A multi-node deep buried tunnel wireless sensor network monitoring system

Abstract

The present invention discloses a multi-node deep-buried tunnel wireless sensor network monitoring system, which relates to the technical field of tunnel safety monitoring. When the system is running, by integrating edge monitoring modules, request initiation modules, edge node acquisition modules, multiple verification modules and decision feedback modules, a highly intelligent monitoring mechanism is constructed, which can calculate the self-check deviation factor Eem in real time, and trigger events under abnormal conditions based on the request initiation module, adjust the acquisition frequency in time, and respond to emergencies quickly. Detailed data of adjacent nodes are obtained through the cloud, and a comprehensive status assessment is provided by analyzing the trend index Enu of adjacent nodes to reduce false alarms and omissions. The present invention makes maintenance response more efficient, improves the accuracy and real-time performance of data acquisition, and also enhances the early warning capability of potential abnormalities in tunnel structures, avoiding structural instability and operational safety problems caused by monitoring lags.

Description

Technical Field

[0001] The invention relates to the technical field of tunnel safety monitoring, in particular to a multi-node deep-buried tunnel wireless sensor network monitoring system. Background Art

[0002] In infrastructure safety monitoring, multi-node deep buried tunnel wireless sensor network monitoring is an important intelligent monitoring method. IoT technology uses various sensors and data acquisition devices to achieve real-time monitoring and data analysis of large infrastructure. In specific application scenarios, the long-term stability of deep buried tunnel structures is crucial, and tunnel safety monitoring has become an important area of ​​infrastructure safety management.

[0003] At present, most tunnel monitoring systems rely on fixed-time data collection, that is, sensors collect data at intervals within a predetermined time. Although this method can provide regular monitoring data, in the event of emergencies, including a sharp increase in tunnel displacement and vibration, the limitation of fixed collection frequency may cause the system to fail to capture important data at critical moments and fail to provide instant high-frequency data analysis. The limitations of this model make it impossible for the monitoring system to respond quickly in the early stages of a problem, and it is easy to miss early warning signals, thus failing to effectively ensure the safety of the tunnel. Summary of the invention

[0004] In view of the shortcomings of the prior art, the present invention provides a multi-node deep-buried tunnel wireless sensor network monitoring system, which can capture important data at the critical moment of the early stage of an accident and provide instant high-frequency data analysis.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: a multi-node deep-buried tunnel wireless sensor network monitoring system, including an edge monitoring module, a request initiation module, an edge node acquisition module, a multiple verification module and a decision feedback module;

[0006] The edge monitoring module collects data in real time and performs calculations of the self-check mechanism through multiple node integrated self-check mechanisms to obtain the edge node self-check deviation factor Eem;

[0007] The request initiation module matches the edge node self-check deviation factor Eem with the preset request initiation threshold QZ, obtains the edge node request initiation judgment result, and carries the location of the current node according to the edge node request initiation judgment result;

[0008] The edge node acquisition module obtains the operating status information of the neighboring nodes of the requesting node through the cloud, and analyzes and counts the surface microstress wave change information, tunnel wall sound wave spectrum change information and groundwater level change rate of the neighboring nodes to obtain the edge detection trend index Enu within a fixed period of the neighboring nodes;

[0009] The multi-verification module fits the edge node self-check deviation factor Eem and the edge detection trend index Enu to obtain the node evaluation coefficient Emu, and matches it with the preset node abnormality evaluation threshold JZ to obtain the secondary verification evaluation result of the current request node;

[0010] The decision feedback module generates a notification and prompt message according to the secondary verification evaluation result, and sends the notification and prompt message to the relevant maintenance personnel to handle the abnormal node.

[0011] Preferably, the edge monitoring module includes a data acquisition unit and a self-checking calculation unit;

[0012] The data acquisition unit collects detection data in the tunnel from multiple sensor nodes, each node is configured with multiple data acquisition sensors, collects real-time surface stress value Sreal(t), instantaneous sound wave frequency Areal(t) and real-time groundwater level height Wreal(t) detection data, and caches the real-time collected detection data through the buffer mechanism inside the sensor;

[0013] The self-check calculation unit performs a self-check mechanism on the cached detection data through the self-check mechanism integrated in the sensor node, performs statistical analysis on the cached detection data, obtains the surface microstress fluctuation S, the tunnel wall acoustic wave spectrum fluctuation A and the groundwater level change rate W, and then establishes the detection formula of the sensor node, and performs calculation to obtain the edge node self-check deviation factor Eem;

[0014] The surface microstress fluctuation S is obtained by the following statistical analysis formula:

[0015] ;

