CN116432297A - Method for collecting heterogeneous data of intelligent building facilities - Google Patents
Method for collecting heterogeneous data of intelligent building facilities Download PDFInfo
- Publication number
- CN116432297A CN116432297A CN202310688484.2A CN202310688484A CN116432297A CN 116432297 A CN116432297 A CN 116432297A CN 202310688484 A CN202310688484 A CN 202310688484A CN 116432297 A CN116432297 A CN 116432297A
- Authority
- CN
- China
- Prior art keywords
- data
- heterogeneous
- types
- data processing
- heterogeneous data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 29
- 238000012545 processing Methods 0.000 claims abstract description 207
- 238000010276 construction Methods 0.000 claims description 56
- 230000000875 corresponding effect Effects 0.000 claims description 45
- 238000012549 training Methods 0.000 claims description 8
- 230000002596 correlated effect Effects 0.000 claims description 7
- 238000012360 testing method Methods 0.000 claims description 7
- 238000012795 verification Methods 0.000 claims description 7
- 238000010219 correlation analysis Methods 0.000 claims description 5
- 238000010801 machine learning Methods 0.000 claims description 4
- 238000013468 resource allocation Methods 0.000 claims description 4
- 238000006243 chemical reaction Methods 0.000 abstract description 14
- 230000010354 integration Effects 0.000 abstract description 9
- 230000000694 effects Effects 0.000 abstract description 5
- 238000009960 carding Methods 0.000 abstract description 3
- 238000013480 data collection Methods 0.000 abstract 2
- 238000004458 analytical method Methods 0.000 description 6
- 238000004364 calculation method Methods 0.000 description 5
- 238000009435 building construction Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000013507 mapping Methods 0.000 description 3
- 238000006317 isomerization reaction Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000003672 processing method Methods 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000003828 downregulation Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 230000003827 upregulation Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/13—Architectural design, e.g. computer-aided architectural design [CAAD] related to design of buildings, bridges, landscapes, production plants or roads
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/08—Construction
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- Software Systems (AREA)
- Computer Hardware Design (AREA)
- General Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Business, Economics & Management (AREA)
- Medical Informatics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Computational Mathematics (AREA)
- Civil Engineering (AREA)
- Mathematical Analysis (AREA)
- Architecture (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Optimization (AREA)
- Economics (AREA)
- Structural Engineering (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a collection method of building intelligent facility heterogeneous data, which relates to the technical field of data processing, and comprises the steps of selecting main data types based on M data structure types, analyzing the relativity of M-1 heterogeneous data types and the main data types, building a data processing model to process M data sets of a collected target building, combining the data sets corresponding to the main data types as data collection integrated processing results, solving the technical problems that the building data cannot be communicated due to the existence of multiple heterogeneous data types, the integration and carding are difficult, the collection processing effect of the building data and the advancing of the digital intelligent building are affected, respectively constructing data processing modules according to the relativity of different heterogeneous data types in the target building, and carrying out targeted conversion processing of different heterogeneous data to unify the data types so as to maximize the data collection processing effects.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a method for collecting heterogeneous data of an intelligent building facility.
Background
Because the multiparty departments are used for design, construction and installation in the building construction process, the facilities of the building construction are diversified, the difference of data types cannot be avoided when data acquisition is carried out, and the acquisition and processing of the data are limited.
The prior art has no targeted collection and processing method of heterogeneous data, and the building data cannot be communicated due to the existence of various heterogeneous data types, so that the integration and the carding are difficult, the collection and processing effects of the building data are affected, and the advancement of digital intelligent buildings is affected.
Disclosure of Invention
The application provides a collection method of building intelligent facility heterogeneous data, which is used for solving the technical problems that the existing technology does not have a targeted collection processing method of heterogeneous data, the building data cannot be communicated due to the existence of various heterogeneous data types, the integration and the carding are difficult, the collection and processing effects of the building data are affected, and the advancement of a digital intelligent building are affected.
In view of the above problems, the present application provides a method for collecting heterogeneous data of a building intelligent facility.
In a first aspect, the present application provides a method for collecting heterogeneous data of a building intelligent facility, where the method includes:
obtaining M data structure types existing in a target building, wherein the M data structure types are heterogeneous;
Selecting a target data structure type in the M data structure types as a main data type, and obtaining M-1 heterogeneous data types;
according to the record of data acquisition processing of the building type of the target building in a preset time range, analyzing the correlation between the M-1 heterogeneous data types and the main data types to obtain M-1 correlation coefficients;
constructing a data processing model for carrying out integrated processing on heterogeneous data of the M-1 heterogeneous data types according to the M-1 correlation coefficients, wherein the data processing model comprises M-1 data processing modules corresponding to the M-1 heterogeneous data types, and the constructed data quantity and accuracy of the M-1 data processing modules are positively correlated with the sizes of the M-1 correlation coefficients;
collecting data of the M data structure types in the target building to obtain M data sets;
inputting M-1 heterogeneous data sets corresponding to the M-1 heterogeneous data types into the data processing model to obtain M-1 heterogeneous data processing result sets, and combining the data sets corresponding to the main data types to obtain a data acquisition integrated processing result of the target building.
