Gas pipe network construction site safety monitoring method and related equipment

Abstract

The invention belongs to the technical field of safety monitoring, and discloses a gas pipe network construction site safety monitoring method and related equipment, and the method comprises the steps: collecting the original inclination angle data of different positions of a construction region in real time through distributed monitoring nodes, and carrying out the filtering processing of the original inclination angle data through a moving average filtering algorithm, the differential settlement value is calculated according to the filtered inclination angle data of the adjacent monitoring nodes, the differential settlement value and the dynamic settlement difference threshold value are compared to determine the settlement risk and give an early warning, real-time monitoring of gas pipe network construction site ground settlement is achieved, and the settlement monitoring accuracy is improved.

Description

Technical Field

[0001] This application relates to the technical field of safety monitoring, and more particularly, to a method and related equipment for safety monitoring at the construction site of a gas pipeline network. Background Art

[0002] In the field of urban infrastructure construction and maintenance, especially in the renovation projects of old gas pipeline networks, trenchless construction technology has increasingly become a mainstream construction option because it can significantly reduce the ground excavation area and thus minimize the interference with the daily life of urban residents. For example, when trenchless technologies such as pipe jacking and directional drilling are used for underground tunneling operations, although they have many advantages, their construction processes will still inevitably cause disturbances to the surrounding strata, and this kind of disturbance is the potential root cause of ground settlement. Ground settlement not only may directly threaten the structural safety of existing buildings and underground pipelines, but also poses a risk that cannot be ignored for newly laid gas pipeline networks. Therefore, during the entire trenchless construction period, continuous and real-time monitoring and early warning of ground settlement are of crucial significance for ensuring construction safety and preventing engineering accidents.

[0003] Currently, for the monitoring of ground settlement, in practice, it mainly relies on traditional manual observation methods or uses high-precision settlement sensors for data collection. However, traditional manual observation methods have inherent limitations. Their work efficiency is relatively low, the monitoring period is long, and it is difficult to reflect the whole process of settlement changes in real time and dynamically. Therefore, there are obvious deficiencies in rapid early warning response and it is difficult to meet the requirements of modern engineering safety management. Although high-precision settlement sensors can achieve continuous monitoring to a certain extent, in the context of the complex urban underground environment where infrastructure such as power cables and communication cables are densely distributed, the resulting complex electromagnetic environment is extremely likely to interfere with high-precision sensors, thereby causing data drift and false alarm phenomena, which will undoubtedly seriously affect the accuracy and reliability of the early warning system. Furthermore, the electromagnetic interference in the urban environment is not constant but has a certain degree of randomness, which makes traditional simple filtering methods ineffective in eliminating the influence of noise and difficult to completely eliminate the interference of noise. In addition, the sensors themselves inevitably have measurement errors, and the combined effect of these factors further exacerbates the uncertainty of the monitoring data.

[0004] In view of the above problems, the existing technology urgently needs to be improved. Summary of the Invention

[0005] The purpose of this application is to provide a method and related equipment for safety monitoring at the construction site of a gas pipeline network, which can improve the accuracy of settlement monitoring.

[0006] In a first aspect, the present application provides a method for safety monitoring at the construction site of a gas pipeline network, which is used for settlement monitoring at the construction site of trenchless renovation of an old gas pipeline network. The steps of this method include: A1. Obtain the original tilt angle data collected by multiple monitoring nodes; the multiple monitoring nodes are spaced along the ground on both sides of the trenchless construction line of the gas pipeline network, and each monitoring node is equipped with at least two tilt sensors with perpendicular directions to sense the tilt angles of the ground in different directions so as to obtain the original tilt angle data; A2. Use a moving average filtering algorithm to filter the original tilt angle data to obtain the filtered tilt angle data of each monitoring node; A3. Calculate the tilt angle difference between adjacent monitoring nodes based on the filtered tilt angle data of adjacent monitoring nodes to obtain a differential settlement value; A4. Compare the differential settlement value with a dynamic settlement difference threshold to determine the settlement risk and send a warning message in a timely manner; the dynamic settlement difference threshold is dynamically adjusted based on historical monitoring data, geological conditions, and the degree of pipeline network aging.

[0007] This method uses distributed monitoring nodes to collect the original tilt angle data at different positions in the construction area in real time, uses a moving average filtering algorithm to filter the original tilt angle data, calculates the differential settlement value based on the filtered tilt angle data of adjacent monitoring nodes, compares the differential settlement value with the dynamic settlement difference threshold to determine the settlement risk and give a warning, realizes real-time monitoring of ground settlement at the construction site of the gas pipeline network, and improves the accuracy of settlement monitoring.

[0008] Preferably, step A1 includes: A101. Obtain the geographical location information of each monitoring node; A102. Based on the geographical location information of each monitoring node, use differential GPS positioning technology to calibrate each monitoring node to obtain the calibrated geographical location information of each monitoring node; A103. According to the calibrated geographical location information of each monitoring node, determine the relative position relationship between each monitoring node, and based on the relative position relationship, combine the Kalman filtering algorithm and the historical tilt angle data of each tilt sensor to correct the attitude of the tilt sensor of each monitoring node to obtain the corrected tilt angle data of each tilt sensor as the original tilt angle data.

[0009] Through attitude correction, the deviation of tilt angle data caused by factors such as sensor installation error and uneven ground can be eliminated or reduced, and more accurate original tilt angle data can be obtained, thereby providing a more reliable data basis for subsequent settlement calculation and risk warning, and improving the effectiveness and accuracy of the entire safety monitoring method.

[0010] Preferably, step A103 includes: B1. Calculate the distance between any two monitoring nodes according to the calibrated geographical location information of each monitoring node, and construct the topological relationship between the monitoring nodes based on the distance to obtain the monitoring node topology graph; B2. Sequentially use each monitoring node as the target monitoring node and execute: B201. Based on the monitoring node topology graph, screen several adjacent monitoring nodes directly adjacent to the target monitoring node, and obtain the historical tilt angle data of the tilt sensors of each adjacent monitoring node within a preset time window; B202. For the tilt sensor of the target monitoring node, use the Kalman filter algorithm to fuse the historical tilt angle data of the target monitoring node and each adjacent monitoring node, and perform attitude correction on the tilt sensor of the target monitoring node to obtain the corrected tilt angle data of the tilt sensor of the target monitoring node as the original tilt angle data of the target monitoring node; during the attitude correction process, dynamically adjust the noise covariance matrix of the Kalman filter according to the vibration frequency range of the construction equipment to reduce the influence of the construction equipment vibration on the attitude correction.

[0011] Through the above steps, in the attitude correction link, this solution effectively utilizes the spatial layout information of the monitoring nodes and reduces the interference of vibration noise at the construction site, thereby improving the quality of the original tilt angle data and providing a more reliable data basis for subsequent settlement analysis and early warning.

[0012] Preferably, step B202 includes: Obtain the tilt angle data collected by the tilt sensors of the target monitoring node and each adjacent monitoring node within a preset time window, calculate the variance value of the tilt angle data of each monitoring node between the target monitoring node and each adjacent monitoring node, and obtain the tilt angle data variance value of the tilt sensors of the target monitoring node and each adjacent monitoring node; According to the vibration frequency range of the construction equipment and each tilt angle data variance value, determine the weight ratio of the influence of the construction equipment vibration on the tilt sensors of the target monitoring node and each adjacent monitoring node; Based on the weight ratio, dynamically adjust the process noise covariance matrix and the measurement noise covariance matrix of the Kalman filter to reduce the influence of the construction equipment vibration on the attitude correction and obtain the adjusted Kalman filter; Use the adjusted Kalman filter to fuse the historical tilt angle data of the target monitoring node and each adjacent monitoring node, and perform attitude correction on the tilt sensor of the target monitoring node to obtain the corrected tilt angle data of the tilt sensor of the target monitoring node as the original tilt angle data of the target monitoring node.

[0013] Preferably, step A2 includes: A201. Obtain the original tilt angle data within a preset time window, calculate the spectrum of the original tilt angle data, and determine the proportion of spectrum energy corresponding to the vibration frequency range of the construction equipment; A202. Adjust the sliding window size of the moving average filtering algorithm according to the proportion of spectrum energy; when the proportion of spectrum energy exceeds the preset threshold, reduce the sliding window size; when the proportion of spectrum energy does not exceed the preset threshold, increase the sliding window size; A203. Filter the original tilt angle data using the adjusted moving average filtering algorithm to obtain the filtered tilt angle data of each monitoring node.

[0014] Preferably, step A3 includes: A301. Obtain the filtered tilt angle data of adjacent monitoring nodes within a preset time window, and calculate the mean and standard deviation of the filtered tilt angle data of each adjacent monitoring node; A302. According to the mean and standard deviation of the filtered tilt angle data of each adjacent monitoring node, calculate the confidence interval of the filtered tilt angle data of each adjacent monitoring node according to the formula: [mean - K * standard deviation, mean + K * standard deviation], where K is the confidence coefficient; A303. Determine whether the filtered tilt angle data of adjacent monitoring nodes is within the corresponding confidence interval; A304. If there is filtered tilt angle data of adjacent monitoring nodes that is not within the corresponding confidence interval, perform rejection processing on the filtered tilt angle data that is not within the confidence interval, and calculate the tilt angle difference between adjacent monitoring nodes based on the remaining filtered tilt angle data to obtain the differential settlement value; A305. If the filtered tilt angle data of all adjacent monitoring nodes is within the corresponding confidence interval, directly calculate the tilt angle difference between adjacent monitoring nodes based on the filtered tilt angle data of all adjacent monitoring nodes to obtain the differential settlement value.

