Distributed cooperative measurement method and system of load sensor array

Abstract

The invention discloses a distributed cooperative measurement method and system for a load sensor array, and relates to the technical field of sensor networks and distributed measurement, and the method comprises the steps: executing the finite element simulation of a target structure, recognizing a stress key region, a boundary region and a load type, and building a structure and demand feature library; deploying a three-layer nested sensor array according to the feature library, configuring a communication protocol, and collecting data after global clock synchronization to establish a time sequence data set; and finally, outputting a load measurement result according to a cooperative authentication result through cooperative processing. The technical problems that in the prior art, a multivariable distributed measurement mechanism is lacked, intelligent sensors are insufficient in cooperation, structural load information is difficult to comprehensively and accurately obtain, and the requirements for accurate measurement and effective management and control of structural loads cannot be met are solved, and the purposes of comprehensively and accurately obtaining the structural load information and effectively managing and controlling the structural loads are achieved. And the technical effects of accurate measurement and effective management and control of the structural load are achieved.

Description

Technical Field

[0001] The present invention relates to the field of sensor networks and distributed measurement technology, and in particular to a distributed collaborative measurement method and system for a load sensor array. Background Art

[0002] Structural load measurement is crucial for the safe operation of equipment and the assessment of the stability of large structures. Existing technologies often use traditional centralized sensors or single-variable measurement devices for load detection, which have played a certain role in monitoring stable working conditions or simple structures. However, with the increasing requirements for complex structures and dynamic load monitoring, these methods have exposed limitations: due to the lack of multivariable distributed measurement capabilities and the inadequate coordination mechanism of intelligent sensors, they are unable to fully capture multi-dimensional load information in different areas, resulting in one-sided and inaccurate data, making it difficult to meet the needs of accurate assessment and effective management and control. Summary of the Invention

[0003] The present application provides a distributed collaborative measurement method and system for a load sensor array, which is used to solve the technical problems that the existing technology lacks a multivariable distributed measurement mechanism and insufficient collaboration of intelligent sensors, making it difficult to obtain structural load information comprehensively and accurately, and unable to meet the needs of precise measurement and effective control of structural loads.

[0004] The first aspect of the present application provides a distributed collaborative measurement method for a load sensor array, the method comprising: performing finite element simulation of a target structure, identifying key stress areas, boundary areas, and load types based on the finite element simulation results, and establishing a structural feature library and a demand feature library; deploying a sensor array based on the structural feature library and the demand feature library, the sensor array being a three-layer nested array, including a base layer node, a relay layer node, and a global layer node; after configuring a communication protocol for the sensor array, completing network topology initialization; after configuring global clock synchronization, activating the sensor array to collect node data and establish a time series data set; performing collaborative processing of the time series data set under the three-layer nested array based on the network topology, and outputting load measurement results based on the collaborative authentication results.

[0005] The second aspect of the present application provides a distributed collaborative measurement system for a load sensor array, the system comprising: a feature library construction module for performing finite element simulation of a target structure, identifying key stress areas, boundary areas, and load types based on the finite element simulation results, and establishing a structural feature library and a demand feature library; a sensor array deployment module for deploying a sensor array based on the structural feature library and the demand feature library, the sensor array being a three-layer nested array, including a base layer node, a relay layer node, and a global layer node; a network topology initialization module for completing network topology initialization after configuring a communication protocol for the sensor array; a timing data set construction module for activating the sensor array for node data acquisition and establishing a timing data set after configuring global clock synchronization; a load measurement result acquisition module for performing collaborative processing of timing data sets under the three-layer nested array according to the network topology, and outputting load measurement results based on collaborative authentication results.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages: This application establishes a structural feature library and a demand feature library by performing finite element simulation of the target structure, and accordingly deploys a three-layer nested intelligent sensor array consisting of base layer, relay layer, and global layer nodes. After network topology initialization and global clock synchronization, the collection is activated to establish a time series data set, and then through the collaborative processing and data authentication of the three-layer nodes, multi-variable distributed measurement is realized, thereby accurately obtaining the load information of the structure, making the load measurement results more accurate and reliable, achieving the comprehensive and accurate acquisition of structural load information, and meeting the technical effect of accurate measurement and effective control of structural loads. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0008] Figure 1 It is a flow chart of the distributed collaborative measurement method of the load sensor array provided in an embodiment of the present application.

[0009] Figure 2 It is a structural diagram of a distributed collaborative measurement system of a load sensor array provided in an embodiment of the present application.

[0010] Description of the accompanying symbols: feature library construction module 1, sensor array deployment module 2, network topology initialization module 3, time series data set construction module 4, load measurement result acquisition module 5. DETAILED DESCRIPTION

[0011] The present application provides a distributed collaborative measurement method and system for a load sensor array, which is used to solve the technical problems that the existing technology lacks a multivariable distributed measurement mechanism and insufficient collaboration of intelligent sensors, making it difficult to obtain structural load information comprehensively and accurately, and unable to meet the needs of precise measurement and effective control of structural loads.

[0012] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only some of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0013] It should be noted that the terms "first", "second", etc. in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products or devices.

[0014] Example 1, as Figure 1 As shown, a distributed collaborative measurement method of a load sensor array, wherein the method includes: Step A100: Execute finite element simulation of the target structure, identify key stress areas, boundary areas, and load types based on the finite element simulation results, and establish a structural feature library and a demand feature library.

[0015] In the embodiments of this application, finite element simulation refers to a simulation analysis of a target structure, which can provide results such as the critical stress areas, boundary areas, and load types of the structure. A structural feature library is a library established based on structural information such as the critical stress areas and boundary areas identified in the finite element simulation results. A demand feature library is a library established based on information such as the load types identified in the finite element simulation results.

[0016] Specifically, when performing finite element simulation of a target structure, comprehensive parameter collection is first required for the target structure, such as the main beam of a large bridge or the load-bearing frame of heavy machinery. This includes the structure's material properties, geometric dimensions, and actual connection methods, such as bolting and welding. Based on these parameters, a three-dimensional geometric model of the target structure is first constructed, and then discretized into a number of finite element units using meshing techniques. For critical areas with complex loads, such as near bridge supports and at the corners of machinery, a fine mesh with a unit size of 5cm×5cm can be used. A coarser mesh with a unit size of 20cm×20cm is used in non-critical areas to achieve a balance between computational accuracy and efficiency.

[0017] Then comes the construction of the mechanical calculation model: first, define the constraints, such as the fixed hinge support constraints at both ends of the bridge main beam to limit horizontal and vertical displacements; the fixed constraints connecting the mechanical frame to the ground to limit three-dimensional displacements. Then, set the load conditions according to the actual working scenario of the structure, including static loads, such as the deadweight of the bridge and the constant load of the machinery; dynamic loads, such as the impact load of vehicles passing over the bridge and the periodic loads during mechanical operation, and clarify the size, direction and position of each load. At the same time, introduce material constitutive relations (such as linear elastic models and elastic-plastic models) to describe the deformation law of the material after being subjected to force, set the iterative accuracy of the solver (such as the convergence error 1e-6) and the time step (set to 0.01s for dynamic simulation) to ensure that the mechanical calculation model can truly reflect the mechanical response of the structure.

