Building structure data acquisition method based on edge network
By deploying edge computing nodes and sensor correlation matrices in building structures and optimizing data transmission strategies, the problems of data latency and high load in traditional building structure monitoring are solved, achieving efficient, low-power real-time response and analysis.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-30
- Publication Date
- 2026-03-31
AI Technical Summary
In traditional building structure monitoring, the direct uploading of massive amounts of sensor data to the cloud system for processing leads to large delays, an inability to respond to emergency anomalies in a timely manner, and high computing and storage loads on the cloud system, resulting in a decline in the performance of the monitoring system.
An edge network-based data acquisition method is adopted, which deploys edge computing nodes in the building to perform data aggregation, preprocessing and feature extraction. The sensor correlation matrix is used to optimize the data transmission strategy, prioritizing the transmission of abnormal data to the cloud and delaying the transmission of normal data. Combined with dynamic network status monitoring and priority reallocation, the data can be processed locally and intelligently filtered.
By reducing network bandwidth usage and data transmission latency, the system improves real-time response capabilities to structural anomalies, reduces cloud computing and storage load, and builds an efficient, low-power, and high-real-time building structure data acquisition and analysis system.
Smart Images
Figure CN121771191A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of data acquisition and processing technology, specifically to a method for acquiring building structure data based on edge networks. Background Technology
[0002] Building structure monitoring refers to the technical means of using various sensors and testing instruments to regularly or in real-time monitor buildings or structures (such as bridges, buildings, tunnels, etc.) to assess their health status and safety performance. For example, vibration sensors, strain gauges, accelerometers, temperature and humidity monitors, and other equipment are installed at key parts of the building structure (such as support points, connection points, and areas of concentrated stress) to collect physical parameters. All sensor data is centralized to a central server, control room, or data center for data analysis and early warning processing to assess the safety status of the building structure.
[0003] However, in traditional building structure monitoring, massive amounts of sensor data are directly reported to the cloud system for processing and analysis, which often introduces significant delays and makes it impossible to respond to emergency anomalies in a timely manner. Furthermore, the cloud system needs to process large amounts of data from multiple sensors simultaneously, which places high demands on database storage, data processing, and computing capabilities, leading to a decline in the performance of the entire monitoring system or even the emergence of processing bottlenecks. Summary of the Invention
[0004] To address the problems existing in the prior art, the present invention aims to provide a building structure data acquisition method based on edge networks, which can realize the local processing and filtering of data by sinking computing power to the network edge, thereby improving the real-time response capability of building monitoring systems to structural anomalies.
[0005] To achieve the above objectives, the present invention provides the following technical solution: a building structure data acquisition method based on edge networks, comprising:
[0006] Multiple types of sensors are deployed in key parts of the building structure, and edge computing nodes are deployed in the building to be monitored. The sensors deployed in the building transmit data to the edge computing nodes in a unified manner to form data aggregation.
[0007] The sensor data is preprocessed by using wavelet transform or other noise reduction techniques to remove high-frequency noise and by using data compression algorithms to reduce the amount of data in order to obtain usable data.
[0008] Key features are extracted from the available data, and abnormal available data is identified. Abnormal data is marked as abnormal data and transmitted to the cloud system in a priority collection manner. The remaining normal available data is marked as normal data and transmitted to the cloud system in a delayed collection manner.
[0009] Based on the building structure drawings and the physical deployment locations of the sensors, a sensor association matrix is constructed. Each row and column of the sensor association matrix corresponds to one sensor, and the elements in the matrix represent the association weight between two sensors. After the edge computing node receives abnormal data, it performs priority collection on the relevant normal data that was originally collected with a lag.
[0010] In some implementations, when extracting key features from available data, the key features include the increase or decrease of the same data, and a corresponding baseline value and warning threshold are preset for each type of data, and the increase or decrease is obtained by subtracting the monitored data from the baseline value; The increase or decrease of the same received data is compared with the warning threshold corresponding to that data. When the increase or decrease of the data is within the warning threshold, the available data is marked as normal data; when the increase or decrease of the data exceeds the warning threshold, the available data is marked as abnormal data.
