Construction early warning method and device based on multi-source data space-time association

Abstract

The application provides a construction early warning method and device based on multi-source data space-time correlation, belongs to the technical field of data processing, and aims at the problems of insufficient instantaneity and effectiveness of current construction prediction. First construction state data of a target construction area is acquired, space-time correlation of the first construction state data is performed based on a first generation timestamp and a first generation space position of the first construction state data, a plurality of first construction space-time data groups are obtained, a key construction risk source of the target construction area is identified based on the first construction space-time data groups, a first risk identification result is obtained, and in the case that the first risk identification result is that there is a risk, an alarm prompt information is generated and the alarm prompt information is displayed in the target construction area. Construction risk identification can be performed according to the space-time correlation of multi-source heterogeneous data of the construction site, prediction of future risks is realized, and the instantaneity, reliability and foresight of construction early warning are significantly improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to a construction early warning method and device based on multi-source data space-time association. BACKGROUND

[0002] Construction early warning is an important link in building engineering management, aiming to identify potential risks in advance and take measures to ensure engineering safety, progress controllability and quality standard through real-time monitoring, data analysis and information transmission. Modern construction early warning has shifted from traditional experience judgment to data-driven intelligent management, realizing risk pre-control through Internet of Things, big data analysis and closed-loop mechanism. In the process of bridge engineering construction, there are a large number of construction machinery and equipment, construction personnel and complex and changeable construction environment in the construction site. How to real-time acquire and process these heterogeneous data and associate and map them with bridge engineering to realize accurate monitoring and control of the construction process is a technical problem to be solved. Specifically, the heterogeneous data of the construction site includes working state data of mechanical equipment, behavior trajectory data of construction personnel, environmental meteorological and geological data, etc. These data are from different sources, have various formats, and have large data volume and high real-time requirements. How to realize fast access, analysis and fusion of heterogeneous data and realize construction risk early warning and construction risk control of the construction site is a technical problem to be solved. SUMMARY

[0003] In view of the above problems, the embodiments of the present application provide a construction early warning method and device based on multi-source data space-time association, electronic equipment and readable storage medium, so as to overcome the above problems or at least partially solve the above problems.

[0004] In the first aspect, the embodiments of the present application provide a construction early warning method based on multi-source data space-time association, which comprises:

[0005] obtaining first construction state data of a target construction area; the first construction state data includes construction machinery and equipment state data, construction personnel trajectory data, construction environment meteorological data and construction geological data;

[0006] based on the first generation time stamp and the first generation space position of the first construction state data, performing space-time association on the first construction state data to obtain a plurality of first construction space-time data groups;

[0007] based on the first construction space-time data group, performing risk identification on a key construction risk source of the target construction area to obtain a first risk identification result;

[0008] in the case that the first risk identification result is that there is a risk, generating an alarm prompt information and displaying the alarm prompt information in the target construction area.

[0009] Optionally, based on the first construction spatio-temporal data set, a risk identification is performed on a key construction risk source of the target construction area to obtain a first risk identification result, including:

[0010] Based on the first construction spatio-temporal data set, a digital twin image sequence of the target construction area in a continuous historical time period is generated;

[0011] The digital twin image sequence is input into a pre-trained spatio-temporal prediction neural network model, and the digital twin image sequence is processed by the spatio-temporal prediction neural network model to output a predicted digital twin image at a future time;

[0012] The predicted digital twin image is processed by a semantic segmentation model to obtain a pixel-level construction element classification map, the construction element including personnel, mechanical equipment, and high-risk area;

[0013] The construction element classification map is logically matched with a pre-set safety rule library, and if the matching result violates the safety rule, an early warning information corresponding to the future time is generated.

[0014] Optionally, the spatio-temporal prediction neural network model is a PredRNN network; the PredRNN network models the spatio-temporal evolution characteristics in the digital twin image sequence through a spatio-temporal long short-term memory unit and a cross-layer bidirectional recursive structure contained therein; a training target of the PredRNN network is guided by a composite loss function, the composite loss function including a reconstruction loss between predicted image pixels and real image pixels, and a perception loss calculated based on feature map differences of a feature extraction network.

[0015] Optionally, based on the first construction state data, a first generation timestamp and a first generation spatial position are determined, and a plurality of first construction spatio-temporal data sets are obtained by spatio-temporal association of the first construction state data, including:

[0016] Based on the first generation timestamps of the construction mechanical equipment state data, the construction personnel trajectory data, the construction environment meteorological data, and the construction geological data, a first generation time difference between the construction mechanical equipment state data, the construction personnel trajectory data, the construction environment meteorological data, and the construction geological data is determined;

[0017] Based on the first generation spatial positions of the construction mechanical equipment state data, the construction personnel trajectory data, the construction environment meteorological data, and the construction geological data, a first spatial Euclidean distance between the construction mechanical equipment state data, the construction personnel trajectory data, the construction environment meteorological data, and the construction geological data is determined;

[0018] determine, based on the first generation time difference and the first spatial Euclidean distance, in combination with the first construction state data, that a first data mapping with spatiotemporal correlation exists;

[0019] In the case that each of the first data mappings contains the same construction mechanical equipment state data, construction personnel trajectory data, construction environment meteorological data or construction geological data, the first data mappings are fused to obtain a plurality of first construction spatiotemporal data groups.

[0020] Optionally, the determining, based on the first generation time difference and the first spatial Euclidean distance, in combination with the first construction state data, that a first data mapping with spatiotemporal correlation exists, comprises:

[0021] In the case that the first generation time difference is less than or equal to a first time threshold, and the first spatial Euclidean distance between the two data corresponding to the first generation time difference is less than or equal to a first distance threshold, a first data mapping between the two first construction state data corresponding to the first generation time difference is generated.

[0022] Optionally, the determining, based on the first generation time difference and the first spatial Euclidean distance, in combination with the first construction state data, that a first data mapping with spatiotemporal correlation exists, comprises:

[0023] Based on the first generation time difference and the first spatial Euclidean distance between the two data corresponding to the first generation time difference, a first continuous correlation degree between the construction mechanical equipment state data, the construction personnel trajectory data, the construction environment meteorological data and the construction geological data is calculated.

[0024] In the case that the first continuous correlation degree is greater than or equal to a first threshold, a first data mapping between the two first construction state data corresponding to the first continuous correlation degree is generated.

[0025] Optionally, the spatiotemporal correlation of the first construction state data based on the first generation time stamp and the first generation spatial position of the first construction state is to obtain a plurality of first construction spatiotemporal data groups, comprising:

[0026] In the case that the first construction state data has data missing at a first time, the data acquisition frequency at the first time is determined.