[0018] After the mechanical calculation model is constructed, simulations are performed to numerically iterate and solve the equilibrium equations of each finite element, outputting data such as stress distribution, strain, and displacement under different operating conditions. This data is used to directly identify critical stress areas (i.e., areas where stress concentration exceeds 80% of the material's allowable stress); boundary areas (i.e., transition areas where constraints exist and stress gradients suddenly change); and load types (i.e., stable stress under static loads and periodic stress fluctuations over time under dynamic loads). This provides a quantitative basis for the subsequent establishment of a structural feature library and a demand feature library.

[0019] For example, when simulating a certain type of crane boom, it can be found that the stress value of the middle section of the boom under the rated load is 200MPa, which is much higher than the 50MPa at both ends. Based on this, the middle section is identified as the key stress area; at the same time, the connection between the boom and the base is marked as a boundary area due to stress concentration and a displacement change rate 40% higher than other areas. Based on the data output by the simulation, the load type is further analyzed. If the load of the boom at the moment of lifting in the simulation increases sharply from 0 to 1.2 times the rated value within 0.3 seconds, and is accompanied by high-frequency fluctuations, it is determined that there is an impact load; and in the uniform lifting stage, the load fluctuation amplitude is stable within ±5%, which is determined to be a static load. Through this type of analysis, the type and characteristic parameters of the load on the target structure are clarified.

[0020] Subsequently, the above identification results are integrated to establish a structural feature library and a demand feature library. The structural feature library contains the three-dimensional coordinates of key stress-bearing areas, stress thresholds, connection forms of boundary areas, and strain thresholds. The demand feature library determines measurement requirements based on load type. For example, impact loads require a sampling frequency of 1kHz to capture transient changes, while static loads require a sampling frequency of 0.1kHz to ensure data stability. It also includes indicator thresholds for abnormal monitoring such as overload and rapid change. For example, the overload threshold for impact loads is set at 200kN, and the rapid change threshold is set at 50kN / ms.

[0021] By performing finite element simulation of the target structure, identifying key stress areas, boundary areas and load types and establishing a corresponding feature library, the foundation is laid for the scientific deployment of the sensor array, achieving the effect of improving the targetedness of load measurement and the validity of data.

[0022] Step A200: deploying a sensor array according to the structural feature library and the demand feature library. The sensor array is a three-layer nested array, including a basic layer node, a relay layer node, and a global layer node.

[0023] Optionally, before deploying the sensor array, the core information of the structural feature library and the demand feature library must be clarified. The structural feature library records the three-dimensional coordinates and mechanical characteristics of the target structure's key stress-bearing areas and boundary areas. The demand feature library specifies measurement requirements based on static loads and impact loads. For example, impact loads require a sampling frequency of 1kHz to capture transient changes, while static loads require a sampling frequency of 0.1kHz to balance accuracy and energy consumption. It also includes thresholds for abnormal monitoring such as overload and rapid changes.

[0024] Based on the structural feature library, the base layer nodes are preferentially deployed in the key stress areas and boundary areas. Taking the robotic arm as an example, its joints are the key areas, and every 10cm 2 Deploy one base layer node to ensure that subtle stress changes can be captured; the deployment density in the boundary area is every 20cm 2 1, covering the stress gradient transition range, so that the collected data can directly reflect the load state of the core parts of the structure. These nodes integrate lightweight algorithms to meet the real-time preprocessing requirements of the demand feature library.

[0025] The deployment of relay-layer nodes is based on the regional divisions and base-layer node distribution in the structural feature library. If the structural feature library divides the robotic arm into three independent force-bearing zones, with 40-60 base-layer nodes in each zone, relay-layer nodes are deployed at the geometric center of each zone, ensuring a communication distance of no more than 8 meters to all base-layer nodes within that zone. Each relay-layer node is responsible for receiving and initially integrating data from 20-30 base-layer nodes, thus avoiding data delays caused by long communication distances while efficiently processing load information within the zone.

[0026] Global-layer nodes must be deployed across the entire target structure, with their number determined by the relay-layer distribution. For example, if each of the three regions of a robotic arm has a relay-layer node, the global-layer node can be deployed near the robotic arm's control system, ensuring a communication delay of no more than 30ms with each relay-layer node. This allows for rapid aggregation of data across regions, meeting the real-time requirements of global collaborative analysis while also meeting the global data integration accuracy requirement of <2% in the required feature library.

[0027] If sensor array deployment relies on manual experience, it can lead to insufficient nodes in key areas and redundant nodes in non-key areas, resulting in low data collection efficiency and missing core information. However, based on the deployment method of the structural feature library and the demand feature library described above, nodes at the base layer, relay layer, and global layer can accurately cover the required areas, and the configuration matches the load measurement requirements, enabling nodes at each layer to form efficient collaboration in data collection, transmission, and analysis.

[0028] By deploying a three-layer nested sensor array consisting of basic layer, relay layer, and global layer nodes in a hierarchical manner based on the structural feature library and the demand feature library, we ensure reasonable node coverage and adaptive configuration, thereby achieving the effect of improving the targeted load measurement and system coordination efficiency.

[0029] Step A300: After configuring the communication protocol for the sensor array, the network topology initialization is completed.

[0030] In one embodiment of the present application, when configuring the communication protocol for the sensor array, first, the adaptation protocol is selected in combination with the functional requirements of the three-layer nodes. Between the basic layer nodes and the relay layer nodes, it is necessary to transmit locally pre-processed time series data with high real-time requirements, and the delay needs to be <20ms. A low-power wireless protocol based on IEEE802.15.4 is adopted, and the communication rate is set to 250kbps. At the same time, a conflict avoidance mechanism is configured to ensure that data uploads of 30 basic layer nodes can be completed within 100ms. Between the relay layer nodes and the global layer nodes, it is necessary to aggregate the data of multiple nodes in the area, and the LoRaWAN protocol is used to balance the transmission distance and bandwidth. The spreading factor is set to 10 and the transmission rate is 50kbps to ensure that the communication packet loss rate between the relay layer and the global layer nodes is <1% within a range of 100 meters.

[0031] After completing the communication protocol configuration, the network topology initialization starts from the base layer node. The base layer node will actively send a broadcast signal to detect other base layer nodes and connectable relay layer nodes in the surrounding area, and screen and establish a first-level connection table based on signal strength and communication stability. The first-level connection table is the core communication connection information recorded by each base layer node. Specifically, it contains the communication addresses of 3-5 adjacent base layer nodes, which are used for local information sharing between nodes, and the communication address of 1 main relay layer node, which serves as the main data upload channel. To ensure transmission reliability, when the signal strength of the main relay layer node is lower than -85dBm, the first-level connection table will automatically switch to the pre-stored backup relay layer node address. This redundant design ensures that the base layer data can be stably uploaded to the relay layer.

[0032] After the base layer completes the establishment of the first-level connection table, the relay layer node initiates the next step of initialization. The relay layer node collects the first-level connection table information of all base layer nodes under its jurisdiction and integrates it to generate a second-level connection table. The second-level connection table not only contains the communication relationship between the relay layer and the base layer nodes under its jurisdiction, but also records the addresses of the two highest-priority adjacent relay layer nodes through interaction with other relay layer nodes. These addresses are used for cross-regional data collaboration. When data anomalies occur within a region and require correlation analysis, the relay layer can quickly forward information through the adjacent addresses in the second-level connection table, achieving time consistency analysis and data linkage between regions.