[0016] Where Sreal(t) represents the surface stress value at time point t, Tw represents the time window, represents the peak value of the surface stress value Sreal(t) within the time window Tw, Represents the valley value of the surface stress value Sreal(t) within the time window Tw;

[0017] The tunnel wall acoustic wave spectrum fluctuation A is obtained by the following statistical analysis formula:

[0018] ;

[0019] Where Areal(t) represents the frequency of the sound wave at time point t, represents the peak value of the sound wave frequency Areal(t) within the time window Tw, Represents the valley value of the sound wave frequency Areal(t) within the time window Tw;

[0020] The groundwater level change rate W is obtained by the following statistical analysis formula:

[0021] ;

[0022] Where Wreal(t) represents the groundwater level at time point t. represents the groundwater level Wreal(t) recorded at the beginning of the time window Tw, Represents the groundwater level Wreal(t) recorded at the end of the time window Tw.

[0023] Preferably, the node self-check deviation factor Eem is obtained by the following calculation formula:

[0024] ;

[0025] Where S0, A0 and W0 represent the preset reference values ​​of surface microstress fluctuation S, tunnel wall acoustic wave spectrum fluctuation A and groundwater level change rate W, respectively.

[0026] Preferably, the request initiation module includes a deviation determination unit and a request generation unit;

[0027] The deviation judgment unit matches the edge node self-check deviation factor Eem with the preset request initiation threshold QZ to obtain the edge node request initiation judgment result;

[0028] The edge node request initiation judgment result is obtained by the following matching method:

[0029] ;

[0030] When the Rq return signal is 1, the edge node request initiation judgment result is obtained as the initiation request result;

[0031] When the Rq return signal is 0, the edge node request initiation judgment result is obtained as a result of not initiating the request;

[0032] When the edge node request initiation judgment result is an initiation request result, the request generation unit carries the location of the current node, including GPS coordinates, node number and sensor node unique identification information, which is used for subsequent positioning of the sensor node and related personnel to go to handle it.

[0033] Preferably, the edge node acquisition module includes a cloud data acquisition unit and a trend analysis unit;

[0034] The cloud data acquisition unit obtains the GPS coordinates and node number of the requesting node through the cloud, searches for the GPS coordinates and node numbers of the stored adjacent nodes through the cloud, obtains a number of adjacent nodes to form an adjacent node list Njd, and synchronously sends a data collection request according to the adjacent node list Njd to obtain the monitoring data within the time window Tw, including the surface stress value Sreal(t), the instantaneous sound wave frequency Areal(t) and the real-time groundwater level height Wreal(t) detection data, to form an adjacent node list detection data group Jds;

[0035] By executing the self-check mechanism on the adjacent node list detection data group Jds, the adjacent node detection fluctuation data group Bz={S1,...,Sn,A1,...,An,W1,...,Wn} is obtained, which specifically includes the surface microstress fluctuation S, tunnel wall acoustic wave spectrum fluctuation A and groundwater level change rate W of adjacent nodes 1 to n.

[0036] Preferably, the trend analysis unit obtains the adjacent node detection fluctuation data group Bz within the adjacent node time window Tw and performs statistical analysis to obtain the surface microstress wave change △Si, the tunnel wall sound wave spectrum change △Ai and the groundwater level change △Wi of each adjacent node;

[0037] The average fluctuation of node i is obtained by statistically analyzing the variation of ground microstress wave △Si, the variation of tunnel wall acoustic wave spectrum △Ai and the variation of groundwater level △Wi;

[0038] Then, the trend matrix Mtrend of the adjacent node list detection data group Jds is established through the average fluctuation amount △Xi, and the fluctuation amount of each adjacent node in the time window Tw is counted to obtain the edge detection trend index Enu.

[0039] Preferably, the surface microstress wave variation ΔSi is obtained by the following calculation formula:

[0040] ;

[0041] Where △Si represents the change of surface microstress wave of the ith adjacent node, n represents the length of the adjacent node detection fluctuation data group Bz, S(t)i represents the surface microstress fluctuation of the ith adjacent node at time point t within the time window Tw, and Si,avg represents the average value of the surface microstress fluctuation of the ith adjacent node within the time window Tw;

[0042] The tunnel wall acoustic wave spectrum change △Ai is obtained by the following calculation formula:

[0043] ;

[0044] Wherein, △Ai represents the change in the tunnel wall acoustic wave spectrum of the ith adjacent node, A(t)i represents the fluctuation of the tunnel wall acoustic wave spectrum of the ith adjacent node at time point t within the time window Tw, and Ai,avg represents the average value of the fluctuation of the tunnel wall acoustic wave spectrum of the ith adjacent node within the time window Tw;

[0045] The groundwater level change ΔWi is obtained by the following calculation formula:

[0046] ;

[0047] Where △Wi represents the groundwater level change of the ith adjacent node, W(t)i represents the groundwater level change rate of the ith adjacent node at time point t within the time window Tw, and Wi,avg represents the average value of the groundwater level change rate of the ith adjacent node within the time window Tw;

[0048] The average fluctuation △Xi is Obtain calculation formula;

[0049] The trend matrix Mtrend={△X1, △X2, △X3, ..., △Xn};

[0050] The edge detection trend index Enu is obtained by Obtain the calculation formula.