[0027] Based on the data acquisition frequency, the missing data at the first time is filled to obtain second construction state data.

[0028] The second construction state data is spatio-temporally associated based on the first generation time stamp and the first generation space position of the second construction state data, to obtain a plurality of first construction spatio-temporal data groups.

[0029] Optionally, the alarm prompt information is a digital twin image marked with a high-risk area identifier to mark the construction risk source with risks.

[0030] Optionally, the method further comprises:

[0031] In a case where the first risk identification result is no risks, construction progress of the target construction area is acquired;

[0032] In a case where the construction progress is less than the planned progress, construction progress warning information is generated and displayed.

[0033] In a second aspect, an embodiment of the present application provides a construction warning device based on spatio-temporal association of multi-source data, the device comprising:

[0034] A first acquisition module is configured to acquire first construction state data of a target construction area; the first construction state data comprises construction mechanical equipment state data, construction personnel trajectory data, construction environment meteorological data, and construction geological data;

[0035] A spatio-temporal association module is configured to perform spatio-temporal association on the first construction state data based on a first generation time stamp and a first generation space position of the first construction state data, to obtain a plurality of first construction spatio-temporal data groups;

[0036] A risk identification module is configured to perform risk identification on a key construction risk source of the target construction area based on the first construction spatio-temporal data groups, to obtain a first risk identification result;

[0037] A first generation module is configured to generate alarm prompt information and display the alarm prompt information in the target construction area in a case where the first risk identification result is risks.

[0038] Optionally, the risk identification module comprises:

[0039] A generation sub-module is configured to generate a digital twin image sequence of the target construction area in a continuous historical time period based on the first construction spatio-temporal data groups;

[0040] An input-output sub-module is configured to input the digital twin image sequence into a pre-trained spatio-temporal prediction neural network model, process the digital twin image sequence by the spatio-temporal prediction neural network model, and output a predicted digital twin image at a future time point.

[0041] ​An image processing submodule is configured to process the predicted digital twin image by using a semantic segmentation model to obtain a pixel-level construction element classification map, wherein the construction elements include personnel, mechanical equipment, and high-risk areas.

[0042] A logic matching submodule is configured to perform logic matching between the construction element classification map and a preset safety rule library, and generate early warning information corresponding to a future time if the matching result violates a safety rule.

[0043] Optionally, the spatio-temporal prediction neural network model is a PredRNN network. The PredRNN network models the spatio-temporal evolution characteristics in the digital twin image sequence by using a spatio-temporal long short-term memory unit and a cross-layer bidirectional recursive structure contained therein. The training target of the PredRNN network is guided by a composite loss function, which includes a reconstruction loss between predicted image pixels and real image pixels and a perception loss calculated based on the difference between feature maps of a feature extraction network.

[0044] Optionally, the spatio-temporal correlation module includes:

[0045] A first determination submodule is configured to determine a first generation time difference between the construction mechanical equipment state data, the construction personnel trajectory data, the construction environment meteorological data, and the construction geological data based on first generation time stamps of the construction mechanical equipment state data, the construction personnel trajectory data, the construction environment meteorological data, and the construction geological data.

[0046] A second determination submodule is configured to determine a first spatial Euclidean distance between the construction mechanical equipment state data, the construction personnel trajectory data, the construction environment meteorological data, and the construction geological data based on first generation spatial positions of the construction mechanical equipment state data, the construction personnel trajectory data, the construction environment meteorological data, and the construction geological data.

[0047] A third determination submodule is configured to determine a first data mapping with spatio-temporal correlation based on the first generation time difference and the first spatial Euclidean distance in combination with the first construction state data.

[0048] A data fusion submodule is configured to fuse each of the first data mappings in a case where the same construction mechanical equipment state data, construction personnel trajectory data, construction environment meteorological data, or construction geological data is contained in each of the first data mappings to obtain a plurality of first construction spatio-temporal data groups.

[0049] Optionally, the third determination submodule includes:

[0050] The first generation unit is configured to generate a first data mapping between two first construction state data corresponding to the first generation time difference, in a case that the first generation time difference is less than or equal to a first time threshold, and a first spatial Euclidean distance between the two data corresponding to the first generation time difference is less than or equal to a first distance threshold.

[0051] Optionally, the third determination sub-module comprises:

[0052] The calculation unit is configured to calculate a first continuous correlation degree between the construction machinery equipment state data, the construction personnel trajectory data, the construction environment meteorological data and the construction geological data, based on the first generation time difference and a first spatial Euclidean distance between the two data corresponding to the first generation time difference.

[0053] The second generation unit is configured to generate a first data mapping between two first construction state data corresponding to the first continuous correlation degree, in a case that the first continuous correlation degree is greater than or equal to a first threshold.

[0054] Optionally, the space-time correlation module comprises:

[0055] The fourth determination sub-module is configured to determine a data acquisition frequency at a first time, in a case that the first construction state data has data missing at the first time.

[0056] The data completion sub-module is configured to complete the missing data at the first time based on the data acquisition frequency, to obtain second construction state data.

[0057] The space-time correlation sub-module is configured to perform space-time correlation on the second construction state data based on a first generation time stamp and a first generation spatial position of the second construction state data, to obtain a plurality of first construction space-time data groups.

[0058] Optionally, the alarm prompt information is a digital twin image for marking a construction risk source with risk using a high-risk area identifier.

[0059] Optionally, the apparatus further comprises:

[0060] The second acquisition module is configured to acquire a construction progress of the target construction area, in a case that the first risk identification result indicates that there is no risk.

[0061] The second generation module is configured to generate construction progress warning information and display the construction progress warning information, in a case that the construction progress is less than a planned progress.

[0062] In a third aspect, an electronic device is provided, which includes a memory, a processor, and a computer program stored in the memory. The processor executes the computer program to implement the construction early warning method based on multi-source data spatio-temporal correlation according to any one of the above.

[0063] In a fourth aspect, a readable storage medium is provided, which stores a program or instructions. When the program or instructions are executed by a processor, the construction early warning method based on multi-source data spatio-temporal correlation according to any one of the above is implemented.