[0033] The secondary connection table of the relay layer is eventually uploaded to the global layer node, which then constructs a global topology map based on it. The global topology map is a panoramic presentation of the connection relationship of the entire sensor network. It not only includes the communication addresses of all base layer and relay layer nodes, but also marks the real-time status of each node, such as key parameters such as online duration and signal stability. To ensure the accuracy of the topology, the global topology map is updated every 5 seconds. If a node fails to communicate three times in a row, it will be temporarily removed from the map to avoid invalid data transmission and occupying resources. Through the global topology map, the global layer nodes can clearly grasp the connection logic of the entire network, providing a complete network structure basis for the collaborative processing of time series data sets by the subsequent three-layer nodes.

[0034] By configuring an adaptive communication protocol for the three-layer nested sensor array and dynamically initializing the network topology, efficient connection and data transmission of each node are ensured, thereby improving the system communication reliability and real-time data transmission.

[0035] Step A400: After configuring global clock synchronization, activate the sensor array to collect node data and establish a time series data set.

[0036] Specifically, when configuring global clock synchronization, the global layer node is used as the time reference source, and a time synchronization signal is sent to each relay layer node through a preset high-precision synchronization protocol, that is, the precise time protocol based on IEEE 1588. After receiving the signal, the relay layer node forwards the signal to the base layer node under its jurisdiction, forming a hierarchical synchronization link of the global layer-relay layer-base layer.

[0037] During the synchronization process, the global layer node sends a timestamp every 10 milliseconds. After receiving it, the relay layer node calibrates the local clock to ensure that the time deviation from the global benchmark does not exceed 5 microseconds. Then, the synchronization signal is forwarded to the base layer node at the same frequency. The base layer node compares the received timestamp with the local clock and adjusts the crystal oscillator frequency to control the time deviation within ±10 microseconds.

[0038] At the same time, the system will monitor the synchronization status of each node in real time. If the synchronization deviation of a node exceeds 20 microseconds for three consecutive times, the global layer node will trigger the forced synchronization mechanism, resend the high-precision time reference and record the synchronization exception log to ensure that the collection time of all nodes in the basic layer, relay layer and global layer remains consistent, providing a reliable time reference for the subsequent construction of time series data sets and the time series consistency analysis of the relay layer.

[0039] Next, the sensor array is activated to collect node data to establish a time series data set. A front-end event perception unit is configured for each sensor node. During collection, event perception is performed based on the unit to establish node perception results. The adaptive collection window is updated through the shared perception results of the first-level neighborhood set, and then the window is used to continue collection. The specific steps are described in detail in A410-A440.

[0040] By configuring global clock synchronization to ensure the time consistency of each node, combined with the node data collection after activating the sensor array, a time series data set is established to provide basic data for time alignment for subsequent three-layer collaborative processing.

[0041] Step A500: executing collaborative processing of time series data sets in a three-layer nested array according to the network topology, and outputting load measurement results according to collaborative authentication results.

[0042] Specifically, the collaborative processing of time series data sets under the three-layer nested array includes base layer preprocessing to establish the first abnormal load identification, relay layer analysis to establish the second abnormal load identification, map structure construction based on network topology and structural feature library, global layer collaborative analysis and combining multiple information for data collaborative authentication and output of load measurement results. The specific steps are detailed in A510-A550.

[0043] Furthermore, step A500 in the method provided in the embodiment of the present application includes: A510: Perform local node preprocessing of the time series data set at the base layer node to establish a first abnormal load identifier.

[0044] A520: Upload the time series data set and the first abnormal load identifier to the relay layer node, perform load analysis within the region, and establish a second abnormal load identifier.

[0045] A530: Establish a graph structure based on the network topology and the structural feature library, where the nodes of the graph structure are sensor nodes, and the edges of the graph structure represent information channels or structural couplings between any two nodes.

[0046] A540: Coupling the graph structure to a global layer node, performing global collaborative analysis, and establishing a global collaborative analysis result.

[0047] A550: Performs data collaborative authentication based on the global collaborative analysis results, the first abnormal load identifier, the second abnormal load identifier, and the time series data set, and outputs the load measurement results.

[0048] In the embodiment of the present application, the graph structure is established based on the network topology and structural feature library, the nodes are sensor nodes, and the edges represent the information channel or structural coupling between any two nodes.

[0049] Specifically, first, the base layer node preprocesses the time series data set through a lightweight algorithm, configures abnormal indicators such as overload, rapid change, and non-stationary fluctuation, and identifies the abnormal preprocessing results based on the indicators to establish the first abnormal load identification. The specific steps are described in detail in A511-A513.

[0050] Next, the relay layer node obtains the communication relationship with the base layer node and calls the timing data set to establish a local perception subgraph. Based on this graph and the first abnormal load identifier, the timing consistency of the nodes in the area is analyzed to identify synchronization, propagation, and isolated anomalies, and then the second abnormal load identifier is established. The specific steps are described in detail in A521-A523.

[0051] Then, when building a graph structure based on the network topology and structural feature library, all nodes in the sensor array, including the base layer, relay layer, and global layer nodes, are first used as vertices of the graph structure. Each vertex is assigned attributes such as the node number, the layer it belongs to, and the coordinates of the monitoring area. Next, the information channels between nodes are determined based on the network topology: the communication relationships between each node are extracted from the network topology, such as the communication link between the base layer node and the relay layer node that governs it, and the adjacent communication paths between relay layer nodes. If two nodes have stable data transmission, that is, a communication success rate ≥90% and a delay ≤50ms, they are connected by a communication edge in the graph. The edge weight is set according to the communication quality, such as 0.8-1.0 for high signal strength and 0.3-0.7 for weaker signal strength.

[0052] At the same time, the structural coupling relationship is supplemented based on the structural feature library: the structural connection information of the sensor node location is obtained from the structural feature library. For example, whether two nodes are located on the same force flow transmission path can be determined by the stress distribution of finite element simulation. Whether they belong to adjacent critical stress areas or boundary areas can be determined. If there is a physical structural force transmission relationship, they are connected by a structural edge. The edge weight is set according to the strength of the structural coupling. For example, the weight of the node with direct rigid connection is 0.9, and that of the node with indirect transmission is 0.4-0.8. In the final graph structure, each edge can represent either the information channel or the structural coupling independently, or it can combine the properties of both to fully present the communication connection and structural relationship between sensor nodes.

[0053] Afterwards, the graph structure is coupled to the global layer node to perform global collaborative analysis to establish the global collaborative analysis results, including parsing the graph structure to construct a double verification graph containing a structural subgraph constructed through structural connection relationships and a similar subgraph constructed based on node feature similarity. Based on this graph, collaborative analysis is performed with the central node as the starting point, and the global collaborative analysis results are output based on all the double verification graphs. The specific steps are described in detail in A541-A542.