[0051] Preferably, the multiple verification module includes a fitting calculation unit and a fitting matching unit;

[0052] The fitting calculation unit fits the edge node self-detection deviation factor Eem and the edge detection trend index Enu, obtains the node evaluation coefficient Emu, and evaluates the trend status of the current request node and the adjacent nodes;

[0053] The node evaluation coefficient Emu is obtained by the following calculation formula:

[0054] ;

[0055] In the formula, Indicates the preset weight value.

[0056] Preferably, the fitting and matching unit matches the node evaluation coefficient Emu with a preset node abnormality evaluation threshold JZ to obtain a secondary verification evaluation result of the current request node;

[0057] The secondary verification evaluation result is obtained by the following matching method:

[0058] ;

[0059] When the Rcheck return signal is 1, the secondary verification evaluation result is abnormal;

[0060] When the Rcheck return signal is 0, the secondary verification evaluation result is obtained as no abnormality.

[0061] Preferably, the decision feedback module generates a notification and prompt message according to the secondary verification evaluation result, and sends the notification and prompt message to the relevant maintenance personnel to handle the abnormal node;

[0062] When the secondary verification evaluation result is an abnormality, a pending notification and prompt content of the node is generated and sent to the pending list of the inspection and maintenance personnel. The notification and prompt content include the GPS coordinates of the abnormal node, the node number, the unique identification information of the sensor node and the abnormality type; the abnormality types include surface microstress wave anomalies, tunnel wall acoustic wave spectrum anomalies and groundwater level anomalies.

[0063] The present invention provides a multi-node deep-buried tunnel wireless sensor network monitoring system, which has the following beneficial effects:

[0064] (1) When the system is running, a highly intelligent monitoring mechanism is constructed by integrating the edge monitoring module, the request initiation module, the edge node collection module, the multiple verification module and the decision feedback module, which solves the response lag problem caused by the traditional monitoring system that relies on collecting data at fixed time intervals. First, the edge monitoring module can calculate the self-check deviation factor Eem in real time, and trigger events based on the request initiation module under abnormal conditions, so that the system can adjust the collection frequency in time and respond quickly to emergencies. Secondly, the edge node collection module further obtains detailed data of adjacent nodes through the cloud, and provides a comprehensive status assessment by analyzing the trend index Enu of adjacent nodes. The multiple verification module ensures accurate identification of abnormal nodes by fitting Eem and Enu, reducing false alarms and omissions. The decision feedback module provides maintenance personnel with detailed abnormality notifications and handling suggestions, making maintenance responses more efficient, and also enhancing the early warning capability of potential abnormalities in the tunnel structure, avoiding structural instability and operational safety problems caused by monitoring lag.

[0065] (2) The integrity and continuity of key stress, acoustic wave and water level data are ensured through a buffering mechanism. At the same time, the self-checking calculation unit performs in-depth statistical analysis on the cached data, accurately calculates the stress fluctuations, acoustic wave spectrum changes and water level change rates of the nodes, and combines these fluctuations to establish the self-checking deviation factor Eem, which further optimizes the recognition efficiency of node anomalies. In particular, through accurate deviation judgment and automatic request generation, the system can carry location information to initiate requests in a timely manner, ensuring that maintenance personnel can quickly locate abnormal sensor nodes. The system improves the depth of data processing and strengthens the system's response speed to node anomalies, thereby effectively shortening the time between anomaly occurrence and processing, and improving the overall safety and operation and maintenance efficiency of the tunnel.

[0066] (3) By locating and collecting stress waves, acoustic spectrum and groundwater level multi-dimensional data of adjacent nodes, the operating status of the tunnel environment is fully understood. At the same time, the trend analysis unit generates the trend matrix Mtrend and calculates the edge detection trend index Enu based on the monitoring fluctuation data of adjacent nodes, which can accurately reflect the comprehensive state changes of adjacent nodes. The advantage of this system is that through the integrated analysis of adjacent nodes, it not only strengthens the judgment of single node anomalies, but also provides more global trend monitoring, avoids misjudgment caused by local information, and effectively improves the comprehensiveness of tunnel structure monitoring and the accuracy of early warning.