[0011] In some implementations, performing priority acquisition transmission specifically includes: after available data is marked as abnormal data, canceling the preprocessing operation performed on the data received by this sensor at the edge computing node, and immediately uploading the sensor's raw data to the cloud system.
[0012] In some implementations, the delayed acquisition method specifically includes: after available data is marked as normal data, the normal data is stored in the local storage hard drive of the edge computing node to form a data batch. The normal data stored locally is set to be packaged and uploaded to the cloud system in batches at regular intervals. The network status between the edge computing node and the cloud system is monitored so that if poor network quality or congestion is detected during the transmission process, the edge computing node can pause the data transmission to the cloud system and continue to store the data in the local hard drive. The transmission will resume after the network is restored. In addition, normal data is not transmitted to the cloud system at the same time as abnormal data. The normal data is transmitted to the PTZ system only after there is no abnormal data in the transmission list.
[0013] In some implementations, the specific methods for constructing the sensor association matrix include: obtaining building structure drawings, digitally recording the specific locations of the sensors in the drawings, labeling the structural components to which each sensor belongs, and obtaining the physical distance between each sensor. Based on the actual distance of each sensor in the drawings, two sensors with shorter distances indicate a stronger connection, while those with greater distances indicate a weaker connection. Based on the physical connection of the components to which the sensors belong, a sensor association matrix is constructed.
[0014] In some implementations, historical data continuously collected on edge nodes is used to continuously correct or update the correlation weights between sensors using statistical correlation analysis or machine learning methods, thereby enabling dynamic updating of the correlation relationships in the sensor management matrix.
[0015] In some implementations, the specific way to determine whether normal data is associated with abnormal data is to first set an association threshold. When the association weight of two sensors exceeds the threshold, it is determined that there is a strong data correlation between the two sensors. This normal data is then judged as associated data with abnormal data, and this normal data is changed to be transmitted to the cloud system in a priority acquisition mode.
[0016] In some implementations, the priority of normal data that is changed to be transmitted in the upload list in the priority collection mode is arranged after that of abnormal data. When network congestion is detected, a priority reallocation strategy is executed so as to cancel the judgment of performing priority collection on some normal data. The priority reallocation strategy includes obtaining a list of normal data that has been changed to the priority collection method for transmission, comparing the list of normal data with the baseline value, and setting a comparison threshold. When the proportion of data with the same baseline value in the normal data that has been changed to the priority collection method for transmission reaches the comparison threshold, the priority collection and upload of this normal data is canceled, and the original delayed collection and upload is restored.
[0017] In some implementations, when the amount of normal data remaining on the edge computing node reaches a preset threshold, a data reduction strategy is executed to reduce the amount of data storage and processing in the edge computing node. The data reduction strategy includes obtaining a list of data from the normal data generated by the same sensor, transferring data in the normal data that has the same value as the benchmark to the recycle bin, removing the normal data in the recycle bin from the list uploaded to the cloud system, and permanently deleting the contents of the recycle bin within a preset time.
[0018] The present invention further provides a computer-readable storage medium storing a computer program that is executed by a processor to implement the above-described edge network-based building structure data acquisition method.
[0019] The technical solution provided by this invention has the following advantages compared with the prior art: Firstly, this invention constructs a three-tiered collaborative data processing architecture consisting of sensors, edge computing nodes, and a cloud system platform. By deploying edge computing nodes on-site to aggregate and preprocess sensor data in their respective areas, computing power is pushed down to the network edge, enabling localized data processing and intelligent filtering. This significantly reduces network bandwidth usage and data transmission latency, improves the system's real-time response to structural anomalies, and lowers the cloud computing and storage load. As a result, it can build a highly efficient, low-power, and real-time building structure data acquisition and analysis system.
[0020] Secondly, this invention modifies the collection priority of normal data by constructing a sensor correlation matrix. When abnormal problems occur in the building structure, it can capture the correlation data of abnormal data and combine it with the correlated normal data to more accurately reconstruct the on-site situation, helping the backend to comprehensively analyze the root cause of abnormal events, the affected area, and possible structural health hazards. Attached Figure Description
[0021] Figure 1 This is a flowchart illustrating the building structure data acquisition method based on edge networks according to the present invention. Detailed Implementation
[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0023] It is understood that the term "a" should be understood as "at least one" or "one or more", that is, in one embodiment, the number of an element can be one, while in another embodiment, the number of the element can be multiple, and the term "a" should not be understood as a limitation on the number.