In a second aspect, the present application provides a system for collecting heterogeneous data of a building intelligent facility, the system comprising:
the data structure type acquisition module is used for acquiring M data structure types existing in a target building, wherein the M data structure types are heterogeneous;
the main data type determining module is used for selecting a target data structure type in the M data structure types as a main data type and obtaining M-1 heterogeneous data types;
the correlation analysis module is used for carrying out data acquisition processing records within a preset time range according to the building type of the target building, analyzing the correlation between the M-1 heterogeneous data types and the main data types and obtaining M-1 correlation coefficients;
the model construction module is used for constructing a data processing model for carrying out integrated processing on heterogeneous data of the M-1 heterogeneous data types according to the M-1 correlation coefficients, wherein the data processing model comprises M-1 data processing modules corresponding to the M-1 heterogeneous data types, and the constructed data quantity and accuracy of the M-1 data processing modules are positively correlated with the sizes of the M-1 correlation coefficients;
The data acquisition module is used for acquiring the data of the M data structure types in the target building to obtain M data sets;
the data processing module is used for inputting M-1 heterogeneous data sets corresponding to the M-1 heterogeneous data types into the data processing model to obtain M-1 heterogeneous data processing result sets, and combining the data sets corresponding to the main data types to obtain a data acquisition integrated processing result of the target building.
One or more technical solutions provided in the present application have at least the following technical effects or advantages:
according to the method for acquiring the heterogeneous data of the building intelligent facility, M data structure types existing in a target building are acquired, the target data structure types are selected to serve as main data types, M-1 heterogeneous data types are acquired, data acquisition processing records are carried out in a preset time range according to the building types of the target building, the correlation between the M-1 heterogeneous data types and the main data types is analyzed, M-1 correlation coefficients are acquired, a data processing model for carrying out integrated processing on the heterogeneous data of the M-1 heterogeneous data types is constructed, data of the M data structure types in the target building are acquired, M data sets are acquired, M-1 heterogeneous data sets corresponding to the M-1 heterogeneous data types are screened and input into the data processing model, M-1 heterogeneous data processing result sets are acquired, and the data acquisition integrated processing results of the target building are combined with the corresponding data sets of the main data types.
Drawings
Fig. 1 is a schematic flow chart of a method for collecting heterogeneous data of an intelligent building facility;
fig. 2 is a schematic diagram of a correlation analysis flow of M-1 heterogeneous data types and a main data type in a method for collecting heterogeneous data of a building intelligent facility;
FIG. 3 is a schematic diagram of a data processing model construction flow in a method for collecting heterogeneous data of a building intelligent facility;
fig. 4 is a schematic structural diagram of a system for collecting heterogeneous data of an intelligent building facility.
Reference numerals illustrate: the system comprises a data structure type acquisition module 11, a main data type determination module 12, a correlation analysis module 13, a model construction module 14, a data acquisition module 15 and a data processing module 16.
Detailed Description
According to the method, M data structure types existing in a target building are acquired, main data types are selected, based on records of acquisition processing, correlation between M-1 heterogeneous data types and the main data types is analyzed, M-1 correlation coefficients are obtained, a data processing model is constructed, M data sets of the target building are acquired, model analysis is conducted to obtain M-1 heterogeneous data processing result sets, and the data sets corresponding to the main data types are combined to serve as data acquisition integration processing results.
Embodiment one: as shown in fig. 1, the present application provides a method for collecting heterogeneous data of a building intelligent facility, where the method includes:
step S100: obtaining M data structure types existing in a target building, wherein the M data structure types are heterogeneous;
specifically, as the multi-party departments are used for designing, constructing and installing in the building construction process, the facilities of the building construction are diversified, so that data cannot be communicated due to the isomerism of data during building data acquisition, and the follow-up data processing limitation exists.
Specifically, the target building is a building to be subjected to facility management, the coverage data source is collected in the target building, and the heterogeneous data source is determined, for example, the heterogeneous data source can be obtained according to historical collection data or facility configuration. And identifying the data structure types of the heterogeneous data sources, for example, heterogeneous data types such as data in different operating systems, including chart data, linked list data, array data and the like, and determining M data structure types of the heterogeneous data sources, wherein the M data structure types are initial acquisition data states to be subjected to type conversion.
Step S200: selecting a target data structure type in the M data structure types as a main data type, and obtaining M-1 heterogeneous data types;
specifically, based on the M data structure types, one of the M data structure types is selected as the master data type, where the master data type is a unified data format for performing heterogeneous data type conversion, for example, a data structure type with the most extensive data size among the M data structure types may be used as the master data type, and the M-1 heterogeneous data types other than the master data structure type among the M data structure types may be determined, where the M-1 heterogeneous data types are targets for performing data type conversion. By way of example, the main data types in the target building are CGR and DWG in bim data, and other heterogeneous data types may include images of some facilities in the building, size data of the equipment, operation data of the pipeline, and other heterogeneous data in data formats, such as building space information, building element information, and electromechanical equipment pipeline information.