[0015] Preferably, step A304 includes: If there is filtered tilt angle data of adjacent monitoring nodes that is not within the corresponding confidence interval, identify the abnormal type of the filtered tilt angle data that is not within the confidence interval based on a preset abnormal detection rule library; the abnormal types include random noise type abnormality, sudden interference type abnormality, and sensor failure type abnormality; For random noise type abnormality, use the wavelet threshold filtering algorithm based on adaptive threshold to smooth the corresponding filtered tilt angle data to obtain the smoothed tilt angle data to replace the original filtered tilt angle data; For sudden interference type anomalies, the linear interpolation method based on data in adjacent time windows is used to interpolate and repair the corresponding filtered tilt angle data, obtaining the interpolated and repaired tilt angle data to replace the original filtered tilt angle data; For sensor failure type anomalies, the tilt angle data of the corresponding monitoring nodes are marked as invalid and excluded from subsequent differential settlement calculations, and at the same time, a sensor failure warning message is sent to the monitoring center; Based on the processed filtered tilt angle data, the tilt angle difference between adjacent monitoring nodes is calculated to obtain the differential settlement value.

[0016] Preferably, step A4 includes: A401. Obtain historical monitoring data within a preset time period, and extract a subset of historical settlement data that matches the geological conditions and pipeline network aging degree of the current construction area; A402. For the subset of historical settlement data, use time series analysis methods to predict the settlement change trend within a future time window, obtaining predicted settlement values; A403. According to the geological survey report of the current construction area, determine geological parameters, and combine the material, service life, and repair records of the pipeline network to evaluate the structural strength and remaining life of the pipeline network, obtaining pipeline network risk assessment parameters; A404. Based on the predicted settlement values and pipeline network risk assessment parameters, use the fuzzy logic reasoning method to evaluate the settlement risk level of the current construction area, and dynamically adjust the settlement difference threshold according to the settlement risk level; the higher the risk level, the smaller the settlement difference threshold; A405. Compare the differential settlement value with the dynamic settlement difference threshold. If the differential settlement value exceeds the dynamic settlement difference threshold, it is determined that there is a settlement risk, and a warning message including the risk location, settlement risk level, and recommended treatment measures is generated and sent to the monitoring center.

[0017] In a second aspect, the present application provides a safety monitoring device for the construction site of a gas pipeline network, which is used to monitor the settlement of the construction site of the trenchless renovation of an old gas pipeline network. The device includes: A monitoring data acquisition module, which is used to acquire the original tilt angle data collected by multiple monitoring nodes; the multiple monitoring nodes are distributed at intervals on both sides of the ground along the trenchless construction line of the gas pipeline network, and each monitoring node is configured with at least two tilt sensors with perpendicular directions to sense the tilt angles of the ground in different directions to obtain the original tilt angle data; A filtering processing module, which is used to perform filtering processing on the original tilt angle data by using a moving average filtering algorithm to obtain the filtered tilt angle data of each monitoring node; Differential settlement calculation module, which is used to calculate the differential inclination angle between adjacent monitoring nodes based on the filtered inclination angle data of adjacent monitoring nodes, so as to obtain the differential settlement value; Real-time warning module, which is used to compare the differential settlement value with the dynamic settlement difference threshold to determine the settlement risk and send warning information in a timely manner; the dynamic settlement difference threshold is dynamically adjusted based on historical monitoring data, geological conditions and the degree of pipeline network aging.

[0018] In a third aspect, the present application provides a safety monitoring system for the construction site of a gas pipeline network, which is used to monitor the settlement of the construction site of the trenchless renovation of an old gas pipeline network. The system includes multiple monitoring nodes, a mobile monitoring device, a cloud server and a monitoring center; all the monitoring nodes are communicatively connected to the mobile monitoring device, and both the mobile monitoring device and the monitoring center are communicatively connected to the cloud server; The multiple monitoring nodes are spaced apart along the ground on both sides of the trenchless construction line of the gas pipeline network. Each monitoring node is equipped with at least two tilt sensors perpendicular to each other in direction to sense the tilt angles of the ground in different directions so as to obtain the original tilt angle data; The mobile monitoring device is used to send the original tilt angle data collected by the multiple monitoring nodes to the cloud server; The cloud server is used to execute: Obtain the original tilt angle data collected by multiple monitoring nodes; Use the moving average filtering algorithm to filter the original tilt angle data to obtain the filtered tilt angle data of each monitoring node; Calculate the differential inclination angle between adjacent monitoring nodes based on the filtered inclination angle data of adjacent monitoring nodes to obtain the differential settlement value; Compare the differential settlement value with the dynamic settlement difference threshold to determine the settlement risk and send a warning message to the monitoring center in a timely manner; the dynamic settlement difference threshold is dynamically adjusted based on historical monitoring data, geological conditions and the degree of pipeline network aging.

[0019] Beneficial effects: A safety monitoring method and related equipment for the construction site of a gas pipeline network provided by the present application utilize distributed monitoring nodes to collect the original tilt angle data at different positions in the construction area in real time, use the moving average filtering algorithm to filter the original tilt angle data, calculate the differential settlement value based on the filtered tilt angle data of adjacent monitoring nodes, compare the differential settlement value with the dynamic settlement difference threshold to determine the settlement risk and give a warning, realizing the real-time monitoring of the ground settlement at the construction site of the gas pipeline network and improving the accuracy of settlement monitoring. Description of the Drawings

[0020] Figure 1Flowchart of a safety monitoring method for the construction site of a gas pipeline network provided by an embodiment of the present application.

[0021] Figure 2 Schematic structural diagram of a safety monitoring device for the construction site of a gas pipeline network provided by an embodiment of the present application.

[0022] Figure 3 Schematic structural diagram of a safety monitoring system for the construction site of a gas pipeline network provided by an embodiment of the present application.

[0023] Label description: 1. Monitoring data acquisition module; 2. Filtering processing module; 3. Differential settlement calculation module; 4. Real-time warning module; 100. Monitoring node; 200. Mobile monitoring device; 300. Cloud server; 400. Monitoring center. Specific implementation manners

[0024] The following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only some, but not all, embodiments of the present application. The components of the embodiments of the present application described and illustrated herein can be arranged and designed in various different configurations. Therefore, the detailed description of the embodiments of the present application provided herein is not intended to limit the scope of the present application claimed, but merely represents selected embodiments of the present application. All other embodiments obtained by those skilled in the art based on the embodiments of the present application without creative efforts fall within the scope of protection of the present application.

[0025] It should be noted that: similar reference numerals and letters denote similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of the present application, the terms "first", "second", etc. are only used for descriptive distinction and cannot be understood as indicating or implying relative importance.

[0026] Please refer to Figure 1 , a safety monitoring method for the construction site of a gas pipeline network provided by the present application, which is used for settlement monitoring of the non-excavation renovation construction site of an old gas pipeline network. The steps of this method include: A1. Obtain the original tilt angle data collected by multiple monitoring nodes; the multiple monitoring nodes are distributed at intervals on both sides of the ground along the non-excavation construction line of the gas pipeline network. Each monitoring node is configured with at least two tilt sensors perpendicular to each other in direction to sense the tilt angles of the ground in different directions to obtain the original tilt angle data; A2. Perform filtering processing on the original tilt angle data by using a moving average filtering algorithm to obtain the filtered tilt angle data of each monitoring node; A3. Calculate the tilt angle difference between adjacent monitoring nodes based on the filtered tilt angle data of adjacent monitoring nodes to obtain the differential settlement value; A4. Compare the differential settlement value with the dynamic settlement difference threshold to determine the settlement risk and issue a warning message in a timely manner; the dynamic settlement difference threshold is dynamically adjusted based on historical monitoring data, geological conditions, and the degree of pipeline network aging.

[0027] In step A1, multiple monitoring nodes are arranged at intervals along both sides of the ground of the trenchless construction line of the gas pipeline network. Each monitoring node is configured with at least two tilt sensors perpendicular to each other in different directions to sense the tilt angles of the ground in different directions, thereby obtaining the original tilt angle data. The layout of the monitoring nodes and the configuration of the sensors are designed to comprehensively collect the ground deformation information at the construction site.

[0028] In step A2, the original tilt angle data is filtered using a moving average filtering algorithm. The adoption of the moving average filtering algorithm aims to smooth the original data, reduce noise interference, and thus obtain more accurate filtered tilt angle data.

