weights corresponding to each of the second adjusted three-dimensional component elements, wherein the first key-value relationship feature is used to characterize the authentication information of the attention index to which the maximum discrimination weight belongs among the two discrimination weights corresponding to the first adjusted three-dimensional component element, and the second key-value relationship feature is used to characterize the authentication information of the attention index to which the maximum discrimination weight belongs among the two discrimination weights corresponding to the second adjusted three-dimensional component element; discriminant analysis is performed on the x first key-value relationship features to obtain x first discrimination weights, and discriminant analysis is performed on the x second key-value relationship features to obtain x second discrimination weights; the first adjustment model annotation information is obtained by using the first basic model annotation information and the x first discrimination weights, and the second adjustment model annotation information is obtained by using the second basic model annotation information and the x second discrimination weights.

[0099] In the above-mentioned application scenarios, the data processing system needs to continuously update and optimize the model annotation information to ensure that it is consistent with the actual state of the 3D component elements. In some optional technical solutions, this process involves identifying the operation and maintenance trends of the 3D component elements and carefully adjusting the basic model annotation information.

[0100] First, the data processing system identifies the operation and maintenance trends of the x first adjusted three-dimensional component elements and the x second adjusted three-dimensional component elements. The purpose of this step is to obtain two discrimination weights corresponding to each three-dimensional component element, which respectively characterize the possibility that they belong to or do not belong to the element vector of the cross state of the power line layout. These discrimination weights are calculated by complex algorithms, taking into account multiple factors such as the spatial position, shape, and relationship with other components of the component.

[0101] Next, the system determines the key-value relationship feature of each adjusted three-dimensional component element based on these discrimination weights. This feature is actually a certification information, which indicates which of the two discrimination weights (the weight of belonging to or not belonging to the cross state of the power line layout) of the corresponding component element is larger, thus indicating which state the component element is more likely to be in.

[0102] Then, the data processing system performs discriminant analysis on these key-value relationship features to obtain the specific discriminant weight of each component element. This process further refines the evaluation of the status of each component element and provides a more accurate basis for the subsequent adjustment of model annotation information.

[0103] Finally, the system uses these judgment weights and the original basic model annotation information to obtain the adjusted model annotation information. This step ensures the real-time and accuracy of the model annotation information, so that it can better reflect the current status of the three-dimensional component elements.

[0104] In this way, the data processing system can accurately process the three-dimensional model data of the power system building. It not only takes into account the properties of the component elements themselves, but also combines their trends and state changes in the actual operation and maintenance environment. This processing method greatly improves the accuracy and practicality of the model annotation information, and provides strong data support for the operation and maintenance management of the power system. At the same time, this technical solution also demonstrates the flexibility and efficiency of the data processing system in processing complex three-dimensional model data, and contributes an important force to the digital transformation of the power industry.

[0105] In some further exemplary design ideas, the first basic model annotation information is adjusted according to the x first adjusted three-dimensional component elements to obtain first adjusted model annotation information, and the second basic model annotation information is adjusted according to the x second adjusted three-dimensional component elements to obtain second adjusted model annotation information, including: adjusting the x first adjusted three-dimensional component elements according to the x first adjusted three-dimensional component elements, the x second adjusted three-dimensional component elements, the first adjustment model annotation information and the second adjustment model annotation information to obtain x third adjusted three-dimensional component elements, and adjusting the x first adjusted three-dimensional component elements according to the x second adjusted three-dimensional component elements, the first adjustment model annotation information and the second adjustment model annotation information to obtain x third adjusted three-dimensional component elements. Adjust the three-dimensional component elements to obtain x fourth adjusted three-dimensional component elements; adjust the first adjustment model annotation information according to the x third adjusted three-dimensional component elements to obtain third adjustment model annotation information, and adjust the second adjustment model annotation information according to the x fourth adjusted three-dimensional component elements to obtain fourth adjustment model annotation information; determine the first model annotation information according to the third adjustment model annotation information, and determine the second model annotation information according to the fourth adjustment model annotation information, the first model annotation information is the adjustment model annotation information obtained when the number of cyclic adjustments of the first basic model annotation information is not less than the set number, and the second model annotation information is the adjustment model annotation information obtained when the number of cyclic adjustments of the second basic model annotation information is not less than the set number.

[0106] In this embodiment, the data processing system uses an iterative optimization method to continuously adjust and improve the model annotation information. This method gradually improves the accuracy and adaptability of the model annotation information through multiple cycles of adjustment.

[0107] For example, the data processing system first uses the adjusted three-dimensional component elements (i.e., the first adjusted three-dimensional component element and the second adjusted three-dimensional component element) and the previous round of adjustment model annotation information (i.e., the first adjustment model annotation information and the second adjustment model annotation information) to perform a new round of adjustments. The purpose of this round of adjustments is to obtain more accurate three-dimensional component elements, i.e., the third adjusted three-dimensional component element and the fourth adjusted three-dimensional component element.