[0054] Finally, during data collaborative verification, the global collaborative analysis results are first cross-checked with the first and second abnormal load identifiers to verify whether the anomalies identified at different levels are logically related. For example, whether an overload anomaly identified at the base layer manifests as a propagation anomaly in the regional timing consistency analysis at the relay layer and corresponds to a structural jump anomaly or similar island anomaly identified at the global layer. Subsequently, the time series dataset is retrieved and the raw data, such as timestamps and load values, corresponding to each anomaly identifier are verified to confirm the temporal synchronization of the anomaly occurrence (based on global clock synchronization results) and data continuity. For example, whether the load changes before and after the anomaly conform to the mechanical laws in the structural feature library are consistent. For conflicting anomalies, such as a base layer anomaly not supported by the relay layer or global analysis, a secondary verification is performed using detailed features in the time series data, such as sampling frequency and fluctuation trends, to eliminate misjudgments caused by single-node errors or local interference. Finally, all information that passes collaborative verification is integrated to form a load measurement result that includes normal load data, abnormal load types confirmed at multiple levels, and detailed information about their occurrence, ensuring the accuracy and reliability of the output results.

[0055] Through the hierarchical collaborative processing of three-layer nested arrays, namely base layer preprocessing, relay layer regional analysis, global layer collaborative analysis, and multi-information collaborative authentication of global collaborative results, two-level anomaly identification, and time series data sets, the technical effect of improving the accuracy and reliability of load measurement results is achieved.

[0056] Furthermore, step A510 in the method provided in the embodiment of the present application includes: A511: Use lightweight algorithms integrated in base layer nodes to perform time series data set preprocessing and establish preprocessing results.

[0057] A512: Configure abnormal indicators associated with the base layer nodes, including overload indicators, rapid change indicators, and non-stationary fluctuation indicators.

[0058] A513: Perform abnormal identification on the corresponding preprocessing result according to the abnormal indicator, and establish the first abnormal load identifier.

[0059] Optionally, when the base layer node performs local preprocessing on the time series data set, it calls the integrated simplified Kalman filter algorithm to first initialize the original load data of 100 sampling points per second, and set the initial state vector, which contains the current load estimate and error covariance. The initial load estimate takes the average of the first 5 sampling points, and the error covariance is set to 0.1 to reflect the initial uncertainty.

[0060] After entering the prediction phase, the algorithm uses the previous load value and the preset state transition matrix, along with structural dynamic characteristics, such as a load change rate upper limit of 5kN / ms, to predict the current load estimate and error range. For example, based on 100kN at time t-1, the load at time t is predicted to be 102kN±0.5kN. The update phase then begins, comparing the current actual sampled value (e.g., 103kN at time t) with the predicted value. The predicted value is corrected using the Kalman gain, and the prediction error is calculated based on the measurement noise variance (set to 0.8 to filter out high-frequency interference). This results in a filtered load value, removing instantaneous high-frequency noise and matching the overall trend.

[0061] Next, the algorithm applies sliding window smoothing to the filtered sequence, with a window size of five sampling points. The mean of each window is calculated as the smoothed value at that moment, making the data curve smoother. Finally, key eigenvalues ​​are extracted: by traversing the smoothed sequence, peak values ​​exceeding the values ​​of three adjacent points and valley values ​​below the values ​​of three adjacent points are identified, and the slope of change between consecutive peak values ​​is calculated. The final preprocessing result includes the filtered value, smoothed value, peak value, valley value, and slope of change. This not only preserves the core trend but also compresses the data volume to reduce the space occupied.

[0062] Subsequently, specific abnormality indicators were configured for the base-layer nodes based on the target structure's stress characteristics and the monitoring standards in the required feature library. The overload indicator was set at 1.2 times the allowable material load in the area. For example, if the material's allowable load was 160kN, the overload indicator would be 192kN. The rapid change indicator was set at 50kN / ms; any load change exceeding this value per unit time was considered abnormal. The non-stationary fluctuation indicator was set at a load fluctuation exceeding ±10% at three consecutive sampling points. For example, if the baseline value was 100kN, the fluctuation would exceed the range of 90kN-110kN.

[0063] Next, the preprocessing results were compared with the aforementioned abnormality indicators one by one to identify anomalies. For example, if the load at a certain moment in the preprocessing results reached 200kN, exceeding the overload indicator of 192kN, this would be marked as an overload anomaly. A load increase from 180kN to 195kN within 0.02 seconds, resulting in a calculated rapid change of 750kN / s, exceeding 50kN / ms, would be marked as a rapid change anomaly. Furthermore, if the loads at three consecutive sampling points were 100kN, 112kN, and 98kN, and the fluctuation range was greater than ±10% within ±12%, this would be marked as a non-stationary fluctuation anomaly. These anomaly types and occurrence times were integrated to establish the first abnormal load identification.

[0064] Through lightweight algorithm preprocessing of basic layer nodes, targeted abnormal indicator configuration and abnormal identification, the first abnormal load identification is established, which achieves the effect of reducing data transmission volume and improving the timeliness and accuracy of abnormal identification.

[0065] Furthermore, step A520 in the method provided in the embodiment of the present application includes: A521: Obtain the communication relationship between the relay layer nodes and the base layer nodes, call the time series data set according to the communication relationship, and establish a local perception subgraph.

[0066] A522: Based on the local perception subgraph and the first abnormal load identifier, a timing consistency analysis of the nodes in the area is performed to establish a timing consistency anomaly. The timing consistency anomaly includes a synchronization anomaly, a propagation anomaly, and an isolation anomaly.

[0067] A523: Create a second abnormal payload identifier based on the timing consistency exception.

[0068] Specifically, the relay layer node first obtains the communication relationship with the base layer node. This relationship is based on the secondary connection table constructed when the network topology is initialized in step A300. It clarifies the range of base layer nodes under the jurisdiction of each relay layer node. For example, a relay layer node connects to 20-30 base layer nodes. At the same time, it records the communication signal strength between each node (for example, normal communication requires a signal strength ≥-70dBm) and the data transmission frequency 30 times per second. Based on this communication relationship, the relay layer node calls the time series data set of the corresponding base layer node. These data sets have been pre-processed by the base layer and contain timestamps and load values ​​accurate to microseconds. Then, these base layer nodes are used as subgraph nodes, with communication relationships as edges. Edges with high signal strength are given high weights, such as 0.8-1.0, to construct a local perception subgraph, which intuitively presents the connection and data transmission status of nodes in the area.

[0069] Next, based on the local perception subgraph and the first abnormal load identifier uploaded by the base layer, the relay layer node performs a timing consistency analysis within the region. The identification of synchronization anomalies is based on the global clock synchronization standard. If the timestamp deviation between a base layer node and more than 80% of the nodes in the region exceeds the propagation characteristics of the load in the structural feature library, for example, according to the finite element simulation results, the theoretical time for the load to propagate from node A to the adjacent node B is 20 milliseconds. If the propagation delay in the actual data reaches 100 milliseconds and occurs more than 3 times in a row, it is marked as a propagation anomaly. An isolated anomaly refers to the first anomaly identifier of a base layer node. For example, if the overload anomaly is significantly different from that of more than 3 neighboring nodes and cannot be explained by the communication link, if the signals of the surrounding nodes are normal but there are no anomalies, for example, if a node reports an overload 3 times in a row, while the load of the adjacent node is stable within the normal range, it is determined to be an isolated anomaly.

[0070] Finally, based on the identified synchronization anomalies, propagation anomalies, and isolated anomalies, the relay layer nodes integrate them to form a second abnormal load identification. This identification includes the anomaly type, such as synchronization anomaly; the node number involved, such as base layer node 5; the severity of the anomaly, such as a time deviation of 50 microseconds; and associated first anomaly identification information, such as previous overload anomalies at the node, providing a more comprehensive description of the load status within the region.