[0029] In step A3, based on the filtered tilt angle data of adjacent monitoring nodes, the tilt angle difference between adjacent monitoring nodes is calculated to obtain the differential settlement value. The differential settlement value can reflect the relative settlement degree between adjacent monitoring points and is a key indicator for evaluating the ground settlement risk.

[0030] In step A4, the differential settlement value is compared with the dynamic settlement difference threshold to determine whether there is a settlement risk and issue a warning message when necessary. The dynamic settlement difference threshold is not fixed but is dynamically adjusted based on historical monitoring data, geological conditions, and the degree of pipeline network aging. The setting of the dynamic threshold enables the warning system to flexibly adjust the warning standard according to the actual situation.

[0031] Specifically, the working principle of the proposed safety monitoring method for the gas pipeline network construction site is as follows: At the construction site of trenchless renovation of old urban gas pipelines, ground settlement is an important safety risk factor. To effectively monitor the settlement situation, first, a number of monitoring nodes are arranged at intervals along the ground on both sides of the construction line. The tilt sensors installed on each monitoring node can collect the tilt angle data of the ground in real time, and these data reflect the deformation state of the ground. The collected original tilt angle data are then processed by a moving average filtering algorithm to eliminate or reduce the interference of random noise and improve the accuracy of the data. The filtered tilt angle data are used to calculate the differential settlement value between adjacent monitoring nodes. The differential settlement value can more directly reflect the unevenness of ground settlement and highlight the differential settlement that may pose a threat to the safety of the gas pipeline network. Finally, the calculated differential settlement value is compared with the dynamic settlement difference threshold. The dynamic settlement difference threshold is dynamically adjusted according to factors such as historical monitoring data, geological conditions, and the aging degree of the pipeline network, and can more accurately reflect the current settlement risk level. When the differential settlement value exceeds the dynamic settlement difference threshold, the system determines that there is a settlement risk and issues a warning message in a timely manner, thus ensuring the construction safety. Through the above steps, this method constructs a complete ground settlement monitoring and warning system and realizes the effective monitoring of the safety status of the gas pipeline network construction site.

[0032] In some preferred embodiments, step A1 includes: A101. Obtain the geographical location information of each monitoring node; A102. Based on the geographical location information of each monitoring node, use differential GPS positioning technology to calibrate each monitoring node to obtain the calibrated geographical location information of each monitoring node; A103. According to the calibrated geographical location information of each monitoring node, determine the relative position relationship between each monitoring node, and based on the relative position relationship, combined with the Kalman filtering algorithm and the historical tilt angle data of each tilt sensor, correct the attitude of the tilt sensor of each monitoring node to obtain the corrected tilt angle data of each tilt sensor as the original tilt angle data.

[0033] Among them, for the problem that the original tilt angle data may be interfered by various factors, such as inaccurate geographical location information of the monitoring nodes or incorrect postures of the tilt sensors themselves, these factors will cause errors in the original tilt angle data, thereby affecting the accuracy and reliability of subsequent settlement monitoring. To improve the accuracy of the original tilt angle data, in step A101, the geographical location information of each monitoring node is first obtained, which lays a foundation for subsequent positioning calibration and attitude correction. In step A102, the differential GPS positioning technology is used to calibrate the geographical locations of each monitoring node. The differential GPS technology can effectively improve the positioning accuracy, reduce the geographical location error, and provide a more accurate position reference for subsequent attitude correction based on the position information. Step A103 is the key step. Among them, the calibrated geographical location information is used to determine the relative position relationship between the monitoring nodes, and thus the spatial topology structure between the monitoring nodes is established. Subsequently, the Kalman filter algorithm is used, combined with the historical tilt angle data of each tilt sensor and the relative position relationship between the monitoring nodes, to correct the attitude of the tilt sensors of each monitoring node. As an effective state estimation algorithm, the Kalman filter algorithm can fuse multi-source information, reduce noise interference, and improve the accuracy and stability of attitude estimation. Through attitude correction, the deviation of the tilt angle data caused by factors such as sensor installation errors and uneven ground can be eliminated or reduced, and more accurate original tilt angle data can be obtained, thereby providing a more reliable data basis for subsequent settlement calculation and risk warning, and improving the effectiveness and accuracy of the entire safety monitoring method.

[0034] Specifically, at the construction site of the trenchless renovation of the gas pipeline network, multiple monitoring nodes are distributed at intervals along the ground on both sides of the trenchless construction line of the gas pipeline network. Each monitoring node is equipped with at least two tilt sensors with perpendicular directions to sense the tilt angles of the ground in different directions. In step A101, the initial geographical location information of each monitoring node can be obtained by using measurement tools such as a handheld GPS device or a total station. Further, in step A102, in order to improve the accuracy of the geographical location information, differential GPS positioning technology is adopted. A differential GPS system usually includes a reference station and one or more mobile stations. The reference station is set up at a known precise coordinate point, continuously receives GPS satellite signals, compares the received signals with the known coordinates, and calculates the GPS positioning error. Then, the reference station sends these error correction messages to the mobile station through a wireless data link. While receiving GPS satellite signals, the mobile station also receives the error correction messages from the reference station and uses this information to correct its own GPS positioning result, thereby obtaining a higher-precision positioning result. Thus, the geographical location information of each monitoring node is calibrated. In step A103, the relative position relationship between each monitoring node is determined according to the calibrated geographical location information. For example, the distance and azimuth angle between any two monitoring nodes can be calculated. The Kalman filter algorithm is applied to attitude correction. A Kalman filter is a recursive filter that, based on the dynamic model and measurement model of the system, uses the state equation and observation equation of the system to optimally estimate the system state. In this solution, the attitude of the tilt sensor is regarded as the system state, and the measured value of the tilt sensor and the theoretical tilt angle value calculated based on the relative position relationship are regarded as the observed values. The Kalman filter algorithm fuses the historical tilt angle data and the current measurement data, continuously corrects the estimation of the tilt sensor attitude, and thus obtains the corrected tilt angle data.

[0035] In some possible implementation manners, step A103 includes: B1. According to the calibrated geographical location information of each monitoring node, calculate the distance between any two monitoring nodes, construct the topological relationship between the monitoring nodes according to the distance, and obtain the monitoring node topology graph; B2. Sequentially use each monitoring node as the target monitoring node and execute: B201. Based on the monitoring node topology graph, screen a number of adjacent monitoring nodes directly adjacent to the target monitoring node, and obtain the historical tilt angle data of the tilt sensors of each adjacent monitoring node within a preset time window (the size of the preset time window can be set according to actual needs); For the tilt sensor of the target monitoring node, using the Kalman filtering algorithm, fusing the historical tilt angle data of the target monitoring node and each adjacent monitoring node, performing attitude correction on the tilt sensor of the target monitoring node, obtaining the corrected tilt angle data of the tilt sensor of the target monitoring node, and using it as the original tilt angle data of the target monitoring node; during the attitude correction process, according to the vibration frequency range of the construction equipment, dynamically adjust the noise covariance matrix of the Kalman filter to reduce the influence of the construction equipment vibration on the attitude correction.

[0036] Among them, in step B1, the construction of the monitoring node topology map can be achieved through the following method: First, according to the calibrated geographical location information of each monitoring node, calculate the geographical distance between any two monitoring nodes, such as the Euclidean distance. Thus, based on a preset distance threshold, determine whether the monitoring nodes are adjacent. Specifically, if the distance between two monitoring nodes is less than or equal to the distance threshold, it is determined as adjacent, and a connection is established in the topology map. Further, the monitoring node topology map can be stored using the data structure of an adjacency matrix or an adjacency list to represent the connection relationship between the monitoring nodes.

[0037] Among them, in step B201, the screening of adjacent monitoring nodes is completed based on the monitoring node topology map constructed in step B1. Specifically, for each target monitoring node, by querying its adjacency relationship in the topology map, the adjacent monitoring nodes directly connected to it can be quickly screened out.

[0038] Among them, in step B202, the Kalman filtering algorithm is used to fuse the historical tilt angle data of the target monitoring node and the adjacent monitoring nodes. The Kalman filter includes two main steps: prediction and update. In the prediction step, based on the state equation of the system and the state estimate value at the previous moment, predict the state value and covariance matrix at the current moment. In the update step, use the measurement value at the current moment to correct the predicted state value to obtain the optimal state estimate value at the current moment. In this solution, the state vector of the Kalman filter can be the tilt angle of the target monitoring node, and the measurement vector can include the tilt angle data of the target monitoring node and the adjacent monitoring nodes. By fusing the data of adjacent nodes, the accuracy and robustness of the target node attitude correction can be improved. In addition, the dynamic adjustment of the noise covariance matrix is the key link to reduce the influence of construction equipment vibration. Specifically, the vibration frequency ranges of different types of construction equipment can be pre-calibrated. During the attitude correction process, real-time monitor the vibration situation at the construction site. If it is detected that the vibration frequency falls within the preset construction equipment vibration frequency range, increase the process noise covariance matrix and measurement noise covariance matrix of the Kalman filter to reduce the influence of vibration noise on the filtering result. Conversely, if the vibration is small or the frequency is not within the construction equipment vibration frequency range, reduce the noise covariance matrix to improve the sensitivity and response speed of the filter.