Step S300: according to the record of data acquisition processing of the building type of the target building in a preset time range, analyzing the correlation between the M-1 heterogeneous data types and the main data types to obtain M-1 correlation coefficients;
Further, as shown in fig. 2, according to the record of data acquisition processing performed on the building type of the target building within the preset time range, the correlation between the M-1 heterogeneous data types and the master data types is analyzed, and step S300 of the present application further includes:
step S310: acquiring the target building category and building scale as a target building type;
step S320: acquiring a plurality of acquisition records of heterogeneous data acquisition of a plurality of buildings of the target building type in a past preset time range, wherein each acquisition record comprises data whether the data of the M data structure types are acquired or not;
step S330: acquiring M-1 data amount information of the M-1 heterogeneous data types in a past preset time range of a plurality of buildings of the target building type;
step S340: according to the plurality of acquisition records, analyzing M-1 first correlation coefficients of the M-1 heterogeneous data types and the main data types;
step S350: calculating the ratio of each data volume information to the sum of the M-1 data volume information according to the M-1 data volume information to obtain M-1 second correlation coefficients;
Step S360: and calculating and obtaining the M-1 correlation coefficients according to the M-1 first correlation coefficients and the M-1 second correlation coefficients.
Further, according to the plurality of acquisition records, analyzing the M-1 first correlation coefficients of the M-1 heterogeneous data types and the master data type, the step S340 further includes:
step S341: acquiring the number of times that the main data type and the data of the M-1 heterogeneous data types are acquired simultaneously in the plurality of acquisition records, and acquiring M-1 times information;
step S342: and respectively calculating the ratio of the M-1 times information to the number of the plurality of acquisition records to obtain the M-1 first correlation coefficients.
Specifically, for the target building, determining a plurality of buildings of the same type as the target building to perform reference analysis, calling a data acquisition record in the preset time range, respectively performing analysis statistics of the same-frequency acquisition times and the acquisition data volume ratio for the main data type and the M-1 heterogeneous data types, determining a first correlation coefficient and a second correlation coefficient of the M-1 heterogeneous data types, and comprehensively evaluating and determining the M-1 correlation coefficients.
Specifically, the target building is subjected to directional application analysis, and the building category of the target building, such as industrial building, commercial building and the like, is determined, and the differences of different building categories due to the differences of built-in equipment and application systems exist in the corresponding included data types; and collecting coverage areas of the target building, determining the magnitude of various built-in components, and taking the building scale and the building scale as the target building type, wherein the building scale is positively correlated with the generated data quantity of the target building.
Further, the preset time range is set, namely, a self-set time period for data acquisition and calling is set. And taking the target building type as a standard, determining a plurality of buildings of the same type which are consistent with the target building type, and calling a plurality of acquisition records of heterogeneous data acquisition of the plurality of buildings based on the preset time range, wherein the plurality of acquisition records are in one-to-one correspondence with the plurality of buildings. And taking the M data structure types as identification standards, carrying out type identification judgment and identification of the contained data on each acquisition record, for example, for each acquisition record, identifying the acquired data structure type as 1, and identifying the data structure type not acquired as 0.
And respectively analyzing first correlation coefficients of the M-1 heterogeneous data types and the main data types, specifically, taking any one of the main data types and the M-1 heterogeneous data types as identification information, counting the same-frequency occurrence times based on the acquisition records, and determining the same-frequency acquisition times of the various heterogeneous data types and the main data types as M-1 times information which corresponds to and is marked with the main data types and the heterogeneous data types. And calculating the ratio of the number of times information corresponding to each heterogeneous data type to the number of the plurality of acquisition records based on the M-1 number of times information, namely the ratio of the same-frequency acquisition times to the total acquisition times of the plurality of acquisition records, and taking the ratio as the first correlation coefficient.
And respectively analyzing second correlation coefficients of the M-1 heterogeneous data types and the main data types, specifically, adding the M-1 data amount information, determining total data amount information of the M-1 heterogeneous data types, calculating the ratio of each data amount information in the M-1 data information to the total data amount information, and integrating the second correlation coefficients of the M-1 data information as the second correlation coefficients, wherein the ratio of each data amount information in the M-1 data information to the total data amount information is used as the second correlation coefficient.
The first correlation coefficient and the second correlation coefficient are indexes for measuring the degree of correlation with the main data type based on different information dimensions, and overall correlation coefficient determination is performed based on the combination of the first correlation coefficient and the second correlation coefficient so as to maximize the measurement accuracy of the correlation of guarantee determination. And respectively extracting the first correlation coefficient and the second correlation coefficient corresponding to each heterogeneous data type map based on the M-1 first correlation coefficients and the M-1 second correlation coefficients to perform comprehensive calculation, wherein the larger the number of times that each heterogeneous data type and the main data type are simultaneously acquired, the larger the acquired data volume, the larger the correlation of the heterogeneous data type is, and the more important the correlation of the heterogeneous data type is. For example, taking the product of the first correlation coefficient and the second correlation coefficient as the correlation coefficient corresponding to the heterogeneous data type, determining and mapping the M-1 correlation coefficients corresponding to the M-1 heterogeneous data types, wherein the M-1 correlation coefficients are the construction basis of the subsequent data processing modules.
Step S400: constructing a data processing model for carrying out integrated processing on heterogeneous data of the M-1 heterogeneous data types according to the M-1 correlation coefficients, wherein the data processing model comprises M-1 data processing modules corresponding to the M-1 heterogeneous data types, and the constructed data quantity and accuracy of the M-1 data processing modules are positively correlated with the sizes of the M-1 correlation coefficients;
Further, as shown in fig. 3, according to the M-1 correlation coefficients, a data processing model for performing integrated processing on the M-1 heterogeneous data types is constructed, and step S400 of the present application further includes:
step S410: performing resource allocation of a construction model according to the M-1 correlation coefficients to obtain M-1 resource coefficients;
step S420: constructing a first data processing module in the data processing model according to a first resource coefficient of a first heterogeneous data type in the M-1 heterogeneous data types;
step S430: and constructing M-2 data processing modules corresponding to other M-2 heterogeneous data types according to other M-2 resource coefficients, and obtaining the data processing model.