[0071] By acquiring communication relationships to establish a local perception subgraph, analyzing temporal consistency to identify synchronization, propagation, and isolated anomalies, and establishing a second abnormal load identifier, the comprehensiveness and accuracy of load anomaly analysis within the region are improved, providing reliable regional-level data support for global collaborative processing.

[0072] Furthermore, step A540 in the method provided in the embodiment of the present application includes: A541: Perform structural analysis on the graph structure to construct a node-centered dual verification graph, wherein the dual verification graph includes a structural subgraph and a similarity subgraph. The structural subgraph is constructed through structural connection relationships, and the similarity subgraph is constructed based on node feature similarity.

[0073] A542: Perform collaborative analysis based on the double verification graph with the central node as the starting point, and output a global collaborative analysis result based on the entire double verification graph.

[0074] Specifically, after receiving the graph structure, the global-level nodes first perform structural analysis. During this analysis, attribute information is extracted for all nodes in the graph, such as their level, monitoring area, load measurement range, and edge association types, such as information channels or structural coupling. Combined with the target structural connectivity relationships in the structural feature library, such as the rigid connections between critical and boundary areas and the force flow transmission paths, edges related to the physical connections of the structure are selected to construct a structural subgraph. For example, sensor nodes belonging to the same critical stress area and rigidly connected by bolts are connected by edges. Edge weights are set based on the connection strength, with a weight of 0.9 for rigid connections and 0.5 for flexible connections.

[0075] At the same time, the characteristic vector of each node is calculated, including the sampling frequency, historical load mean, anomaly recognition threshold, etc. The node feature similarity is calculated using the cosine similarity algorithm, and the nodes with similarity ≥ 0.7 are connected with edges to construct a similar subgraph. The weight of the edge is the similarity value, and finally a double verification graph centered on each node is formed.

[0076] After completing the construction of the double verification graph, a collaborative analysis is performed based on the graph with the central node as the starting point. By analyzing the double verification graph of each central node, the collaborative analysis results of all nodes are summarized to form a global collaborative analysis result. The specific steps are described in detail in A542-1-A542-5.

[0077] By analyzing the graph structure and constructing a double verification graph, combined with collaborative analysis initiated by the central node and summarizing the results, the accuracy and comprehensiveness of the global collaborative analysis are improved, providing a reliable global basis for the certification of load measurement results.

[0078] Furthermore, step A542 in the method provided in the embodiment of the present application includes: A542-1: Taking the central node as a starting point, determine multi-order neighbors based on the structural subgraph and the similar subgraph respectively, and establish a structural neighborhood set and a similar neighborhood set.

[0079] A542-2: Use the structural neighborhood set to perform force flow consistency analysis between nodes and establish structural jump anomalies.

[0080] A542-3: Use the similar neighborhood set to perform group deviation analysis and establish similar island anomalies.

[0081] A542-4: Configure the load consistency score based on the structural jump anomaly and configure the similarity deviation score based on the similar island anomaly.

[0082] A542-5: Establish joint anomaly confidence based on load consistency score and similarity deviation score to output global collaborative analysis results.

[0083] In the embodiment of the present application, force flow consistency analysis is to use the structural neighborhood set to perform force flow consistency analysis between nodes to establish the analysis process of structural jump anomalies. Group deviation analysis is to establish the analysis process of similar island anomalies.

[0084] Specifically, taking the central node as the starting point, in the structural subgraph, multi-order neighbors are determined based on the structural connection relationship, including nodes directly connected to the central node through structural edges, as well as nodes indirectly connected through these directly connected nodes, and these nodes are integrated to form a structural neighborhood set; at the same time, in the similarity subgraph, based on the similarity of node features, such as sampling features, monitoring area attributes, etc., directly similar nodes with similar features to the central node and indirectly similar nodes associated with these directly similar nodes are determined, and integrated to form a similar neighborhood set.

[0085] Next, when analyzing the force flow consistency between nodes using the structural neighborhood set, the load transfer status of each node within the neighborhood is compared with the force flow transfer patterns of the target structure recorded in the structural feature library, such as the continuity and attenuation characteristics of force flow transfer in different regions. If there is a significant discontinuity between the load transfer status of a node and that of adjacent nodes, such as if the load change at a node along a normal force flow path has no reasonable correlation with the load change at adjacent nodes, it is determined to be a structural jump anomaly.

[0086] When using similar neighborhood sets for group deviation analysis, focus on the common load characteristics of nodes within similar neighborhoods, such as load fluctuation range and change trend. If the load characteristics of a node are significantly different from those of most nodes in the neighborhood, for example, under similar working conditions, the load fluctuation of the node significantly exceeds the normal fluctuation range of other nodes, and this difference cannot be explained by differences in node characteristics, it is determined to be a similar island anomaly.

[0087] When configuring the load consistency score, the core consideration is the ratio of the deviation value of the structural jump anomaly to the corresponding threshold, strictly adhering to the principle that the larger the deviation, the lower the score. First, set a threshold for the structural jump anomaly. For example, based on the force flow transmission rules in the structural feature library, the threshold for a certain area is set to 12kN. This means that any deviation between the actual load and the theoretical value exceeding 12kN is considered a structural jump anomaly. The score is calculated as: load consistency score = 1 - (deviation value / threshold value), and the result must be constrained to fall within the range of 0-1. That is, when the deviation value is ≥ the threshold, the score is 0; when the deviation value is ≤ 0, the score is 1. For example, if the deviation value is 6kN and does not reach the threshold, the score is 1 - (6 / 12) = 0.5; if the deviation value is 12kN, it just reaches the threshold, and the score is 1 - (12 / 12) = 0; if the deviation value is 15kN and exceeds the threshold, the score is also 0. If the global synchronization threshold of 10 microseconds is fully met, it is considered a synchronization anomaly. The propagation anomaly reflects the logic that the score decreases as the deviation increases.

[0088] When configuring the similarity deviation score, the ratio of the deviation value of similar island anomalies to the corresponding threshold is calculated. Similarly, the greater the deviation, the lower the score. First, set a threshold for similar island anomalies. For example, based on the similarity of node features within a similar neighborhood, set a threshold of 20 kN for a certain area. This means that any node load that deviates from the mean of the similar neighborhood by more than 20 kN is considered a similar island anomaly. The score is calculated as: Similarity Deviation Score = 1 - (Deviation Value / Threshold Value). The result is constrained to a range of 0-1 (0 for deviations ≥ threshold; 1 for deviations ≤ 0). For example, if the deviation value is 10 kN (below the threshold), the score is 1 - (10 / 20) = 0.5; if the deviation value is 20 kN (just above the threshold), the score is 1 - (20 / 20) = 0; and if the deviation value is 30 kN (above the threshold), the score is 0. This clearly reflects the trend that the score decreases as the deviation increases.