[0039] Specifically, this solution aims to address the problem that the spatial relationship and the vibration interference of construction equipment are not fully considered during the attitude correction process. First, a monitoring node topology map is constructed through step B1 to determine the spatial proximity relationship between monitoring nodes. The construction of this topology map enables the effective utilization of the spatial layout information of the monitoring nodes. Then, in step B2, for each monitoring node, attitude correction is performed. In step B201, based on the previously constructed topology map, the monitoring nodes adjacent to the target monitoring node are selected as neighboring monitoring nodes, and the historical tilt angle data of these neighboring monitoring nodes are obtained. Thus, during the attitude correction process, spatial correlation is considered, and the information of neighboring nodes is used to assist in improving the attitude correction accuracy of the target node. In step B202, the Kalman filter algorithm is used to fuse the historical tilt angle data of the target monitoring node and the neighboring monitoring nodes for attitude correction. The introduction of the Kalman filter algorithm realizes multi-sensor data fusion and improves the accuracy and reliability of attitude correction. Furthermore, step B202 also emphasizes that during the attitude correction process, the noise covariance matrix of the Kalman filter is dynamically adjusted according to the vibration frequency range of the construction equipment. This dynamic adjustment mechanism can effectively reduce the negative impact of the construction equipment vibration on the attitude correction and make the attitude correction result more accurate and reliable. Through the above steps, this solution effectively utilizes the spatial layout information of the monitoring nodes in the attitude correction link and reduces the interference of the vibration noise at the construction site, thereby improving the quality of the original tilt angle data and providing a more reliable data basis for subsequent settlement analysis and early warning.

[0040] In some embodiments, step B202 includes: Obtain the tilt angle data collected by the tilt sensors of the target monitoring node and each neighboring monitoring node within a preset time window, calculate the variance value of the tilt angle data of each monitoring node among the target monitoring node and each neighboring monitoring node, and obtain the variance value of the tilt angle data of the tilt sensors of the target monitoring node and each neighboring monitoring node; According to the vibration frequency range of the construction equipment and each variance value of the tilt angle data, determine the weight proportion of the tilt sensors of the target monitoring node and each neighboring monitoring node affected by the construction equipment vibration; Based on the weight proportion, dynamically adjust the process noise covariance matrix and the measurement noise covariance matrix of the Kalman filter to reduce the impact of the construction equipment vibration on the attitude correction, and obtain the adjusted Kalman filter; Use the adjusted Kalman filter to fuse the historical tilt angle data of the target monitoring node and each neighboring monitoring node, perform attitude correction on the tilt sensor of the target monitoring node, and obtain the corrected tilt angle data of the tilt sensor of the target monitoring node as the original tilt angle data of the target monitoring node.

[0041] Among them, the preset time window refers to the preset time length used to intercept the tilt angle data collected by the monitoring nodes. Specifically, time units such as seconds and minutes can be adopted. For example, it can be set to 10 seconds, 30 seconds, 1 minute, etc. The length of the time window can be adjusted according to the actual application scenario and data processing requirements. In some embodiments, the time window can be set to cover the vibration cycle duration of the construction equipment, so as to effectively capture the influence of vibration on the tilt angle data.

[0042] Among them, the variance value of the tilt angle data is a statistic used to quantify the degree of dispersion of the tilt angle data, reflecting the fluctuation amplitude of the data relative to the mean value. In specific implementation, the sample variance formula can be used for calculation, that is, for the tilt angle data within the preset time window, calculate the sum of the squares of the differences between each data point and the mean value, and then divide by the number of data points minus one. The larger the variance value, the greater the volatility of the tilt angle data, and the higher the possibility of being affected by the vibration of the construction equipment.

[0043] Among them, the vibration frequency range of the construction equipment refers to the vibration frequency range generated by the main construction equipment at the construction site during operation, which can be obtained by on-site measurement, parameters provided by equipment manufacturers or referring to relevant engineering materials, etc. Determining the vibration frequency range of the construction equipment in advance provides a basis for dynamically adjusting the parameters of the Kalman filter according to the frequency range subsequently.

[0044] Among them, the weight ratio refers to the relative ratio of the degree of influence of the target monitoring node and each adjacent monitoring node by the vibration of the construction equipment. The determination of the weight ratio comprehensively considers the vibration frequency range of the construction equipment and the variance value of the tilt angle data. In specific implementation, the correlation between the variance value of the tilt angle data of each monitoring node and the vibration frequency range of the construction equipment can be analyzed first. For example, if the variance value of the tilt angle data of a certain monitoring node fluctuates greatly within the vibration frequency range of the construction equipment, it is considered that this monitoring node is greatly affected by the vibration and should be given a higher weight ratio. The specific value of the weight ratio can be determined by methods such as expert experience, experimental analysis or machine learning.

[0045] Among them, the process of dynamically adjusting the process noise covariance matrix and the measurement noise covariance matrix of the Kalman filter means that, according to the result of the weight ratio, the values of the process noise covariance matrix and the measurement noise covariance matrix in the Kalman filter are adjusted in real time. The process noise covariance matrix characterizes the intensity of the system process noise, and the measurement noise covariance matrix characterizes the intensity of the sensor measurement noise. The higher the weight ratio, the greater the influence of the vibration of the construction equipment, and the corresponding numerical value of the noise covariance matrix should be increased to reduce the confidence level of the Kalman filter in the data affected by vibration and reduce the influence of vibration noise on the attitude correction result. On the contrary, the lower the weight ratio, the numerical value of the noise covariance matrix can be appropriately reduced to improve the attitude correction accuracy of the Kalman filter. The adjustment of the noise covariance matrix can adopt various methods such as linear adjustment and non-linear adjustment. For example, the adjustment amount of the noise covariance matrix can be set to be proportional to the weight ratio.

[0046] Among them, the Kalman filter is an optimal estimation algorithm for estimating the system state. In this application, the Kalman filter is used to fuse the historical tilt angle data of the target monitoring node and each adjacent monitoring node to correct the attitude of the tilt sensor of the target monitoring node. The Kalman filtering algorithm is based on the system state equation and the measurement equation, and uses the state estimate value of the previous moment and the measurement value of the current moment to recursively estimate the optimal state value of the current moment. Through the Kalman filtering algorithm, random noise can be effectively filtered out, and the accuracy and stability of attitude correction can be improved.

[0047] Among them, attitude correction refers to correcting the attitude of the tilt sensor to eliminate or reduce the influence of factors such as the sensor's own error and installation error on the tilt angle data, and improve the accuracy of the tilt angle data. In this application, the purpose of attitude correction is to obtain more accurate tilt angle data, providing a reliable data basis for subsequent settlement analysis and safety warning.

[0048] Specifically, in the solution of this application, first, the variance value of the tilt angle data of each monitoring node is calculated to evaluate the degree of influence of each monitoring node by vibration. Then, in combination with the vibration frequency range of the construction equipment, the weight ratio of each monitoring node is determined, and the noise covariance matrix of the Kalman filter is dynamically adjusted based on the weight ratio. Thus, when the Kalman filter performs data fusion, it can adaptively adjust the confidence level of different source data according to the degree of influence of each monitoring node by vibration, reduce the influence of vibration noise on the attitude correction result, and obtain a more accurate attitude correction result. Therefore, the attitude correction method provided by the solution of this application can more effectively suppress the vibration noise of the construction equipment, improve the attitude correction accuracy, and thus enhance the reliability of settlement monitoring.

[0049] As a preferred embodiment, the solution of this application is specifically implemented as follows: The preset time window is set to 10 seconds; Obtain the tilt angle data collected by the target monitoring node and its neighboring monitoring nodes in the past 10 seconds; Calculate the variance value of the tilt angle data of each monitoring node in the past 10 seconds; Pre-obtain the vibration frequency range of the main construction equipment at the construction site; According to the variance value of the tilt angle data of each monitoring node and the vibration frequency range of the construction equipment, determine the weight ratio of each monitoring node; for example, it can be set that the weight ratio is proportional to the variance value and normalized so that the sum of the weight ratios of all monitoring nodes is 1; Based on the weight ratio, dynamically adjust the process noise covariance matrix Q and the measurement noise covariance matrix R of the Kalman filter. For example, the diagonal elements of the Q and R matrices can be multiplied by the weight ratio coefficient, and the higher the weight ratio, the larger the corresponding noise covariance matrix element value.

[0050] Use the adjusted Kalman filter to fuse the historical tilt angle data of the target monitoring node and each neighboring monitoring node, perform attitude correction, and obtain the corrected tilt angle data of the tilt sensor of the target monitoring node.

[0051] Through the above technical solutions, the present application can dynamically adjust the noise covariance matrix of the Kalman filter according to the vibration frequency range of the construction equipment and the variance value of the tilt angle data of each monitoring node, so that the Kalman filter can adaptively reduce the influence of the construction equipment vibration noise during the attitude correction process, obtain a more accurate attitude correction result, and effectively improve the accuracy and reliability of the tilt sensor attitude correction in the construction vibration environment.