Further, according to the first resource coefficient of the first heterogeneous data type of the M-1 heterogeneous data types, a first data processing module in the data processing model is constructed, and step S420 of the present application further includes:
step S421: acquiring a preset construction data amount for constructing the M-1 data processing module;
step S422: according to the ratio of the first resource coefficient to the average value of the M-1 resource coefficients, adjusting the preset construction data quantity to obtain a first construction data quantity;
Step S423: acquiring a first heterogeneous data set of a sample according to the first construction data volume and the data of the first heterogeneous data type acquired by a plurality of buildings of the building type in a past preset time range, wherein the first heterogeneous data set of the sample comprises first heterogeneous data of the first construction data volume;
step S424: converting the first heterogeneous data of the sample in the first heterogeneous data set of the sample into the data of the main data type to obtain a first heterogeneous data processing result set of the sample;
step S425: and constructing the first data processing module by adopting the first heterogeneous data set of the sample and the first heterogeneous data processing result set of the sample as construction data.
Further, the first data processing module is constructed by using the first heterogeneous data set of samples and the first heterogeneous data processing result set of samples as construction data, and step S425 of the present application further includes:
step S4251: acquiring a preset accuracy requirement and a preset accuracy requirement range of the M-1 data processing module;
step S4252: according to the ratio of the first resource coefficient to the average value of the M-1 resource coefficients, adjusting the preset accuracy requirement within the preset accuracy requirement range to obtain a first accuracy requirement;
Step S4253: according to the first heterogeneous data type, based on machine learning, constructing a network architecture of the first data processing module, wherein input data of the first data processing module are first heterogeneous data, and output data are first heterogeneous data processing results;
step S4254: and according to the first accuracy requirement, adopting the first heterogeneous data set of the sample and the first heterogeneous data processing result set of the sample as construction data, performing supervision training, verification and test on the first data processing module, and obtaining the first data processing module under the condition of meeting the preset accuracy requirement.
Specifically, based on the M-1 correlation coefficients, M-1 data processing modules are respectively constructed and used for processing and converting the M-1 heterogeneous data types, the M-1 data processing modules are integrated to generate the data processing model, and the data processing model is used as a self-built auxiliary analysis tool for data processing.
Specifically, based on the sizes of the M-1 correlation coefficients, configuring construction resources of an adaptive amount, wherein the larger the correlation coefficient is, the larger the correlation and the importance of the heterogeneous data type are, and correspondingly, more calculation resources are required to be configured for data processing. Calculating coefficient sums of the M-1 correlation coefficients, taking the ratio of each correlation coefficient to the coefficient sum as a corresponding resource coefficient of a data processing module of each heterogeneous data type, and determining the M-1 resource coefficients.
Further, based on the M-1 heterogeneous data types, extracting a heterogeneous data type to be subjected to module construction as the first heterogeneous data type, matching the first resource coefficient in the M-1 resource coefficients, and constructing the first data processing module.
Specifically, an initial construction data amount for constructing the M-1 data processing modules is determined and used as the preset construction data amount, the preset construction data amount is adapted and adjusted according to the M-1 resource coefficients, and the actual construction data amount of each data processing module is determined. And carrying out average value calculation on the M-1 resource coefficients, calculating the ratio of the first resource coefficient to the M-1 resource coefficients, and taking the ratio as amplitude modulation of the preset construction data quantity, for example, taking the product of the ratio and the preset construction data quantity as the adjusted first construction data quantity.
Further, based on a plurality of collection records of a plurality of buildings of the building type in the past within the preset time range, the first construction data size is used as a sample demand calling size, data conforming to the first heterogeneous data type is called, and data integration is conducted to form a first heterogeneous data set of the sample. And further performing data type conversion on the first heterogeneous data set of the sample by taking the main data type as a conversion standard, completing conversion of the sample data type through manual conversion, acquiring a first heterogeneous data processing result set of the sample, mapping the first heterogeneous data processing result set of the sample with the first heterogeneous data set of the sample, and completing construction of the first data processing module by taking the first heterogeneous data processing result set of the sample as construction data.
Specifically, the preset accuracy requirement of the M-1 data processing module and the preset accuracy requirement range are configured in a self-defined manner based on processing requirements, namely, initial limiting conditions, such as conversion integrity, conversion accuracy and the like, used for constructing the M-1 data processing module are configured to be corresponding measurement intervals, such as 90% -100%. And (3) moderately adjusting the resource coefficient based on the resource coefficient, and determining the actual accuracy requirement and the actual accuracy requirement range. And carrying out average value calculation on the M-1 resource coefficients, taking the ratio of the first resource coefficient to the coefficient average value as amplitude modulation, and adjusting the preset accuracy requirement and the preset accuracy requirement range within the preset accuracy requirement range, for example, carrying out up-regulation on the ratio of more than 1, carrying out down-regulation on the ratio of less than 1, wherein the first accuracy requirement is the accuracy requirement matched with the first heterogeneous data type.