[0089] Finally, to establish the joint anomaly confidence score, a weighted summation is used to integrate the load consistency score and similarity deviation score. The weights can be set based on the priority of structural importance and feature similarity, for example, 0.5 for each. For example, the load consistency score of 0.3 and the similarity deviation score of 0.2 for a node are first obtained. The joint anomaly confidence score is then calculated using the formula: load consistency score × weight + similarity deviation score × weight, resulting in 0.3 × 0.5 + 0.2 × 0.5 = 0.25. The result is then compared with a preset significant anomaly threshold (e.g., 0.3). If the score is below the threshold (e.g., 0.25 < 0.3), the anomaly is considered non-significant. If the score is above or equal to the threshold (e.g., 0.4 > 0.3 for a node), the anomaly is considered significant. Finally, the results of all nodes are integrated to form a global collaborative analysis result.

[0090] By determining multi-order neighbors to establish sets, identifying two-dimensional anomalies, configuring quantitative scores, and building joint anomaly confidence levels, the comprehensiveness and accuracy of global collaborative analysis are improved, providing a reliable global judgment basis for load measurement results.

[0091] Furthermore, step A400 in the method provided in the embodiment of the present application includes: A410: A front-end event sensing unit is configured in each sensor node of the sensor array.

[0092] A420: When executing node data collection, event perception is performed based on the front-end event perception unit to establish a node perception result.

[0093] A430: Configure a primary neighborhood set of the sensor node, share node perception results according to the primary neighborhood set, and update the adaptive acquisition window.

[0094] A440: Continuously collect data using the adaptive collection window to establish a time series data set.

[0095] In this embodiment, the pre-event sensing unit is a unit configured within each sensor node in the sensor array. It senses events during node data collection and generates node perception results, which serve as a basis for subsequent sharing of these results across the primary neighborhood and updating the adaptive acquisition window. The adaptive acquisition window, used to continuously collect data, dynamically adjusts parameters such as the sampling frequency based on the perceived events.

[0096] In one embodiment, before activating the sensor array for node data collection, a pre-event sensing unit is first deployed at each sensor node (including base-layer and relay-layer nodes). This unit integrates a simple signal discrimination module to monitor sudden changes in load signals in real time, such as the 50kN / ms rate of change in the rapid-change indicator and the sudden amplitude change in the non-stationary fluctuation indicator. This allows for rapid response to potential load events without relying on complex calculations. Conventional sensors often collect data at a fixed frequency, such as 100 times per second. This can easily generate redundant data during stable load conditions or lose critical information during transient changes due to excessive sampling intervals. Pre-event sensing units, however, can detect event signs in advance, providing a basis for dynamically adjusting collection strategies.

[0097] When performing node data collection, the front-end event perception unit (FEP) first activates, continuously monitoring the dynamic signals of the local load in real time. It calculates the instantaneous value of the load, the change per unit time, and the smoothness of the signal fluctuations, such as whether the fluctuation amplitude exceeds ±5% for 10 consecutive sampling points. If any metric exceeds a preset threshold, such as a rapid change metric exceeding 50kN / ms, a load exceeding the overload threshold of 200kN, or a fluctuation amplitude exceeding ±5% for three consecutive sampling points, a perception response is immediately triggered. The precise time of the event is recorded, along with a globally synchronized timestamp, the current load value, the specific metric type exceeding the threshold, and details of the change, such as an increase from 290kN to 360kN in 0.05 seconds. This generates a node perception result and uniformly marks it as a potential anomaly. This design eliminates the need for complex data processing, relying solely on a lightweight algorithm for preliminary judgment. This allows for rapid generation of perception results, providing immediate basis for subsequent information sharing within the first-level neighborhood set and adaptive acquisition window adjustment. This ensures rapid detection of load anomalies while avoiding response delays caused by complex analysis.

[0098] Subsequently, a first-level neighborhood set is configured for each sensor node. This is based on the first-level connection table constructed during network topology initialization in step A300, which determines each node's three to five neighboring nodes, such as the neighboring base-layer nodes of a base-layer node. Each node uses this set to share its own node perception results. For example, when a base-layer node senses a rapidly changing load, it sends the result to the other four nodes in its first-level neighborhood. Upon receiving the result, the neighboring nodes compare their own perception results. If two or more nodes also sense a similar signal, it is considered a significant event in the area, requiring improved acquisition accuracy. If only a single node senses the signal, it is considered a localized interference, and regular acquisition is maintained. Based on this shared analysis, the node automatically updates its adaptive acquisition window: under normal conditions, the window is set to 100ms per sampling, but is adjusted to 10ms per sampling when a significant event occurs in the area. In the event of localized interference, the window remains at 100ms but the monitoring duration is extended, ensuring that critical data is captured while minimizing invalid acquisitions.

[0099] Finally, using the updated adaptive collection window, each node continues to perform data collection: the base layer nodes record the real-time value of the load at the adjusted frequency, such as generating 100 data points per second at 10ms / time, and the relay layer nodes synchronously collect the summary signal in the area. All collected data are accompanied by a timestamp after global clock synchronization, and are finally integrated into a time series data set containing time series, signal strength, and event markers.

[0100] If sensor data is collected using a fixed window, transient loads may be missed or data redundancy may occur during steady-state conditions, resulting in time series data sets that either lack key information or have a high proportion of invalid data. However, by leveraging pre-event awareness, neighborhood sharing, and adaptive window adjustment, the collection strategy can dynamically match load variation characteristics, ensuring that time series data contains complete details of abnormal events while avoiding the accumulation of redundant information.

[0101] By configuring a front-end event perception unit for the sensor node, combining the shared perception results of the first-level neighborhood set to update the adaptive acquisition window, and then continuously acquiring and establishing a time series data set, the pertinence and efficiency of data acquisition are improved, and a high-quality time series data set is constructed.

[0102] Furthermore, step A500 in the method provided in the embodiment of the present application includes: A610: Abnormal load identification is performed based on the load measurement result, and abnormal load identification attention is established.

[0103] A620: Call the historical load database, perform verification based on the historical load database and the abnormal load identifier, and update the abnormal load identifier.

[0104] Optionally, when abnormal load identification is performed based on the load measurement results after collaborative certification, it is necessary to combine the load change characteristics in the measurement data, such as numerical value, change rate, and fluctuation stability, and compare them with the normal load range of the area in the structural feature library, such as the normal fluctuation range under static conditions and the peak upper limit under dynamic conditions, to identify load conditions that exceed the normal range. For example, when the load in a key area continuously exceeds the allowable range of the material, or changes drastically in a short period of time, or the fluctuation amplitude far exceeds the normal level during stable operation, these conditions will be marked as abnormal and prioritized according to the potential impact of the abnormality (such as whether it involves core load-bearing components and whether it may cause structural damage), forming a list of abnormal load identifications that require special attention.

[0105] When calling the historical load database, the database must contain historical load records of the target structure under different working conditions, abnormal cases that occurred in the past, and processing results, such as the frequency, duration, and final impact of similar abnormalities. Compare each abnormality in the abnormal load identification watch list with historical data: if the characteristics of the current abnormality (such as the occurrence area, load change pattern) are highly consistent with a certain type of abnormality that has appeared many times in history, and past cases show that it has limited impact on the structure, it can be confirmed as a common abnormality; if the current abnormality has never appeared in the historical records, or the characteristics are significantly different from the known abnormalities, it is determined to be a new abnormality. Update the abnormal load identification based on the comparison results, and supplement the historical correlation information or new tags of the abnormality. Examples of verification and update of different types of abnormalities are shown in Table 1. Using matching analysis of historical data, the attributes of the current abnormality can be clarified to avoid misjudging interference as serious abnormalities, and at the same time identify new abnormalities to increase attention.