[0052] In some embodiments, step A2 includes: A201. Obtain the original tilt angle data within the preset time window, calculate the spectrum of the original tilt angle data, and determine the spectrum energy ratio corresponding to the vibration frequency range of the construction equipment; A202. Adjust the sliding window size of the moving average filtering algorithm according to the spectrum energy ratio; when the spectrum energy ratio exceeds the preset threshold, reduce the sliding window size; when the spectrum energy ratio does not exceed the preset threshold, increase the sliding window size; A203. Filter the original tilt angle data using the adjusted moving average filtering algorithm to obtain the filtered tilt angle data of each monitoring node.

[0053] Among them, in step A201, the original tilt angle data is collected by a tilt sensor. Specifically for spectrum calculation, the fast Fourier transform algorithm can be used to process the original tilt angle data within a preset time window to obtain the spectrum distribution. The vibration frequency range of the construction equipment can be preset according to the type and working mode of the construction equipment. For example, the vibration frequency range of an excavator is usually 10 - 30 Hz. The determination method of the spectrum energy ratio is to calculate the total energy of the frequency components corresponding to the vibration frequency range of the construction equipment in the spectrum, and then compare this total energy with the total energy of the entire spectrum to obtain the spectrum energy ratio.

[0054] Among them, in step A202, a preset threshold is used to judge the intensity level of the vibration interference. The preset threshold can be set according to the actual application scenario and the requirements for the filtering effect. For example, the preset threshold can be set to 0.6. The adjustment strategy for the sliding window size is that when the spectrum energy ratio exceeds the preset threshold, it indicates that the vibration interference is strong, and the sliding window size is reduced. The reduction amount is calculated according to the excess amount by which the spectrum energy ratio exceeds the preset threshold. For example, it is calculated according to a preset reduction amount calculation formula, or corresponding reduction amounts are preset for different excess amount ranges, and the corresponding reduction amount is determined according to the actual excess amount range into which it falls; when the spectrum energy ratio does not exceed the preset threshold, it indicates that the vibration interference is weak, and the sliding window size is increased. The increase amount is calculated according to the shortfall amount by which the spectrum energy ratio is less than the preset threshold. For example, it is calculated according to a preset increase amount calculation formula, or corresponding increase amounts are preset for different shortfall amount ranges, and the corresponding increase amount is determined according to the actual shortfall amount range into which it falls; the sliding window size is adjusted by subtracting the reduction amount or adding the increase amount from the preset size.

[0055] Among them, in step A203, the moving average filtering algorithm uses the adjusted sliding window size to filter the original tilt angle data. Specifically, for the original tilt angle data sequence of each monitoring node, the moving average filtering algorithm smooths the data and reduces noise interference by calculating the average value of the data within the sliding window.

[0056] Specifically, in the method for safety monitoring of the construction site of a gas pipeline network, in order to more effectively eliminate the influence of the vibration of construction equipment on the tilt angle data, the filtering process is improved. First, in step A201, by performing spectral analysis on the original tilt angle data within a preset time window, the frequency components and energy proportion of the vibration interference of the construction equipment are quantified. Thus, the degree of vibration interference at the construction site can be effectively evaluated. Then, in step A202, based on the spectral energy proportion, the size of the sliding window of the moving average filtering algorithm is adaptively adjusted. When strong vibration interference is detected, the sliding window is reduced, and the over-smoothing of the settlement signal by the filtering algorithm is reduced, and the sensitivity of the settlement information is maintained. On the contrary, when the vibration interference is weak, the sliding window is increased, and the ability to filter out random noise is enhanced, and the data smoothness is improved. Finally, in step A203, the adjusted moving average filtering algorithm is used for filtering, and the pertinence and effectiveness of the filtering are improved. Through this adaptive filtering method, the vibration interference of the construction equipment is effectively suppressed during the filtering process, the ground settlement information is extracted more accurately, and the accuracy and reliability of the settlement monitoring are improved, providing more reliable data support for subsequent settlement risk early warning.

[0057] In some embodiments, A3 includes: A301. Obtain the filtered tilt angle data of adjacent monitoring nodes within a preset time window, and calculate the mean and standard deviation of the filtered tilt angle data of each adjacent monitoring node; A302. According to the mean and standard deviation of the filtered tilt angle data of each adjacent monitoring node, calculate the confidence interval of the filtered tilt angle data of each adjacent monitoring node according to the formula: [mean - K * standard deviation, mean + K * standard deviation], where K is the confidence coefficient; A303. Determine whether the filtered tilt angle data of adjacent monitoring nodes is within the corresponding confidence interval; A304. If there is filtered tilt angle data of adjacent monitoring nodes that is not within the corresponding confidence interval, then perform elimination processing on the filtered tilt angle data that is not within the confidence interval, and calculate the tilt angle difference between adjacent monitoring nodes based on the remaining filtered tilt angle data to obtain the differential settlement value; A305. If the filtered tilt angle data of all adjacent monitoring nodes is within the corresponding confidence interval, then directly calculate the tilt angle difference between adjacent monitoring nodes based on the filtered tilt angle data of all adjacent monitoring nodes to obtain the differential settlement value.

[0058] Among them, in step A301, adjacent monitoring nodes are pre-determined. For example, the two monitoring nodes with the closest physical distance in the monitoring line can be defined as adjacent monitoring nodes. The filtered tilt angle data is obtained by processing the original tilt angle data with a moving average filtering algorithm. The mean and standard deviation of the combined filtered tilt angle data of two adjacent monitoring nodes are calculated. The mean reflects the average level of the filtered tilt angle data within a preset time window, and the standard deviation reflects the degree of dispersion of these data relative to the mean.

[0059] Among them, in step A302, the confidence coefficient K is a preset value used to control the width of the confidence interval. The larger the K value, the wider the confidence interval and the higher the tolerance for the data to be judged as normal, and vice versa. The K value can be adjusted according to the actual application scenario and the requirements for data reliability. For example, the K value can be taken as 1.96 corresponding to a 95% confidence level, or 2.58 corresponding to a 99% confidence level.

[0060] Among them, in step A303, the judgment process is to compare the filtered tilt angle data of each adjacent monitoring node with the corresponding confidence interval calculated in step A302.

[0061] Among them, for the data elimination process in step A304, the data not within the confidence interval can be directly removed from the dataset and will no longer participate in the subsequent differential settlement value calculation.

[0062] Among them, in step A305, directly calculating the difference based on the filtered tilt angle data of all adjacent monitoring nodes means that when all data are judged to be normal, no data is eliminated, and all data is directly used for subsequent calculations.

[0063] Specifically, for the accuracy and reliability issues of differential settlement value calculation, an abnormal data elimination method based on the confidence interval is proposed. First, in step A301, the filtered tilt angle data from adjacent monitoring nodes within a certain time range is obtained. Then, the statistical characteristics of these data, including the mean and standard deviation, are calculated to evaluate the fluctuation range of the data. Further, in step A302, a confidence interval is constructed using the mean and standard deviation, which represents the range where normal data should fall. Thus, in step A303, each data point is checked to determine whether it falls within the confidence interval. For those data points outside the confidence interval, they are identified as outliers and eliminated in step A304. Subsequently, the differential settlement value is calculated based on the remaining normal data. On the contrary, if in step A305, all data are within the confidence interval, then all data is directly used to calculate the differential settlement value. In this way, the influence of abnormal data on the differential settlement value calculation can be effectively reduced, and the accuracy of the settlement monitoring results can be improved.

[0064] In some possible embodiments, step A304 includes: If the filtered tilt angle data of adjacent monitoring nodes are not within the corresponding confidence intervals, based on a preset anomaly detection rule library, identify the anomaly types of the filtered tilt angle data that are not within the confidence intervals; the anomaly types include random noise type anomalies, sudden interference type anomalies, and sensor failure type anomalies; For random noise type anomalies, use a wavelet threshold filtering algorithm based on an adaptive threshold to smooth the corresponding filtered tilt angle data to obtain smoothed tilt angle data to replace the original filtered tilt angle data; For sudden interference type anomalies, use a linear interpolation method based on adjacent time window data to interpolate and repair the corresponding filtered tilt angle data to obtain interpolated and repaired tilt angle data to replace the original filtered tilt angle data; For sensor failure type anomalies, mark the tilt angle data of the corresponding monitoring node as invalid, exclude it from subsequent differential settlement calculations, and send a sensor failure warning message to the monitoring center; Based on the processed filtered tilt angle data, calculate the tilt angle difference between adjacent monitoring nodes to obtain the differential settlement value.

[0065] Among them, the construction method of the anomaly detection rule library is to pre-analyze various situations that may cause anomalies in tilt angle data, such as environmental noise interference, construction equipment vibration, sensor self-drift or damage, etc. For each situation, set the corresponding anomaly type and discrimination rules. The discrimination rules may include but are not limited to: data mutation amplitude threshold, data duration exceeding the confidence interval, data fluctuation frequency characteristics, etc. In practical applications, the anomaly detection rule library can be implemented by methods such as expert systems, decision trees, or neural networks.