[0064] Specific beneficial effects are that:

[0065] The embodiments of the present application obtain first construction state data of a target construction area. The first construction state data includes construction mechanical equipment state data, construction personnel trajectory data, construction environment meteorological data, and construction geological data. Based on a first generation timestamp and a first generation spatial position of the first construction state data, the first construction state data is spatio-temporally correlated to obtain a plurality of first construction spatio-temporal data groups. Based on the first construction spatio-temporal data groups, a key construction risk source of the target construction area is identified to obtain a first risk identification result. In the case that the first risk identification result is that there is a risk, an alarm prompt information is generated and the alarm prompt information is displayed in the target construction area. The correlation between the above multi-source heterogeneous data can be established according to the spatio-temporal correlation of the multi-source heterogeneous data of the construction site, and the risk identification is performed according to the correlated construction spatio-temporal data groups, which can improve the instantaneity and reliability of the construction early warning to a certain extent. In addition, by introducing the digital twin image prediction method based on the spatio-temporal prediction neural network, the predictive risk identification of the future construction state can be realized, the early warning information is generated, and the foresight and initiative of the construction safety control are significantly improved. BRIEF DESCRIPTION OF DRAWINGS

[0066] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the description of the embodiments of the present application will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0067] Figure 1 is a flowchart of a construction early warning method based on multi-source data spatio-temporal correlation provided by the embodiments of the present application;

[0068] Figure 2 is a flowchart of another construction early warning method based on multi-source data spatio-temporal correlation provided by the embodiments of the present application;

[0069] Figure 3 is a logic block diagram of a construction early warning device based on multi-source data space-time association provided by an embodiment of the present application.

[0070] Figure 4 is a schematic diagram of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0071] Exemplary embodiments of the present application will be described in greater detail below with reference to the accompanying drawings. Although exemplary embodiments of the present application are shown in the drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided so that the present application can be more thoroughly understood and so that the scope of the present application can be completely conveyed to those skilled in the art.

[0072] Referring to Figure 1 , Figure 1 is a flowchart of a construction early warning method based on multi-source data space-time association provided by an embodiment of the present application. The method can include:

[0073] Step 101, acquiring first construction state data of a target construction area; the first construction state data includes construction mechanical equipment state data, construction personnel trajectory data, construction environment meteorological data, and construction geological data.

[0074] In an embodiment of the present application, the first construction state data of the target construction area can include construction mechanical equipment state data, construction personnel trajectory data, construction environment meteorological data, and construction geological data, and other heterogeneous data. The construction mechanical equipment state data can be obtained through the equipment operation log, the construction personnel trajectory data can be obtained through the positioner, the construction environment meteorological data can be determined according to the current area weather data or obtained according to the sensor (such as temperature, humidity, dust sensor), and the construction geological data can be obtained by geological analysis of the target construction area before construction. In this way, the first construction state data of the target construction area can be obtained. After obtaining the first construction state data, the first construction state data can be preprocessed, and the obviously redundant data (i.e. there is no other type of data at the moment of generation of this data), and the obviously erroneous data can be deleted.

[0075] Step 102, based on the first generation time stamp and the first generation space position of the first construction state data, the first construction state data is space-time associated to obtain a plurality of first construction space-time data groups.

[0076] In the embodiments of the present application, each first construction state data can have a first generation time stamp and a first generation space position as metadata of the first construction state data and be hard associated with the first construction state data. The first construction state data can be spatio-temporally associated according to the first generation time stamp and the first generation space position, so that a plurality of first construction spatio-temporal data groups can be obtained. For example, the first construction state data corresponding to the first generation time stamp can be associated in the case that the first generation time stamp is similar and the first generation space position is close. The spatio-temporal association degree can be calculated through the first generation time stamp and the first generation space position, so that whether the corresponding data has spatio-temporal association can be determined according to the spatio-temporal association degree.

[0077] In step 103, a key construction risk source of the target construction area is identified based on the first construction spatio-temporal data group, and a first risk identification result is obtained.

[0078] In the embodiments of the present application, the key construction risk source of the target construction area can be identified according to the first construction spatio-temporal data group, so that the first risk identification result is obtained. The risk identification can be performed according to the values of the first spatio-temporal data group. If more than half of the first construction state data included in the first construction spatio-temporal data group corresponding to the key construction risk source deviates from the normal value by a large range, it can be considered that the key construction risk source has a risk. Or, if the value change of the continuous two first construction spatio-temporal data groups corresponding to the key construction risk source exceeds a preset range, it can be considered that the key construction risk source has a risk. The first risk identification result includes two results of having a risk and not having a risk.

[0079] In step 104, an alarm prompt information is generated and displayed in the target construction area in the case that the first risk identification result is having a risk.

[0080] In the embodiments of the present application, the alarm prompt information can be generated and displayed in the target construction area in the case that the first risk identification result is having a risk. The alarm prompt information can be in the form of voice, text, image, animation, etc.

[0081] In the embodiment of the present application, by acquiring first construction state data of a target construction area; the first construction state data includes construction mechanical equipment state data, construction personnel trajectory data, construction environment meteorological data and construction geological data, based on the first generation timestamp and the first generation space position of the first construction state data, the first construction state data is spatio-temporally associated to obtain a plurality of first construction space-time data groups, based on the first construction space-time data group, a key construction risk source of the target construction area is identified to obtain a first risk identification result, in the case that the first risk identification result is that there is a risk, an alarm prompt information is generated and the alarm prompt information is displayed in the target construction area, the correlation between the above multi-source heterogeneous data can be established according to the spatio-temporal correlation of the multi-source heterogeneous data of the construction site, and the risk identification is performed according to the associated construction space-time data group, which can improve the instantaneity and reliability of the construction early warning to a certain extent.

[0082] Referring to Figure 2 , Figure 2 The flowchart of another construction early warning method based on multi-source data spatio-temporal correlation provided by the embodiment of the present application can include:

[0083] Step 201, acquiring first construction state data of a target construction area; the first construction state data includes construction mechanical equipment state data, construction personnel trajectory data, construction environment meteorological data and construction geological data.

[0084] In the embodiment of the present application, the implementation content of this step can refer to the embodiment content of step 101, which will not be repeated here.

[0085] Step 202, based on the first generation timestamp and the first generation space position of the first construction state data, the first construction state data is spatio-temporally associated to obtain a plurality of first construction space-time data groups.

[0086] In the embodiment of the present application, the implementation content of this step can refer to the embodiment content of step 102, which will not be repeated here.

[0087] Optionally, in step 202, the following substep can be included:

[0088] Substep 2021, in the case that the first construction state data has data missing at a first time, determining the data acquisition frequency at the first time.