[0024] This invention provides a method for acquiring building structure data based on edge networks, such as... Figure 1 As shown, the method includes the following steps: Step 1: Deploy multiple types of sensors in key parts of the building structure (such as beams, columns, cantilever, or connection nodes) to collect physical quantity data such as vibration, strain, displacement, temperature, and humidity in real time. Deploy edge computing nodes (such as industrial control computers, embedded gateways, or edge servers) within the building to be monitored. The sensors deployed within the building will transmit data to the edge computing nodes in a unified manner to form data aggregation. Step two involves preprocessing the sensor data. Specifically, filtering algorithms (such as Kalman filtering and low-pass filtering) are used to smooth the continuous data, wavelet transform or other noise reduction techniques are used to remove high-frequency noise, and data compression algorithms (such as LZW and differential coding) are used to reduce the amount of data to obtain usable data. Step 3: Extract key features from available data, identify abnormal available data, mark abnormal data as anomalous data, and transmit it to the cloud system in a priority collection manner, while mark the remaining normal available data as normal data and transmit it to the cloud system in a delayed collection manner. In step three, when extracting key features from the available data, these key features include the magnitude of increase or decrease of the same data. A corresponding baseline value and warning threshold are preset for each type of data. The magnitude of increase or decrease is obtained by subtracting the monitored data from the baseline value, which represents the ideal value of the monitored data. The magnitude of increase or decrease of the received data is compared with the corresponding warning threshold. When the magnitude of increase or decrease is within the warning threshold, it indicates that the data continuously monitored by the sensor has not changed significantly, or the change is within a reasonable range, and the available data is marked as normal data. Conversely, when the magnitude of increase or decrease exceeds the warning threshold, it indicates that the data continuously monitored by the sensor has experienced an unexpected magnitude of increase or decrease, and the available data is marked as abnormal data.
[0025] For example, a vibration sensor is installed on a main beam in a building structure, and this sensor collects vibration values (unit: m / s) at regular intervals. 2 The system has a pre-set warning threshold of ±0.3 m / s for this vibration data. 2 The edge computing node is configured to receive continuously available data, with the sensor continuously collecting data at a speed of 0.1 m / s. 2 0.15m / s 2 0.2m / s 2 and 0.15m / s 2 The baseline value for the vibration of the main beam is set at 0.1 m / s². 2 By subtracting the collected data from the baseline value, the increase or decrease in the collected data can be obtained as follows: 0 m / s 2 0.05m / s 2 0.1m / s 2 and 0.05m / s 2 Since the increase or decrease in the received available data is within the warning threshold, the edge computing node marks the received data as normal data.
[0026] The transmission of data using the priority acquisition method specifically includes: after available data is marked as abnormal data, canceling the preprocessing operation of the data received by this sensor at the edge computing node, that is, removing the processing cycle of the data judged as available data, and directly uploading the raw data of the sensor to the cloud system immediately, so that the cloud system can quickly obtain abnormal information, perform real-time alarms, conduct in-depth analysis and initiate subsequent emergency response. The delayed data acquisition method specifically includes: after available data is marked as normal data, the normal data is stored on the local hard drive of the edge computing node to form a data batch. The normal data stored locally is set to be packaged and uploaded to the cloud system in batches at regular intervals. The network status between the edge computing node and the cloud system is monitored. If poor network quality or congestion is detected during the transmission process, the edge computing node can pause the data transmission to the cloud system and continue to store the data on the local hard drive. The transmission will resume after the network is restored, ensuring that the data is not lost due to delayed upload. In addition, normal data is not transmitted to the cloud system at the same time as abnormal data. Instead, the transmission of normal data to the PTZ system is performed only after there is no abnormal data in the transmission list. By extracting and comparing key features from available data, continuous data collected by sensors is classified according to its deviation from preset benchmark values and warning thresholds. Data with fluctuations within the threshold range is marked as normal data, while data exceeding the warning threshold is marked as abnormal data. This achieves data diversion management. In practical applications, this approach enables edge computing nodes to capture abnormal data changes in real time and upload them directly to the cloud system. This ensures that the cloud can quickly obtain early warning information for real-time alarms, in-depth analysis, and subsequent emergency response, thereby improving the sensitivity and response speed of the monitoring system to abnormal events in building structures. On the other hand, normal data that does not significantly deviate from the benchmark value is collected with a delay strategy. It is first stored on the local hard drive of the edge computing node and then uploaded to the cloud in batches after the abnormal data transmission with normal network conditions and high priority is completed. This not only effectively alleviates the network congestion and transmission delay problems that may be caused by the upload of a large amount of normal data, but also achieves reasonable scheduling of data resources and avoids mutual interference between ordinary data and key abnormal data during transmission.