[0067] (4) By performing weighted fitting on the edge node self-check deviation factor Eem and the edge detection trend index Enu, the node's comprehensive evaluation coefficient Emu is obtained, thereby combining the self-check information of a single node with the trend analysis results of adjacent nodes, improving the comprehensiveness and accuracy of the overall evaluation. The fitting and matching unit generates a secondary verification evaluation result by matching the node evaluation coefficient Emu with the node abnormality evaluation threshold JZ, ensuring that only nodes that are actually abnormal will trigger maintenance requests. The special advantage is that it effectively filters out false alarms and unnecessary maintenance requests through the secondary verification mechanism, reducing resource waste, while ensuring that maintenance actions are only targeted at truly abnormal nodes, effectively improving the response efficiency and accuracy of the monitoring system. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 The present invention is a block diagram of a multi-node deep-buried tunnel wireless sensor network monitoring system. DETAILED DESCRIPTION

[0069] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0070] Example 1

[0071] The present invention provides a multi-node deep buried tunnel wireless sensor network monitoring system, please refer to Figure 1, including edge monitoring module, request initiation module, edge node collection module, multiple verification module and decision feedback module.

[0072] The edge monitoring module integrates self-checking mechanisms through multiple nodes, collects data in real time, calculates the self-checking mechanism, and obtains the edge node self-checking deviation factor Eem.

[0073] The request initiation module matches the edge node self-check deviation factor Eem with the preset request initiation threshold QZ, obtains the edge node request initiation judgment result, and carries the location of the current node according to the edge node request initiation judgment result;

[0074] The edge node acquisition module obtains the operating status information of the neighboring nodes of the requesting node through the cloud, and analyzes and counts the surface microstress wave change information, tunnel wall acoustic wave spectrum change information and groundwater level change rate of the neighboring nodes, and obtains the edge detection trend index Enu within a fixed period of the neighboring nodes;

[0075] The multi-verification module fits the edge node self-check deviation factor Eem and the edge detection trend index Enu to obtain the node evaluation coefficient Emu, and matches it with the preset node anomaly evaluation threshold JZ to obtain the secondary verification evaluation result of the current request node;

[0076] The decision feedback module generates notification and prompt messages based on the secondary verification evaluation results, and sends the notification and prompt messages to the relevant maintenance personnel to handle the abnormal nodes.

[0077] In this embodiment, by integrating the edge monitoring module, the request initiation module, the edge node acquisition module, the multiple verification module and the decision feedback module, a highly intelligent monitoring mechanism is constructed to solve the problem of response lag caused by relying on fixed time intervals to collect data in the traditional monitoring system. First, the edge monitoring module can calculate the self-check deviation factor Eem in real time, and trigger events based on the request initiation module under abnormal conditions, so that the system can adjust the acquisition frequency in time and respond quickly to emergencies. Secondly, the edge node acquisition module further obtains detailed data of adjacent nodes through the cloud, and provides a comprehensive status assessment by analyzing the trend index Enu of adjacent nodes. The multiple verification module ensures accurate identification of abnormal nodes and reduces false alarms and omissions by fitting Eem and Enu. The decision feedback module provides maintenance personnel with detailed abnormal notifications and processing suggestions, making maintenance responses more efficient. Compared with traditional fixed-time acquisition, this system not only improves the accuracy and real-time performance of data acquisition, but also enhances the early warning capability of potential abnormalities in tunnel structures, avoiding structural instability and operational safety problems caused by monitoring lag.

[0078] Example 2

[0079] This example is an explanation based on Example 1. Figure 1 ,Specifically: the edge monitoring module includes a data acquisition unit and a ,self-checking calculation unit;

[0080] The data acquisition unit collects detection data in the tunnel from multiple sensor nodes. Each node is equipped with multiple data acquisition sensors to collect real-time surface stress value Sreal(t), instantaneous sound wave frequency Areal(t) and real-time groundwater level height Wreal(t) detection data, and caches the real-time collected detection data through the buffer mechanism inside the sensor;

[0081] The self-check calculation unit performs a self-check mechanism on the cached detection data through the self-check mechanism integrated in the sensor node, performs statistical analysis on the cached detection data, obtains the surface microstress fluctuation S, the tunnel wall acoustic wave spectrum fluctuation A and the groundwater level change rate W, and then establishes the detection formula of the sensor node and performs calculation to obtain the edge node self-check deviation factor Eem;

[0082] The surface microstress fluctuation S is obtained by the following statistical analysis formula:

[0083] ;

[0084] Where Sreal(t) represents the surface stress value at time point t, Tw represents the time window, represents the peak value of the surface stress value Sreal(t) within the time window Tw, Represents the valley value of the surface stress value Sreal(t) within the time window Tw;