[0089] In the embodiments of the present application, since any signal collection time point can correspond to multiple different first construction state data, and the collection sources of each type of construction state data are usually different, the first construction state data at the same time point can be missing. Whether the first construction state data is missing can be determined according to the signal collection time point. If the first construction state data is missing at the first time point, the data collection frequency at the first time point can be determined. The data collection frequency can be the data collection frequency corresponding to the data type of the missing data.

[0090] In the sub-step 2022, the missing data at the first time point is filled based on the data collection frequency, and second construction state data is obtained.

[0091] In the embodiments of the present application, if the data collection frequency is greater than or equal to the preset frequency, the long short-term memory network can be used to predict the missing construction state data based on the existing construction state data, so as to fill the missing data at the first time point; if the data collection frequency is less than the preset frequency, the cubic spline interpolation method can be used to fill the missing data. After filling the data, the second construction state data can be obtained.

[0092] In the sub-step 2023, the second construction state data is spatio-temporally associated based on the first generation time stamp and the first generation space position of the second construction state data, and a plurality of first construction spatio-temporal data groups are obtained.

[0093] In the embodiments of the present application, the second construction state data after data filling can be spatio-temporally associated based on the first generation time stamp and the first generation space position of the second construction state data, and a plurality of first construction spatio-temporal data groups are obtained. The specific implementation content of spatio-temporal association can refer to the embodiment content of step 102 about spatio-temporal association, which will not be described here.

[0094] In the embodiments of the present application, by determining the data collection frequency at the first time point when the first construction state data is missing at the first time point, filling the missing data at the first time point based on the data collection frequency, obtaining the second construction state data, and spatio-temporally associating the second construction state data based on the first generation time stamp and the first generation space position of the second construction state data, a plurality of first construction spatio-temporal data groups are obtained. In the case that the first construction state data is missing, the first construction state data can be filled, and the first construction spatio-temporal data groups can be obtained based on the second construction state data after filling, which can improve the data integrity and accuracy of the first construction spatio-temporal data groups to a certain extent.

[0095] Sub-step 2024, based on the first generation time stamp of the construction machinery equipment state data, the construction personnel trajectory data, the construction environment meteorological data and the construction geological data, determine the first generation time difference between the construction machinery equipment state data, the construction personnel trajectory data, the construction environment meteorological data and the construction geological data.

[0096] In the embodiment of the application, the first generation time difference between the construction machinery equipment state data, the construction personnel trajectory data, the construction environment meteorological data and the construction geological data can be determined according to the first generation time stamp of the construction machinery equipment state data, the construction personnel trajectory data, the construction environment meteorological data and the construction geological data, and the calculation method is shown in the following formula 1:

[0097] (Formula 1);

[0098] In formula 1, represents the first generation time difference, and is the first generation time stamp corresponding to any two data in the construction machinery equipment state data, the construction personnel trajectory data, the construction environment meteorological data and the construction geological data, respectively. Generally, the first generation time difference can only be calculated between different data types.

[0099] Sub-step 2025, based on the first generation space position of the construction machinery equipment state data, the construction personnel trajectory data, the construction environment meteorological data and the construction geological data, determine the first space Euclidean distance between the construction machinery equipment state data, the construction personnel trajectory data, the construction environment meteorological data and the construction geological data.

[0100] In the embodiment of the application, the first space Euclidean distance between the construction machinery equipment state data, the construction personnel trajectory data, the construction environment meteorological data and the construction geological data can be calculated according to the first generation space position of the construction machinery equipment state data, the construction personnel trajectory data, the construction environment meteorological data and the construction geological data, and the calculation method is shown in the following formula 2:

[0101] (Formula 2);

[0102] In formula 2, is the first space Euclidean distance between the data and the data in the construction machinery equipment state data, the construction personnel trajectory data, the construction environment meteorological data and the construction geological data , the first generation space position of the data is , and the first generation space position of the data Generally, the first spatial Euclidean distance can be defined to be calculated only between different data types.

[0103] In substep 2026, based on the first generation time difference and the first spatial Euclidean distance, the first data mapping with spatio-temporal correlation is determined in combination with the first construction state data.

[0104] In the embodiments of the present application, the first data mapping with spatio-temporal correlation can be determined in combination with the first construction state data according to the first generation time difference and the first spatial Euclidean distance. That is, when the first generation time difference is less than a preset time length and the first spatial Euclidean distance is less than a preset distance, the two first construction state data corresponding to the first generation time difference or the first spatial Euclidean distance can be associated to obtain the first data mapping.

[0105] Optionally, in substep 2026, the following substeps can be included:

[0106] In substep A1, when the first generation time difference is less than or equal to a first time threshold value and the first spatial Euclidean distance between the two data corresponding to the first generation time difference is less than or equal to a first distance threshold value, the first data mapping between the two first construction state data corresponding to the first generation time difference is generated.

[0107] In the embodiments of the present application, if the first generation time difference is less than or equal to a first time threshold value and the first spatial Euclidean distance between the two data corresponding to the first generation time difference is less than or equal to a first distance threshold value, the first data mapping between the two first construction state data corresponding to the first generation time difference can be generated.

[0108] In the embodiments of the present application, by generating the first data mapping between the two first construction state data corresponding to the first generation time difference when the first generation time difference is less than or equal to a first time threshold value and the first spatial Euclidean distance between the two data corresponding to the first generation time difference is less than or equal to a first distance threshold value, the first data mapping of the self-check of the two first construction state data can be generated in a threshold judgment manner, which can improve the data accuracy of the first data mapping to a certain extent.

[0109] In substep A2, based on the first generation time difference and the first spatial Euclidean distance between the two data corresponding to the first generation time difference, the first continuous correlation degree between the construction machinery equipment state data, the construction personnel trajectory data, the construction environment meteorological data and the construction geological data is calculated.

[0110] In the embodiments of the present application, the continuous correlation degree is defined as the degree of spatio-temporal correlation of two data. The continuous correlation degree function is shown in the following formula 3:

[0111] (Formula 3);

[0112] In the above Formula 3, is a first continuous correlation degree, is a time weight factor, is a time attenuation coefficient, is a space attenuation coefficient, is a first generation time difference, is a first spatial Euclidean distance between two data corresponding to the first generation time difference. According to Formula 3, the first continuous correlation degree between the construction mechanical equipment state data, the construction personnel trajectory data, the construction environment meteorological data and the construction geological data can be calculated in combination with the first generation time difference and the first spatial Euclidean distance between two data corresponding to the first generation time difference.