[0066] Among them, for the processing of random noise type anomalies, a wavelet threshold filtering algorithm based on an adaptive threshold is adopted. Among them, the determination method of the adaptive threshold is to calculate based on the estimated noise level value of the tilt angle data within a preset time window. For example, the median absolute deviation algorithm can be used to estimate the noise level and dynamically adjust the wavelet threshold according to the noise level. The selection of the wavelet basis function can be based on the noise characteristics in the actual application scenario. For example, for scenarios with obvious high-frequency noise, a wavelet basis function with better high-frequency filtering effect can be selected.

[0067] Among them, for the processing of sudden interference type anomalies, a linear interpolation method based on adjacent time window data is adopted. Specifically, for the adjacent time window data used in linear interpolation, the data within a certain number of time windows before and after the abnormal data point can be selected. For example, the data within 3 time windows before and after are selected for linear interpolation. When performing interpolation calculation, a linear regression method can be used to fit a linear model based on the selected adjacent time window data and use the linear model to predict the value of the abnormal data point.

[0068] Among them, for the processing of sensor failure type anomalies, the specific way to mark as invalid is to set an anomaly flag bit in the data storage structure. After the data is marked as invalid, the subsequent data processing flow will automatically skip the invalid data. Excluding from the subsequent differential settlement calculation means that when calculating the tilt angle difference between adjacent monitoring nodes, if the tilt angle data of a certain monitoring node is marked as invalid, the data of this monitoring node is not used for the difference calculation. The way to send a sensor failure warning message is to send a warning message containing sensor failure information to the monitoring center through a wireless communication network. The warning message includes information such as the time, location, and sensor number of the failure.

[0069] Specifically, after obtaining the filtered tilt angle data of the monitoring node, first determine whether the data is within the corresponding confidence interval. If the data is within the confidence interval, the data is considered normal and directly used for the subsequent differential settlement calculation. If the data is not within the confidence interval, start the anomaly detection process. The anomaly detection process first queries the anomaly detection rule library and identifies the anomaly type to which the data belongs according to the preset discrimination rules. If the identification result is a random noise type anomaly, an adaptive threshold wavelet threshold filtering algorithm is used to smooth the data. The smoothed data replaces the original filtered tilt angle data and is used for the subsequent differential settlement calculation. If the identification result is a sudden interference type anomaly, a linear interpolation method based on adjacent time window data is used to interpolate and repair the data. The interpolated and repaired data replaces the original filtered tilt angle data and is used for the subsequent differential settlement calculation. If the identification result is a sensor failure type anomaly, the tilt angle data of this monitoring node is marked as invalid and excluded from the subsequent differential settlement calculation. At the same time, the system generates a sensor failure warning message and sends it to the monitoring center to prompt the maintenance personnel to handle it in time. After completing the processing of abnormal data, based on the processed filtered tilt angle data, calculate the tilt angle difference between adjacent monitoring nodes to obtain the differential settlement value.

[0070] In some specific embodiments, the anomaly detection rule library may adopt the following rules: When the filtered tilt angle data has a change amplitude exceeding a preset first threshold compared with the data in the two time windows before and after it, it is determined as a sudden interference type anomaly; When the filtered tilt angle data exceeds the confidence interval for five consecutive time windows, but the data fluctuation amplitude is less than a preset second threshold, it is determined as a random noise type anomaly; When the filtered tilt angle data exceeds the confidence interval for ten consecutive time windows, and the data value remains zero or constant, it is determined as a sensor failure type anomaly. In the adaptive threshold wavelet threshold filtering algorithm, the wavelet basis function is selected as the db4 wavelet basis function, the decomposition level is set to 3 levels, and the adaptive threshold calculation formula is: threshold = noise level estimated value * threshold coefficient, where the noise level estimated value is calculated using the median absolute deviation algorithm, and the threshold coefficient is set according to actual engineering experience, for example, set to 3. In the linear interpolation method based on adjacent time window data, the data of 3 time windows before and after the abnormal data point are selected for linear interpolation, and the linear regression model is obtained by fitting using the least squares method. The sensor failure warning information is sent to the mobile phone of the maintenance personnel in the monitoring center via text message. By adopting the above abnormal data processing method, the quality of the tilt angle data can be effectively improved, the influence of abnormal data on the settlement monitoring results can be reduced, and the reliability and accuracy of the safety monitoring of the gas pipeline network construction site can be ensured.

[0071] In some embodiments, step A4 includes: A401. Obtain historical monitoring data within a preset time period, and extract a subset of historical settlement data that matches the geological conditions and pipeline network aging degree of the current construction area; A402. For the subset of historical settlement data, adopt time series analysis methods to predict the settlement change trend within the future time window and obtain the predicted settlement value; A403. According to the geological survey report of the current construction area, determine the geological parameters, and combine the material, service life, and repair records of the pipeline network to evaluate the structural strength and remaining life of the pipeline network, and obtain the pipeline network risk assessment parameters; A404. Based on the predicted settlement value and the pipeline network risk assessment parameters, use the fuzzy logic reasoning method to evaluate the settlement risk level of the current construction area, and dynamically adjust the settlement difference threshold according to the settlement risk level; The higher the risk level, the smaller the settlement difference threshold; A405. Compare the differential settlement value with the dynamic settlement difference threshold. If the differential settlement value exceeds the dynamic settlement difference threshold (indicating that at least one differential settlement value exceeds the dynamic settlement difference threshold), it is determined that there is a settlement risk, and a warning information including the risk location, settlement risk level, and recommended treatment measures is generated and sent to the monitoring center.

[0072] Among them, in step A401, obtaining historical monitoring data within a preset period and extracting a subset of historical settlement data that matches the geological conditions and pipeline network aging degree of the current construction area means that in order to make the adjustment of the dynamic settlement difference threshold more in line with the actual situation, the solution obtains historical monitoring data and extracts a subset of historical settlement data that matches the geological conditions and pipeline network aging degree of the current construction area, ensuring that the subsequent threshold adjustment is based on historical data similar to the current working conditions, excluding the interference of irrelevant historical data, and improving the pertinence and accuracy of the threshold adjustment. Among them, the historical monitoring data can include, but is not limited to: settlement monitoring data of the prior construction area, geological exploration reports, pipeline network detection reports, etc. The preset period can be set according to actual monitoring needs, such as set to the last year, the last half year or the last three months, etc. The geological conditions can include parameters such as soil type, stratigraphic structure, and groundwater level. The pipeline network aging degree can be evaluated according to factors such as the service life, material, and corrosion degree of the pipeline network. The matching method can adopt methods such as similarity calculation and cluster analysis to screen out historical settlement data similar to the geological conditions and pipeline network aging degree of the current construction area.

[0073] Among them, in step A402, for the subset of historical settlement data, a time series analysis method is used to predict the settlement change trend within a future time window and obtain a predicted settlement value, which means that in order to be able to predict the future settlement change trend, the solution uses a time series analysis method for the subset of historical settlement data to predict the settlement change trend within a future time window and obtain a predicted settlement value, predicting in advance the possible settlement situation in the future, providing a forward-looking basis for dynamically adjusting the settlement difference threshold, so that the threshold adjustment is not only based on the current state, but also a kind of anticipation and response to future risks. Among them, the time series analysis method can include, but is not limited to: ARIMA model, Prophet model or LSTM neural network, etc. The future time window can be set according to the early warning lead requirement, such as set to the next week, the next three days or the next 24 hours, etc. The predicted settlement value can be the average settlement value, the maximum settlement value or the settlement change rate within the future time window.

[0074] Among them, in step A403, according to the geological survey report of the current construction area, geological parameters are determined. Combining the material, service life and repair records of the pipeline network, the structural strength and remaining life of the pipeline network are evaluated to obtain pipeline network risk assessment parameters. This means that in order to more comprehensively evaluate the risk, the solution also determines geological parameters according to the geological survey report of the current construction area, and combines the material, service life and repair records of the pipeline network to evaluate the structural strength and remaining life of the pipeline network, obtaining pipeline network risk assessment parameters. The potential risks of the pipeline network are comprehensively evaluated from two dimensions of geological conditions and the pipeline network's own state, making the risk assessment more comprehensive and objective, and providing a more reliable risk reference for subsequent threshold adjustment. Among them, geological parameters can include geomechanical parameters such as the compression modulus, internal friction angle, and cohesion of the soil. The materials of the pipeline network can include steel pipes, cast iron pipes, PE pipes, etc. The service life can be calculated starting from the laying date of the pipeline network. Repair records can include information such as the number of leaks, repair methods, and repair times of the pipeline network. The pipeline network risk assessment parameters can be a comprehensive score or grade representing the structural strength and remaining life of the pipeline network. For example, the weighted sum of the structural strength and remaining life of the pipeline network can be calculated to obtain the pipeline network risk assessment parameters.