Further, for the first heterogeneous data type, a network architecture of the first data processing module is constructed based on machine learning, for example, the first data processing module is a multi-layer fully-connected neural network model, and includes an input recognition layer, a decision processing layer and a regular output layer. Preferably, different models are selected according to different data types, for example, pictures can be according to convolutional neural networks; general data may be selected based on one skilled in the art based on a feed forward neural network, etc. Mapping and associating the first heterogeneous data set with the first heterogeneous data processing result of the sample, determining a hierarchy identification node and a hierarchy decision node, taking the hierarchy identification node and the hierarchy decision node as construction data, dividing the construction data based on a preset dividing proportion, and determining a training set, a verification set and a test set, wherein the preset dividing proportion is a self-defined adjustable data dividing proportion. Performing supervised training on the first data processing module based on the training set to generate a trained first data processing module, further performing verification test on the first data processing module based on the verification set and the test set, judging whether the output accuracy of the model meets the first accuracy requirement, and if so, determining the constructed first data processing module; if the output accuracy of the model meets the first accuracy requirement, the division of the constructed data is carried out again, and training, verification and testing are carried out again until the output accuracy of the model meets the first accuracy requirement.
Further, continuing to adjust the preset construction data amount, the preset accuracy requirement and the preset accuracy requirement range based on the other M-2 resource coefficients, performing module construction based on the corresponding heterogeneous data types, and selecting the fit model types to obtain M-2 data processing modules corresponding to other M-2 heterogeneous data types, wherein the construction steps of the data processing modules are the same, the specific construction data and the types are different, and integrating the constructed M-1 data processing modules to generate the data processing model, so that the data processing accuracy can be effectively improved, and the data processing efficiency can be ensured. And the data processing model is used for carrying out targeted conversion of the heterogeneous data, so that the isomerisation of the heterogeneous data is realized, and the acquisition, arrangement and subsequent use of the building data are facilitated.
Step S500: collecting data of the M data structure types in the target building to obtain M data sets;
step S600: inputting M-1 heterogeneous data sets corresponding to the M-1 heterogeneous data types into the data processing model to obtain M-1 heterogeneous data processing result sets, and combining the data sets corresponding to the main data types to obtain a data acquisition integrated processing result of the target building.
Further, inputting the M-1 heterogeneous data sets corresponding to the M-1 heterogeneous data types into the data processing model to obtain M-1 heterogeneous data processing result sets, where step S600 further includes:
step S610: inputting M-1 heterogeneous data sets corresponding to the M-1 heterogeneous data types into an M-1 data processing module in the data processing model to obtain M-1 heterogeneous data processing result sets, wherein data in each heterogeneous data processing result set is the data of the main data type;
step S620: integrating the M-1 heterogeneous data processing result sets and the data sets corresponding to the main data types to obtain the data acquisition integrated processing result.
Specifically, data acquisition is synchronously performed on each acquisition source in the target building, and the M data sets corresponding to the M data structure types are acquired. And extracting M-1 heterogeneous data sets corresponding to M-1 heterogeneous data types based on the M data sets, inputting the M-1 heterogeneous data sets into the data processing model, respectively traversing the M-1 data processing modules for matching, determining the data processing modules corresponding to the heterogeneous data sets, inputting data, converting the main data types as conversion standards to complete data type conversion, realizing data de-isomerization, acquiring M-1 heterogeneous data processing result sets consistent with the main data types, and outputting based on the data processing model. And integrating and regulating the M-1 heterogeneous data processing result sets and the data sets corresponding to the main data types, and taking the data sets as the data acquisition and integration processing results. The data acquisition integrated processing result has data type uniformity, and is convenient for subsequent building data acquisition, arrangement and use.
Embodiment two: based on the same inventive concept as the method for collecting heterogeneous data of a building intelligent facility in the foregoing embodiment, as shown in fig. 4, the present application provides a system for collecting heterogeneous data of a building intelligent facility, the system comprising:
the data structure type acquisition module 11 is used for acquiring M data structure types existing in a target building, wherein the M data structure types are heterogeneous;
a main data type determining module 12, where the main data type determining module 12 is configured to select a target data structure type from the M data structure types as a main data type, and obtain M-1 heterogeneous data types;
the correlation analysis module 13 is used for carrying out data acquisition processing records within a preset time range according to the building type of the target building, analyzing the correlation between the M-1 heterogeneous data types and the main data types and obtaining M-1 correlation coefficients;
the model construction module 14 is configured to construct a data processing model for performing integrated processing on heterogeneous data of the M-1 heterogeneous data types according to the M-1 correlation coefficients, where the data processing model includes M-1 data processing modules corresponding to the M-1 heterogeneous data types, and the constructed data amount and accuracy of the M-1 data processing modules are positively correlated with the magnitude of the M-1 correlation coefficients;
The data acquisition module 15 is used for acquiring data of the M data structure types in the target building to obtain M data sets;
the data processing module 16 is configured to input M-1 heterogeneous data sets corresponding to the M-1 heterogeneous data types into the data processing model, obtain M-1 heterogeneous data processing result sets, and combine the data sets corresponding to the main data types to obtain a data acquisition and integration processing result of the target building.