[0113] Sub-step A3, in a case where the first continuous correlation degree is greater than or equal to a first threshold value, generating a first data mapping between two first construction state data corresponding to the first continuous correlation degree.

[0114] In the embodiments of the present application, if the calculated first continuous correlation degree is greater than or equal to the first threshold value, a first data mapping between two first construction state data corresponding to the first continuous correlation degree can be generated.

[0115] In the embodiments of the present application, by calculating the first continuous correlation degree between the construction mechanical equipment state data, the construction personnel trajectory data, the construction environment meteorological data and the construction geological data based on the first generation time difference and the first spatial Euclidean distance between two data corresponding to the first generation time difference, and in a case where the first continuous correlation degree is greater than or equal to the first threshold value, generating a first data mapping between two first construction state data corresponding to the first continuous correlation degree, the continuous correlation degree can be used as a judgment index for spatio-temporal correlation, and the data accuracy of the first data mapping can be improved to a certain extent.

[0116] Sub-step 2027, in a case where each of the first data mappings contains the same construction mechanical equipment state data, construction personnel trajectory data, construction environment meteorological data or construction geological data, fusing each of the first data mappings to obtain a plurality of first construction spatio-temporal data groups.

[0117] In the embodiments of the present application, the method of data matching can be used to determine whether the same construction machinery equipment state data, construction personnel trajectory data, construction environment meteorological data or construction geological data are contained in each first data mapping. If yes, the first data mappings containing the same construction machinery equipment state data, construction personnel trajectory data, construction environment meteorological data or construction geological data can be fused to obtain a plurality of first construction space-time data groups. Only two types of first construction state data are contained in the first data mappings, after one fusion, data mappings containing three types of first construction state data can be obtained, and after two fusions, a first construction space-time data group containing four types of first construction state data can be obtained.

[0118] In the embodiments of the present application, the method described in sub-steps 2024 to 2027 and their sub-steps can be used as an embodiment of sub-step 2023 to perform space-time association on the second construction state data by using the same method.

[0119] In the embodiments of the present application, by determining the first generation time difference between the construction machinery equipment state data, the construction personnel trajectory data, the construction environment meteorological data and the construction geological data based on the first generation time stamp of the construction machinery equipment state data, the construction personnel trajectory data, the construction environment meteorological data and the construction geological data, determining the first spatial Euclidean distance between the construction machinery equipment state data, the construction personnel trajectory data, the construction environment meteorological data and the construction geological data based on the first generation spatial position of the construction machinery equipment state data, the construction personnel trajectory data, the construction environment meteorological data and the construction geological data, determining the first data mapping with space-time association based on the first generation time difference and the first spatial Euclidean distance combined with the first construction state data, in the case that the same construction machinery equipment state data, construction personnel trajectory data, construction environment meteorological data or construction geological data are contained in each first data mapping, fusing each first data mapping to obtain a plurality of first construction space-time data groups, the first data mapping can be generated according to the first generation time difference and the first spatial Euclidean distance between each construction state data, and then the first construction space-time data group is generated by using the data fusion method, which can improve the data accuracy and reliability of the first construction space-time data group to a certain extent.

[0120] Step 203, based on the first construction space-time data group, risk identification is performed on the key construction risk source of the target construction area to obtain a first risk identification result.

[0121] In the embodiments of the present application, the implementation content of the present step can refer to the implementation content of the embodiment of step 103, which will not be described here.

[0122] Optionally, in step 203, the following sub-steps can be included:

[0123] In substep 2031, based on the first construction spatiotemporal data set, a digital twin image sequence of the target construction area in a continuous historical time period is generated. wherein, represents the i th time point before the time point .

[0124] In substep 2032, the digital twin image sequence is input into a pre-trained spatiotemporal prediction neural network model, the digital twin image sequence is processed by the spatiotemporal prediction neural network model, and a predicted digital twin image at a future time point is output. .

[0125] In substep 2033, the predicted digital twin image is processed by a semantic segmentation model to obtain a pixel-level construction element classification map, the construction elements including personnel, mechanical equipment, and high-risk areas.

[0126] In substep 2034, the construction element classification map is logically matched with a pre-set safety rule library, and if the matching result violates the safety rule, warning information corresponding to the future time point is generated.

[0127] In an embodiment of the present application, the safety rule library can include:

[0128] Mechanical collision risk: simulating that a person suddenly approaches within the swinging range of a tower crane load.

[0129] Area intrusion risk: simulating that a person enters a dynamically changing high-risk operation area (such as a pouring area).

[0130] Environment-derived risk: simulating that the wind speed suddenly increases during a specific work procedure (such as high-altitude operation).

[0131] Equipment failure risk: simulating that the pressure of a hydraulic pump slowly abnormally rises to a critical value.

[0132] Optionally, the spatiotemporal prediction neural network model is a PredRNN network; the PredRNN network models the spatiotemporal evolution features in the digital twin image sequence through a spatiotemporal long short-term memory unit (ST-LSTM) and a cross-layer bidirectional recursive structure contained therein; and a training target of the PredRNN network is guided by a composite loss function, the composite loss function including a reconstruction loss between predicted image pixels and real image pixels, and a perception loss calculated based on feature map differences of a feature extraction network.

[0133] In an embodiment of the present application, the composite loss function can be represented as wherein​​L represents a composite loss function, Table represents a reconstruction loss, represents a perceptual loss, and is a balance coefficient. The spatio-temporal prediction neural network model can be a PredRNN network; the PredRNN network can model the spatio-temporal evolution characteristics in the digital twin image sequence through the spatio-temporal long short-term memory unit (ST-LSTM) and the cross-layer bidirectional recursive structure contained therein, which can improve the prediction performance of the spatio-temporal prediction neural network model, thereby improving the stability of the predicted digital twin image to a certain extent.

[0134] In the embodiments of the present application, by generating a digital twin image sequence of the target construction area in a continuous historical time period based on the first construction spatio-temporal data set, inputting the digital twin image sequence into a pre-trained spatio-temporal prediction neural network model, processing the digital twin image sequence by the spatio-temporal prediction neural network model, outputting a predicted digital twin image at a future time, processing the predicted digital twin image by a semantic segmentation model to obtain a pixel-level construction element classification map, the construction elements including personnel, mechanical equipment, and high-risk areas, logically matching the construction element classification map with a pre-set safety rule library, if the matching result violates the safety rule, generating early warning information corresponding to the future time, the predictive risk identification of the construction state at the future time can be realized, and the early warning information is generated, which significantly improves the forward-looking and initiative of construction safety control.