[0075] Among them, in step A404, based on the predicted settlement value and the pipeline network risk assessment parameters, using the fuzzy logic inference method, the settlement risk level of the current construction area is evaluated, and according to the settlement risk level, the settlement difference threshold is dynamically adjusted; the higher the risk level, the smaller the settlement difference threshold. This means that on this basis, the solution evaluates the settlement risk level of the current construction area based on the predicted settlement value and the pipeline network risk assessment parameters, and dynamically adjusts the settlement difference threshold according to the settlement risk level. The introduction of the fuzzy logic inference method can effectively handle the uncertainty and fuzziness in the predicted settlement value and the pipeline network risk assessment parameters, making the evaluation of the risk level more in line with the actual situation. And, dynamically adjusting the settlement difference threshold according to the settlement risk level realizes the linkage adjustment of the threshold and the risk level. When the risk level is higher, the settlement difference threshold is smaller, and vice versa, ensuring that in high-risk situations, the system can more sensitively capture settlement changes and issue early warnings in a timely manner. Among them, the fuzzy logic inference method can adopt Mamdani fuzzy inference or Takagi-Sugeno fuzzy inference. The fuzzy rule base can be constructed according to expert experience or historical data. The settlement risk level can be divided into multiple levels such as low risk, medium risk, and high risk. Dynamically adjusting the settlement difference threshold can adopt methods such as linear functions, non-linear functions, or look-up tables to establish a mapping relationship between the settlement risk level and the settlement difference threshold.

[0076] Among them, in step A405, the differential settlement value is compared with the dynamic settlement difference threshold. If the differential settlement value exceeds the dynamic settlement difference threshold, it is determined that there is a settlement risk, and a warning message including the risk location, the settlement risk level, and the recommended treatment measures is generated and sent to the monitoring center. That is to say, finally, the solution compares the differential settlement value with the dynamic settlement difference threshold, and when the differential settlement value exceeds the dynamic settlement difference threshold, it is determined that there is a settlement risk, and a warning message including the risk location, the settlement risk level, and the recommended treatment measures is generated and sent to the monitoring center, completing the determination of the settlement risk and the generation of the warning message, providing timely, accurate, and comprehensive information support for subsequent risk handling. The warning message includes the risk location, the settlement risk level, and the recommended treatment measures, making the warning message more instructive, facilitating the monitoring center to quickly locate the risk area and take corresponding treatment measures. Among them, the risk location can be determined by the geographical location information of the monitoring nodes. The recommended treatment measures may include, but are not limited to: stopping construction, strengthening the pipeline network, adjusting construction parameters, etc. The warning message can be sent to the monitoring center through various methods such as text messages, emails, and audible and visual alarms.

[0077] Specifically, this solution first obtains the historical monitoring data within a preset time window and extracts a subset of historical settlement data that matches the geological conditions and pipeline network aging degree of the current construction area. This ensures that the subsequent adjustment of the dynamic settlement difference threshold is based on historical data similar to the current working conditions, eliminates the interference of irrelevant historical data, and improves the pertinence and accuracy of threshold adjustment. Then, for the extracted subset of historical settlement data, a time series analysis method is used to predict the settlement change trend within the future time window, obtaining the predicted settlement value, predicting in advance the possible settlement situation in the future, providing a forward-looking basis for dynamically adjusting the settlement difference threshold, and enabling the threshold adjustment to anticipate and respond to future risks. Further, the solution also determines geological parameters based on the geological survey report of the current construction area, and combines the material, service life, and repair records of the pipeline network to evaluate the structural strength and remaining life of the pipeline network, obtaining pipeline network risk assessment parameters, comprehensively evaluating the potential risks of the pipeline network from two dimensions of geological conditions and the pipeline network's own state, making the risk assessment more comprehensive and objective, and providing a more reliable risk reference for subsequent threshold adjustment. On this basis, the solution uses the fuzzy logic inference method to evaluate the settlement risk level of the current construction area based on the predicted settlement value and pipeline network risk assessment parameters, and dynamically adjusts the settlement difference threshold according to the settlement risk level. The introduction of the fuzzy logic inference method can effectively handle the uncertainty and fuzziness in the predicted settlement value and pipeline network risk assessment parameters, making the evaluation of the risk level more in line with the actual situation, realizing the linkage adjustment of the threshold and the risk level, and ensuring that the system can more sensitively capture settlement changes and issue early warnings in high-risk situations. Finally, the solution compares the differential settlement value with the dynamic settlement difference threshold, and when the differential settlement value exceeds the dynamic settlement difference threshold, it determines that there is a settlement risk, generates a warning message including the risk location, settlement risk level, and recommended treatment measures, and sends it to the monitoring center, completing the determination of the settlement risk and the generation of the warning message, providing timely information support for subsequent risk handling.

[0078] As a preferred embodiment, the solution of this application is specifically implemented as follows: At the non-excavation reconstruction construction site of the old gas pipeline network, first, the system obtains the historical monitoring data of the past three months, including settlement data, geological reports, and pipeline network inspection reports. Then, since the soil type in the current construction area is sandy soil, the pipeline network material is cast iron, and the service life is 30 years, a subset of the historical monitoring data is extracted, which contains the settlement data during the 25-35-year service period of the cast iron pipeline network under sandy soil geological conditions. Next, for this subset of historical settlement data, the ARIMA time series model is used to predict the settlement change trend in the next week, and the predicted settlement value is obtained. At the same time, according to the geological survey report of the current construction area, the soil compression modulus is determined to be 5 MPa, the internal friction angle is 30 degrees, and the cohesion is 20 kPa. Combining with the pipeline network material being cast iron, the service life being 30 years, and the repair record showing 2 leaks in the past five years, the structural strength and remaining life of the pipeline network are evaluated to obtain the pipeline network risk assessment parameters. For example, the risk assessment parameter is 70 points (on a 100-point scale, the higher the score, the higher the risk). After that, based on the predicted settlement value and the pipeline network risk assessment parameter of 70 points, the fuzzy logic reasoning method is used to evaluate the settlement risk level of the current construction area as medium risk. According to the medium risk level, the settlement difference threshold is dynamically adjusted to 5 mm. Finally, the differential settlement value is monitored in real time. When the differential settlement value at a certain location is monitored to be 6 mm, exceeding the dynamic settlement difference threshold of 5 mm, it is determined that there is a settlement risk at this location, and a warning message containing the risk location (monitoring node number), settlement risk level (medium risk), and recommended treatment measures (slow down the construction speed) is generated and immediately sent to the monitoring center.

[0079] Through the above technical solution, the present application can more reasonably and effectively dynamically adjust the settlement difference threshold, making the determination of settlement risk more accurate and reliable, thereby providing a strong guarantee for the safety monitoring of the non-excavation reconstruction construction site of the old gas pipeline network.

[0080] In some preferred embodiments, step A404 includes: Determine the fuzzy membership functions of the predicted settlement value and the pipeline network risk assessment parameters, and based on the fuzzy membership functions, convert the predicted settlement value and the pipeline network risk assessment parameters into corresponding fuzzy sets; According to the fuzzy sets of the predicted settlement value and the pipeline network risk assessment parameters, and the fuzzy rules in the fuzzy rule base, use the fuzzy inference engine for fuzzy inference to obtain the fuzzy set of the settlement risk level of the current construction area; the fuzzy rule base contains multiple fuzzy rules, and each fuzzy rule defines the settlement risk level corresponding to the combination of the predicted settlement value and the pipeline network risk assessment parameters of different fuzzy sets; Perform defuzzification processing on the fuzzy set of the settlement risk level to obtain a clear settlement risk level value. According to the settlement risk level value, according to the formula: settlement difference threshold = basic threshold - settlement risk level value * adjustment coefficient, dynamically adjust the settlement difference threshold.

[0081] Among them, determining the fuzzy membership functions of the predicted settlement value and the pipeline network risk assessment parameters, and converting the predicted settlement value and the pipeline network risk assessment parameters into corresponding fuzzy sets based on the fuzzy membership functions means that in order for the predicted settlement value and the pipeline network risk assessment parameters to be processed by the fuzzy logic system, it is necessary to pre-determine the fuzzy membership functions for these parameters. The role of the fuzzy membership function is to map the exact numerical values of these parameters to the membership values of the fuzzy sets in the [0, 1] interval. Specifically, it can be achieved by using triangular membership functions, trapezoidal membership functions, Gaussian membership functions, etc., so as to reflect the membership degrees of different numerical values in the fuzzy concept. For example, for the predicted settlement value, three fuzzy sets of "low", "medium", and "high" can be set, and corresponding membership functions are designed for each fuzzy set. When a specific predicted settlement value is input, the membership degrees of it belonging to the three fuzzy sets of "low", "medium", and "high" are calculated through the membership function.

[0082] Among them, according to the fuzzy sets of the predicted settlement value and the pipeline network risk assessment parameters, and the fuzzy rules in the fuzzy rule base, using the fuzzy inference engine to perform fuzzy inference to obtain the fuzzy set of the settlement risk level in the current construction area means that after the fuzzy processing of the parameters is completed, it is necessary to establish a fuzzy rule base and a fuzzy inference engine to evaluate the settlement risk level. The fuzzy rule base stores multiple fuzzy rules, and each rule is an IF-THEN statement, which describes the relationship between different fuzzy set combinations of the predicted settlement value and the pipeline network risk assessment parameters and the settlement risk level. The fuzzy inference engine then matches the rules in the fuzzy rule base according to the current input fuzzy sets of the predicted settlement value and the pipeline network risk assessment parameters, and uses a fuzzy inference algorithm (such as the MIN-MAX inference algorithm) to perform inference calculations, and finally obtains the fuzzy set of the settlement risk level. The design of the fuzzy rule base can be based on expert experience, historical data, or simulation models, etc. For example, a fuzzy rule can be: IF the predicted settlement value is high AND the pipeline network risk assessment parameter is high THEN the settlement risk level is high.