[0108] After obtaining a new round of adjusted 3D component elements, the data processing system continues to use these elements to further adjust the previous round of adjusted model annotation information, thereby obtaining the third adjusted model annotation information and the fourth adjusted model annotation information. This process is actually an iterative optimization process. Through continuous cyclic adjustment, the system can gradually approach the most realistic model annotation information.

[0109] It is worth noting that this iterative optimization process is not infinite. When the number of cyclic adjustments reaches a set threshold, the system stops iterating and determines the final model annotation information, namely the first model annotation information and the second model annotation information. These two annotation information are the best results obtained by the system after multiple cyclic adjustments, and they can more accurately reflect the actual status of the 3D model of the power system building.

[0110] Through this iterative optimization method, the data processing system can significantly improve the accuracy and reliability of model annotation information. This not only helps to improve the quality of the three-dimensional model of the power system building, but also provides a more solid data foundation for the subsequent power system operation and maintenance management. At the same time, this technical solution also demonstrates the powerful ability and flexibility of the data processing system in processing complex three-dimensional model data, providing strong technical support for the digital transformation of the power industry.

[0111] It can be seen that through multiple cycles of adjustment and iterative optimization, the data processing system can continuously improve and correct the model annotation information to make it closer to the actual situation. The implementation of this technical solution not only improves the accuracy of the model annotation information, but also provides a strong guarantee for the safe and stable operation of the power system.

[0112] In other possible embodiments, the extracting model attributes of the first BIM electrical construction model data to obtain the first BIM state attribute vector, and extracting model attributes of the second BIM electrical construction model data to obtain the second BIM state attribute vector, include: extracting model attributes of the first BIM electrical construction model data to obtain the first BIM state attribute vector and the third BIM state attribute vector, and extracting model attributes of the second BIM electrical construction model data to obtain the second BIM state attribute vector and the fourth BIM state attribute vector, the characteristic dimension of the first BIM state attribute vector is different from the characteristic dimension of the third BIM state attribute vector, and the characteristic dimension of the second BIM state attribute vector is different from the characteristic dimension of the fourth BIM state attribute vector.

[0113] Furthermore, after obtaining the first operation and maintenance fault risk viewpoint between the first BIM electrical construction model data and the second BIM electrical construction model data, the method further includes: constructing element sampling for the third BIM state attribute vector according to the first operation and maintenance fault risk viewpoint, the characteristic dimension of the first BIM state attribute vector, the characteristic dimension of the third BIM state attribute vector and the third BIM state attribute vector to obtain u BIM electrical construction model data subsets corresponding to the u first multi-mode three-dimensional component elements, and constructing element sampling for the fourth BIM state attribute vector according to the first operation and maintenance fault risk viewpoint, the characteristic dimension of the second BIM state attribute vector, the characteristic dimension of the fourth BIM state attribute vector and the fourth BIM state attribute vector to obtain v BIM electrical construction model data subsets corresponding to the v second multi-mode three-dimensional component elements; constructing element sampling for the fourth BIM state attribute vector according to the u BIM electrical construction model data subsets and the v BIM electrical construction model The u BIM electric construction model data subsets are adjusted according to the u BIM electric construction model data subsets to obtain u model data adjustment subsets, and the v BIM electric construction model data subsets are adjusted according to the u BIM electric construction model data subsets and the v BIM electric construction model data subsets to obtain v model data adjustment subsets; the key data clusters of the u model data adjustment subsets are detected to obtain u key data clusters, and according to the model fine-grained comparison result between the third BIM electric construction model data and the fourth BIM electric construction model data, the v model data adjustment subsets are fine-grained optimized to obtain v target model data subsets; distribution feature mapping processing is performed on the v target model data subsets to obtain v distribution mapping features corresponding to the v target model data subsets; state jump analysis is performed in combination with the u key data clusters and the v distribution mapping features to obtain a second operation and maintenance fault risk viewpoint between the first BIM electric construction model data and the second BIM electric construction model data.

[0114] Based on this embodiment, the data processing system needs to conduct in-depth analysis and processing of BIM (Building Information Modeling) electrical construction model data to assess the risk of operation and maintenance failures. In this embodiment, the system adopts a more detailed and complex data processing process.

[0115] First, the data processing system extracts model attributes from the first BIM electrical construction model data and the second BIM electrical construction model data. This step is similar to the previous embodiment, but the difference is that the system not only extracts the first BIM state attribute vector and the second BIM state attribute vector, but also additionally extracts the third BIM state attribute vector and the fourth BIM state attribute vector. The characteristic dimensions of these state attribute vectors are different, providing more comprehensive model information.

[0116] After obtaining these state attribute vectors, the system uses the first operation and maintenance fault risk viewpoint and the characteristic dimensions of each state attribute vector to construct element sampling for the third BIM state attribute vector and the fourth BIM state attribute vector. The purpose of this step is to extract the most relevant part of the operation and maintenance fault risk from the huge BIM electric construction model data to form multiple BIM electric construction model data subsets.

[0117] Next, the system uses the extracted BIM electrical construction model data subsets to adjust each other to eliminate noise and outliers in the data and obtain a more accurate model data adjustment subset. This step helps improve the accuracy of subsequent analysis.

[0118] Then, the system detects key data clusters and performs fine-grained optimization processing on the model data adjustment subset. Key data clusters refer to data sets that have a significant impact on the risk assessment of operation and maintenance failures, and fine-grained optimization processing is to further improve the accuracy and availability of data.

[0119] After completing the above processing, the system performs distribution feature mapping on the target model data subset and converts it into distribution mapping features that are easier to analyze. These features can intuitively reflect the distribution of BIM electric construction model data and the potential trend of operation and maintenance failure risks.

[0120] Finally, the system combines key data clusters and distribution mapping features to perform state transition analysis to derive a second view of the operation and maintenance failure risk between the first BIM electric power construction model data and the second BIM electric power construction model data. This view is derived after comprehensively considering multiple factors and data features, so it has higher accuracy and reliability.

[0121] Through this embodiment, the data processing system can more deeply explore the potential information in the BIM electric construction model data and provide a more accurate risk assessment for the operation and maintenance management of the power system. This not only helps to improve the safety and stability of the power system, but also provides more effective decision-making support for operation and maintenance personnel. At the same time, this meticulous data processing method also demonstrates the powerful ability and flexibility of the data processing system in processing complex three-dimensional model data. In this way, by introducing more state attribute vectors and more complex data processing processes, the accuracy and comprehensiveness of the operation and maintenance fault risk assessment are improved. The implementation of this technical solution provides a strong guarantee for the safe and stable operation of the power system and promotes the process of digital transformation in the power industry.

[0122] In some more preferred embodiments, the distribution feature mapping processing is performed on the v target model data subsets to obtain v distribution mapping features corresponding to the v target model data subsets, including: obtaining the relative relationship features of each model unit in the first target model data subset among the v target model data subsets; and feature enhancement is performed on the relative relationship features and influence coefficients of each model unit in the first target model data subset to obtain the distribution mapping features of the first target model data subset.

[0123] In this embodiment, the data processing system needs to process a large amount of three-dimensional model data and extract useful information from it. Specifically, the data processing system performs distribution feature mapping processing on the target model data subset to obtain corresponding distribution mapping features.

[0124] First, the data processing system obtains the first of the v target model data subsets, namely the first target model data subset. This subset contains multiple model units, which may be various components in the power system building, such as transformers, switchgear, lines, etc. Each model unit has its unique attributes and position in space.

[0125] Next, the data processing system analyzes the relative relationship characteristics of each model unit in the first target model data subset. The relative relationship characteristics describe the spatial position relationship, connection relationship, and other possible relationships between model units. For example, a transformer may be located between two switchgears, or a line may connect two different devices. These relative relationship characteristics are the key to understanding the architectural structure and working principle of the power system.

[0126] After obtaining the relative relationship features, the data processing system performs feature enhancement on these features and influence coefficients. The influence coefficient reflects the degree of influence of different relative relationships on the performance of the power system. Through feature enhancement, the system can more accurately capture those relative relationships that have a significant impact on the performance of the power system, thereby highlighting their performance in the distribution mapping features.

[0127] Finally, the distribution mapping feature of the first target model data subset after feature enhancement processing is generated. This feature not only contains the relative relationship between model units, but also reflects the degree of influence of these relationships on the performance of the power system. Such distribution mapping features provide strong data support for subsequent power system operation and maintenance management, fault prediction, etc.

[0128] Through this embodiment, the data processing system can more deeply mine the useful information in the three-dimensional model data of the power system building, and provide a more accurate data basis for the stable operation and optimization of the power system. This not only improves the safety and reliability of the power system, but also provides a more effective decision-making basis for operation and maintenance personnel. At the same time, this optimized data processing method also demonstrates the advancement and efficiency of the data processing system in processing complex three-dimensional model data.

[0129] In some independent embodiments, the method further includes: extracting model attributes from a first BIM electric construction model data sample to obtain a first BIM state attribute vector sample, and extracting model attributes from a second BIM electric construction model data sample to obtain a second BIM state attribute vector sample; using a trend simulation processing algorithm to identify operation and maintenance trends for the first BIM state attribute vector sample to obtain trend simulation weights of each first multi-mode three-dimensional component element in the first BIM state attribute vector sample, and using the trend simulation processing algorithm to identify operation and maintenance trends for the second BIM state attribute vector sample to obtain trend simulation weights of each second multi-mode three-dimensional component element in the second BIM state attribute vector sample, the trend simulation weight being used to characterize a confidence coefficient for predicting the operation and maintenance failure risk of the BIM electric construction model data sample; constructing element sampling for the first BIM state attribute vector sample based on the trend simulation weights of each first multi-mode three-dimensional component element to obtain x first multi-mode three-dimensional component elements, and constructing element sampling for the first BIM state attribute vector sample based on the trend simulation weights of each second multi-mode three-dimensional component element. The second BIM state attribute vector sample is used to construct element sampling to obtain x second multi-mode three-dimensional component elements; the x first multi-mode three-dimensional component elements and the x second multi-mode three-dimensional component elements are identified for operation and maintenance trends by using a BIM annotation classification network to obtain first model annotation information corresponding to the x first multi-mode three-dimensional component elements and second model annotation information corresponding to the x second multi-mode three-dimensional component elements, wherein the first model annotation information is used to indicate mining u first multi-mode three-dimensional component elements in the x first multi-mode three-dimensional component elements that have a power line layout intersection state with the x second multi-mode three-dimensional component elements, and the second model annotation information is used to enhance v second multi-mode three-dimensional component elements in the x second multi-mode three-dimensional component elements that have a power line layout intersection state with the x first multi-mode three-dimensional component elements; the trend simulation processing algorithm is debugged according to the trend simulation weights of the respective first multi-mode three-dimensional component elements and the trend simulation weights of the respective second multi-mode three-dimensional component elements, and the BIM annotation classification network is debugged according to the first model annotation information and the second model annotation information.

[0130] In the application scenario of power system building 3D model data processing, the data processing system performs a series of complex operations to optimize the operation and maintenance management of the power building information model (BIM). The following will describe these operation steps and their beneficial effects in detail.

[0131] First, the data processing system extracts model attributes from the first BIM electric construction model data sample and the second BIM electric construction model data sample to obtain the first BIM state attribute vector sample and the second BIM state attribute vector sample respectively. The purpose of this step is to capture the attribute information of each component in the model and provide basic data for subsequent analysis.

[0132] Next, the system uses the trend simulation processing algorithm to identify the operation and maintenance trends of the extracted BIM status attribute vector samples. In this process, the algorithm analyzes the attribute change trend of each multi-mode three-dimensional component element and assigns it a trend simulation weight. This weight is actually a confidence coefficient that characterizes the reliability of the operation and maintenance failure risk prediction for the component element.

[0133] Subsequently, the system samples the component elements in the BIM state attribute vector sample according to these trend simulation weights. The purpose of sampling is to select representative component elements for more in-depth analysis. Through sampling, the system obtains x first multi-mode three-dimensional component elements and x second multi-mode three-dimensional component elements.

[0134] The data processing system then uses the BIM annotation classification network to identify the operation and maintenance trends of these 2x component elements. The purpose of this step is to add annotation information to these component elements to indicate the specific relationship between them. In this example, this annotation information is used to identify the first group of component elements that have a power line layout intersection with the second group of component elements, and to enhance the second group of component elements that have the same intersection with the first group of component elements.

[0135] Finally, the system debugs the trend simulation processing algorithm and BIM annotation classification network based on the previously calculated trend simulation weights and the newly added model annotation information. The purpose of this step is to optimize the performance of these two tools so that they can more accurately predict the operation and maintenance trends of the power building model and identify the relationship between key component elements.

[0136] With this design, firstly, by extracting and analyzing the attribute information in the BIM model, the system can have a more comprehensive understanding of the operation and maintenance status of the power building, thereby timely discovering potential risk points. Secondly, by using the trend simulation processing algorithm and the BIM annotation classification network, the system can accurately predict the operation and maintenance trends of component elements and identify the complex relationships between key components, which is crucial for the stable operation of the power system. Finally, by continuously optimizing the performance of these two tools, the data processing system can provide more efficient and accurate support for the operation and maintenance management of the power system, thereby ensuring the safety and stability of power supply.

[0137] In some other independent embodiments, debugging the trend simulation processing algorithm based on the trend simulation weights of the respective first multimodal three-dimensional component elements and the trend simulation weights of the respective second multimodal three-dimensional component elements includes: constructing a first algorithm debugging error based on the difference between model annotation information used to characterize whether the first multimodal three-dimensional component elements in the first BIM state attribute vector sample have abnormal thermal characteristics and the trend simulation weights of the first multimodal three-dimensional component elements in the first BIM state attribute vector sample; and constructing a second algorithm debugging error based on the difference between model annotation information used to characterize whether the second multimodal three-dimensional component elements in the second BIM state attribute vector sample have abnormal thermal characteristics and the trend simulation weights of the second multimodal three-dimensional component elements in the second BIM state attribute vector sample; and debugging the trend simulation processing algorithm based on the first algorithm debugging error and the second algorithm debugging error.

[0138] Based on this embodiment, the data processing system debugs the trend simulation processing algorithm according to the trend simulation weights of each first multi-mode three-dimensional component element and the trend simulation weights of each second multi-mode three-dimensional component element calculated in the previous step. This debugging process is to improve the accuracy and efficiency of the algorithm, so as to better predict the operation and maintenance trend of the BIM electric construction model data sample.

[0139] For example, the data processing system firstly identifies whether the first multi-mode three-dimensional component element in the first BIM state attribute vector sample has abnormal thermal characteristics based on the model annotation information. Then, the system compares the annotation information with the trend simulation weight of the first multi-mode three-dimensional component element and analyzes the difference between them. This difference reflects the accuracy of the algorithm in predicting the operation and maintenance trend of these component elements. Based on this difference, the data processing system constructs a first algorithm debugging error.

[0140] Similarly, the data processing system will also perform similar processing on the second multi-mode three-dimensional component elements in the second BIM state attribute vector sample. The system compares the model annotation information that characterizes whether these component elements have abnormal thermal characteristics with their trend simulation weights to construct a second algorithm debugging error.

[0141] After constructing the first algorithm debugging error and the second algorithm debugging error, the data processing system uses these two errors to debug the trend simulation processing algorithm. The goal of debugging is to minimize these two errors so that the prediction results of the algorithm are closer to the actual operation and maintenance trend.

[0142] In this way, the data processing system can significantly improve the accuracy and efficiency of the trend simulation processing algorithm. This not only helps to more accurately predict the operation and maintenance trends of the power building model and timely discover potential failure risks, but also provides more reliable support for the stable operation of the power system. At the same time, this debugging method also has good versatility and flexibility, and can be widely used in different types of BIM power construction model data processing, providing strong technical support for the intelligent management of the power system.

[0143] In some other independent embodiments, debugging the BIM annotation classification network based on the first model annotation information and the second model annotation information includes: constructing a third algorithm debugging error based on the difference between the first model annotation information and the prior model annotation information, and constructing a fourth algorithm debugging error based on the difference between the second model annotation information and the prior model annotation information; wherein the prior model annotation information is model annotation information for characterizing whether the power line layout crossing state of the x first multi-mode three-dimensional component elements and the x second multi-mode three-dimensional component elements has abnormal thermal characteristics; and debugging the BIM annotation classification network based on the third algorithm debugging error and the fourth algorithm debugging error.

[0144] Based on this embodiment, the data processing system will debug the BIM annotation classification network according to the first model annotation information and the second model annotation information. This process is intended to improve the network's recognition accuracy of specific annotations in the three-dimensional model of the power system building, thereby optimizing the efficiency and accuracy of operation and maintenance management.

[0145] For example, the data processing system will first compare the first model annotation information with the prior model annotation information. The prior model annotation information is formed based on historical data and expert knowledge, and is used to characterize whether the cross-state of the power line layout of x first multi-mode three-dimensional component elements and x second multi-mode three-dimensional component elements has abnormal thermal characteristics. The purpose of this comparison is to find the difference between the current annotation of the network and the actual situation. This difference reflects the accuracy of the BIM annotation classification network in identifying the characteristics of these elements. Through comparison, the system will construct a third algorithm debugging error, which quantifies the deficiencies of the network in annotating the first set of component elements.

[0146] Similarly, the system will also perform similar comparison processing on the second model annotation information, and build a fourth algorithm debugging error by comparing it with the prior model annotation information. This error reflects the performance gap of the BIM annotation classification network when processing the second group of component elements.

[0147] After constructing the third and fourth algorithm debugging errors, the data processing system will use these two errors to debug the BIM annotation classification network. The debugging process includes adjusting the parameters of the network, optimizing the algorithm, etc., with the aim of minimizing these two errors and improving the accuracy of the network's annotation of component element features.

[0148] Through such a debugging process, the data processing system can significantly improve the performance of the BIM annotation classification network, enabling it to more accurately identify key component elements and their characteristics in the three-dimensional model of the power system building. This not only helps to promptly discover possible problems in the model, such as abnormal thermal characteristics, but also provides more effective support for the stable operation of the power system. In addition, this debugging method can also enhance the adaptability of the data processing system to different scenarios and needs, and improve its timeliness in the operation and maintenance management of the power system.

[0149] Further, Figure 2 Schematic diagram of a data processing system 200 provided in an embodiment of the present application. Figure 2 The data processing system 200 shown includes a processor 210, and the processor 210 can call and run a computer program from a memory 230 to implement the method in the embodiment of the present application.

[0150] Alternatively, if Figure 2 As shown, the data processing system 200 may further include a memory 230. The processor 210 may call and run a computer program from the memory 230 to implement the method in the embodiment of the present application.

[0151] The memory 230 may be a separate device independent of the processor 210 , or may be integrated into the processor 210 .

[0152] Alternatively, if Figure 2 As shown, the data processing system 200 may further include a transceiver 220. The processor 210 may control the transceiver 220 to interact with other devices. Specifically, the transceiver 220 may send information or data to other devices, or receive information or data sent by other devices.

[0153] Optionally, the data processing system 200 may implement the corresponding processes corresponding to the storage engine or components in the storage engine (such as a processing module) or a device deployed with a storage engine in each method of the embodiments of the present application, which will not be described in detail here for the sake of brevity.

[0154] It should be understood that the processor of the embodiment of the present application may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method embodiment can be completed by an integrated logic circuit of hardware in the processor or an instruction in the form of software. The above processor may be a general-purpose processor, a digital signal processor (Digital Signal Processor, DSP), an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), a field programmable gate array (Field Programmable Gate Array, FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The steps of the method disclosed in the embodiment of the present application can be directly embodied as a hardware decoding processor to be executed, or the hardware and software modules in the decoding processor are combined and executed. The software module may be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.

[0155] It can be understood that the memory in the embodiments of the present application can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories. Among them, the non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory can be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct RAM bus random access memory (DR RAM). It should be noted that the memory of the systems and methods described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0156] It should be understood that the above-mentioned memory is exemplary but not restrictive. For example, the memory in the embodiments of the present application may also be static random access memory (static RAM, SRAM), dynamic random access memory (dynamic RAM, DRAM), synchronous dynamic random access memory (synchronous DRAM, SDRAM), double data rate synchronous dynamic random access memory (double data rate SDRAM, DDR SDRAM), enhanced synchronous dynamic random access memory (enhanced SDRAM, ESDRAM), synchronous link dynamic random access memory (synch link DRAM, SLDRAM) and direct memory bus random access memory (Direct Rambus RAM, DR RAM), etc. That is to say, the memory in the embodiments of the present application is intended to include but not limited to these and any other suitable types of memory.

[0157] Based on the above, a computer-readable storage medium is provided, on which a computer program is stored, and the computer program implements the above method when running.

[0158] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations.

Claims

1. A method for processing three-dimensional model data of a power system building using BIM, characterized in that: The method comprises at least the following steps: Extracting model attributes from the first BIM electric construction model data to obtain a first BIM state attribute vector, and extracting model attributes from the second BIM electric construction model data to obtain a second BIM state attribute vector; Performing operation and maintenance trend identification on the first BIM state attribute vector to obtain trend simulation weights of each first multi-mode three-dimensional component element in the first BIM state attribute vector, and performing operation and maintenance trend identification on the second BIM state attribute vector to obtain trend simulation weights of each second multi-mode three-dimensional component element in the second BIM state attribute vector, wherein the trend simulation weights are used to characterize confidence coefficients for operation and maintenance fault risk prediction of BIM electric construction model data; According to the trend simulation weights of the respective first multimodal three-dimensional component elements, constructing element sampling on the first BIM state attribute vector to obtain x first multimodal three-dimensional component elements; and according to the trend simulation weights of the respective second multimodal three-dimensional component elements, constructing element sampling on the second BIM state attribute vector to obtain x second multimodal three-dimensional component elements; Performing operation and maintenance trend identification on the x first multi-mode three-dimensional component elements and the x second multi-mode three-dimensional component elements, obtaining first model annotation information corresponding to the x first multi-mode three-dimensional component elements and second model annotation information corresponding to the x second multi-mode three-dimensional component elements, wherein the first model annotation information is used to indicate mining u first multi-mode three-dimensional component elements in the x first multi-mode three-dimensional component elements that have a power line layout intersection state with the x second multi-mode three-dimensional component elements, and the second model annotation information is used to enhance v second multi-mode three-dimensional component elements in the x second multi-mode three-dimensional component elements that have a power line layout intersection state with the x first multi-mode three-dimensional component elements; An operation and maintenance failure risk prediction is performed in combination with the u first multi-mode three-dimensional component elements and the v second multi-mode three-dimensional component elements to obtain a first operation and maintenance failure risk view between the first BIM electric power construction model data and the second BIM electric power construction model data.

2. The method according to claim 1, characterized in that The step of constructing element sampling on the first BIM state attribute vector according to the trend simulation weights of the first multimodal three-dimensional component elements to obtain x first multimodal three-dimensional component elements, and constructing element sampling on the second BIM state attribute vector according to the trend simulation weights of the second multimodal three-dimensional component elements to obtain x second multimodal three-dimensional component elements includes: Determine x according to preset parameters; According to the trend simulation weights of each first multimodal three-dimensional component element and in accordance with the descending arrangement rule of the trend simulation weights, the first x first multimodal three-dimensional component elements in the first BIM state attribute vector are determined as the x first multimodal three-dimensional component elements; and according to the trend simulation weights of each second multimodal three-dimensional component element and in accordance with the descending arrangement rule of the trend simulation weights, the first x second multimodal three-dimensional component elements in the second BIM state attribute vector are determined as the x second multimodal three-dimensional component elements.

3. The method according to claim 2, characterized in that The determining x according to the preset parameters includes: determining x according to the preset parameters and the characteristic dimension of the first BIM state attribute vector, or determining x according to the preset parameters and the characteristic dimension of the second BIM state attribute vector.

4. The method according to any one of claims 1 to 3, characterized in that The performing operation and maintenance trend identification on the x first multimodal three-dimensional component elements and the x second multimodal three-dimensional component elements to obtain first model annotation information corresponding to the x first multimodal three-dimensional component elements and second model annotation information corresponding to the x second multimodal three-dimensional component elements includes: Determine first basic model annotation information corresponding to the x first multi-mode three-dimensional component elements, and determine second basic model annotation information corresponding to the x second multi-mode three-dimensional component elements; According to the x first multimodal three-dimensional component elements, the x second multimodal three-dimensional component elements, the first basic model annotation information, and the second basic model annotation information, the x first multimodal three-dimensional component elements are adjusted to obtain x first adjusted three-dimensional component elements; and according to the x first multimodal three-dimensional component elements, the x second multimodal three-dimensional component elements, the first basic model annotation information, and the second basic model annotation information, the x second multimodal three-dimensional component elements are adjusted to obtain x second adjusted three-dimensional component elements; According to the x first adjusted three-dimensional component elements, the first basic model annotation information is adjusted to obtain first adjusted model annotation information, and according to the x second adjusted three-dimensional component elements, the second basic model annotation information is adjusted to obtain second adjusted model annotation information; The first model annotation information is determined according to the first adjustment model annotation information, and the second model annotation information is determined according to the second adjustment model annotation information.

5. The method according to claim 4, characterized in that Before extracting the model attributes of the first BIM electrical construction model data to obtain the first BIM state attribute vector, and extracting the model attributes of the second BIM electrical construction model data to obtain the second BIM state attribute vector, the method further includes: In the case where the feature dimension of the first BIM electrical construction model data and the feature dimension of the second BIM electrical construction model data do not match, converting the first BIM electrical construction model data and the second BIM electrical construction model data to the same feature dimension; The determining of the first basic model annotation information corresponding to the x first multi-mode three-dimensional component elements, and the determining of the second basic model annotation information corresponding to the x second multi-mode three-dimensional component elements, comprises: Determine the first basic model annotation information according to the feature members augmented during the dimensional conversion of the first BIM electrical construction model data, wherein the model query vector corresponding to the quantized value of the feature members augmented in the x first multi-mode three-dimensional component elements in the first basic model annotation information is a annotation keyword; Based on the feature members augmented during the dimensional conversion of the second BIM electrical construction model data, the second basic model annotation information is determined, in which the model query vector corresponding to the quantized value of the feature members augmented in the x second multi-modal three-dimensional component elements is the annotation keyword.

6. The method according to claim 4, characterized in that The method of adjusting the x first multimodal three-dimensional component elements according to the x first multimodal three-dimensional component elements, the x second multimodal three-dimensional component elements, the first basic model annotation information, and the second basic model annotation information to obtain x first adjusted three-dimensional component elements, and adjusting the x second multimodal three-dimensional component elements according to the x first multimodal three-dimensional component elements, the x second multimodal three-dimensional component elements, the first basic model annotation information, and the second basic model annotation information to obtain x second adjusted three-dimensional component elements includes: Using the first basic model annotation information and the x first multimodal three-dimensional component elements, x first spatial structure relationship vectors are obtained; and using the second basic model annotation information and the x second multimodal three-dimensional component elements, x second spatial structure relationship vectors are obtained; The x first spatial structure relationship vectors are subjected to intra-class feature focusing processing by using the first intra-class feature focusing branch, and the x second spatial structure relationship vectors and the x processing results generated by the first intra-class feature focusing branch are subjected to the first inter-class feature focusing branch to obtain the x first adjusted three-dimensional component elements, and the x second spatial structure relationship vectors are subjected to intra-class feature focusing processing by using the second intra-class feature focusing branch, and the x first spatial structure relationship vectors and the x processing results generated by the second intra-class feature focusing branch are subjected to the second inter-class feature focusing branch to obtain the x second adjusted three-dimensional component elements.

7. The method according to claim 4, characterized in that The first basic model annotation information is adjusted according to the x first adjusted three-dimensional component elements to obtain first adjusted model annotation information, and the second basic model annotation information is adjusted according to the x second adjusted three-dimensional component elements to obtain second adjusted model annotation information, including: Performing operation and maintenance trend identification on the x first adjusted three-dimensional component elements to obtain two discrimination weights corresponding to each of the x first adjusted three-dimensional component elements, and performing operation and maintenance trend identification on the x second adjusted three-dimensional component elements to obtain two discrimination weights corresponding to each of the x second adjusted three-dimensional component elements, the two discrimination weights being used to characterize the discrimination weight of the element vector belonging to the cross state of the power line layout and the discrimination weight of the element vector not belonging to the cross state of the power line layout; Determining x first key-value relationship features corresponding to the x first adjusted three-dimensional component elements based on the two discrimination weights corresponding to the respective first adjusted three-dimensional component elements, and determining x second key-value relationship features corresponding to the x second adjusted three-dimensional component elements based on the two discrimination weights corresponding to the respective second adjusted three-dimensional component elements, wherein the first key-value relationship feature is used to characterize authentication information of an attention index to which a maximum discrimination weight belongs among the two discrimination weights corresponding to the first adjusted three-dimensional component elements, and the second key-value relationship feature is used to characterize authentication information of an attention index to which a maximum discrimination weight belongs among the two discrimination weights corresponding to the second adjusted three-dimensional component elements; Performing discriminant analysis on the x first key-value relationship features to obtain x first discriminant weights, and performing discriminant analysis on the x second key-value relationship features to obtain x second discriminant weights; The first adjustment model annotation information is obtained by using the first basic model annotation information and the x first discrimination weights, and the second adjustment model annotation information is obtained by using the second basic model annotation information and the x second discrimination weights.

8. The method according to claim 4, characterized in that The first basic model annotation information is adjusted according to the x first adjusted three-dimensional component elements to obtain first adjusted model annotation information, and the second basic model annotation information is adjusted according to the x second adjusted three-dimensional component elements to obtain second adjusted model annotation information, including: According to the x first adjusted three-dimensional component elements, the x second adjusted three-dimensional component elements, the first adjustment model annotation information, and the second adjustment model annotation information, the x first adjusted three-dimensional component elements are adjusted to obtain x third adjusted three-dimensional component elements; and according to the x first adjusted three-dimensional component elements, the x second adjusted three-dimensional component elements, the first adjustment model annotation information, and the second adjustment model annotation information, the x second adjusted three-dimensional component elements are adjusted to obtain x fourth adjusted three-dimensional component elements; According to the x third adjusted three-dimensional component elements, adjusting the first adjustment model annotation information to obtain third adjustment model annotation information, and according to the x fourth adjusted three-dimensional component elements, adjusting the second adjustment model annotation information to obtain fourth adjustment model annotation information; The first model annotation information is determined based on the third adjustment model annotation information, and the second model annotation information is determined based on the fourth adjustment model annotation information, the first model annotation information is the adjustment model annotation information obtained when the number of cyclic adjustments of the first basic model annotation information is not less than the set number of times, and the second model annotation information is the adjustment model annotation information obtained when the number of cyclic adjustments of the second basic model annotation information is not less than the set number of times.

9. The method according to any one of claims 1 to 3, characterized in that The extracting model attributes of the first BIM electric construction model data to obtain a first BIM state attribute vector, and extracting model attributes of the second BIM electric construction model data to obtain a second BIM state attribute vector, include: Performing model attribute extraction on the first BIM electrical construction model data to obtain the first BIM state attribute vector and the third BIM state attribute vector, and performing model attribute extraction on the second BIM electrical construction model data to obtain the second BIM state attribute vector and the fourth BIM state attribute vector, wherein the feature dimension of the first BIM state attribute vector is different from the feature dimension of the third BIM state attribute vector, and the feature dimension of the second BIM state attribute vector is different from the feature dimension of the fourth BIM state attribute vector; Wherein, after obtaining the first operation and maintenance failure risk view between the first BIM electric power construction model data and the second BIM electric power construction model data, the method further includes: According to the first operation and maintenance failure risk viewpoint, the characteristic dimension of the first BIM state attribute vector, the characteristic dimension of the third BIM state attribute vector and the third BIM state attribute vector, construct element sampling for the third BIM state attribute vector to obtain u BIM electrical construction model data subsets corresponding to the u first multi-mode three-dimensional component elements, and according to the first operation and maintenance failure risk viewpoint, the characteristic dimension of the second BIM state attribute vector, the characteristic dimension of the fourth BIM state attribute vector and the fourth BIM state attribute vector, construct element sampling for the fourth BIM state attribute vector to obtain v BIM electrical construction model data subsets corresponding to the v second multi-mode three-dimensional component elements; adjusting the u BIM electric power construction model data subsets according to the u BIM electric power construction model data subsets and the v BIM electric power construction model data subsets to obtain u model data adjustment subsets, and adjusting the v BIM electric power construction model data subsets according to the u BIM electric power construction model data subsets and the v BIM electric power construction model data subsets to obtain v model data adjustment subsets; Detecting the key data clusters of the u model data adjustment subsets to obtain u key data clusters, and performing fine-grained optimization processing on the v model data adjustment subsets according to the model fine-grained comparison result between the third BIM electric power construction model data and the fourth BIM electric power construction model data to obtain v target model data subsets; Performing distribution feature mapping processing on the v target model data subsets to obtain v distribution mapping features corresponding to the v target model data subsets; Combining the u key data clusters and the v distribution mapping features to perform state transition analysis, a second operation and maintenance fault risk view between the first BIM electric construction model data and the second BIM electric construction model data; The performing distribution feature mapping processing on the v target model data subsets to obtain v distribution mapping features corresponding to the v target model data subsets includes: Acquire relative relationship features of each model unit in a first target model data subset among the v target model data subsets; The relative relationship characteristics and influence coefficients of each model unit in the first target model data subset are enhanced to obtain the distribution mapping characteristics of the first target model data subset.

10. A data processing system, characterized in that: The method comprises at least one processor and a memory; the memory stores computer-executable instructions; the at least one processor executes the computer-executable instructions stored in the memory, so that the at least one processor executes the method according to any one of claims 1 to 9.

Images