[0135] Step 204, in the case that the first risk identification result is that there is a risk, generating an alarm prompt information and displaying the alarm prompt information in the target construction area.

[0136] In the embodiments of the present application, the implementation content of this step can refer to the embodiment content of step 104, which will not be described here.

[0137] Optionally, the alarm prompt information is a digital twin image in which a high-risk area identifier is used to mark the construction risk source with a risk.

[0138] In the embodiments of the present application, the digital twin image can be used as the alarm prompt information. The digital twin image can be a three-dimensional virtual image corresponding to the target construction area. When the key risk control source is at risk, the key risk control source at risk can be marked in red, and the red color can be displayed in a flashing manner. At the same time, a voice prompt can be given to prompt the construction personnel that there is a construction risk in the target construction area. This way can improve the intuitiveness and referenceability of the alarm prompt information to a certain extent, and facilitate the construction personnel to determine the risk.

[0139] Step 205, in the case that the first risk identification result is no risk, obtaining the construction progress of the target construction area.

[0140] In the embodiment of the present application, if the first risk identification result is no risk, the construction progress of the target construction area can be calculated by the coincidence degree of the model of the current construction body and the construction drawing.

[0141] Step 206, in the case that the construction progress is less than the planned progress, generating construction progress warning information and displaying the construction progress warning information.

[0142] In the embodiment of the present application, if the construction progress is less than the planned progress, construction progress warning information can be generated and displayed. The construction progress warning has less harm to personnel, and a relatively mild warning method can be adopted, for example, using red flashing text, or a construction progress prompt pop-up window information with a confirmation button.

[0143] In the embodiment of the present application, by obtaining the construction progress of the target construction area in the case that the first risk identification result is no risk, and generating construction progress warning information and displaying the construction progress warning information in the case that the construction progress is less than the planned progress, the construction progress can be warned and identified in the case that there is no construction risk in the target construction area, which facilitates the construction personnel to adjust the construction plan in time according to the construction progress warning, and can improve the stability of the construction progress to a certain extent.

[0144] In order to comprehensively and objectively verify the effectiveness, reliability and advancement of the embodiments of the present application, we designed a strict experiment and carried out simulation and field test in the construction process (6 months) of a large-scale cross-river bridge engineering project.

[0145] 1. Experimental setup:

[0146] Data sources: Real-time state data (oil pressure, speed, GPS position) of 10 key construction machines (such as tower crane, pump truck), positioning trajectory data of 50 construction personnel, data of 5 weather stations (wind speed, wind direction, humidity, precipitation) deployed in the construction site and real-time data of geological settlement monitoring points.

[0147] Data preprocessing: The original data was cleaned, including processing of data missing caused by sensor communication interruption, eliminating abnormal jump values caused by signal interference, and time synchronization alignment of multi-source data to the same timestamp.

[0148] Hardware and software platform: deployed on a cloud virtual server, configured with Intel Xeon Gold 6248R CPU, NVIDIA Tesla V100 GPU, and 128GB RAM.

[0149] Software environment: operating system Ubuntu 18.04, development language Python 3.8, deep learning framework PyTorch 1.9, and digital twin engine Unity 3D 2020 LTS.

[0150] 2. Experimental methods and evaluation indicators:

[0151] Test scenarios: covering three typical high-risk stages of bridge construction, including foundation excavation, pier pouring, and beam erection.

[0152] Risk event injection: in cooperation with domain experts, four categories of 20 representative risk scenarios were defined for testing:

[0153] Mechanical collision risk: simulating personnel suddenly approaching within the swing range of the tower crane.

[0154] Regional intrusion risk: simulating personnel entering the dynamically changing high-risk operation area (such as pouring area).

[0155] Environment-derived risk: simulating sudden increase in wind speed during specific processes (such as high-altitude operation).

[0156] Equipment failure risk: simulating slow abnormal increase of hydraulic pump pressure to critical value.

[0157] Evaluation indicators:

[0158] Risk identification accuracy: including precision and F1-Score.

[0159] Timeliness of early warning: average time delay from risk condition establishment to system alarm.

[0160] False alarm rate: frequency of system false alarms in total 500 hours of normal construction data.

[0161] Advanced warning capability: proportion and average advance time of successfully predicted future risk events.

[0162] System performance: average time consumption (ms) and CPU / GPU occupancy rate of model single inference.

[0163] 3. Experimental results:

[0164] Table 1: Comparison of early warning effects

[0165] ;

[0166] 4. Result analysis:

[0167] According to the above table 1, the F1 score of the early warning method of the application is 96.3 (average), which is higher than that of the traditional method and the multi-sensor alarm method, and the average early warning delay is 2.8 minutes, which is lower than that of the traditional method and the multi-sensor alarm method. In terms of false alarm rate, the false alarm rate of the early warning method of the application is 4.5%, which is significantly lower than 19.2% of the traditional method and 11.8% of the multi-sensor alarm method. In terms of advanced early warning success rate, the traditional method and the multi-sensor alarm method cannot perform advanced early warning, while the early warning method of the application can realize advanced early warning with a success rate of 85%. Due to the use of more refined and intelligent early warning method, the upper limit of the time consumption of the early warning method of the application is 500ms, which is higher than that of the traditional method and the multi-sensor alarm method, but considering the overall length of the early warning object, the slight increase in reasoning time will not affect the overall superiority of the application.

[0168] The experimental data fully show that:

[0169] Effectiveness: The application (method A) is significantly better than the comparative method in various core indicators. The high F1-Score proves that the multi-source spatio-temporal correlation model can more comprehensively capture risk characteristics and significantly reduce false negatives and false positives.

[0170] Prospective: The unique advanced early warning function provides an average decision window period of more than 12 minutes for active safety control, which is not achieved by the traditional method.

[0171] Practicality: Although the model complexity is higher, the average reasoning time is controlled within 500 milliseconds, fully meeting the real-time requirements of the construction site, and proving the engineering application value of the application.

[0172] Robustness: In different stages of bridge construction, the application maintains stable high performance and exhibits good environmental adaptability.

[0173] Reference Figure 3 , Figure 3 A logic block diagram of a construction early warning device based on multi-source data spatio-temporal correlation provided by the embodiment of the application, the construction early warning device based on multi-source data spatio-temporal correlation 300 can include:

[0174] The first acquisition module 301 is configured to acquire first construction state data of a target construction area; the first construction state data includes construction mechanical equipment state data, construction personnel trajectory data, construction environment meteorological data and construction geological data;

[0175] The space-time correlation module 302 is configured to perform space-time correlation on the first construction state data based on a first generation time stamp and a first generation space position of the first construction state data, to obtain a plurality of first construction space-time data groups.