[0085] The tunnel wall acoustic wave spectrum fluctuation A is obtained by the following statistical analysis formula:

[0086] ;

[0087] Where Areal(t) represents the frequency of the sound wave at time point t, represents the peak value of the sound wave frequency Areal(t) within the time window Tw, Represents the valley value of the sound wave frequency Areal(t) within the time window Tw;

[0088] The groundwater level change rate W is obtained by the following statistical analysis formula:

[0089] ;

[0090] Where Wreal(t) represents the groundwater level at time point t. represents the groundwater level Wreal(t) recorded at the beginning of the time window Tw, Represents the groundwater level Wreal(t) recorded at the end of the time window Tw.

[0091] The node self-check deviation factor Eem is obtained by the following calculation formula:

[0092] ;

[0093] Where S0, A0 and W0 represent the preset reference values ​​of surface microstress fluctuation S, tunnel wall acoustic wave spectrum fluctuation A and groundwater level change rate W, respectively.

[0094] The request initiation module includes a deviation determination unit and a request generation unit;

[0095] The deviation judgment unit matches the edge node self-check deviation factor Eem with the preset request initiation threshold QZ to obtain the edge node request initiation judgment result;

[0096] The edge node request initiation judgment result is obtained through the following matching method:

[0097] ;

[0098] When the Rq return signal is 1, the edge node request initiation judgment result is obtained as the initiation request result;

[0099] When the Rq return signal is 0, the edge node request initiation judgment result is obtained as a result of not initiating the request;

[0100] When the edge node request initiation judgment result is the initiation request result, the request generation unit carries the location of the current node, including GPS coordinates, node number and sensor node unique identification information, which is used for subsequent positioning of the sensor node and related personnel to go to handle it.

[0101] In this embodiment, the integrity and continuity of key stress, acoustic wave and water level data are ensured through a buffering mechanism. At the same time, the self-checking calculation unit performs in-depth statistical analysis on the cached data, accurately calculates the stress fluctuations, acoustic wave spectrum changes and water level change rates of the nodes, and establishes the self-checking deviation factor Eem based on these fluctuations, thereby further optimizing the recognition efficiency of node anomalies. In particular, through accurate deviation judgment and automatic request generation, the system can initiate requests with location information in a timely manner, ensuring that maintenance personnel can quickly locate abnormal sensor nodes. It improves the depth of data processing and strengthens the system's response speed to node anomalies, thereby effectively shortening the time for anomaly occurrence and processing, and improving the overall safety and operation and maintenance efficiency of the tunnel.

[0102] Example 3

[0103] This example is an explanation based on Example 1. Figure 1,Specifically: the edge node acquisition module includes a cloud data acquisition unit and a ,trend analysis unit;

[0104] The cloud data acquisition unit obtains the GPS coordinates and node number of the requesting node through the cloud, searches for the GPS coordinates and node numbers of the stored adjacent nodes through the cloud, obtains several adjacent nodes to form an adjacent node list Njd, and simultaneously sends a data collection request according to the adjacent node list Njd to obtain the monitoring data within the time window Tw, including the surface stress value Sreal(t), the instantaneous sound wave frequency Areal(t) and the real-time groundwater level height Wreal(t) detection data, to form an adjacent node list detection data group Jds;

[0105] By executing the self-check mechanism on the adjacent node list detection data group Jds, the adjacent node detection fluctuation data group Bz={S1,...,Sn,A1,...,An,W1,...,Wn} is obtained, which specifically includes the surface microstress fluctuation S, tunnel wall acoustic wave spectrum fluctuation A and groundwater level change rate W of adjacent nodes 1 to n.

[0106] The trend analysis unit obtains the adjacent node detection fluctuation data group Bz within the adjacent node time window Tw and performs statistical analysis to obtain the surface microstress wave change △Si, the tunnel wall sound wave spectrum change △Ai and the groundwater level change △Wi of each adjacent node;

[0107] The average fluctuation of node i is obtained by statistically analyzing the variation of ground microstress wave △Si, the variation of tunnel wall acoustic wave spectrum △Ai and the variation of groundwater level △Wi;

[0108] Then, the trend matrix Mtrend of the adjacent node list detection data group Jds is established through the average fluctuation amount △Xi, and the fluctuation amount of each adjacent node in the time window Tw is counted to obtain the edge detection trend index Enu.