[0027] Step 4: Based on the building structure drawings and the physical deployment locations of the sensors, construct a sensor association matrix. Each row and column of the sensor association matrix corresponds to one sensor. The elements in the matrix represent the association weight between two sensors (determined based on physical distance or structural coupling method). After the edge computing node receives abnormal data, it changes the normal data that was originally collected with a lag and performs priority collection. The specific methods for constructing a sensor association matrix include: obtaining building structure drawings, digitally recording the specific locations of sensors in the drawings, labeling the structural components to which each sensor belongs (such as beams, columns, cantilever, connection nodes, etc.), and obtaining the physical distance between each sensor. Based on the actual distance of each sensor in the drawings, two sensors with shorter distances indicate a stronger connection, while those with greater distances indicate a weaker connection. A static sensor association matrix is constructed based on the physical connection of the components to which the sensors belong (such as whether they belong to the same column or the same connection node).
[0028] Meanwhile, by utilizing historical data continuously collected on edge nodes, statistical correlation analysis or machine learning methods (such as correlation coefficients and mutual information indicators within a sliding window) are employed to continuously correct or update the correlation weights between various sensors. This achieves the effect of dynamically updating the correlation relationships of the sensor management matrix, reflecting whether there is a real linkage relationship between the sensors in the current state, rather than simply relying on a fixed matrix model constructed from architectural drawings.
[0029] The specific method for determining whether normal data is correlated with abnormal data is to first set a correlation threshold. When the correlation weight between two sensors exceeds this threshold, it is considered that there is a strong data correlation between the two sensors. This normal data is then identified as correlated with the abnormal data, and the normal data is prioritized for collection before being transmitted to the cloud system. The rationale for this design is that the generation of abnormal data indicates unexpected abnormal changes in the building structure. Abnormal events in building structures are usually not caused by problems in a single structure; it is easy for related structures to experience a chain reaction. Even if the sensors monitoring related structures do not detect abnormal data, they still need to be closely monitored subsequently. More importantly, this design can capture the correlated data of abnormal data, and by combining it with correlated normal data, it can more accurately reconstruct the on-site situation, helping the backend to comprehensively analyze the root cause of the abnormal event, the affected area, and potential structural health hazards.
[0030] Furthermore, when normal data is switched to a priority collection method for transmission to the cloud system, the amount of normal data associated with abnormal data may be significantly higher than the amount of abnormal data. Therefore, in the upload list, the priority of normal data switched to priority collection is placed after abnormal data; that is, abnormal data is uploaded first, followed by associated normal data. Moreover, if network congestion is detected, a priority reallocation strategy is implemented to cancel the priority collection decision for some normal data. The priority reallocation strategy involves obtaining a list of normal data to be transferred using the priority collection method, comparing this list with a baseline value, and setting a comparison threshold, such as 80%. If 80% of the normal data transferred using the priority collection method matches the baseline value, the priority collection and upload of this normal data is canceled, and the original delayed collection and upload method is restored.
[0031] When the amount of normal data remaining on the edge computing nodes reaches a preset threshold, a data reduction strategy is implemented to reduce the amount of data stored and processed on the edge computing nodes. Reaching this threshold indicates that the cloud system's upload network channel is congested for an extended period. Under the condition of prioritizing data transmission, normal data will accumulate on the local storage hard drives of the edge computing nodes. To avoid this problem, the data reduction strategy includes obtaining a list of normal data from the same sensor, moving data with the same baseline value to a recycle bin, removing normal data from the recycle bin from the list uploaded to the cloud system, and permanently deleting the contents of the recycle bin within a preset time. The purpose of delayed deletion in the recycle bin is to prevent this normal data from being subsequently matched as related data to abnormal data, leading to data loss. Furthermore, by reducing redundant data unrelated to anomalies, the risk of data congestion from edge nodes to the cloud can be reduced. The delayed deletion mechanism ensures that background cleanup occurs during network downtime, preventing excessive duplicate data transmission during peak periods and improving overall data transmission efficiency.
[0032] By combining data classification, hierarchical uploading, and dynamic network status monitoring, the system not only improves the real-time performance and accuracy of abnormal data transmission, ensuring that critical abnormal information receives priority access to cloud processing resources, but also effectively preserves and delays normal data transmission. This optimizes the data interaction efficiency between edge computing nodes and the cloud system, providing efficient, intelligent, and resilient support for the overall building health monitoring system. It meets the dual requirements of real-time processing and transmission of large-scale data in a multi-sensor distributed environment, effectively reducing the risk of data loss or slow response due to network congestion or transmission delays. Thus, while ensuring the accuracy of structural safety monitoring, it achieves efficient operation of the monitoring system and intelligent scheduling of the overall data transmission mechanism.
[0033] In summary, this invention aims to design a building structure data acquisition method based on edge networks. Addressing the problem in traditional building structure monitoring where large-scale data transmission from multiple sensors leads to network congestion and processing delays, this invention constructs a three-tiered collaborative data processing architecture consisting of sensors, edge computing nodes, and a cloud system platform. Edge computing nodes deployed on-site aggregate and preprocess sensor data from their respective areas, pushing computing power down to the network edge. This enables localized data processing and intelligent filtering, significantly reducing network bandwidth consumption and data transmission latency, improving the system's real-time response to structural anomalies, and lowering cloud computing and storage load. This results in a highly efficient, low-power, and real-time-enabled building structure data acquisition and analysis system. By constructing a sensor correlation matrix to adjust the acquisition priority of normal data, it can capture correlated data from abnormal data when structural anomalies occur. Combining this with correlated normal data allows for a more accurate reconstruction of the on-site situation, aiding the backend in comprehensively analyzing the root cause of the anomaly, the affected area, and potential structural health hazards. By designing a priority redistribution strategy, when network bandwidth is tight, if most of the normal data collected first is completely consistent with the baseline state, it means that these data do not reflect significant changes. Uploading such a large amount of redundant data will consume a lot of network resources and affect the transmission of truly abnormal data. Therefore, restoring data that is highly consistent with the baseline value to be collected later can free up bandwidth and ensure that truly abnormal and critical data can be uploaded first.
[0034] The processes described above with reference to the flowcharts in the embodiments disclosed in this invention can be implemented as computer software programs. Embodiments of this invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication component, and / or installed from a removable medium. When the computer program is executed by a central processing unit, it performs the functions defined in the methods of this application. It should be noted that the computer-readable medium described above in this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection having one or more conductor segments, a portable computer disk, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory, an optical fiber, a portable compact disk read-only memory, an optical storage device, a magnetic storage device, or any suitable combination thereof. In this application, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wireless segments, wire segments, optical cables, RF, etc., or any suitable combination thereof.
[0035] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.
[0036] Those skilled in the art should understand that the above description is only a specific embodiment of this application, but the protection scope of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the protection scope of this application.
Claims
1. A method for acquiring building structure data based on edge networks, characterized in that, include: Multiple types of sensors are deployed in key parts of the building structure, and edge computing nodes are deployed in the building to be monitored. The sensors deployed in the building transmit data to the edge computing nodes in a unified manner to form data aggregation. The sensor data is preprocessed by using wavelet transform or other noise reduction techniques to remove high-frequency noise and by using data compression algorithms to reduce the amount of data in order to obtain usable data. Key features are extracted from the available data, and abnormal available data is identified. Abnormal data is marked as abnormal data and transmitted to the cloud system in a priority collection manner. The remaining normal available data is marked as normal data and transmitted to the cloud system in a delayed collection manner. Based on the building structure drawings and the physical deployment locations of the sensors, a sensor association matrix is constructed. Each row and column of the sensor association matrix corresponds to one sensor, and the elements in the matrix represent the association weight between two sensors. After the edge computing node receives abnormal data, it performs priority collection on the relevant normal data that was originally collected with a lag.
2. The building structure data acquisition method based on edge networks according to claim 1, characterized in that, When extracting key features from available data, key features include the increase or decrease of the same data, and a corresponding baseline value and warning threshold are preset for each type of data, and the increase or decrease is obtained by subtracting the monitored data from the baseline value; The increase or decrease of the same received data is compared with the warning threshold corresponding to that data. When the increase or decrease of the data is within the warning threshold, the available data is marked as normal data; when the increase or decrease of the data exceeds the warning threshold, the available data is marked as abnormal data.
3. The building structure data acquisition method based on edge networks according to claim 2, characterized in that, The priority acquisition method for data transmission specifically includes: after available data is marked as abnormal data, canceling the preprocessing operation of the data received by this sensor at the edge computing node, and immediately uploading the raw data of the sensor to the cloud system.
4. The building structure data acquisition method based on edge networks according to claim 3, characterized in that, The delayed acquisition method for data transmission specifically includes: after available data is marked as normal data, the normal data is stored on the local storage hard drive of the edge computing node to form a data batch. The normal data stored locally is set to be packaged and uploaded to the cloud system in batches at regular intervals. The network status between the edge computing node and the cloud system is monitored so that if poor network quality or congestion is detected during the transmission process, the edge computing node can pause the data transmission to the cloud system and continue to store the data on the local hard drive. The transmission will resume after the network is restored. In addition, normal data is not transmitted to the cloud system at the same time as abnormal data. The normal data is transmitted to the PTZ system only after there is no abnormal data in the transmission list.
5. The building structure data acquisition method based on edge networks according to claim 1, characterized in that, The specific methods for constructing the sensor association matrix include: obtaining building structure drawings, digitally recording the specific locations of the sensors in the drawings, labeling the structural components to which each sensor belongs, and obtaining the physical distance between each sensor. Based on the actual distance of each sensor in the drawings, two sensors with shorter distances indicate a stronger connection, while those with greater distances indicate a weaker connection. Based on the physical connection of the components to which the sensors belong, the sensor association matrix is constructed.
6. The building structure data acquisition method based on edge networks according to claim 5, characterized in that, By utilizing historical data continuously collected on edge nodes, statistical correlation analysis or machine learning methods are employed to continuously correct or update the correlation weights between various sensors, thereby enabling dynamic updating of the correlation relationships in the sensor management matrix.
7. The building structure data acquisition method based on edge networks according to claim 6, characterized in that, The specific way to determine whether normal data is associated with abnormal data is to first set an association threshold. When the association weight of two sensors exceeds the threshold, it is determined that there is a strong data correlation between the two sensors. This normal data is then judged as associated data with abnormal data, and this normal data is changed to be transmitted to the cloud system in a priority acquisition mode.
8. The building structure data acquisition method based on edge networks according to claim 7, characterized in that, Normal data that has been changed to the priority collection method in the upload list will be prioritized after abnormal data. If network congestion is detected, a priority reallocation strategy will be executed to cancel the priority collection of some normal data. The priority reallocation strategy includes obtaining a list of normal data that has been changed to the priority collection method for transmission, comparing the list of normal data with the baseline value, and setting a comparison threshold. When the proportion of data with the same baseline value in the normal data that has been changed to the priority collection method for transmission reaches the comparison threshold, the priority collection and upload of this normal data is canceled, and the original delayed collection and upload is restored.
9. The method for acquiring building structure data based on edge networks according to claim 8, characterized in that, When the amount of normal data remaining on the edge computing node reaches a preset threshold, a data reduction strategy is executed to reduce the amount of data storage and processing in the edge computing node. The data reduction strategy includes obtaining a list of data from the normal data generated by the same sensor, transferring data in the normal data that has the same value as the benchmark to the recycle bin, removing the normal data in the recycle bin from the list uploaded to the cloud system, and permanently deleting the contents of the recycle bin within a preset time.
10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that is executed by a processor to implement the edge network-based building structure data acquisition method according to any one of claims 1-9.