Further, the system further comprises:
the target building type determining module is used for acquiring the target building type and the building scale as target building types;
the acquisition record acquisition module is used for acquiring a plurality of acquisition records of heterogeneous data acquisition of a plurality of buildings of the target building type in a past preset time range, wherein each acquisition record comprises data of whether the data of the M data structure types are acquired or not;
the data volume information acquisition module is used for acquiring M-1 data volume information of the M-1 heterogeneous data types in a past preset time range of a plurality of buildings of the target building type;
The first correlation coefficient analysis module is used for analyzing M-1 first correlation coefficients of the M-1 heterogeneous data types and the main data types according to the plurality of acquisition records;
the second correlation coefficient acquisition module is used for calculating the ratio of the sum of each data volume information and the M-1 data volume information according to the M-1 data volume information to obtain M-1 second correlation coefficients;
and the correlation coefficient calculation module is used for calculating and obtaining the M-1 correlation coefficients according to the M-1 first correlation coefficients and the M-1 second correlation coefficients.
Further, the system further comprises:
the frequency information acquisition module is used for acquiring the frequency of the simultaneous acquisition of the main data type and the data of the M-1 heterogeneous data types in the plurality of acquisition records to acquire M-1 frequency information;
the ratio calculating module is used for calculating the ratio of the M-1 times information to the number of the plurality of acquisition records respectively to obtain the M-1 first correlation coefficients.
Further, the system further comprises:
the resource allocation module is used for carrying out resource allocation of a construction model according to the M-1 correlation coefficients to obtain M-1 resource coefficients;
the construction module is used for constructing a first data processing module in the data processing model according to a first resource coefficient of a first heterogeneous data type in the M-1 heterogeneous data types;
and the data processing model construction module is used for continuously constructing M-2 data processing modules corresponding to other M-2 heterogeneous data types according to other M-2 resource coefficients to obtain the data processing model.
Further, the system further comprises:
the preset construction data volume acquisition module is used for acquiring preset construction data volumes for constructing the M-1 data processing module;
the data volume adjusting module is used for adjusting the preset construction data volume according to the ratio of the first resource coefficient to the average value of the M-1 resource coefficients to obtain a first construction data volume;
the first heterogeneous data set acquisition module is used for acquiring first heterogeneous data sets of the samples according to the first construction data volume and the data of the first heterogeneous data type acquired by a plurality of buildings of the building type in a past preset time range, wherein the first heterogeneous data sets of the samples comprise first heterogeneous data of the first construction data volume;
The data conversion module is used for converting the first heterogeneous data of the sample in the first heterogeneous data set of the sample into the data of the main data type to obtain a first heterogeneous data processing result set of the sample;
the first data processing module construction module is used for constructing the first data processing module by adopting the sample first heterogeneous data set and the sample first heterogeneous data processing result set as construction data.
Further, the system further comprises:
the accuracy information acquisition module is used for acquiring the preset accuracy requirement and the preset accuracy requirement range of the M-1 data processing module;
the requirement adjustment module is used for adjusting the preset accuracy requirement within the preset accuracy requirement range according to the ratio of the first resource coefficient to the average value of the M-1 resource coefficients to obtain a first accuracy requirement;
the network architecture construction module is used for constructing a network architecture of the first data processing module based on machine learning according to the first heterogeneous data type, wherein input data of the first data processing module are first heterogeneous data, and output data are first heterogeneous data processing results;
And the supervision and training module is used for performing supervision and training, verification and test on the first data processing module by adopting the first heterogeneous data set of the sample and the first heterogeneous data processing result set of the sample as construction data according to the first accuracy requirement, and obtaining the first data processing module under the condition of meeting the preset accuracy requirement.
Further, the system further comprises:
the heterogeneous data processing module is used for inputting M-1 heterogeneous data sets corresponding to the M-1 heterogeneous data types into the M-1 data processing module in the data processing model to obtain M-1 heterogeneous data processing result sets, and data in each heterogeneous data processing result set is the data of the main data type;
and the data integration module is used for integrating the M-1 heterogeneous data processing result sets and the data sets corresponding to the main data types to obtain the data acquisition integration processing result.
The foregoing detailed description of a method for collecting heterogeneous data of a building intelligent facility will be clear to those skilled in the art, and the device disclosed in this embodiment is relatively simple to describe, and the relevant places refer to the method part for description.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (8)
1. A method for collecting heterogeneous data of a building intelligent facility, the method comprising:
obtaining M data structure types existing in a target building, wherein the M data structure types are heterogeneous;
selecting a target data structure type in the M data structure types as a main data type, and obtaining M-1 heterogeneous data types;
according to the record of data acquisition processing of the building type of the target building in a preset time range, analyzing the correlation between the M-1 heterogeneous data types and the main data types to obtain M-1 correlation coefficients;
constructing a data processing model for carrying out integrated processing on heterogeneous data of the M-1 heterogeneous data types according to the M-1 correlation coefficients, wherein the data processing model comprises M-1 data processing modules corresponding to the M-1 heterogeneous data types, and the constructed data quantity and accuracy of the M-1 data processing modules are positively correlated with the sizes of the M-1 correlation coefficients;
Collecting data of the M data structure types in the target building to obtain M data sets;
inputting M-1 heterogeneous data sets corresponding to the M-1 heterogeneous data types into the data processing model to obtain M-1 heterogeneous data processing result sets, and combining the data sets corresponding to the main data types to obtain a data acquisition integrated processing result of the target building.