[0109] The surface microstress wave variation △Si is obtained by the following calculation formula:

[0110] ;

[0111] Where △Si represents the change of surface microstress wave of the ith adjacent node, n represents the length of the adjacent node detection fluctuation data group Bz, S(t)i represents the surface microstress fluctuation of the ith adjacent node at time point t within the time window Tw, and Si,avg represents the average value of the surface microstress fluctuation of the ith adjacent node within the time window Tw;

[0112] The change in the tunnel wall acoustic spectrum △Ai is obtained by the following calculation formula:

[0113] ;

[0114] Wherein, △Ai represents the change in the tunnel wall acoustic wave spectrum of the ith adjacent node, A(t)i represents the fluctuation of the tunnel wall acoustic wave spectrum of the ith adjacent node at time point t within the time window Tw, and Ai,avg represents the average value of the fluctuation of the tunnel wall acoustic wave spectrum of the ith adjacent node within the time window Tw;

[0115] The groundwater level change △Wi is obtained by the following calculation formula:

[0116] ;

[0117] Where △Wi represents the groundwater level change of the ith adjacent node, W(t)i represents the groundwater level change rate of the ith adjacent node at time point t within the time window Tw, and Wi,avg represents the average value of the groundwater level change rate of the ith adjacent node within the time window Tw;

[0118] The average fluctuation △Xi is obtained by Obtain calculation formula;

[0119] Trend matrix Mtrend={△X1,△X2,△X3,...,△Xn};

[0120] Edge detection trend index Enu is obtained by Obtain the calculation formula.

[0121] In this embodiment, by constructing an efficient adjacent node data collection and analysis mechanism, the breadth and accuracy of the data are significantly improved. Through the cloud data acquisition unit, the system can quickly locate and collect stress waves, acoustic spectrum and groundwater level multi-dimensional data of adjacent nodes to ensure a comprehensive grasp of the operating status of the tunnel environment. At the same time, the trend analysis unit generates a trend matrix Mtrend and calculates the edge detection trend index Enu based on the monitoring fluctuation data of adjacent nodes, which can accurately reflect the comprehensive state changes of adjacent nodes. The advantage of this system is that through the integrated analysis of adjacent nodes, it not only strengthens the judgment of single node anomalies, but also provides a more global trend monitoring, avoids misjudgment caused by local information, and effectively improves the comprehensiveness of tunnel structure monitoring and the accuracy of early warning.

[0122] Example 4

[0123] This embodiment is explained in Example 1, please refer to Figure 1 ,Specifically: the multiple verification module includes a fitting calculation unit and a fitting matching unit;

[0124] The fitting calculation unit fits the edge node self-detection deviation factor Eem and the edge detection trend index Enu, obtains the node evaluation coefficient Emu, and evaluates the trend status of the current request node and the adjacent nodes;

[0125] The node evaluation coefficient Emu is obtained by the following calculation formula:

[0126] ;

[0127] In the formula, Indicates the preset weight value.

[0128] The fitting matching unit matches the node evaluation coefficient Emu with the preset node abnormality evaluation threshold JZ to obtain the secondary verification evaluation result of the current request node;

[0129] The secondary verification evaluation results are obtained through the following matching methods:

[0130] ;

[0131] When the Rcheck return signal is 1, the secondary verification evaluation result is abnormal;

[0132] When the Rcheck return signal is 0, the secondary verification evaluation result is obtained as no abnormality.

[0133] The decision feedback module generates notification and prompt messages based on the secondary verification evaluation results, and sends the notification and prompt messages to the relevant maintenance personnel to handle the abnormal nodes;

[0134] When the secondary verification evaluation result is an abnormality, a pending notification and prompt content of the node is generated and sent to the pending list of the inspection and maintenance personnel. The notification and prompt content include the GPS coordinates of the abnormal node, the node number, the unique identification information of the sensor node and the abnormality type; the abnormality types include surface microstress wave anomalies, tunnel wall acoustic wave spectrum anomalies and groundwater level anomalies.

[0135] In this embodiment, the accuracy of abnormal node judgment is further improved through the collaborative work of the fitting calculation unit and the fitting matching unit. The fitting calculation unit obtains the comprehensive evaluation coefficient Emu of the node by weighted fitting the edge node self-check deviation factor Eem and the edge detection trend index Enu, thereby combining the self-check information of a single node with the trend analysis results of adjacent nodes, thereby improving the comprehensiveness and accuracy of the overall evaluation. The fitting matching unit generates a secondary verification evaluation result by matching the node evaluation coefficient Emu with the node abnormality evaluation threshold JZ, ensuring that only nodes that are actually abnormal will trigger maintenance requests. The special advantage is that it effectively filters out false alarms and unnecessary maintenance requests through the secondary verification mechanism, reducing resource waste, while ensuring that maintenance actions are only targeted at truly abnormal nodes, effectively improving the response efficiency and accuracy of the monitoring system.