[0106] By marking anomalies based on measurement results, establishing attention, and calling the historical database to verify and update the identification, the accuracy and pertinence of abnormal load identification are improved, misjudgment is reduced, and new anomalies are identified in a timely manner.

[0107] Table 1: Exception type verification and update status table Exception Type Current measurement feature Historical database matching features Verify the results Updated logo Overload exception 200kN, 10s In a similar area in a certain year, the load was 198kN and 205kN, lasting 8-12s. Historical Re-enactment Overload anomaly (historical recurrence, an average of 2 times per year) Rapid change anomaly 75kN / ms, instantaneous The historical maximum is 60kN / ms, with no similar transient characteristics New anomaly Abnormal rapid change (first occurrence, requires intensive monitoring) Non-stationary fluctuation anomaly Fluctuation of 5 consecutive sampling points ±12% There were three fluctuations of ±10%-15% in a certain year, all of which were interference Interference type Abnormal non-stationary fluctuations (caused by interference, no warning for the time being) Furthermore, step A620 in the method provided in the embodiment of the present application includes: A621: Identify the updated abnormal load identification according to the warning level and configure visual warning anomalies.

[0108] A622: Perform early warning management based on the visual early warning anomaly and early warning level identification results.

[0109] In one embodiment, when processing the updated abnormal load identification, it is necessary to identify the warning level in combination with the characteristics of the abnormality. These characteristics include the type of abnormality (such as overload, rapid change), the area where it occurs (such as the critical stress area or the boundary area), the duration, and the degree of impact it may have on the structure. By comprehensively evaluating these characteristics, the abnormalities are divided into different warning levels, such as a low level reflecting a minor abnormality, a medium level indicating that attention is needed, and a high level indicating that there may be serious risks. At the same time, visual warning abnormalities are configured for these different levels of abnormalities, and abnormal information is presented in an intuitive way, such as using different colors to identify different levels, and using graphics to show the location and range of the abnormality, so that relevant personnel can quickly understand the abnormal situation.

[0110] Afterwards, early warning management is performed based on the configured visual warning anomalies and the identified warning levels. Low-level warnings may simply be recorded internally in the system and alert maintenance personnel to review them regularly. Medium-level warnings require prompt notification to relevant personnel and arrangements for inspections. High-level warnings immediately trigger an emergency response mechanism, notifying decision-makers and initiating emergency response procedures to ensure that relevant personnel can take appropriate measures based on the severity of the warning.

[0111] By identifying and visually configuring the warning levels of abnormal load signs, and then conducting graded warning management based on the results, the warning information is made clear and intuitive, and the response measures are accurate and effective.

[0112] In summary, the distributed collaborative measurement method of the load sensor array provided in the embodiments of the present application has the following technical effects: This application establishes a structural feature library and a demand feature library by performing finite element simulation of the target structure, and deploys a three-layer nested sensor array based on this. After network topology initialization and global clock synchronization, the nodes are activated to collect data, and then the three-layer nodes collaboratively process the time series data set and perform data collaborative authentication to output the load measurement results, making the load measurement more accurate and reliable, achieving comprehensive and accurate acquisition of structural load information, and meeting the technical effect of accurate measurement and effective control of structural loads.

[0113] Example 2, as Figure 2 As shown, based on the same inventive concept as the aforementioned embodiment 1, this embodiment of the present application provides a distributed collaborative measurement system for a load sensor array, the system comprising: The feature library construction module 1 is used to perform finite element simulation of the target structure, identify key stress areas, boundary areas, and load types based on the finite element simulation results, and establish a structural feature library and a demand feature library.

[0114] The sensor array deployment module 2 is used to deploy the sensor array according to the structural feature library and the demand feature library. The sensor array is a three-layer nested array, including a basic layer node, a relay layer node, and a global layer node.

[0115] The network topology initialization module 3 is used to complete the network topology initialization after configuring the communication protocol for the sensor array.

[0116] The time series data set construction module 4 is used to activate the sensor array to collect node data and establish a time series data set after configuring global clock synchronization.

[0117] The load measurement result acquisition module 5 is used to perform collaborative processing of time series data sets under the three-layer nested array according to the network topology, and output the load measurement result according to the collaborative authentication result.

[0118] Furthermore, the load measurement result acquisition module 5 is configured to perform the following steps: Perform local node preprocessing of the time series data set at the base layer node to establish a first abnormal load identifier; upload the time series data set and the first abnormal load identifier to the relay layer node, perform load analysis within the region, and establish a second abnormal load identifier; establish a graph structure based on the network topology and structural feature library, the nodes of the graph structure are sensor nodes, and the edges of the graph structure represent the information channels or structural couplings between any two nodes; couple the graph structure to the global layer node, perform global collaborative analysis, and establish a global collaborative analysis result; perform data collaborative authentication based on the global collaborative analysis result, the first abnormal load identifier, the second abnormal load identifier, and the time series data set, and output the load measurement result.

[0119] Furthermore, the load measurement result acquisition module 5 is configured to perform the following steps: Use the lightweight algorithm integrated in the base layer node to perform time series data set preprocessing and establish the preprocessing results; configure abnormal indicators associated with the base layer nodes, the abnormal indicators include overload indicators, rapid change indicators, and non-stationary fluctuation indicators; perform corresponding preprocessing result abnormality identification based on the abnormal indicators, and establish the first abnormal load identifier.

[0120] Furthermore, the load measurement result acquisition module 5 is configured to perform the following steps: Obtain the communication relationship between the relay layer nodes and the base layer nodes, call the timing data set according to the communication relationship, and establish a local perception subgraph; perform timing consistency analysis of the nodes in the area based on the local perception subgraph and the first abnormal load identifier, and establish timing consistency anomalies. The timing consistency anomalies include synchronization anomalies, propagation anomalies, and isolated anomalies; establish a second abnormal load identifier based on the timing consistency anomaly.

[0121] Furthermore, the load measurement result acquisition module 5 is configured to perform the following steps: The graph structure is structurally analyzed to construct a node-centered double verification graph, which includes a structural subgraph and a similarity subgraph. The structural subgraph is constructed through structural connection relationships, and the similarity subgraph is constructed based on the similarity of node features. Based on the double verification graph, a collaborative analysis is performed with the central node as the starting point, and a global collaborative analysis result is output based on all the double verification graphs.

[0122] Furthermore, the load measurement result acquisition module 5 is configured to perform the following steps: Taking the central node as the starting point, multi-order neighbors based on the structural subgraph and similar subgraph are determined respectively to establish a structural neighborhood set and a similar neighborhood set; the structural neighborhood set is used to perform force flow consistency analysis between nodes to establish a structural jump anomaly; the similar neighborhood set is used to perform group deviation analysis to establish a similar island anomaly; the load consistency score is configured according to the structural jump anomaly, and the similarity deviation score is configured according to the similar island anomaly; a joint anomaly confidence is established based on the load consistency score and the similarity deviation score to output a global collaborative analysis result.