[0176] The risk identification module 303 is configured to perform risk identification on a key construction risk source of the target construction area based on the first construction space-time data group, to obtain a first risk identification result.

[0177] The first generation module 304 is configured to generate an alarm prompt information and display the alarm prompt information in the target construction area in a case where the first risk identification result is that there is a risk.

[0178] Optionally, the space-time correlation module 302 comprises:

[0179] The first determination sub-module is configured to determine a first generation time difference between the construction mechanical equipment state data, the construction personnel trajectory data, the construction environment meteorological data and the construction geological data based on first generation time stamps of the construction mechanical equipment state data, the construction personnel trajectory data, the construction environment meteorological data and the construction geological data.

[0180] The second determination sub-module is configured to determine a first space Euclidean distance between the construction mechanical equipment state data, the construction personnel trajectory data, the construction environment meteorological data and the construction geological data based on first generation space positions of the construction mechanical equipment state data, the construction personnel trajectory data, the construction environment meteorological data and the construction geological data.

[0181] The third determination sub-module is configured to determine a first data mapping with space-time correlation based on the first generation time difference and the first space Euclidean distance in combination with the first construction state data.

[0182] The data fusion sub-module is configured to fuse each of the first data mappings in a case where the same construction mechanical equipment state data, construction personnel trajectory data, construction environment meteorological data or construction geological data is contained between the first data mappings, to obtain a plurality of first construction space-time data groups.

[0183] Optionally, the third determination sub-module comprises:

[0184] The first generation unit is configured to generate a first data mapping between two first construction state data corresponding to the first generation time difference in a case where the first generation time difference is less than or equal to a first time threshold value, and a first space Euclidean distance between the two data corresponding to the first generation time difference is less than or equal to a first distance threshold value.

[0185] Optionally, the third determining sub-module comprises:

[0186] a calculating unit configured to calculate a first continuous correlation degree between the construction machinery equipment state data, the construction personnel trajectory data, the construction environment meteorological data and the construction geological data based on the first generated time difference and a first spatial Euclidean distance between two data corresponding to the first generated time difference;

[0187] a second generating unit configured to generate a first data mapping between two first construction state data corresponding to the first continuous correlation degree in a case where the first continuous correlation degree is greater than or equal to a first threshold.

[0188] Optionally, the space-time correlation module 302 comprises:

[0189] a fourth determining sub-module configured to determine a data acquisition frequency at a first time in a case where the first construction state data at the first time has data missing;

[0190] a data completion sub-module configured to complete the missing data at the first time based on the data acquisition frequency to obtain second construction state data;

[0191] a space-time correlation sub-module configured to perform space-time correlation on the second construction state data based on a first generation time stamp and a first generation spatial position of the second construction state data to obtain a plurality of first construction space-time data groups.

[0192] Optionally, the alarm prompt information is a digital twin image for marking a construction risk source with risk using a high-risk area identifier.

[0193] Optionally, the construction early warning device 300 based on space-time correlation of multi-source data further comprises:

[0194] a second acquiring module configured to acquire a construction progress of the target construction area in a case where the first risk identification result is that there is no risk;

[0195] a second generating module configured to generate construction progress early warning information and display the construction progress early warning information in a case where the construction progress is less than a planned progress.

[0196] The construction early warning device based on multi-source data space-time correlation in the embodiments of the present application can be an electronic device, or a component in an electronic device, such as an integrated circuit or a chip. The electronic device can be a terminal, or other devices other than a terminal. For example, the electronic device can be a mobile phone, a tablet computer, a notebook computer, a palm computer, a vehicle-mounted electronic device, a Mobile Internet Device (MID), an augmented reality (AR) / virtual reality (VR) device, a robot, a wearable device, an ultra-mobile personal computer (UMPC), a netbook, or a personal digital assistant (PDA), and can also be a server, a Network Attached Storage (NAS), a personal computer (PC), and the like. The embodiments of the present application are not limited in this regard.

[0197] The construction early warning device based on multi-source data space-time correlation in the embodiments of the present application can be a device with an operating system. The operating system can be an Android operating system, a Linux operating system, a Windows operating system, or other possible operating systems. The embodiments of the present application are not limited in this regard.

[0198] The construction early warning device based on multi-source data space-time correlation provided in the embodiments of the present application can achieve the method embodiments Figures 1 to 2 The processes implemented by the method embodiments are not repeated here to avoid repetition.

[0199] The embodiments of the present application provide an electronic device, referring to Figure 4 The electronic device 40 includes a processor 401, a memory 402, and a computer program 4021 stored in the memory 402 and executable on the processor 401. When the processor 401 executes the program, the construction early warning method based on multi-source data space-time correlation in the foregoing embodiments is implemented.

[0200] The embodiments of the present application also provide a computer readable storage medium having a computer program / instruction stored thereon. When the computer program / instruction is executed by a processor, the steps in the construction early warning method based on multi-source data space-time correlation disclosed in the embodiments of the present application are implemented.

[0201] The embodiment of the present application further provides a computer program product, which, when running on an electronic device, causes a processor to implement the steps in the construction early warning method based on multi-source data space-time association as disclosed in the embodiment of the present application.

[0202] Each of the embodiments in the specification is described in a progressive manner, and each embodiment focuses on the difference from other embodiments, and the same or similar parts between the embodiments can be referred to each other.

[0203] The embodiments of the present application are described with reference to flowcharts and / or block diagrams of the methods, devices, electronic devices and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of the flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing terminal equipment to produce a machine, so that the instructions executed by the computer or other programmable data processing terminal equipment produce a device for implementing the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in one flow or multiple flows and / or blocks Figure 1 The functions specified in one flow or multiple flows and / or blocks

[0204] These computer program instructions can also be stored in a computer readable memory that can guide the computer or other programmable data processing terminal equipment to work in a specific way, so that the instructions stored in the computer readable memory produce a product including instruction devices, which implement the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in one flow or multiple flows and / or blocks Figure 1 The functions specified in one flow or multiple flows and / or blocks

[0205] These computer program instructions can also be loaded into a computer or other programmable data processing terminal equipment, so that a series of operation steps are performed on the computer or other programmable terminal equipment to produce a computer implemented process, so that the instructions executed on the computer or other programmable terminal equipment provide a process for implementing the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in one flow or multiple flows and / or blocks Figure 1 The functions specified in one flow or multiple flows and / or blocks

[0206] Although the preferred embodiments of the present application have been described, those skilled in the art can make additional changes and modifications to the embodiments once they know the basic inventive concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the embodiments of the present application.