[0136] Those skilled in the art can understand that the above are only preferred examples of the invention and are not intended to limit the invention. Although the invention is described in detail with reference to the above examples, those skilled in the art can still modify the technical solutions recorded in the above examples or replace some of the technical features with equivalents. Any modification, equivalent replacement, etc. made within the spirit and principle of the invention shall be included in the protection scope of the invention.

Claims

1. A multi-node deep buried tunnel wireless sensor network monitoring system, characterized by: It includes edge monitoring module, request initiation module, edge node acquisition module, multiple verification module and decision feedback module; The edge monitoring module collects data in real time and performs calculations of the self-check mechanism through multiple node integrated self-check mechanisms to obtain the edge node self-check deviation factor Eem; The edge monitoring module includes a data acquisition unit and a self-checking calculation unit; The data acquisition unit collects detection data in the tunnel from multiple sensor nodes, each node is configured with multiple data acquisition sensors, collects real-time surface stress value Sreal(t), instantaneous sound wave frequency Areal(t) and real-time groundwater level height Wreal(t) detection data, and caches the real-time collected detection data through the buffer mechanism inside the sensor; The self-check calculation unit performs a self-check mechanism on the cached detection data through the self-check mechanism integrated in the sensor node, performs statistical analysis on the cached detection data, obtains the surface microstress fluctuation S, the tunnel wall acoustic wave spectrum fluctuation A and the groundwater level change rate W, and then establishes the detection formula of the sensor node, and performs calculation to obtain the edge node self-check deviation factor Eem; The surface microstress fluctuation S is obtained by the following statistical analysis formula: ； Where Sreal(t) represents the surface stress value at time point t, Tw represents the time window, represents the peak value of the surface stress value Sreal(t) within the time window Tw, Represents the valley value of the surface stress value Sreal(t) within the time window Tw; The tunnel wall acoustic wave spectrum fluctuation A is obtained by the following statistical analysis formula: ； Where Areal(t) represents the frequency of the sound wave at time point t, It represents the peak value of the sound wave frequency Areal(t) in the time window Tw, Represents the valley value of the sound wave frequency Areal(t) within the time window Tw; The groundwater level change rate W is obtained by the following statistical analysis formula: ； Where Wreal(t) represents the groundwater level at time point t. represents the groundwater level Wreal(t) recorded at the beginning of the time window Tw, represents the groundwater level Wreal(t) recorded at the end of the time window Tw; The node self-check deviation factor Eem is obtained by the following calculation formula: ； Where S0, A0 and W0 represent the preset reference values ​​of surface microstress fluctuation S, tunnel wall acoustic wave spectrum fluctuation A and groundwater level change rate W respectively; The request initiation module matches the edge node self-check deviation factor Eem with the preset request initiation threshold QZ, obtains the edge node request initiation judgment result, and carries the location of the current node according to the edge node request initiation judgment result; The edge node acquisition module obtains the operating status information of the neighboring nodes of the requesting node through the cloud, and analyzes and counts the surface microstress wave change information, tunnel wall sound wave spectrum change information and groundwater level change rate of the neighboring nodes, obtains the surface microstress wave change △Si, tunnel wall sound wave spectrum change △Ai and groundwater level change △Wi of each neighboring node, and then obtains the edge detection trend index Enu within a fixed period of the neighboring node; The surface microstress wave variation ΔSi is obtained by the following calculation formula: ； Where △Si represents the change of surface microstress wave of the ith adjacent node, n represents the length of the adjacent node detection fluctuation data group Bz, S(t)i represents the surface microstress fluctuation of the ith adjacent node at time point t within the time window Tw, and Si,avg represents the average value of the surface microstress fluctuation of the ith adjacent node within the time window Tw; The tunnel wall acoustic wave spectrum change △Ai is obtained by the following calculation formula: ； Wherein, △Ai represents the change in the tunnel wall acoustic wave spectrum of the ith adjacent node, A(t)i represents the fluctuation of the tunnel wall acoustic wave spectrum of the ith adjacent node at time point t within the time window Tw, and Ai,avg represents the average value of the fluctuation of the tunnel wall acoustic wave spectrum of the ith adjacent node within the time window Tw; The groundwater level change ΔWi is obtained by the following calculation formula: ； Where △Wi represents the groundwater level change of the ith adjacent node, W(t)i represents the groundwater level change rate of the ith adjacent node at time point t within the time window Tw, and Wi,avg represents the average value of the groundwater level change rate of the ith adjacent node within the time window Tw; The average fluctuation △Xi is obtained by Obtain calculation formula; The trend matrix Mtrend of the adjacent node list detection data group Jds is established by the average fluctuation amount △Xi, wherein the trend matrix Mtrend={△X1, △X2, △X3, ..., △Xn}; The edge detection trend index Enu is obtained by Obtain calculation formula; The multi-verification module fits the edge node self-check deviation factor Eem and the edge detection trend index Enu to obtain the node evaluation coefficient Emu, and matches it with the preset node abnormality evaluation threshold JZ to obtain the secondary verification evaluation result of the current request node; The multiple verification module includes a fitting calculation unit and a fitting matching unit; The fitting calculation unit fits the edge node self-detection deviation factor Eem and the edge detection trend index Enu, obtains the node evaluation coefficient Emu, and evaluates the trend status of the current request node and the adjacent nodes; The node evaluation coefficient Emu is obtained by the following calculation formula: ； In the formula, Indicates the preset weight value; The decision feedback module generates a notification and prompt message according to the secondary verification evaluation result, and sends the notification and prompt message to the relevant maintenance personnel to handle the abnormal node.