2. The method of claim 1, wherein analyzing the correlation of the M-1 heterogeneous data types and the master data type according to the record of the data acquisition processing of the building type of the target building within a preset time range comprises:
acquiring the target building category and building scale as a target building type;
acquiring a plurality of acquisition records of heterogeneous data acquisition of a plurality of buildings of the target building type in a past preset time range, wherein each acquisition record comprises data whether the data of the M data structure types are acquired or not;
acquiring M-1 data amount information of the M-1 heterogeneous data types in a past preset time range of a plurality of buildings of the target building type;
According to the plurality of acquisition records, analyzing M-1 first correlation coefficients of the M-1 heterogeneous data types and the main data types;
calculating the ratio of each data volume information to the sum of the M-1 data volume information according to the M-1 data volume information to obtain M-1 second correlation coefficients;
and calculating and obtaining the M-1 correlation coefficients according to the M-1 first correlation coefficients and the M-1 second correlation coefficients.
3. The method of claim 2, wherein analyzing M-1 first correlation coefficients of the M-1 heterogeneous data types and the master data type from the plurality of acquisition records comprises:
acquiring the number of times that the main data type and the data of the M-1 heterogeneous data types are acquired simultaneously in the plurality of acquisition records, and acquiring M-1 times information;
and respectively calculating the ratio of the M-1 times information to the number of the plurality of acquisition records to obtain the M-1 first correlation coefficients.
4. The method of claim 1, wherein constructing a data processing model for integrated processing of heterogeneous data of the M-1 heterogeneous data types based on the M-1 correlation coefficients comprises:
Performing resource allocation of a construction model according to the M-1 correlation coefficients to obtain M-1 resource coefficients;
constructing a first data processing module in the data processing model according to a first resource coefficient of a first heterogeneous data type in the M-1 heterogeneous data types;
and constructing M-2 data processing modules corresponding to other M-2 heterogeneous data types according to other M-2 resource coefficients, and obtaining the data processing model.
5. The method of claim 4, wherein constructing a first data processing module within the data processing model based on a first resource coefficient of a first heterogeneous data type of the M-1 heterogeneous data types, comprises:
acquiring a preset construction data amount for constructing the M-1 data processing module;
according to the ratio of the first resource coefficient to the average value of the M-1 resource coefficients, adjusting the preset construction data quantity to obtain a first construction data quantity;
acquiring a first heterogeneous data set of a sample according to the first construction data volume and the data of the first heterogeneous data type acquired by a plurality of buildings of the building type in a past preset time range, wherein the first heterogeneous data set of the sample comprises first heterogeneous data of the first construction data volume;
Converting the first heterogeneous data of the sample in the first heterogeneous data set of the sample into the data of the main data type to obtain a first heterogeneous data processing result set of the sample;
and constructing the first data processing module by adopting the first heterogeneous data set of the sample and the first heterogeneous data processing result set of the sample as construction data.
6. The method of claim 5, wherein constructing the first data processing module using the sample first heterogeneous data set and the sample first heterogeneous data processing result set as construction data comprises:
acquiring a preset accuracy requirement and a preset accuracy requirement range of the M-1 data processing module;
according to the ratio of the first resource coefficient to the average value of the M-1 resource coefficients, adjusting the preset accuracy requirement within the preset accuracy requirement range to obtain a first accuracy requirement;
according to the first heterogeneous data type, based on machine learning, constructing a network architecture of the first data processing module, wherein input data of the first data processing module are first heterogeneous data, and output data are first heterogeneous data processing results;
And according to the first accuracy requirement, adopting the first heterogeneous data set of the sample and the first heterogeneous data processing result set of the sample as construction data, performing supervision training, verification and test on the first data processing module, and obtaining the first data processing module under the condition of meeting the preset accuracy requirement.
7. The method of claim 1, wherein inputting M-1 heterogeneous data sets corresponding to the M-1 heterogeneous data types into the data processing model to obtain M-1 heterogeneous data processing result sets, comprises:
inputting M-1 heterogeneous data sets corresponding to the M-1 heterogeneous data types into an M-1 data processing module in the data processing model to obtain M-1 heterogeneous data processing result sets, wherein data in each heterogeneous data processing result set is the data of the main data type;
integrating the M-1 heterogeneous data processing result sets and the data sets corresponding to the main data types to obtain the data acquisition integrated processing result.