[0207] Finally, it is to be understood that the phraseology or terminology such as "first" and "second" etc. used herein is merely intended to differentiate one entity or operation from another entity or operation, without necessarily requiring or implying any actual such relationship or order between such entities or operations. Moreover, the terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus. An element proceeded by "comprises... a" does not, without more constraints, exclude the existence of additional identical elements in the process, method, article, or apparatus that comprises the element.

[0208] The above describes in detail the construction early warning method and device based on multi-source data space-time correlation provided by the present application. The principles and implementation manners of the present application are described by using specific examples. The above example is only used to help understand the method and core idea of the present application. Meanwhile, for those skilled in the art, the specific implementation manners and application regions can be changed according to the idea of the present application. In conclusion, the content of the specification should not be understood as a limitation of the present application.

Claims

1. A construction early warning method based on spatiotemporal correlation of multi-source data, characterized in that, The method includes: Obtain the first construction status data of the target construction area; the first construction status data includes construction machinery and equipment status data, construction personnel trajectory data, construction environment meteorological data, and construction geological data; Based on the first generation timestamp and the first generation spatial location of the first construction status data, the first construction status data is spatiotemporally correlated to obtain multiple first construction spatiotemporal data groups; Based on the first construction spatiotemporal data set, key construction risk sources in the target construction area are identified to obtain a first risk identification result; including: Based on the first construction spatiotemporal data set, a digital twin image sequence of the target construction area within a continuous historical time period is generated; The digital twin image sequence is input into a pre-trained spatiotemporal prediction neural network model, which processes the digital twin image sequence and outputs a predicted digital twin image for future times. The predicted digital twin image is processed using a semantic segmentation model to obtain a pixel-level construction element classification map, wherein the construction elements include personnel, machinery and equipment, and high-risk areas; The construction element classification map is logically matched with a pre-set safety rule base. If the matching result violates the safety rules, a warning message corresponding to a future time is generated. The spatiotemporal prediction neural network model is a PredRNN network; the PredRNN network models the spatiotemporal evolution features in the digital twin image sequence through its internal spatiotemporal long short-term memory units and cross-layer bidirectional recursive structure; the training objective of the PredRNN network is guided by a composite loss function, which includes the reconstruction loss between predicted image pixels and real image pixels, and the perceptual loss calculated based on the feature map difference of the feature extraction network; If the first risk identification result indicates the existence of a risk, an alarm message is generated and displayed in the target construction area.

2. The method according to claim 1, characterized in that, Based on the first generation timestamp and the first generation spatial location of the first construction status data, spatiotemporal correlation is performed on the first construction status data to obtain multiple first construction spatiotemporal data groups, including: Based on the first generation timestamps of the construction machinery and equipment status data, the construction personnel trajectory data, the construction environment meteorological data, and the construction geological data, a first generation time difference is determined between the construction machinery and equipment status data, the construction personnel trajectory data, the construction environment meteorological data, and the construction geological data; Based on the first generated spatial location of the construction machinery and equipment status data, the construction personnel trajectory data, the construction environment meteorological data, and the construction geological data, a first spatial Euclidean distance is determined between the construction machinery and equipment status data, the construction personnel trajectory data, the construction environment meteorological data, and the construction geological data; Based on the first generation time difference and the first spatial Euclidean distance, combined with the first construction status data, a first data mapping with spatiotemporal correlation is determined; If the first data mappings contain the same construction machinery and equipment status data, construction personnel trajectory data, construction environment meteorological data, or construction geological data, the first data mappings are merged to obtain multiple first construction spatiotemporal data groups.

3. The method according to claim 2, characterized in that, The step of determining a first data mapping with spatiotemporal correlation based on the first generation time difference and the first spatial Euclidean distance, combined with the first construction status data, includes: When the first generation time difference is less than or equal to the first time threshold, and the first spatial Euclidean distance between the two data corresponding to the first generation time difference is less than or equal to the first distance threshold, a first data mapping is generated between the two first construction status data corresponding to the first generation time difference.

4. The method according to claim 3, characterized in that, The step of determining a first data mapping with spatiotemporal correlation based on the first generation time difference and the first spatial Euclidean distance, combined with the first construction status data, includes: Based on the first generation time difference and the first spatial Euclidean distance between the two data corresponding to the first generation time difference, the first continuous correlation degree between the construction machinery and equipment status data, the construction personnel trajectory data, the construction environment meteorological data and the construction geological data is calculated; If the first continuous correlation degree is greater than or equal to the first threshold, a first data mapping is generated between the two first construction status data corresponding to the first continuous correlation degree.

5. The method according to claim 1, characterized in that, Based on the first generation timestamp and the first generation spatial location of the first construction state, the first construction state data is spatiotemporally correlated to obtain multiple first construction spatiotemporal data groups, including: In the case that there is a data gap in the first construction status data at the first moment, the data collection frequency at the first moment shall be determined; Based on the data acquisition frequency, the missing data at the first moment is filled in to obtain the second construction status data; Based on the first generation timestamp and the first generation spatial location of the second construction status data, the second construction status data is spatiotemporally correlated to obtain multiple first construction spatiotemporal data groups.

6. The method according to claim 1, characterized in that, The alarm message is a digital twin image that marks the construction risk sources with high-risk area identifiers.

7. The method according to claim 1, characterized in that, The method further includes: If the first risk identification result indicates that there is no risk, the construction progress of the target construction area is obtained; If the construction progress is less than the planned progress, a construction progress warning message is generated and displayed.

8. A construction early warning device based on spatiotemporal correlation of multi-source data, characterized in that, The device is implemented using the method according to any one of claims 1-7, comprising: The first acquisition module is used to acquire the first construction status data of the target construction area; the first construction status data includes construction machinery and equipment status data, construction personnel trajectory data, construction environment meteorological data, and construction geological data. The spatiotemporal correlation module is used to perform spatiotemporal correlation on the first construction status data based on the first generation timestamp and the first generation spatial location of the first construction status data to obtain multiple first construction spatiotemporal data groups; The risk identification module is used to identify key construction risk sources in the target construction area based on the first construction spatiotemporal data group, and obtain the first risk identification result; The first generation module is used to generate an alarm message and display the alarm message in the target construction area when the first risk identification result indicates that there is a risk.

Images