2. A multi-node deep tunnel wireless sensor network monitoring system according to claim 1, characterized in that: The request initiation module includes a deviation determination unit and a request generation unit; The deviation judgment unit matches the edge node self-check deviation factor Eem with the preset request initiation threshold QZ to obtain the edge node request initiation judgment result; The edge node request initiation judgment result is obtained by the following matching method: ； When the Rq return signal is 1, the edge node request initiation judgment result is obtained as the initiation request result; When the Rq return signal is 0, the edge node request initiation judgment result is obtained as a result of not initiating the request; When the edge node request initiation judgment result is an initiation request result, the request generation unit carries the location of the current node, including GPS coordinates, node number and sensor node unique identification information, which is used for subsequent positioning of the sensor node and related personnel to go to handle it.

3. A multi-node deep tunnel wireless sensor network monitoring system according to claim 1, characterized in that: The edge node acquisition module includes a cloud data acquisition unit and a trend analysis unit; The cloud data acquisition unit obtains the GPS coordinates and node number of the requesting node through the cloud, searches for the GPS coordinates and node numbers of the stored adjacent nodes through the cloud, obtains a number of adjacent nodes to form an adjacent node list Njd, and synchronously sends a data collection request according to the adjacent node list Njd to obtain the monitoring data within the time window Tw, including the surface stress value Sreal(t), the instantaneous sound wave frequency Areal(t) and the real-time groundwater level height Wreal(t) detection data, to form an adjacent node list detection data group Jds; By executing the self-check mechanism on the adjacent node list detection data group Jds, the adjacent node detection fluctuation data group Bz={S1,...,Sn,A1,...,An,W1,...,Wn} is obtained, which specifically includes the surface microstress fluctuation S, tunnel wall acoustic wave spectrum fluctuation A and groundwater level change rate W of adjacent nodes 1 to n.

4. A multi-node deep buried tunnel wireless sensor network monitoring system according to claim 3, characterized in that: The trend analysis unit obtains the adjacent node detection fluctuation data group Bz within the adjacent node time window Tw and performs statistical analysis to obtain the surface microstress wave change △Si, the tunnel wall sound wave spectrum change △Ai and the groundwater level change △Wi of each adjacent node; The average fluctuation of node i is obtained by statistically analyzing the variation of ground microstress wave △Si, the variation of tunnel wall acoustic wave spectrum △Ai and the variation of groundwater level △Wi; Then, the trend matrix Mtrend of the adjacent node list detection data group Jds is established through the average fluctuation amount △Xi, and the fluctuation amount of each adjacent node in the time window Tw is counted to obtain the edge detection trend index Enu.

5. A multi-node deep buried tunnel wireless sensor network monitoring system according to claim 4, characterized in that: The fitting matching unit matches the node evaluation coefficient Emu with the preset node abnormality evaluation threshold JZ to obtain the secondary verification evaluation result of the current request node; The secondary verification evaluation result is obtained by the following matching method: ； When the Rcheck return signal is 1, the secondary verification evaluation result is abnormal; When the Rcheck return signal is 0, the secondary verification evaluation result is obtained as no abnormality.

6. A multi-node deep tunnel wireless sensor network monitoring system according to claim 2, characterized in that: The decision feedback module generates a notification and prompt message according to the secondary verification evaluation result, and sends the notification and prompt message to the relevant maintenance personnel to handle the abnormal node; When the secondary verification evaluation result is an abnormality, a pending notification and prompt content of the node is generated and sent to the pending list of the inspection and maintenance personnel. The notification and prompt content include the GPS coordinates of the abnormal node, the node number, the unique identification information of the sensor node and the abnormality type; the abnormality types include surface microstress wave anomalies, tunnel wall acoustic wave spectrum anomalies and groundwater level anomalies.

Images