[0123] Furthermore, the time series data set construction module 4 is used to perform the following steps: A front-end event sensing unit is configured at each sensor node in the sensor array; when executing node data collection, event perception is performed based on the front-end event sensing unit to establish a node perception result; a first-level neighborhood set of the sensor node is configured, the node perception result is shared according to the first-level neighborhood set, and an adaptive collection window is updated; and continuous data collection is performed using the adaptive collection window to establish a time series data set.

[0124] Furthermore, the load measurement result acquisition module 5 is configured to perform the following steps: Abnormal load identification is performed based on the load measurement result, and abnormal load identification attention is established; a historical load database is called, attention verification is performed according to the historical load database and the abnormal load identification attention, and the abnormal load identification is updated.

[0125] Furthermore, the load measurement result acquisition module 5 is configured to perform the following steps: The updated abnormal load identification is identified by the warning level, and a visual warning abnormality is configured; and the early warning management is performed according to the visual warning abnormality and the warning level identification results.

[0126] The distributed collaborative measurement system of the load sensor array provided in the embodiment of the present invention can execute the distributed collaborative measurement method of the load sensor array provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0127] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other and are not used to limit the scope of protection of the present invention.

[0128] The above specific embodiments do not constitute a limitation to the scope of protection of this application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent replacements and improvements made within the spirit and principles of this application should be included in the scope of protection of this application. In some cases, the actions or steps recorded in this application can be performed in an order different from that in the embodiments and can still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A distributed collaborative measurement method for a load sensor array, characterized in that: The method comprises: Perform finite element simulation of the target structure, identify key stress areas, boundary areas, and load types based on the finite element simulation results, and establish a structural feature library and a demand feature library; Deploying a sensor array according to the structural feature library and the demand feature library, wherein the sensor array is a three-layer nested array including a basic layer node, a relay layer node, and a global layer node; After configuring the communication protocol for the sensor array, the network topology initialization is completed; After configuring global clock synchronization, activate the sensor array to collect node data and build a time series data set; The time series data set collaborative processing under the three-layer nested array is performed according to the network topology, and the load measurement result is output according to the collaborative authentication result.

2. The distributed collaborative measurement method of a load sensor array according to claim 1, wherein: The collaborative processing of the time series data sets under the three-layer nested array is performed according to the network topology, including: Performing local node preprocessing of the time series data set at the base layer node to establish a first abnormal load identifier; Uploading the time series data set and the first abnormal load identifier to a relay layer node, performing load analysis within the region, and establishing a second abnormal load identifier; Establishing a graph structure based on the network topology and structural feature library, wherein the nodes of the graph structure are sensor nodes, and the edges of the graph structure represent information channels or structural couplings between any two nodes; Coupling the graph structure to global layer nodes, performing global collaborative analysis, and establishing global collaborative analysis results; Data collaborative authentication is performed based on the global collaborative analysis results, the first abnormal load identifier, the second abnormal load identifier, and the time series data set, and the load measurement results are output.

3. The distributed collaborative measurement method of a load sensor array according to claim 2, wherein: The performing local node preprocessing of the time series data set at the base layer node to establish a first abnormal load identifier includes: Use lightweight algorithms integrated in the base layer nodes to perform time series data set preprocessing and establish preprocessing results; Configuring abnormal indicators associated with base layer nodes, including overload indicators, rapid change indicators, and non-stationary fluctuation indicators; The corresponding preprocessing result abnormality identification is performed according to the abnormal indicator to establish the first abnormal load identifier.

4. The distributed collaborative measurement method of a load sensor array according to claim 2, wherein: The uploading of the time series data set and the first abnormal load identifier to the relay layer node, performing load analysis within the region, and establishing a second abnormal load identifier includes: Acquire the communication relationship between the relay layer node and the base layer node, call the time series data set according to the communication relationship, and establish a local perception subgraph; Performing a temporal consistency analysis of nodes in the region based on the local perception subgraph and the first abnormal load identifier, and establishing a temporal consistency anomaly, wherein the temporal consistency anomaly includes a synchronization anomaly, a propagation anomaly, and an isolation anomaly; A second abnormal load identifier is established according to the timing consistency abnormality.

5. The distributed collaborative measurement method of a load sensor array according to claim 2, wherein: The coupling of the graph structure to the global layer node, performing global collaborative analysis, and establishing global collaborative analysis results include: Performing structural analysis on the graph structure to construct a node-centered dual verification graph, wherein the dual verification graph includes a structural subgraph and a similarity subgraph, wherein the structural subgraph is constructed through structural connection relationships, and the similarity subgraph is constructed based on node feature similarity; A collaborative analysis is performed based on the double verification graph with the central node as the starting point, and a global collaborative analysis result is output according to the entire double verification graph.

6. The distributed collaborative measurement method of a load sensor array according to claim 5, wherein: The collaborative analysis based on the dual verification graph and taking the central node as the starting point includes: Taking the central node as a starting point, determining multi-order neighbors based on the structural subgraph and the similar subgraph respectively, and establishing a structural neighborhood set and a similar neighborhood set; Using the structural neighborhood set to perform force flow consistency analysis between nodes, a structural jump anomaly is established; Using the similar neighborhood set to perform group deviation analysis and establish similar island anomalies; Configure the load consistency score based on the structural jump anomaly and configure the similarity deviation score based on the similar island anomaly; A joint anomaly confidence level is established based on the load consistency score and similarity deviation score to output the global collaborative analysis results.

7. The distributed collaborative measurement method of a load sensor array according to claim 1, wherein: The activating the sensor array to collect node data and establish a time series data set includes: A front-end event sensing unit is configured in each sensor node of the sensor array; When executing node data collection, event perception is performed based on the front-end event perception unit to establish a node perception result; Configuring a primary neighborhood set of sensor nodes, sharing node perception results based on the primary neighborhood set, and updating an adaptive acquisition window; Continuous data acquisition is performed using the adaptive acquisition window to establish a time series data set.

8. The distributed collaborative measurement method of a load sensor array according to claim 1, wherein: Outputting the load measurement result according to the collaborative authentication result includes: Perform abnormal load identification based on the load measurement result and establish abnormal load identification attention; A historical load database is called, and verification is performed based on the historical load database and the abnormal load identifier, and the abnormal load identifier is updated.

9. The distributed collaborative measurement method of a load sensor array according to claim 8, wherein: The update abnormality payload identifier includes: Identify the updated abnormal load identification by warning level and configure visual warning abnormalities; Early warning management is performed based on the visual early warning anomaly and early warning level identification results.

10. A distributed collaborative measurement system of a load sensor array, characterized in that: A distributed collaborative measurement method for a load sensor array according to any one of claims 1 to 9, the system comprising: The feature library construction module is used to perform finite element simulation of the target structure, identify key stress areas, boundary areas, and load types based on the finite element simulation results, and establish a structural feature library and a required feature library; A sensor array deployment module, configured to deploy a sensor array according to the structural feature library and the demand feature library, wherein the sensor array is a three-layer nested array including a base layer node, a relay layer node, and a global layer node; A network topology initialization module, configured to complete network topology initialization after configuring the communication protocol for the sensor array; The time series data set building module is used to activate the sensor array to collect node data and build a time series data set after configuring global clock synchronization; The load measurement result acquisition module is used to perform collaborative processing of time series data sets under the three-layer nested array according to the network topology, and output the load measurement result according to the collaborative authentication result.

Images