8. A system for collecting heterogeneous data of a building intelligent facility, the system comprising:
The data structure type acquisition module is used for acquiring M data structure types existing in a target building, wherein the M data structure types are heterogeneous;
the main data type determining module is used for selecting a target data structure type in the M data structure types as a main data type and obtaining M-1 heterogeneous data types;
the correlation analysis module is used for carrying out data acquisition processing records within a preset time range according to the building type of the target building, analyzing the correlation between the M-1 heterogeneous data types and the main data types and obtaining M-1 correlation coefficients;
the model construction module is used for constructing a data processing model for carrying out integrated processing on heterogeneous data of the M-1 heterogeneous data types according to the M-1 correlation coefficients, wherein the data processing model comprises M-1 data processing modules corresponding to the M-1 heterogeneous data types, and the constructed data quantity and accuracy of the M-1 data processing modules are positively correlated with the sizes of the M-1 correlation coefficients;
The data acquisition module is used for acquiring the data of the M data structure types in the target building to obtain M data sets;
the data processing module is used for inputting M-1 heterogeneous data sets corresponding to the M-1 heterogeneous data types into the data processing model to obtain M-1 heterogeneous data processing result sets, and combining the data sets corresponding to the main data types to obtain a data acquisition integrated processing result of the target building.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310688484.2A CN116432297B (en) | 2023-06-12 | 2023-06-12 | Method for collecting heterogeneous data of intelligent building facilities |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310688484.2A CN116432297B (en) | 2023-06-12 | 2023-06-12 | Method for collecting heterogeneous data of intelligent building facilities |
Publications (2)
Publication Number | Publication Date |
---|---|
CN116432297A true CN116432297A (en) | 2023-07-14 |
CN116432297B CN116432297B (en) | 2023-09-12 |
Family
ID=87081824
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310688484.2A Active CN116432297B (en) | 2023-06-12 | 2023-06-12 | Method for collecting heterogeneous data of intelligent building facilities |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116432297B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117113508A (en) * | 2023-09-08 | 2023-11-24 | 绥化市麦微科技有限公司 | Building information model BIM data generation method and system |
CN117519948A (en) * | 2023-12-11 | 2024-02-06 | 广东筠诚建筑科技有限公司 | Method and system for realizing computing resource adjustment under building construction based on cloud platform |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114461603A (en) * | 2021-12-30 | 2022-05-10 | 奇安信科技集团股份有限公司 | Multi-source heterogeneous data fusion method and device |
CN114661810A (en) * | 2022-05-24 | 2022-06-24 | 国网浙江省电力有限公司杭州供电公司 | Lightweight multi-source heterogeneous data fusion method and system |
CN115396151A (en) * | 2022-07-25 | 2022-11-25 | 浙江大有实业有限公司杭州科技发展分公司 | Artificial intelligence network-based security event quick response system |
-
2023
- 2023-06-12 CN CN202310688484.2A patent/CN116432297B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114461603A (en) * | 2021-12-30 | 2022-05-10 | 奇安信科技集团股份有限公司 | Multi-source heterogeneous data fusion method and device |
CN114661810A (en) * | 2022-05-24 | 2022-06-24 | 国网浙江省电力有限公司杭州供电公司 | Lightweight multi-source heterogeneous data fusion method and system |
CN115396151A (en) * | 2022-07-25 | 2022-11-25 | 浙江大有实业有限公司杭州科技发展分公司 | Artificial intelligence network-based security event quick response system |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117113508A (en) * | 2023-09-08 | 2023-11-24 | 绥化市麦微科技有限公司 | Building information model BIM data generation method and system |
CN117519948A (en) * | 2023-12-11 | 2024-02-06 | 广东筠诚建筑科技有限公司 | Method and system for realizing computing resource adjustment under building construction based on cloud platform |
CN117519948B (en) * | 2023-12-11 | 2024-04-26 | 广东筠诚建筑科技有限公司 | Method and system for realizing computing resource adjustment under building construction based on cloud platform |
Also Published As
Publication number | Publication date |
---|---|
CN116432297B (en) | 2023-09-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN116432297B (en) | Method for collecting heterogeneous data of intelligent building facilities | |
CN106803799B (en) | Performance test method and device | |
CN111210483B (en) | Simulated satellite cloud picture generation method based on generation of countermeasure network and numerical mode product | |
CN111368911B (en) | Image classification method and device and computer readable storage medium | |
CN106452934B (en) | Method and device for analyzing network performance index change trend | |
CN106612511B (en) | Wireless network throughput evaluation method and device based on support vector machine | |
Martínez-Solaeche et al. | J-PAS: Measuring emission lines with artificial neural networks | |
CN112348052A (en) | Power transmission and transformation equipment abnormal sound source positioning method based on improved EfficientNet | |
CN109088793B (en) | Method and apparatus for detecting network failure | |
Guillaume Blanchet et al. | A new cost‐effective approach to survey ecological communities | |
CN116821685B (en) | Sound signal identification method and system for cruising robot | |
CN110941933A (en) | Complex electromagnetic environment fidelity evaluation model and method based on similar theory | |
CN112783744A (en) | Data detection method, device, equipment and storage medium | |
CN116823172A (en) | Model optimization-based engineering cost assessment method and system | |
Chen et al. | A novel prediction approach for exploring PM2. 5 spatiotemporal propagation based on convolutional recursive neural networks | |
Kafaee et al. | Pile-up correction by genetic algorithm and artificial neural network | |
CN110209575A (en) | Test document generation method, device, electronic equipment and storage medium | |
Ragozin et al. | State Prediction in Compound Control Systems via Time Series: Neural Network Approach | |
CN115392102A (en) | Method and device for establishing energy consumption prediction model and method and system for predicting energy consumption | |
CN112711873A (en) | Method, system and medium for carrying out damage positioning based on lamb wave and neural network | |
CN106227702B (en) | A kind of user requirements analysis method based on spiral iterative model | |
CN117354053B (en) | Network security protection method based on big data | |
Latinopoulos et al. | Trust in IoT-enabled mobility services: Predictive analytics and the impact of prediction errors on the quality of service in bike sharing | |
CN117076454B (en) | Engineering quality acceptance form data structured storage method and system | |
CN111400152B (en) | Data processing method, first server and second server |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |