Building structure monitoring method and system based on multi-source data fusion

Abstract

The invention provides a building structure monitoring method and system based on multi-source data fusion, and the method comprises the steps: collecting the initial data of a building structure, employing the initial data to drive a relation logic model for the physical attributes of the building structure through a digital twin technology, and carrying out the calculation of the initial variable data, according to the method, the state of the building structure is deduced by using the digital twinning technology by obtaining the safety data in the real-time data of the building structure, and the related parameters related to the abnormal state data can be determined by analyzing the abnormal state data of the building structure, so that the safety data in the real-time data of the building structure can be obtained. According to the method, the correlation parameters and the abnormal parameters are analyzed in a combined mode, so that the cause of abnormity is determined, the monitoring model is established, the abnormity possibly occurring in the building structure can be predicted in advance through the detection model, maintenance personnel are helped to determine the maintenance direction and carry out maintenance in advance, and the building structure can be analyzed more comprehensively.

Description

Technical Field

[0001] The present invention relates to the technical field of building structure monitoring, and in particular to a building structure monitoring method and system based on multi-source data fusion. Background Art

[0002] With the acceleration of the current social development, the number of buildings in various places is increasing. The safety of building structures is the factor that people are most concerned about. For example, in construction projects, it is necessary to monitor whether high-rise buildings are safe to ensure the smooth progress of construction.

[0003] At present, the monitoring method of building structures relies on single mechanical equipment, manual inspection and evaluation operations, such as using a theodolite to measure the angle and side length of the building, and then evaluating whether the building is deformed based on these data. This method cannot effectively and accurately monitor the building structure, cannot timely discover potential safety problems of the building structure, and such methods cannot predict the danger of the building in advance. Summary of the invention

[0004] Technical issues solved

[0005] In view of the deficiencies in the prior art, the present invention provides a building structure monitoring method and system based on multi-source data fusion to solve the problems raised in the above background technology.

[0006] Technical Solution

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: A building structure monitoring method based on multi-source data fusion, comprising the following steps:

[0008] Collect and process the initial data of the building structure to obtain the initial variable information and initial state information of the building structure;

[0009] Calculating initial variable information of the building structure based on the digital twin technology, the relational logic model and the real-time monitoring data of the building structure to obtain a deduction result of the digital twin of the building structure;

[0010] Obtain real-time data of building structures and extract safety data from the real-time data, which includes normal parameters and abnormal parameters;

[0011] Obtaining a node corresponding to an abnormal parameter, marking it as an abnormal node, marking the safety data corresponding to the abnormal node as a first parameter, performing a correlation analysis on the abnormal node and the first parameter, and obtaining a correlation parameter;

[0012] The digital twin is used to deduce the abnormal parameters and associated parameters of the abnormal initial point to obtain the deduction results;

[0013] Determine a corresponding relationship set based on the correction relationship between the deduction result and the actual result of the abnormal node, determine a corresponding forward calculation model based on the physical property variables from the abnormal initial point to the abnormal node, and use the corresponding relationship set to optimize the forward calculation model to obtain a monitoring model;

[0014] Monitoring is achieved based on at least one current monitoring data and a monitoring model of the building structure to be detected.

[0015] As a preferred embodiment of this embodiment, the initial variable information of the building structure is calculated based on the digital twin technology, the relational logic model and the real-time monitoring data of the building structure to obtain the deduction result of the digital twin of the building structure, including:

[0016] Through digital twin technology, the physical properties of the building structure are driven by the initial data-driven relational logic model, and the initial variable data are calculated to obtain the deduction results of the digital twin.

[0017] As a preferred embodiment of this embodiment, the process of performing correlation analysis on the abnormal node and the first parameter to obtain the correlation parameter includes:

[0018] Taking the abnormal node as the end point, reversely collect the abnormal parameters of the building structure, and use the abnormal parameters of the abnormal node as the benchmark data;

[0019] Obtain the abnormal parameter fluctuation of the reverse time series of the benchmark data to obtain the evaluation parameters;

[0020] Preset an evaluation threshold, compare the evaluation parameter with the evaluation threshold, and when the evaluation parameter is less than or equal to the evaluation threshold for the first time, mark the corresponding time as the abnormal starting point;

[0021] Obtain all normal parameters from the abnormal starting point to the abnormal point to obtain the second parameter;

[0022] An offset calculation is performed on the second parameter to obtain a correlation parameter.

[0023] As a preferred embodiment of this embodiment, the process of performing offset calculation on the second parameter to obtain the associated parameter includes:

[0024] The abnormal node is taken as the end point, and the second parameter of the abnormal node is taken as the benchmark data;

[0025] Obtain the abnormal parameter fluctuation of the reverse time series of the benchmark data to obtain the evaluation parameters;

[0026] Preset an evaluation threshold, compare the evaluation parameter with the evaluation threshold, and when the evaluation parameter is less than or equal to the evaluation threshold for the first time, mark the corresponding time as the parameter starting point;

[0027] The offset parameter of the second parameter is calculated by the offset function:

[0028] ;

[0029] Among them, P represents the offset parameter, n represents the total amount of data of the second parameter from the starting point of the parameter to the abnormal node, and X n-1 Brother X n Represents the adjacent second parameter;

[0030] By presetting the offset threshold, comparing the deviation parameter of the second parameter with the offset threshold;

[0031] If the offset parameter is greater than or equal to the offset threshold, then the second parameter is an associated parameter;

[0032] If the offset parameter is less than the offset threshold, it means that the second parameter is not a correlation parameter.

[0033] As a preferred embodiment of this embodiment, the process of determining the corresponding relationship set based on the correction relationship between the deduction result and the actual result of the abnormal node includes:

[0034] Compare and analyze the deduction results with the actual results of the abnormal nodes, and modify the deduction results based on the comparison results;

[0035] The monitoring data corresponding to the corrected points in the deduction results and the actual results of the abnormal nodes are extracted to obtain a corresponding relationship set.

[0036] As a preferred embodiment of this embodiment, a building structure monitoring system based on multi-source data fusion is used to implement the above-mentioned building structure monitoring method based on multi-source data fusion, including:

[0037] A deduction and analysis module is used to collect and process the initial data of the building structure to obtain the initial variable information and initial state information of the building structure, and calculate the initial variable information of the building structure based on the digital twin technology, the relational logic model and the real-time monitoring data of the building structure to obtain the deduction result of the digital twin of the building structure;

[0038] The acquisition and tracing module is used to acquire the real-time data of the building structure and extract the safety data in the real-time data. The safety data includes normal parameters and abnormal parameters. The module is also used to acquire the nodes corresponding to the abnormal parameters, mark them as abnormal nodes, mark the safety data corresponding to the abnormal nodes as first parameters, and perform correlation analysis on the abnormal nodes and the first parameters to obtain the correlation parameters.

[0039] A model building module is used to use the digital twin to deduce the abnormal parameters and associated parameters of the abnormal initial point to obtain the deduction results, and determine the corresponding relationship set based on the correction relationship between the deduction results and the actual results of the abnormal nodes, determine the corresponding forward calculation model based on the physical property variables from the abnormal initial point to the abnormal node, and use the corresponding relationship set to optimize the forward calculation model to obtain the monitoring model;

[0040] The monitoring module implements monitoring according to at least one current monitoring data and a monitoring model of the building structure to be detected.

[0041] As a preferred embodiment of this embodiment, the initial variable information of the building structure is calculated based on the digital twin technology, the relational logic model and the real-time monitoring data of the building structure to obtain the deduction result of the digital twin of the building structure, including:

[0042] Through digital twin technology, the physical properties of the building structure are driven by the initial data-driven relational logic model, and the initial variable data are calculated to obtain the deduction results of the digital twin.

[0043] As a preferred embodiment of this embodiment, the process of performing correlation analysis on the abnormal node and the first parameter to obtain the correlation parameter includes:

[0044] Taking the abnormal node as the end point, reversely collect the abnormal parameters of the building structure, and use the abnormal parameters of the abnormal node as the benchmark data;

[0045] Obtain the abnormal parameter fluctuation of the reverse time series of the benchmark data to obtain the evaluation parameters;

[0046] Preset an evaluation threshold, compare the evaluation parameter with the evaluation threshold, and when the evaluation parameter is less than or equal to the evaluation threshold for the first time, mark the corresponding time as the abnormal starting point;

[0047] Obtain all normal parameters from the abnormal starting point to the abnormal point to obtain the second parameter;

[0048] An offset calculation is performed on the second parameter to obtain a correlation parameter.

[0049] As a preferred embodiment of this embodiment, the process of performing offset calculation on the second parameter to obtain the associated parameter includes:

[0050] The abnormal node is taken as the end point, and the second parameter of the abnormal node is taken as the benchmark data;

[0051] Obtain the abnormal parameter fluctuation of the reverse time series of the benchmark data to obtain the evaluation parameters;

[0052] Preset an evaluation threshold, compare the evaluation parameter with the evaluation threshold, and when the evaluation parameter is less than or equal to the evaluation threshold for the first time, mark the corresponding time as the parameter starting point;

[0053] The offset parameter of the second parameter is calculated by the offset function:

[0054] ;

[0055] Among them, P represents the offset parameter, n represents the total amount of data of the second parameter from the starting point of the parameter to the abnormal node, and X n-1 Brother X n Represents the adjacent second parameter;

[0056] By presetting the offset threshold, comparing the deviation parameter of the second parameter with the offset threshold;

[0057] If the offset parameter is greater than or equal to the offset threshold, then the second parameter is an associated parameter;

[0058] If the offset parameter is less than the offset threshold, it means that the second parameter is not a correlation parameter.

[0059] As a preferred embodiment of this embodiment, the process of determining the corresponding relationship set based on the correction relationship between the deduction result and the actual result of the abnormal node includes:

[0060] Compare and analyze the deduction results with the actual results of the abnormal nodes, and modify the deduction results based on the comparison results;

[0061] The monitoring data corresponding to the corrected points in the deduction results and the actual results of the abnormal nodes are extracted to obtain a corresponding relationship set.

[0062] Beneficial Effects

[0063] The present invention provides a building structure monitoring method and system based on multi-source data fusion, which has the following beneficial effects: by collecting the initial data of the building structure, and using the initial data to drive the relational logic model for the physical properties of the building structure through the digital twin technology, the initial variable data is calculated to obtain the deduction result of the digital twin, and at the same time, by obtaining the safety data in the real-time data of the building structure, wherein the safety data includes normal parameters and abnormal parameters, by analyzing the monitoring data corresponding to the abnormal node, the abnormal node and the abnormal starting point are determined, and then the abnormal parameters and associated parameters are determined, and finally the abnormal parameters and associated parameters of the abnormal initial point are deduced through the digital twin to obtain the deduction result, and the actual result of the abnormal node is obtained based on the deduction result and the actual result of the abnormal node. The correction relationship between them is used to determine the corresponding relationship set, the corresponding forward calculation model is determined based on the physical property variables from the abnormal initial point to the abnormal node, the forward calculation model is optimized using the corresponding relationship set to obtain a monitoring model, and the real-time monitoring of the building structure is achieved through the detection model. It can be seen that the present application uses digital twin technology to deduce the state of the building structure, and at the same time, by analyzing the abnormal state data of the building structure, it can determine the associated parameters related to it, and then combine the associated parameters and the abnormal parameters for analysis to determine the cause of the abnormality, and establish a monitoring model. The detection model can be used to predict possible abnormalities in the building structure in advance, thereby helping maintenance personnel to determine the maintenance direction and perform maintenance in advance, and the present application can conduct a more comprehensive analysis of the building structure. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 It is a flow chart of the building structure monitoring method based on multi-source data fusion of the present invention;

[0065] Figure 2 It is a block diagram of the building structure monitoring system based on multi-source data fusion of the present invention. DETAILED DESCRIPTION

[0066] The embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0067] The disclosure below provides many different embodiments or examples to realize different structures of the present invention. In order to simplify the disclosure of the present invention, the parts and settings of specific examples are described below. Of course, they are only examples, and the purpose is not to limit the present invention. In addition, the present invention can repeat reference numbers and / or reference letters in different examples, and this repetition is for the purpose of simplicity and clarity, which itself does not indicate the relationship between the various embodiments and / or settings discussed. In addition, the present invention provides various specific examples of processes and materials, but those of ordinary skill in the art can be aware of the application of other processes and / or the use of other materials.

[0068] like Figure 1 As shown, an embodiment of the present invention provides a building structure monitoring method based on multi-source data fusion, comprising the following steps:

[0069] S1: Collect and process the initial data of the building structure to obtain the initial variable information and initial state information of the building structure;

[0070] Wherein, the real-time data includes building information and environmental information;

[0071] Specifically, the initial data of the building structure can be collected using electronic devices, including sensors, infrared imagers, laser scanners, and infrared imagers.

[0072] In this embodiment, by collecting the initial data of the building structure (the current data of the building structure), these data are the technology for subsequent digital twin deduction, providing detailed data support for the subsequent monitoring and deduction of the building structure.

[0073] The processing of the initial data specifically involves extracting data related to building structure safety from the initial data.

[0074] S2: Calculate the initial variable information of the building structure based on the digital twin technology, the relational logic model and the real-time monitoring data of the building structure to obtain the deduction result of the digital twin of the building structure;

[0075] Specifically, the digital twin technology uses the initial data to drive the relational logic model for the physical properties of the building structure and calculates the initial variable data to obtain the deduction results of the digital twin.

[0076] In this embodiment, after obtaining the deduction result of the digital twin, it also includes: comparing the digital twin deduction result with the initial state information, correcting the digital twin deduction result according to the corresponding comparison result, and correcting and optimizing the relational logic model based on the corrected deduction result.

[0077] S3: Acquire real-time data of the building structure and extract safety data from the real-time data, where the safety data includes normal parameters and abnormal parameters;

[0078] S4: Obtain a node corresponding to the abnormal parameter, mark it as an abnormal node, mark the safety data corresponding to the abnormal node as a first parameter, perform a correlation analysis on the abnormal node and the first parameter, and obtain a correlation parameter;

[0079] Specifically, the process of performing correlation analysis on the abnormal node and the first parameter to obtain the correlation parameter includes:

[0080] S41: taking the abnormal node as the end point, reversely collecting abnormal parameters of the building structure, and taking the abnormal parameters of the abnormal node as the benchmark data;

[0081] S42: Obtain the abnormal parameter fluctuation of the reverse time series of the benchmark data to obtain the evaluation parameter;

[0082] S43: Preset an evaluation threshold, compare the evaluation parameter with the evaluation threshold, and when the evaluation parameter is less than or equal to the evaluation threshold for the first time, mark the corresponding time as the abnormal starting point;

[0083] S44: Acquire all normal parameters from the abnormal starting point to the abnormal point to obtain a second parameter;

[0084] S45: Perform offset calculation on the second parameter to obtain an associated parameter.

[0085] In this embodiment, the process of performing offset calculation on the second parameter to obtain the associated parameter includes:

[0086] The abnormal node is taken as the end point, and the second parameter of the abnormal node is taken as the benchmark data;

[0087] Obtain the abnormal parameter fluctuation of the reverse time series of the benchmark data to obtain the evaluation parameters;

[0088] Preset an evaluation threshold, compare the evaluation parameter with the evaluation threshold, and when the evaluation parameter is less than or equal to the evaluation threshold for the first time, mark the corresponding time as the parameter starting point;

[0089] The offset parameter of the second parameter is calculated by the offset function:

[0090] ;

[0091] Among them, P represents the offset parameter, n represents the total amount of data of the second parameter from the starting point of the parameter to the abnormal node, and X n-1 Brother X n Represents the adjacent second parameter;

[0092] By presetting the offset threshold, comparing the deviation parameter of the second parameter with the offset threshold;

[0093] If the offset parameter is greater than or equal to the offset threshold, then the second parameter is an associated parameter;

[0094] If the offset parameter is less than the offset threshold, it means that the second parameter is not a correlation parameter.

[0095] In this embodiment, after the abnormal node is determined, the abnormal node is used as the end node, and the monitoring data of the reverse time series is collected, and the abnormal node is used as the baseline parameter. Then, the abnormal parameter and the second parameter of the reverse time series are calculated through the baseline parameter, so that the starting point of the abnormal node and the associated parameters with the abnormal parameter can be determined, wherein, by analyzing the amount of time from the abnormal end point to the parameter starting point of the second parameter, the associated parameters of the abnormal parameter can be further accurately analyzed.

[0096] It should be noted that the second parameter includes several groups of different building structure parameter data.

[0097] S5: Use digital twins to deduce the abnormal parameters and associated parameters of the abnormal initial point to obtain the deduction results;

[0098] S6: determining a corresponding relationship set based on the correction relationship between the deduction result and the actual result of the abnormal node, determining a corresponding forward calculation model based on the physical property variables from the abnormal initial point to the abnormal node, and optimizing the forward calculation model using the corresponding relationship set to obtain a monitoring model;

[0099] Specifically, the abnormal parameters and associated parameters of the building structure are certain, and the results are also certain, that is, the parameters and results have a certain correlation in the time series (conforming to the positive problem model, that is, the forward calculation model).

[0100] In this embodiment, the process of determining the corresponding relationship set based on the correction relationship between the deduction result and the actual result of the abnormal node includes:

[0101] Compare and analyze the deduction results with the actual results of the abnormal nodes, and modify the deduction results based on the comparison results;

[0102] The monitoring data corresponding to the corrected points in the deduction results and the actual results of the abnormal nodes are extracted to obtain a corresponding relationship set.

[0103] S7: Implement monitoring according to at least one current monitoring data and monitoring model of the building structure to be detected.

[0104] The present invention collects initial data of the building structure, and uses the initial data to drive the relationship logic model for the physical properties of the building structure through the digital twin technology, calculates the initial variable data to obtain the deduction result of the digital twin, and at the same time obtains the safety data in the real-time data of the building structure, wherein the safety data includes normal parameters and abnormal parameters, analyzes the monitoring data corresponding to the abnormal node, determines the abnormal node and the abnormal starting point, and then determines the abnormal parameters and associated parameters, and finally deduces the abnormal parameters and associated parameters of the abnormal initial point through the digital twin to obtain the deduction result, determines the corresponding relationship set based on the correction relationship between the deduction result and the actual result of the abnormal node, and determines the corresponding relationship set based on the abnormal node. The physical property variables from the normal initial point to the abnormal node determine the corresponding forward calculation model, and the corresponding relationship set is used to optimize the forward calculation model to obtain the monitoring model. The real-time monitoring of the building structure is achieved through the detection model. It can be seen that the present application uses digital twin technology to deduce the state of the building structure. At the same time, by analyzing the abnormal state data of the building structure, the associated parameters related to it can be determined, and then the associated parameters and the abnormal parameters are combined for analysis to determine the cause of the abnormality, and establish a monitoring model. The detection model can predict possible abnormalities in the building structure in advance, thereby helping maintenance personnel to determine the maintenance direction and perform maintenance in advance. In addition, the present application can conduct a more comprehensive analysis of the building structure.

[0105] This embodiment also provides a building structure monitoring system based on multi-source data fusion, which is used to implement the above-mentioned building structure monitoring method based on multi-source data fusion, including:

[0106] A deduction and analysis module is used to collect and process the initial data of the building structure to obtain the initial variable information and initial state information of the building structure, and calculate the initial variable information of the building structure based on the digital twin technology, the relational logic model and the real-time monitoring data of the building structure to obtain the deduction result of the digital twin of the building structure;

[0107] The acquisition and tracing module is used to acquire the real-time data of the building structure and extract the safety data in the real-time data. The safety data includes normal parameters and abnormal parameters. The module is also used to acquire the nodes corresponding to the abnormal parameters, mark them as abnormal nodes, mark the safety data corresponding to the abnormal nodes as first parameters, and perform correlation analysis on the abnormal nodes and the first parameters to obtain the correlation parameters.

[0108] A model building module is used to use the digital twin to deduce the abnormal parameters and associated parameters of the abnormal initial point to obtain the deduction results, and determine the corresponding relationship set based on the correction relationship between the deduction results and the actual results of the abnormal nodes, determine the corresponding forward calculation model based on the physical property variables from the abnormal initial point to the abnormal node, and use the corresponding relationship set to optimize the forward calculation model to obtain the monitoring model;

[0109] The monitoring module implements monitoring according to at least one current monitoring data and a monitoring model of the building structure to be detected.

[0110] Furthermore, the initial variable information of the building structure is calculated based on the digital twin technology, the relational logic model and the real-time monitoring data of the building structure to obtain the deduction result of the digital twin of the building structure, including:

[0111] Through digital twin technology, the physical properties of the building structure are driven by the initial data-driven relational logic model, and the initial variable data are calculated to obtain the deduction results of the digital twin.

[0112] Furthermore, the process of performing correlation analysis on the abnormal node and the first parameter to obtain the correlation parameter includes:

[0113] Taking the abnormal node as the end point, reversely collect the abnormal parameters of the building structure, and use the abnormal parameters of the abnormal node as the benchmark data;

[0114] Obtain the abnormal parameter fluctuation of the reverse time series of the benchmark data to obtain the evaluation parameters;

[0115] Preset an evaluation threshold, compare the evaluation parameter with the evaluation threshold, and when the evaluation parameter is less than or equal to the evaluation threshold for the first time, mark the corresponding time as the abnormal starting point;

[0116] Obtain all normal parameters from the abnormal starting point to the abnormal point to obtain the second parameter;

[0117] An offset calculation is performed on the second parameter to obtain a correlation parameter.

[0118] Furthermore, the process of performing offset calculation on the second parameter to obtain the associated parameter includes:

[0119] The abnormal node is taken as the end point, and the second parameter of the abnormal node is taken as the benchmark data;

[0120] Obtain the abnormal parameter fluctuation of the reverse time series of the benchmark data to obtain the evaluation parameters;

[0121] Preset an evaluation threshold, compare the evaluation parameter with the evaluation threshold, and when the evaluation parameter is less than or equal to the evaluation threshold for the first time, mark the corresponding time as the parameter starting point;

[0122] The offset parameter of the second parameter is calculated by the offset function:

[0123] ;

[0124] Among them, P represents the offset parameter, n represents the total amount of data of the second parameter from the starting point of the parameter to the abnormal node, and X n-1Brother X n Represents the adjacent second parameter;

[0125] By presetting the offset threshold, comparing the deviation parameter of the second parameter with the offset threshold;

[0126] If the offset parameter is greater than or equal to the offset threshold, then the second parameter is an associated parameter;

[0127] If the offset parameter is less than the offset threshold, it means that the second parameter is not a correlation parameter.

[0128] Furthermore, the process of determining the corresponding relationship set based on the correction relationship between the deduction result and the actual result of the abnormal node includes:

[0129] Compare and analyze the deduction results with the actual results of the abnormal nodes, and modify the deduction results based on the comparison results;

[0130] The monitoring data corresponding to the corrected points in the deduction results and the actual results of the abnormal nodes are extracted to obtain a corresponding relationship set.

[0131] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A building structure monitoring method based on multi-source data fusion, characterized in that: The following steps are involved: Collect and process the initial data of the building structure to obtain the initial variable information and initial state information of the building structure; Calculating initial variable information of the building structure based on the digital twin technology, the relational logic model and the real-time monitoring data of the building structure to obtain a deduction result of the digital twin of the building structure; Obtain real-time data of building structures and extract safety data from the real-time data, which includes normal parameters and abnormal parameters; Obtaining a node corresponding to an abnormal parameter, marking it as an abnormal node, marking the safety data corresponding to the abnormal node as a first parameter, performing a correlation analysis on the abnormal node and the first parameter, and obtaining a correlation parameter; The digital twin is used to deduce the abnormal parameters and associated parameters of the abnormal initial point to obtain the deduction results; Determine a corresponding relationship set based on the correction relationship between the deduction result and the actual result of the abnormal node, determine a corresponding forward calculation model based on the physical property variables from the abnormal initial point to the abnormal node, and use the corresponding relationship set to optimize the forward calculation model to obtain a monitoring model; Monitoring is achieved based on at least one current monitoring data and a monitoring model of the building structure to be detected.

2. The building structure monitoring method based on multi-source data fusion according to claim 1 is characterized in that: The initial variable information of the building structure is calculated based on the digital twin technology, the relational logic model and the real-time monitoring data of the building structure to obtain the deduction results of the digital twin of the building structure, including: Through digital twin technology, the physical properties of the building structure are driven by the initial data-driven relational logic model, and the initial variable data are calculated to obtain the deduction results of the digital twin.

3. The building structure monitoring method based on multi-source data fusion according to claim 1 is characterized in that: The process of performing correlation analysis on the abnormal node and the first parameter to obtain the correlation parameter includes: Taking the abnormal node as the end point, reversely collect the abnormal parameters of the building structure, and use the abnormal parameters of the abnormal node as the benchmark data; Obtain the abnormal parameter fluctuation of the reverse time series of the benchmark data to obtain the evaluation parameters; Preset an evaluation threshold, compare the evaluation parameter with the evaluation threshold, and when the evaluation parameter is less than or equal to the evaluation threshold for the first time, mark the corresponding time as the abnormal starting point; Obtain all normal parameters from the abnormal starting point to the abnormal point to obtain the second parameter; An offset calculation is performed on the second parameter to obtain a correlation parameter.

4. The building structure monitoring method based on multi-source data fusion according to claim 3 is characterized in that: The process of performing offset calculation on the second parameter to obtain the associated parameter comprises: The abnormal node is taken as the end point, and the second parameter of the abnormal node is taken as the benchmark data; Obtain the abnormal parameter fluctuation of the reverse time series of the benchmark data to obtain the evaluation parameters; Preset an evaluation threshold, compare the evaluation parameter with the evaluation threshold, and when the evaluation parameter is less than or equal to the evaluation threshold for the first time, mark the corresponding time as the parameter starting point; The offset parameter of the second parameter is calculated by the offset function: ； Where P represents the offset parameter, n represents the total amount of data from the starting point of the second parameter to the abnormal node, and X n-1 Brother X n Represents the adjacent second parameter; By presetting the offset threshold, comparing the deviation parameter of the second parameter with the offset threshold; If the offset parameter is greater than or equal to the offset threshold, then the second parameter is an associated parameter; If the offset parameter is less than the offset threshold, it means that the second parameter is not a correlation parameter.

5. The building structure monitoring method based on multi-source data fusion according to claim 1 is characterized in that: The process of determining the corresponding relationship set based on the correction relationship between the deduction result and the actual result of the abnormal node includes: Compare and analyze the deduction results with the actual results of the abnormal nodes, and modify the deduction results based on the comparison results; The monitoring data corresponding to the corrected points in the deduction results and the actual results of the abnormal nodes are extracted to obtain a corresponding relationship set.

6. A building structure monitoring system based on multi-source data fusion, used to implement the building structure monitoring method based on multi-source data fusion according to any one of claims 1 to 5, characterized in that: include: A deduction and analysis module is used to collect and process the initial data of the building structure to obtain the initial variable information and initial state information of the building structure, and calculate the initial variable information of the building structure based on the digital twin technology, the relational logic model and the real-time monitoring data of the building structure to obtain the deduction result of the digital twin of the building structure; The acquisition and tracing module is used to acquire the real-time data of the building structure and extract the safety data in the real-time data. The safety data includes normal parameters and abnormal parameters. The module is also used to acquire the nodes corresponding to the abnormal parameters, mark them as abnormal nodes, mark the safety data corresponding to the abnormal nodes as first parameters, and perform correlation analysis on the abnormal nodes and the first parameters to obtain the correlation parameters. A model building module is used to use the digital twin to deduce the abnormal parameters and associated parameters of the abnormal initial point to obtain the deduction results, and determine the corresponding relationship set based on the correction relationship between the deduction results and the actual results of the abnormal nodes, determine the corresponding forward calculation model based on the physical property variables from the abnormal initial point to the abnormal node, and use the corresponding relationship set to optimize the forward calculation model to obtain the monitoring model; The monitoring module implements monitoring according to at least one current monitoring data and a monitoring model of the building structure to be detected.

7. The building structure monitoring system based on multi-source data fusion according to claim 6 is characterized in that: The initial variable information of the building structure is calculated based on the digital twin technology, the relational logic model and the real-time monitoring data of the building structure to obtain the deduction results of the digital twin of the building structure, including: Through digital twin technology, the physical properties of the building structure are driven by the initial data-driven relational logic model, and the initial variable data are calculated to obtain the deduction results of the digital twin.

8. The building structure monitoring method based on multi-source data fusion according to claim 6 is characterized in that: The process of performing correlation analysis on the abnormal node and the first parameter to obtain the correlation parameter includes: Taking the abnormal node as the end point, reversely collect the abnormal parameters of the building structure, and use the abnormal parameters of the abnormal node as the benchmark data; Obtain the abnormal parameter fluctuation of the reverse time series of the benchmark data to obtain the evaluation parameters; Preset an evaluation threshold, compare the evaluation parameter with the evaluation threshold, and when the evaluation parameter is less than or equal to the evaluation threshold for the first time, mark the corresponding time as the abnormal starting point; Obtain all normal parameters from the abnormal starting point to the abnormal point to obtain the second parameter; An offset calculation is performed on the second parameter to obtain a correlation parameter.

9. The building structure monitoring method based on multi-source data fusion according to claim 8 is characterized in that: The process of performing offset calculation on the second parameter to obtain the associated parameter comprises: The abnormal node is taken as the end point, and the second parameter of the abnormal node is taken as the benchmark data; Obtain the abnormal parameter fluctuation of the reverse time series of the benchmark data to obtain the evaluation parameters; Preset an evaluation threshold, compare the evaluation parameter with the evaluation threshold, and when the evaluation parameter is less than or equal to the evaluation threshold for the first time, mark the corresponding time as the parameter starting point; The offset parameter of the second parameter is calculated by the offset function: ； Among them, P represents the offset parameter, n represents the total amount of data of the second parameter from the starting point of the parameter to the abnormal node, and X n-1 Brother X n Represents the adjacent second parameter; By presetting the offset threshold, comparing the deviation parameter of the second parameter with the offset threshold; If the offset parameter is greater than or equal to the offset threshold, then the second parameter is an associated parameter; If the offset parameter is less than the offset threshold, it means that the second parameter is not a correlation parameter.

10. The building structure monitoring method based on multi-source data fusion according to claim 9 is characterized in that: The process of determining the corresponding relationship set based on the correction relationship between the deduction result and the actual result of the abnormal node includes: Compare and analyze the deduction results with the actual results of the abnormal nodes, and modify the deduction results based on the comparison results; The monitoring data corresponding to the corrected points in the deduction results and the actual results of the abnormal nodes are extracted to obtain a corresponding relationship set.

Images