CN116028886A - BIM-based data processing method, system and cloud platform - Google Patents

Abstract

The invention provides a BIM-based data processing method, a BIM-based data processing system and a cloud platform, and relates to the technical field of data processing and BIM. According to the building information model key data description vector, a corresponding first sub-model key data description vector and a corresponding first component key data description vector are mined; performing key data supplementing operation on the first sub-model key data description vector based on the first component key data description vector to form a second sub-model key data description vector, and performing key data supplementing operation on the first component key data description vector based on the first sub-model key data description vector to form a second component key data description vector; and carrying out sub-model anomaly identification operation based on the second sub-model key data description vector and the second component key data description vector. Based on the above, the reliability of the sub-model anomaly identification can be improved to some extent.

Description

BIM-based data processing method, system and cloud platform

Technical Field

The invention relates to the technical field of data processing and BIM, in particular to a BIM-based data processing method, a BIM-based data processing system and a cloud platform.

Background

The building information model (Building Information Modeling) is constructed by using various relevant information data of a building engineering project as a model basis, and simulating real information of a building by digital information. The method has the eight characteristics of information completeness, information relevance, information consistency, visualization, coordination, simulation, optimality and diagonability. Therefore, the application value of the building information model is higher, so that the application scene is more. Based on some requirements, it is necessary to perform anomaly analysis on the building information model to determine whether the included sub-model is abnormal or the degree of the anomaly, however, in the prior art, there is a problem that the reliability of anomaly analysis is not high.

Disclosure of Invention

Accordingly, the present invention is directed to a method, a system, and a cloud platform for processing data based on BIM, so as to improve reliability of sub-model anomaly identification to a certain extent.

In order to achieve the above purpose, the embodiment of the present invention adopts the following technical scheme:

a BIM-based data processing method, comprising:

performing key data mining operation on building information model data to be processed to mine building information model key data description vectors corresponding to the building information model data to be processed, wherein the building information model data to be processed comprises building information sub-models to be processed;

According to the building information model key data description vector, a first sub-model key data description vector corresponding to the building information sub-model to be processed is dug out, and a first component key data description vector corresponding to a sub-model component is dug out, wherein the building information sub-model to be processed comprises the sub-model component;

performing key data supplementing operation on the first sub-model key data description vector based on the first component key data description vector to form a corresponding second sub-model key data description vector, and performing key data supplementing operation on the first component key data description vector based on the first sub-model key data description vector to form a corresponding second component key data description vector;

and carrying out sub-model anomaly identification operation on the to-be-processed building information sub-model included in the to-be-processed building information model data based on the second sub-model key data description vector and the second component key data description vector.

In some preferred embodiments, in the above BIM-based data processing method, the step of mining a first sub-model key data description vector corresponding to the to-be-processed building information sub-model according to the building information model key data description vector, and mining a first component key data description vector corresponding to the included sub-model component includes:

Carrying out key data analysis processing on the building information model key data description vector so as to analyze the sub-model suspected distribution position of the building information sub-model to be processed in the building information model key data description vector;

the construction information model key data description vector corresponding to the sub-model suspected distribution position is excavated;

and carrying out key data analysis processing on the key data description vector of the local building information model so as to analyze a first sub-model key data description vector corresponding to the sub-model of the building information to be processed and a first component key data description vector corresponding to the included sub-model component.

In some preferred embodiments, in the above BIM-based data processing method, the step of mining a first sub-model key data description vector corresponding to the to-be-processed building information sub-model according to the building information model key data description vector, and mining a first component key data description vector corresponding to the included sub-model component includes:

utilizing a key data mining sub-network included in the target building information model anomaly identification neural network to conduct key data mining operation on the building information model key data description vector, and outputting a local building information model key data description vector corresponding to the building information sub-model to be processed;

Loading the key data description vector of the local building information model to be loaded into a sub-model analysis sub-network included in the target building information model anomaly identification neural network, and analyzing a first sub-model key data description vector corresponding to the building information sub-model to be processed by using the sub-model analysis sub-network;

and loading the key data description vector of the local building information model to load the component analysis sub-network included in the target building information model anomaly identification neural network, and analyzing the first component key data description vector corresponding to the sub-model component by using the component analysis sub-network.

In some preferred embodiments, in the above BIM-based data processing method, the step of performing a sub-model anomaly identification operation on the sub-model of the building information to be processed included in the building information to be processed model data based on the second sub-model key data description vector and the second component key data description vector includes:

performing key data analysis operation on the second sub-model key data description vector by utilizing a sub-model analysis sub-network included in the target building information model anomaly identification neural network to analyze sub-model type description data corresponding to the building information sub-model to be processed and sub-model distribution data corresponding to the building information sub-model to be processed, and performing combined processing on the sub-model type description data and the sub-model distribution data to form sub-model description data corresponding to the building information sub-model to be processed;

Performing key data analysis operation on the second component key data description vector according to the component analysis sub-network included in the target building information model abnormality identification neural network to analyze component identification description data corresponding to each of a second number of component abnormality description data, and performing combination processing on the second number of component identification description data to form component description data corresponding to the sub-model component, wherein the second number of component abnormality description data belongs to an analysis result corresponding to the component analysis sub-network;

and carrying out sub-model abnormality recognition operation on the to-be-processed building information sub-model in the to-be-processed building information model data based on the sub-model description data and the component description data.

In some preferred embodiments, in the above BIM-based data processing method, the sub-model member has a third number; the step of combining the second number of component identification description data to form component description data corresponding to the sub-model component includes:

classifying the component identification description data corresponding to the second number of component abnormality description data based on the sub-model components corresponding to the second number of component abnormality description data to form a corresponding third number of component classification clusters, wherein each component classification cluster corresponds to one sub-model component; respectively carrying out data aggregation operation on the component identification description data in each component classification cluster to form component description data corresponding to each sub-model component;

The step of performing sub-model anomaly identification operation on the to-be-processed building information sub-model in the to-be-processed building information model data based on the sub-model description data and the component description data includes:

and carrying out sub-model anomaly identification operation on the to-be-processed building information sub-model in the to-be-processed building information model data based on the sub-model description data and the component description data corresponding to each sub-model component.

In some preferred embodiments, in the above BIM-based data processing method, the step of performing a sub-model anomaly identification operation on the sub-model of the building information to be processed in the building information to be processed model data based on the sub-model description data and the component description data corresponding to each of the sub-model components includes:

determining component abnormality identification data corresponding to the sub-model description data;

marking the to-be-processed building information sub-model in the to-be-processed building information model data as an abnormal building information sub-model under the condition that component description data corresponding to the component abnormality identification data exists in component description data corresponding to each sub-model component; and marking the to-be-processed building information sub-model in the to-be-processed building information model data as a non-abnormal building information sub-model under the condition that the component description data corresponding to each sub-model component does not have the component description data corresponding to the component abnormality identification data.

In some preferred embodiments, in the above BIM-based data processing method, the second number of component identification description data includes component distribution data and an abnormality evaluation parameter corresponding to each component abnormality description data; the step of performing sub-model anomaly identification operation on the sub-model of the building information to be processed in the building information model data to be processed based on the sub-model description data and the component description data corresponding to each sub-model component includes:

extracting a third number of component importance characterization coefficients corresponding to the sub-model description data, wherein each component importance characterization coefficient corresponds to one sub-model component;

extracting component identification abnormality description data corresponding to each sub-model component, and determining abnormality evaluation parameters corresponding to each component identification abnormality description data in the component description data corresponding to each sub-model component;

according to the third number of component importance characterization coefficients, carrying out fusion operation on abnormality evaluation parameters corresponding to each component identification abnormality description data so as to output target abnormality evaluation parameters corresponding to the building information model data to be processed;

Marking the building information submodel to be processed in the building information model data to be an abnormal building information submodel under the condition that the target abnormality evaluation parameter exceeds a pre-configured reference abnormality evaluation parameter; and marking the to-be-processed building information sub-model in the to-be-processed building information model data as a non-abnormal building information sub-model under the condition that the target abnormality evaluation parameter does not exceed the reference abnormality evaluation parameter.

In some preferred embodiments, in the above BIM-based data processing method, the BIM-based data processing method further includes:

extracting exemplary building information model data, exemplary sub-model real data of an exemplary building information sub-model included in the exemplary building information model data, exemplary component real data corresponding to an exemplary sub-model component, and sub-model abnormality real data of the exemplary building information sub-model in the exemplary building information model data, the exemplary building information sub-model including the exemplary sub-model component;

performing key data mining operation on the exemplary building information model data, and mining out an exemplary building information model key data description vector corresponding to the exemplary building information model data;

Mining the exemplary building information model key data description vector according to a first sub-model analysis sub-network included in the first building information model anomaly identification neural network to output an exemplary first sub-model key data description vector of the exemplary building information sub-model, and mining the exemplary building information model key data description vector according to a first component analysis sub-network included in the first building information model anomaly identification neural network to output an exemplary first component key data description vector corresponding to the exemplary sub-model component;

performing a critical data supplementing operation on the exemplary first sub-model critical data description vector based on the exemplary first component critical data description vector by using the first sub-model analysis sub-network to form a corresponding exemplary second sub-model critical data description vector, and performing a critical data supplementing operation on the exemplary first component critical data description vector based on the exemplary first sub-model critical data description vector by using the first component analysis sub-network to form a corresponding exemplary second component critical data description vector;

Analyzing the exemplary sub-model description data corresponding to the exemplary building information sub-model based on the exemplary second sub-model key data description vector, and analyzing the exemplary component description data corresponding to the exemplary sub-model component based on the exemplary second component key data description vector;

analyzing and outputting sub-model abnormality analysis data corresponding to an exemplary building information sub-model in the exemplary building information model data based on the exemplary sub-model description data and the exemplary component description data;

and performing network optimization on the first building information model anomaly identification neural network based on the example sub-model real data, the example component real data, the sub-model anomaly real data, the example sub-model description data, the example component description data and the sub-model anomaly analysis data to form a target building information model anomaly identification neural network.

The embodiment of the invention also provides a BIM-based data processing system, which comprises:

the key data mining module is used for carrying out key data mining operation on building information model data to be processed so as to mine building information model key data description vectors corresponding to the building information model data to be processed, wherein the building information model data to be processed comprises a building information sub-model to be processed;

The description vector mining module is used for mining a first sub-model key data description vector corresponding to the building information sub-model to be processed according to the building information model key data description vector, mining a first component key data description vector corresponding to a included sub-model component, and the building information sub-model to be processed comprises the sub-model component;

a description vector supplementing module, configured to perform a key data supplementing operation on the first sub-model key data description vector based on the first component key data description vector to form a corresponding second sub-model key data description vector, and perform a key data supplementing operation on the first component key data description vector based on the first sub-model key data description vector to form a corresponding second component key data description vector;

and the sub-model abnormality recognition module is used for carrying out sub-model abnormality recognition operation on the to-be-processed building information sub-model included in the to-be-processed building information model data based on the second sub-model key data description vector and the second component key data description vector.

The embodiment of the invention also provides a BIM-based data processing cloud platform, which comprises a processor and a memory, wherein the memory is used for storing a computer program, and the processor is used for executing the computer program so as to realize the BIM-based data processing method.

The BIM-based data processing method, system and cloud platform provided by the embodiment of the invention can be used for mining the key data description vector of the building information model; mining corresponding first sub-model key data description vectors and first component key data description vectors according to the building information model key data description vectors; performing key data supplementing operation on the first sub-model key data description vector based on the first component key data description vector to form a second sub-model key data description vector, and performing key data supplementing operation on the first component key data description vector based on the first sub-model key data description vector to form a second component key data description vector; and carrying out sub-model anomaly identification operation based on the second sub-model key data description vector and the second component key data description vector. Based on the above, the basis for performing the operation of identifying the abnormal sub-model includes not only the second sub-model key data description vector but also the second component key data description vector, so that the basis is more sufficient, and the second sub-model key data description vector and the second component key data description vector are obtained based on the corresponding complementary operation of the key data, so that the information of the second sub-model key data description vector and the second component key data description vector is more sufficient, thereby improving the reliability of identifying the abnormal sub-model to a certain extent, and further improving the defects of the prior art.

In order to make the above objects, features and advantages of the present invention more comprehensible, preferred embodiments accompanied with figures are described in detail below.

Drawings

FIG. 1 is a block diagram of a BIM-based data processing system according to an embodiment of the present invention.

Fig. 2 is a flowchart illustrating steps included in a BIM-based data processing method according to an embodiment of the present invention.

FIG. 3 is a schematic diagram of modules included in a BIM-based data processing system according to an embodiment of the present invention.

Detailed Description

For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are only some embodiments of the present invention, but not all embodiments of the present invention. The components of the embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations.

Thus, the following detailed description of the embodiments of the invention, as presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

As shown in fig. 1, an embodiment of the present invention provides a BIM-based data processing cloud platform. Wherein, the BIM-based data processing cloud platform may include a memory and a processor.

In detail, the memory and the processor are electrically connected directly or indirectly to realize transmission or interaction of data. For example, electrical connection may be made to each other via one or more communication buses or signal lines. The memory may store at least one software functional module (computer program) that may exist in the form of software or firmware. The processor may be configured to execute an executable computer program stored in the memory, thereby implementing the BIM-based data processing method provided by the embodiment of the present invention.

It is to be appreciated that in some embodiments, the Memory may be, but is not limited to, random access Memory (Random Access Memory, RAM), read Only Memory (ROM), programmable Read Only Memory (Programmable Read-Only Memory, PROM), erasable Read Only Memory (Erasable Programmable Read-Only Memory, EPROM), electrically erasable Read Only Memory (Electric Erasable Programmable Read-Only Memory, EEPROM), and the like. The processor may be a general purpose processor including a central processing unit (Central Processing Unit, CPU), a network processor (Network Processor, NP), a System on Chip (SoC), etc.; but also Digital Signal Processors (DSPs), application Specific Integrated Circuits (ASICs), field Programmable Gate Arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

It will be appreciated that in some embodiments, the BIM-based data processing cloud platform may be a server with data processing capabilities.

With reference to fig. 2, the embodiment of the invention further provides a data processing method based on BIM, which can be applied to the data processing cloud platform based on BIM. The method steps defined by the flow related to the BIM-based data processing method can be realized by the BIM-based data processing cloud platform. The specific flow shown in fig. 2 will be described in detail.

And step S110, carrying out key data mining operation on the building information model data to be processed, and mining out building information model key data description vectors corresponding to the building information model data to be processed.

In the embodiment of the invention, the BIM-based data processing cloud platform can perform key data mining operation on the building information model data to be processed so as to mine out the building information model key data description vector corresponding to the building information model data to be processed. The building information model data to be processed comprises a building information sub-model to be processed (i.e. an object of anomaly identification; illustratively, performing a key data mining operation on the building information model data to be processed may mean performing feature space mapping processing on the building information model data to form a corresponding building information model key data description vector, i.e. representing discrete data by using continuous vectors).

Step S120, according to the description vector of the key data of the building information model, a description vector of the key data of the first sub-model corresponding to the sub-model of the building information to be processed is mined, and a description vector of the key data of the first component corresponding to the included sub-model component is mined.

In the embodiment of the invention, the BIM-based data processing cloud platform can mine a first sub-model key data description vector corresponding to the to-be-processed building information sub-model according to the building information model key data description vector, and mine a first component key data description vector corresponding to the included sub-model component. The building information sub-model to be processed comprises the sub-model component (that is, the sub-model component belongs to the constituent part of the building information sub-model to be processed).

Step S130, performing a key data supplementing operation on the first sub-model key data description vector based on the first component key data description vector to form a corresponding second sub-model key data description vector, and performing a key data supplementing operation on the first component key data description vector based on the first sub-model key data description vector to form a corresponding second component key data description vector.

In the embodiment of the present invention, the BIM-based data processing cloud platform may perform a key data supplementing operation on the first sub-model key data description vector based on the first component key data description vector to form a corresponding second sub-model key data description vector, and perform a key data supplementing operation on the first component key data description vector based on the first sub-model key data description vector to form a corresponding second component key data description vector.

And step S140, performing sub-model anomaly identification operation on the to-be-processed building information sub-model included in the to-be-processed building information model data based on the second sub-model key data description vector and the second component key data description vector.

In the embodiment of the invention, the BIM-based data processing cloud platform may perform sub-model anomaly identification operation on the to-be-processed building information sub-model included in the to-be-processed building information model data based on the second sub-model key data description vector and the second component key data description vector (to obtain corresponding anomaly identification data; for example, may perform sub-model anomaly identification operation on the second sub-model key data description vector and the second component key data description vector by using a target building information model anomaly identification neural network).

Based on the above, the basis for performing the operation of identifying the abnormal sub-model includes not only the second sub-model key data description vector but also the second component key data description vector, so that the basis is more sufficient, and the second sub-model key data description vector and the second component key data description vector are obtained based on the corresponding complementary operation of the key data, so that the information of the second sub-model key data description vector and the second component key data description vector is more sufficient, thereby improving the reliability of identifying the abnormal sub-model to a certain extent, and further improving the defects of the prior art.

It may be appreciated that, in some embodiments, the step of mining the first sub-model key data description vector corresponding to the to-be-processed building information sub-model according to the building information model key data description vector and mining the first component key data description vector corresponding to the included sub-model component may further include the following specific implementation contents:

carrying out key data analysis processing on the building information model key data description vector so as to analyze the sub-model suspected distribution position of the building information sub-model to be processed in the building information model key data description vector;

In the building information model key data description vector, a local building information model key data description vector corresponding to the sub-model suspected distribution position is dug (that is, vector parameters corresponding to the sub-model suspected distribution position in the building information model key data description vector can be extracted to form the local building information model key data description vector in a combined mode;

the local building information model key data description vector is subjected to key data analysis processing to analyze a first sub-model key data description vector corresponding to the building information sub-model to be processed and a first component key data description vector corresponding to a sub-model component included (illustratively, on one hand, in a sub-model analysis sub-network included in a target building information model abnormality identification neural network, the local building information model key data description vector can be subjected to key information decimation by using a plurality of filter units, and the first sub-model key data description vector corresponding to the building information sub-model to be processed is output, on the other hand, the local building information model key data description vector can be loaded to a component analysis sub-network included in the target building information model abnormality identification neural network, the component analysis sub-network can be utilized to analyze the first component key data description vector corresponding to the sub-model component, specifically, the local building information model key data description vector can be subjected to key information decimation by using a plurality of filter units in the component analysis sub-network included in the target building information model abnormality identification neural network, the first component key data description vector corresponding to the sub-model to be processed is output, the component analysis sub-model key data description vector corresponding to the local building information model is extracted, and the component analysis sub-model key data description vector is subjected to be used for the global analysis sub-network operation of the component description vector, therefore, the integration of global information and local information is realized, key data of the building information model data to be processed are comprehensively mined, and the reliability of subsequent processing is improved.

It may be appreciated that, in some embodiments, the step of performing the key data analysis processing on the building information model key data description vector to analyze the sub-model suspected distribution position of the to-be-processed building information sub-model in the building information model key data description vector may further include the following specific implementation matters:

performing a key data mining operation on the building information model key data description vector to analyze and output a first number of sub-model possible distribution positions corresponding to the building information model key data description vector, and analyzing and outputting sub-model possibility evaluation parameters corresponding to each of the first number of sub-model possible distribution positions (illustratively, the possibility that a to-be-processed building information sub-model may appear in each of the sub-model possible distribution positions may be identified, so that the possibility that the to-be-processed building information sub-model appears in the first number of sub-model possible distribution positions may be marked as sub-model possibility evaluation parameters corresponding to each of the first number of sub-model possible distribution positions;

Marking the sub-model possible distribution positions with the maximum corresponding sub-model possible evaluation parameters in the first number of sub-model possible distribution positions to form a first sub-model suspected distribution position; and performing a distribution position update operation on the first sub-model suspected distribution position to form a corresponding sub-model suspected distribution position (for example, a specific implementation manner of the distribution position update operation may be to correct the first sub-model suspected distribution position according to pre-configured reference data, where the reference data may be information such as a sub-model size, a sub-model shape, and the like configured in advance according to information such as experience, and the specific implementation manner is not limited).

It may be appreciated that, in some embodiments, the step of performing a key data supplementing operation on the first sub-model key data description vector based on the first component key data description vector to form a corresponding second sub-model key data description vector, and performing a key data supplementing operation on the first component key data description vector based on the first sub-model key data description vector to form a corresponding second component key data description vector may further include the following specific implementation matters:

Performing key data description vector comparison operation on the first sub-model key data description vector and the first component key data description vector to form description vector correlation data between the first sub-model key data description vector and the first component key data description vector;

mining a corresponding sub-model supplementary key data description vector from the first component key data description vector according to the description vector correlation data, performing key data supplementary operation on the first sub-model key data description vector according to the sub-model supplementary key data description vector to form a second sub-model key data description vector corresponding to the first sub-model key data description vector (illustratively, a rank vector parameter exchanging operation can be performed on the first component key data description vector first, then a multiplication operation can be performed on a result corresponding to the exchanging operation and the description vector correlation data to obtain the sub-model supplementary key data description vector, and then the sub-model supplementary key data description vector and the first sub-model key data description vector can be overlapped to form the second sub-model key data description vector);

And mining a corresponding component supplementing key data description vector from the first sub-model key data description vector according to the description vector correlation data, and performing key data supplementing operation on the first component key data description vector according to the component supplementing key data description vector to form a second component key data description vector corresponding to the first component key data description vector (illustratively, a permutation operation of row and column vector parameters can be performed on the description vector correlation data first, then, a multiplication operation can be performed on a result corresponding to the permutation operation and the first sub-model key data description vector to obtain the component supplementing key data description vector, and then, key data superposition can be performed on the component supplementing key data description vector and the first component key data description vector to form the second component key data description vector).

It may be appreciated that, in some embodiments, the step of performing the key data description vector comparison operation on the first sub-model key data description vector and the first component key data description vector to form description vector correlation data between the first sub-model key data description vector and the first component key data description vector may further include the following specific implementation matters:

The method comprises the steps of respectively carrying out vector dimension integration operation on the first sub-model key data description vector and the first component key data description vector to form an integrated first sub-model key data description vector corresponding to the first sub-model key data description vector and an integrated first component key data description vector corresponding to the first component key data description vector (illustratively, the vector dimension integration operation can be stretching or compressing vector dimensions so that the vector dimension of the integrated first sub-model key data description vector is equal to the vector dimension of the integrated first component key data description vector, thereby facilitating subsequent processing, and in other examples, the component-related key data description vector can be firstly mined from the first sub-model key data description vector, and the sub-model-related key data description vector can be mined from the first component key data description vector;

And multiplying the integrated first sub-model key data description vector and the integrated first component key data description vector to output description vector correlation data between the first sub-model key data description vector and the first component key data description vector.

It may be appreciated that, in some embodiments, the step of mining the first sub-model key data description vector corresponding to the to-be-processed building information sub-model according to the building information model key data description vector and mining the first component key data description vector corresponding to the included sub-model component may further include the following specific implementation contents:

performing key data mining operation on the building information model key data description vector by utilizing a key data mining sub-network included in the target building information model anomaly identification neural network, and outputting a local building information model key data description vector corresponding to the building information sub-model to be processed (illustratively, the target building information model anomaly identification neural network is formed by network optimization);

loading the key data description vector of the local building information model to be loaded into a sub-model analysis sub-network included in the target building information model anomaly identification neural network, and analyzing a first sub-model key data description vector corresponding to the building information sub-model to be processed by using the sub-model analysis sub-network; and loading the key data description vector of the local building information model to a component analysis sub-network included in the target building information model anomaly identification neural network, and analyzing a first component key data description vector corresponding to the sub-model component by using the component analysis sub-network.

It may be appreciated that, in some embodiments, the step of performing a sub-model anomaly identification operation on the sub-model of the building information to be processed included in the building information model data to be processed based on the second sub-model key data description vector and the second component key data description vector may further include the following specific implementation matters:

performing key data analysis operation on the second sub-model key data description vector by using a sub-model analysis sub-network included in the target building information model anomaly identification neural network to analyze sub-model type description data corresponding to the building information sub-model to be processed and sub-model distribution data corresponding to the building information sub-model to be processed, and performing combined processing on the sub-model type description data and the sub-model distribution data to form sub-model description data corresponding to the building information sub-model to be processed (illustratively, the sub-model type can be used for reflecting model types of the building information sub-model to be processed, such as bridges, buildings and the like, and the sub-model distribution data can be used for reflecting distribution coordinates, sizes and the like of the building information sub-model to be processed);

Performing key data analysis operation on the second component key data description vector according to a component analysis sub-network included in the target building information model anomaly identification neural network to analyze component identification description data corresponding to each of a second number of component anomaly description data, and performing combination processing on the second number of component identification description data to form component description data corresponding to the sub-model component, wherein the second number of component anomaly description data belongs to an analysis result corresponding to the component analysis sub-network (illustratively, the component identification description data at least can comprise component distribution data and anomaly evaluation parameters of the corresponding sub-model component, and the component distribution data is distribution coordinates of the sub-model component, etc.);

and carrying out sub-model abnormality recognition operation on the to-be-processed building information sub-model in the to-be-processed building information model data based on the sub-model description data and the component description data.

It will be appreciated that, in some embodiments, the number of the sub-model components may be a third number, based on which the step of performing a combination process on the second number of component identification description data to form component description data corresponding to the sub-model components may further include the following specific implementation matters:

Classifying the component identification description data corresponding to the second number of component abnormality description data based on the sub-model components corresponding to the second number of component abnormality description data to form a corresponding third number of component classification clusters, wherein each component classification cluster corresponds to one sub-model component; and respectively carrying out data aggregation operation on the component identification description data in each component classification cluster to form component description data corresponding to each sub-model component (illustratively, the component identification description data in the component classification cluster can be spliced to form the component description data corresponding to the sub-model component).

On the basis, the step of performing sub-model anomaly identification operation on the to-be-processed building information sub-model in the to-be-processed building information model data based on the sub-model description data and the component description data may further include the following specific execution contents:

and carrying out sub-model anomaly identification operation on the to-be-processed building information sub-model in the to-be-processed building information model data based on the sub-model description data and the component description data corresponding to each sub-model component.

It may be appreciated that, in some embodiments, the step of performing the sub-model anomaly identification operation on the sub-model of the building information to be processed in the building information model data based on the sub-model description data and the component description data corresponding to each of the sub-model components may further include the following specific implementation matters:

determining component abnormality identification data corresponding to the sub-model description data (illustratively, component abnormality identification data corresponding to different sub-model description data is generally different, because different building information sub-models to be processed or different sub-model description data cause different conditions of abnormality of the building information sub-models to be processed, the abnormality may be structural mismatch, etc.);

marking the to-be-processed building information sub-model in the to-be-processed building information model data as an abnormal building information sub-model under the condition that component description data corresponding to the component abnormality identification data exists in component description data corresponding to each sub-model component; and marking the to-be-processed building information sub-model in the to-be-processed building information model data as a non-abnormal building information sub-model under the condition that the component description data corresponding to each sub-model component does not have the component description data corresponding to the component abnormality identification data.

It may be appreciated that, in some embodiments, the second number of component identification description data may include component distribution data and abnormality evaluation parameters corresponding to each component abnormality description data, based on which the step of performing a sub-model abnormality recognition operation on the to-be-processed building information sub-model in the to-be-processed building information model data based on the sub-model description data and the component description data corresponding to each sub-model component may further include the following specific implementation matters:

extracting a third number of component importance characterization coefficients corresponding to the sub-model description data, wherein each component importance characterization coefficient corresponds to one sub-model component;

extracting component identification abnormality description data corresponding to each sub-model component (one or more component abnormality description data corresponding to each sub-model component can comprise component identification non-abnormality description data and component identification abnormality description data), and determining an abnormality evaluation parameter corresponding to each component identification abnormality description data in the component description data corresponding to each sub-model component;

According to the third number of component importance characterization coefficients, carrying out fusion operation on the abnormality evaluation parameters corresponding to each component identification abnormality description data (namely carrying out weighting treatment on the abnormality evaluation parameters so as to consider importance differences among different components when determining abnormality), so as to output target abnormality evaluation parameters corresponding to the building information model data to be processed;

marking the building information submodel to be processed in the building information model data to be an abnormal building information submodel under the condition that the target abnormality evaluation parameter exceeds a pre-configured reference abnormality evaluation parameter; and marking the to-be-processed building information sub-model in the to-be-processed building information model data as a non-abnormal building information sub-model under the condition that the target abnormality evaluation parameter does not exceed the reference abnormality evaluation parameter.

It will be appreciated that in some embodiments, the BIM-based data processing may further include a step of optimizing the formation of the target building information model anomaly identification neural network, and the step may further include the following specific implementation:

Extracting exemplary building information model data, exemplary sub-model real data (i.e., real conditions, i.e., whether an abnormality exists) of an exemplary building information sub-model included in the exemplary building information model data, exemplary component real data corresponding to an exemplary sub-model component, and sub-model abnormality real data of the exemplary building information sub-model in the exemplary building information model data, the exemplary building information sub-model including the exemplary sub-model component;

performing key data mining operation on the exemplary building information model data to mine and output an exemplary building information model key data description vector corresponding to the exemplary building information model data; mining the exemplary building information model key data description vector according to a first sub-model analysis sub-network included in the first building information model anomaly identification neural network to output an exemplary first sub-model key data description vector of the exemplary building information sub-model, and mining the exemplary building information model key data description vector according to a first component analysis sub-network included in the first building information model anomaly identification neural network to output an exemplary first component key data description vector corresponding to the exemplary sub-model component; performing a critical data supplementing operation on the exemplary first sub-model critical data description vector based on the exemplary first component critical data description vector by using the first sub-model analysis sub-network to form a corresponding exemplary second sub-model critical data description vector, and performing a critical data supplementing operation on the exemplary first component critical data description vector based on the exemplary first sub-model critical data description vector by using the first component analysis sub-network to form a corresponding exemplary second component critical data description vector; analyzing the exemplary sub-model description data corresponding to the exemplary building information sub-model based on the exemplary second sub-model key data description vector, and analyzing the exemplary component description data corresponding to the exemplary sub-model component based on the exemplary second component key data description vector; and analyzing and outputting sub-model abnormality analysis data corresponding to an exemplary building information sub-model in the exemplary building information model data (i.e., analyzing and evaluating results, i.e., whether the exemplary building information sub-model is abnormal) based on the exemplary sub-model description data and the exemplary component description data;

And performing network optimization processing (i.e. optimizing corresponding network parameters or network weights) on the first building information model anomaly identification neural network based on the example sub-model real data, the example component real data, the sub-model anomaly real data, the example sub-model description data, the example component description data and the sub-model anomaly analysis data to form a corresponding target building information model anomaly identification neural network.

Wherein, it can be understood that, in some embodiments, the step of performing network optimization processing on the first building information model anomaly identification neural network to form a corresponding target building information model anomaly identification neural network based on the exemplary sub-model real data, the exemplary component real data, the sub-model anomaly real data, the exemplary sub-model description data, the exemplary component description data and the sub-model anomaly analysis data may further include the following specific implementation matters:

calculating and outputting corresponding first dimension error parameters (i.e., errors between real conditions and estimated conditions of the neural network with respect to sub-model description data) based on the exemplary sub-model real data and the exemplary sub-model description data, and calculating and outputting corresponding second dimension error parameters (i.e., errors between real conditions and estimated conditions of the neural network with respect to component description data) based on the exemplary component real data and the exemplary component description data, and calculating and outputting corresponding third dimension error parameters (i.e., errors between real conditions and estimated conditions of the neural network with respect to whether there is an anomaly in the sub-model) based on the sub-model anomaly real data and the sub-model anomaly analysis data;

Optimizing the first sub-model analysis sub-network and the first component analysis sub-network according to the first dimension error parameter, the second dimension error parameter and the third dimension error parameter to form a sub-model analysis sub-network and a component analysis sub-network (for example, the first dimension error parameter, the second dimension error parameter and the third dimension error parameter can be weighted and summed to obtain a target error parameter, and then optimizing the first sub-model analysis sub-network and the first component analysis sub-network based on the target error parameter, namely optimizing network parameters of the first sub-model analysis sub-network and the first component analysis sub-network to obtain a corresponding sub-model analysis sub-network and a corresponding component analysis sub-network);

based on the sub-model analysis sub-network and the component analysis sub-network, a corresponding target building information model anomaly identification neural network is formed (the target building information model anomaly identification neural network may include the sub-model analysis sub-network and the component analysis sub-network).

With reference to fig. 3, an embodiment of the present invention further provides a BIM-based data processing system, which is applicable to the above-mentioned BIM-based data processing cloud platform. Wherein the BIM-based data processing system may include the following software functional modules:

The key data mining module is used for carrying out key data mining operation on building information model data to be processed so as to mine building information model key data description vectors corresponding to the building information model data to be processed, wherein the building information model data to be processed comprises a building information sub-model to be processed;

the description vector mining module is used for mining a first sub-model key data description vector corresponding to the building information sub-model to be processed according to the building information model key data description vector, mining a first component key data description vector corresponding to a included sub-model component, and the building information sub-model to be processed comprises the sub-model component;

a description vector supplementing module, configured to perform a key data supplementing operation on the first sub-model key data description vector based on the first component key data description vector to form a corresponding second sub-model key data description vector, and perform a key data supplementing operation on the first component key data description vector based on the first sub-model key data description vector to form a corresponding second component key data description vector;

And the sub-model abnormality recognition module is used for carrying out sub-model abnormality recognition operation on the to-be-processed building information sub-model included in the to-be-processed building information model data based on the second sub-model key data description vector and the second component key data description vector.

In summary, the BIM-based data processing method, system and cloud platform provided by the invention can be used for mining the key data description vector of the building information model; mining corresponding first sub-model key data description vectors and first component key data description vectors according to the building information model key data description vectors; performing key data supplementing operation on the first sub-model key data description vector based on the first component key data description vector to form a second sub-model key data description vector, and performing key data supplementing operation on the first component key data description vector based on the first sub-model key data description vector to form a second component key data description vector; and carrying out sub-model anomaly identification operation based on the second sub-model key data description vector and the second component key data description vector. Based on the above, the basis for performing the operation of identifying the abnormal sub-model includes not only the second sub-model key data description vector but also the second component key data description vector, so that the basis is more sufficient, and the second sub-model key data description vector and the second component key data description vector are obtained based on the corresponding complementary operation of the key data, so that the information of the second sub-model key data description vector and the second component key data description vector is more sufficient, thereby improving the reliability of identifying the abnormal sub-model to a certain extent, and further improving the defects of the prior art.

The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims (10)

1. A BIM-based data processing method, comprising:

performing key data mining operation on the building information model data to be processed to mine building information model key data description vectors corresponding to the building information model data to be processed, wherein the building information model data to be processed comprises a building information sub-model to be processed;

according to the building information model key data description vector, a first sub-model key data description vector corresponding to the building information sub-model to be processed is dug out, and a first component key data description vector corresponding to a sub-model component is dug out, wherein the building information sub-model to be processed comprises the sub-model component;

performing key data supplementing operation on the first sub-model key data description vector based on the first component key data description vector to form a corresponding second sub-model key data description vector, and performing key data supplementing operation on the first component key data description vector based on the first sub-model key data description vector to form a corresponding second component key data description vector;

And carrying out sub-model anomaly identification operation on the to-be-processed building information sub-model included in the to-be-processed building information model data based on the second sub-model key data description vector and the second component key data description vector.

2. The BIM-based data processing method of claim 1, wherein the step of mining a first sub-model key data description vector corresponding to the sub-model of the building information to be processed and mining a first component key data description vector corresponding to the included sub-model component according to the building information model key data description vector includes:

carrying out key data analysis processing on the building information model key data description vector so as to analyze the sub-model suspected distribution position of the building information sub-model to be processed in the building information model key data description vector;

the construction information model key data description vector corresponding to the sub-model suspected distribution position is excavated;

and carrying out key data analysis processing on the key data description vector of the local building information model so as to analyze a first sub-model key data description vector corresponding to the sub-model of the building information to be processed and a first component key data description vector corresponding to the included sub-model component.

3. The BIM-based data processing method of claim 1, wherein the step of mining a first sub-model key data description vector corresponding to the sub-model of the building information to be processed and mining a first component key data description vector corresponding to the included sub-model component according to the building information model key data description vector includes:

utilizing a key data mining sub-network included in the target building information model anomaly identification neural network to conduct key data mining operation on the building information model key data description vector, and outputting a local building information model key data description vector corresponding to the building information sub-model to be processed;

loading the key data description vector of the local building information model to be loaded into a sub-model analysis sub-network included in the target building information model anomaly identification neural network, and analyzing a first sub-model key data description vector corresponding to the building information sub-model to be processed by using the sub-model analysis sub-network;

and loading the key data description vector of the local building information model to load the component analysis sub-network included in the target building information model anomaly identification neural network, and analyzing the first component key data description vector corresponding to the sub-model component by using the component analysis sub-network.

4. The BIM-based data processing method of claim 3, wherein the step of performing a sub-model abnormality recognition operation on the sub-model of the building information to be processed included in the building information to be processed model data based on the second sub-model key data description vector and the second component key data description vector includes:

performing key data analysis operation on the second sub-model key data description vector by utilizing a sub-model analysis sub-network included in the target building information model anomaly identification neural network to analyze sub-model type description data corresponding to the building information sub-model to be processed and sub-model distribution data corresponding to the building information sub-model to be processed, and performing combined processing on the sub-model type description data and the sub-model distribution data to form sub-model description data corresponding to the building information sub-model to be processed;

performing key data analysis operation on the second component key data description vector according to the component analysis sub-network included in the target building information model abnormality identification neural network to analyze component identification description data corresponding to each of a second number of component abnormality description data, and performing combination processing on the second number of component identification description data to form component description data corresponding to the sub-model component, wherein the second number of component abnormality description data belongs to an analysis result corresponding to the component analysis sub-network;

And carrying out sub-model abnormality recognition operation on the to-be-processed building information sub-model in the to-be-processed building information model data based on the sub-model description data and the component description data.

5. The BIM-based data processing method of claim 4, wherein the submodel component has a third number; the step of combining the second number of component identification description data to form component description data corresponding to the sub-model component includes:

classifying the component identification description data corresponding to the second number of component abnormality description data based on the sub-model components corresponding to the second number of component abnormality description data to form a corresponding third number of component classification clusters, wherein each component classification cluster corresponds to one sub-model component; respectively carrying out data aggregation operation on the component identification description data in each component classification cluster to form component description data corresponding to each sub-model component;

the step of performing sub-model anomaly identification operation on the to-be-processed building information sub-model in the to-be-processed building information model data based on the sub-model description data and the component description data includes:

And carrying out sub-model anomaly identification operation on the to-be-processed building information sub-model in the to-be-processed building information model data based on the sub-model description data and the component description data corresponding to each sub-model component.

6. The BIM-based data processing method of claim 5, wherein the step of performing a sub-model abnormality recognition operation on the sub-model of the building information to be processed in the building information model data to be processed based on the sub-model description data and the component description data corresponding to each of the sub-model components includes:

determining component abnormality identification data corresponding to the sub-model description data;

marking the to-be-processed building information sub-model in the to-be-processed building information model data as an abnormal building information sub-model under the condition that component description data corresponding to the component abnormality identification data exists in component description data corresponding to each sub-model component; and marking the to-be-processed building information sub-model in the to-be-processed building information model data as a non-abnormal building information sub-model under the condition that the component description data corresponding to each sub-model component does not have the component description data corresponding to the component abnormality identification data.

7. The BIM-based data processing method of claim 5, wherein the second number of component identification description data includes component distribution data and an abnormality evaluation parameter corresponding to each component abnormality description data; the step of performing sub-model anomaly identification operation on the sub-model of the building information to be processed in the building information model data to be processed based on the sub-model description data and the component description data corresponding to each sub-model component includes:

extracting a third number of component importance characterization coefficients corresponding to the sub-model description data, wherein each component importance characterization coefficient corresponds to one sub-model component;

extracting component identification abnormality description data corresponding to each sub-model component, and determining abnormality evaluation parameters corresponding to each component identification abnormality description data in the component description data corresponding to each sub-model component;

according to the third number of component importance characterization coefficients, carrying out fusion operation on abnormality evaluation parameters corresponding to each component identification abnormality description data so as to output target abnormality evaluation parameters corresponding to the building information model data to be processed;

Marking the building information submodel to be processed in the building information model data to be an abnormal building information submodel under the condition that the target abnormality evaluation parameter exceeds a pre-configured reference abnormality evaluation parameter; and marking the to-be-processed building information sub-model in the to-be-processed building information model data as a non-abnormal building information sub-model under the condition that the target abnormality evaluation parameter does not exceed the reference abnormality evaluation parameter.

8. The BIM-based data processing method of claim 3, wherein the BIM-based data processing method further includes:

extracting exemplary building information model data, exemplary sub-model real data of an exemplary building information sub-model included in the exemplary building information model data, exemplary component real data corresponding to an exemplary sub-model component, and sub-model abnormality real data of the exemplary building information sub-model in the exemplary building information model data, the exemplary building information sub-model including the exemplary sub-model component;

performing key data mining operation on the exemplary building information model data, and mining out an exemplary building information model key data description vector corresponding to the exemplary building information model data;

Mining the exemplary building information model key data description vector according to a first sub-model analysis sub-network included in the first building information model anomaly identification neural network to output an exemplary first sub-model key data description vector of the exemplary building information sub-model, and mining the exemplary building information model key data description vector according to a first component analysis sub-network included in the first building information model anomaly identification neural network to output an exemplary first component key data description vector corresponding to the exemplary sub-model component;

performing a critical data supplementing operation on the exemplary first sub-model critical data description vector based on the exemplary first component critical data description vector by using the first sub-model analysis sub-network to form a corresponding exemplary second sub-model critical data description vector, and performing a critical data supplementing operation on the exemplary first component critical data description vector based on the exemplary first sub-model critical data description vector by using the first component analysis sub-network to form a corresponding exemplary second component critical data description vector;

Analyzing the exemplary sub-model description data corresponding to the exemplary building information sub-model based on the exemplary second sub-model key data description vector, and analyzing the exemplary component description data corresponding to the exemplary sub-model component based on the exemplary second component key data description vector;

analyzing and outputting sub-model abnormality analysis data corresponding to an exemplary building information sub-model in the exemplary building information model data based on the exemplary sub-model description data and the exemplary component description data;

and performing network optimization on the first building information model anomaly identification neural network based on the example sub-model real data, the example component real data, the sub-model anomaly real data, the example sub-model description data, the example component description data and the sub-model anomaly analysis data to form a target building information model anomaly identification neural network.

9. A BIM-based data processing system, comprising:

the key data mining module is used for carrying out key data mining operation on building information model data to be processed so as to mine building information model key data description vectors corresponding to the building information model data to be processed, wherein the building information model data to be processed comprises a building information sub-model to be processed;

The description vector mining module is used for mining a first sub-model key data description vector corresponding to the building information sub-model to be processed according to the building information model key data description vector, mining a first component key data description vector corresponding to a included sub-model component, and the building information sub-model to be processed comprises the sub-model component;

a description vector supplementing module, configured to perform a key data supplementing operation on the first sub-model key data description vector based on the first component key data description vector to form a corresponding second sub-model key data description vector, and perform a key data supplementing operation on the first component key data description vector based on the first sub-model key data description vector to form a corresponding second component key data description vector;

and the sub-model abnormality recognition module is used for carrying out sub-model abnormality recognition operation on the to-be-processed building information sub-model included in the to-be-processed building information model data based on the second sub-model key data description vector and the second component key data description vector.

10. A BIM-based data processing cloud platform, comprising a processor and a memory, the memory for storing a computer program, the processor for executing the computer program to implement the BIM-based data processing method of any of claims 1 to 9.

Images