[0083] Among them, defuzzifying the fuzzy set of the settlement risk level to obtain a clear settlement risk level value means that the settlement risk level obtained by fuzzy inference is a fuzzy set. In order to be able to use it for subsequent threshold adjustment calculations, it is necessary to perform defuzzification to convert the fuzzy set into a clear numerical value. Specifically, it can be achieved by using defuzzification methods such as the centroid method, the maximum membership degree method, or the weighted average method. For example, when using the centroid method, calculate the abscissa value of the centroid of the fuzzy set area as the settlement risk level value after defuzzification.

[0084] Among them, according to the settlement risk level value, the settlement difference threshold is dynamically adjusted according to the formula: settlement difference threshold = basic threshold - settlement risk level value * adjustment coefficient. That is, after obtaining the clear settlement risk level value, this value is used to dynamically adjust the settlement difference threshold. The adjustment formula adopts a linear decreasing model, that is, the higher the settlement risk level value, the smaller the settlement difference threshold, and vice versa. The basic threshold in the formula can be preset according to engineering experience or specifications, and the adjustment coefficient is used to control the amplitude of threshold adjustment. Through this formula, the adaptive adjustment of the settlement difference threshold with the settlement risk level can be realized, thereby improving the sensitivity and accuracy of settlement risk warning.

[0085] Specifically, this solution focuses on the implementation method of step A404, that is, how to use the fuzzy logic reasoning method to evaluate the settlement risk level and dynamically adjust the settlement difference threshold. During the settlement monitoring process, first, the predicted settlement value and the pipeline network risk assessment parameters are transformed into fuzzy sets through the fuzzy membership function, so that the original precise numerical information is transformed into a form that fuzzy logic can understand and process. Thus, the fuzzy expression of risk parameters is realized, laying a foundation for subsequent fuzzy reasoning. Then, the fuzzy inference engine performs fuzzy reasoning based on these fuzzy sets and referring to the pre-set fuzzy rule base to obtain the fuzzy set of the settlement risk level in the current construction area. The introduction of the fuzzy rule base makes the risk assessment process more standardized and intelligent, and can effectively integrate expert experience and knowledge. Further, through defuzzification, the fuzzy risk level is transformed into a clear numerical value for subsequent threshold adjustment. Finally, using the provided threshold adjustment formula, the settlement difference threshold is dynamically calculated according to the defuzzified settlement risk level value. This threshold adjustment formula can flexibly adjust the threshold size according to the high or low risk level, realizing the dynamic adaptive adjustment of the settlement difference threshold. This dynamic threshold adjustment mechanism enables the settlement warning system to adjust the sensitivity according to the actual risk situation, avoiding unnecessary frequent warnings while ensuring safety, and improving the practicability and effectiveness of the safety monitoring method. Compared with the manual observation method and the traditional fixed threshold method in the background technology, this solution can more accurately and reasonably evaluate the settlement risk and dynamically adjust the warning threshold, improving the accuracy and reliability of the warning.

[0086] Through the above technical solution, this application realizes the accurate evaluation of the settlement risk level and the dynamic adjustment of the settlement difference threshold, enabling the settlement warning system to adaptively adjust the warning sensitivity according to the actual risk situation in the construction area. Thereby, the intelligent level of the safety monitoring of the non-excavation renovation construction site of the old gas pipeline network is improved, providing strong support for ensuring construction safety.

[0087] Reference Figure 2, this application further proposes a safety monitoring device for the construction site of a gas pipeline network, which is used for settlement monitoring of the construction site of non-excavation renovation of old gas pipeline networks. The device includes: A monitoring data acquisition module 1, which is used to acquire the original tilt angle data collected by multiple monitoring nodes; the multiple monitoring nodes are distributed at intervals on both sides of the ground along the non-excavation construction line of the gas pipeline network, and each monitoring node is equipped with at least two tilt sensors perpendicular to each other in direction to sense the tilt angles of the ground in different directions so as to obtain the original tilt angle data (the specific process refers to step A1 in the previous text); A filtering processing module 2, which is used to perform filtering processing on the original tilt angle data by using a moving average filtering algorithm to obtain the filtered tilt angle data of each monitoring node (the specific process refers to step A2 in the previous text); A differential settlement calculation module 3, which is used to calculate the tilt angle difference between adjacent monitoring nodes according to the filtered tilt angle data of adjacent monitoring nodes to obtain the differential settlement value (the specific process refers to step A3 in the previous text); A real-time warning module 4, which is used to compare the differential settlement value with the dynamic settlement difference threshold to determine the settlement risk and issue a warning message in a timely manner; the dynamic settlement difference threshold is dynamically adjusted based on historical monitoring data, geological conditions and the aging degree of the pipeline network (the specific process refers to step A4 in the previous text).

[0088] Reference Figure 3 , this application further proposes a safety monitoring system for the construction site of a gas pipeline network, which is used for settlement monitoring of the construction site of non-excavation renovation of old gas pipeline networks. The system includes multiple monitoring nodes 100, a mobile monitoring device 200, a cloud server 300 and a monitoring center 400; the monitoring nodes 100 are all communicatively connected to the mobile monitoring device 200, and both the mobile monitoring device 200 and the monitoring center 400 are communicatively connected to the cloud server 300; The multiple monitoring nodes 100 are distributed at intervals on both sides of the ground along the non-excavation construction line of the gas pipeline network, and each monitoring node 100 is equipped with at least two tilt sensors perpendicular to each other in direction to sense the tilt angles of the ground in different directions so as to obtain the original tilt angle data; The mobile monitoring device 200 is used to send the original tilt angle data collected by the multiple monitoring nodes 100 to the cloud server 300; The cloud server 300 is used to execute: Acquire the original tilt angle data collected by the multiple monitoring nodes 100 (the specific process refers to step A1 in the previous text); Perform filtering processing on the original tilt angle data by using a moving average filtering algorithm to obtain the filtered tilt angle data of each monitoring node (the specific process refers to step A2 in the previous text); Calculate the tilt angle difference between adjacent monitoring nodes based on the filtered tilt angle data of adjacent monitoring nodes to obtain the differential settlement value (for the specific process, refer to step A3 in the previous text). Compare the differential settlement value with the dynamic settlement difference threshold to determine the settlement risk and send a warning message to the monitoring center 400 in a timely manner; the dynamic settlement difference threshold is dynamically adjusted based on historical monitoring data, geological conditions, and the degree of pipeline network aging (for the specific process, refer to step A4 in the previous text).

[0089] In this document, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations.

[0090] The above are only examples of the present application and are not used to limit the protection scope of the present application. For those skilled in the art, various changes and modifications can be made to the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A gas pipeline network construction site safety monitoring method, used for monitoring the settlement of the old gas pipeline network trenchless transformation construction site, characterized in that: The steps of the method include: A1. Acquire the original tilt angle data collected by multiple monitoring nodes; the multiple monitoring nodes are spaced apart on the ground on both sides of the trenchless construction line of the gas pipeline network, and each monitoring node is equipped with at least two tilt sensors perpendicular to each other to sense the tilt angle of the ground in different directions to obtain the original tilt angle data; A2. The original tilt angle data is filtered using a sliding average filtering algorithm to obtain the filtered tilt angle data of each monitoring node; A3. According to the filtered tilt angle data of adjacent monitoring nodes, the tilt angle difference between adjacent monitoring nodes is calculated to obtain the differential settlement value; A4. Compare the differential settlement value with the dynamic settlement difference threshold to determine the settlement risk and issue early warning information in a timely manner; the dynamic settlement difference threshold is dynamically adjusted based on historical monitoring data, geological conditions and the degree of aging of the pipeline network.

2. A method for safety monitoring of a gas pipeline network construction site according to claim 1, characterized in that: Step A1 includes: A101 obtains the geographical location information of each monitoring node; A102. Based on the geographical location information of each monitoring node, each monitoring node is calibrated using differential GPS positioning technology to obtain the calibrated geographical location information of each monitoring node; A103. According to the calibrated geographic location information of each monitoring node, the relative position relationship between the monitoring nodes is determined, and based on the relative position relationship, combined with the Kalman filter algorithm and the historical tilt angle data of each tilt sensor, the tilt sensor of each monitoring node is subjected to posture correction to obtain the corrected tilt angle data of each tilt sensor as the original tilt angle data.

3. A gas pipeline network construction site safety monitoring method according to claim 2, characterized in that: Step A103 includes: B1. Calculate the distance between any two monitoring nodes according to the calibrated geographic location information of each monitoring node, construct a topological relationship between the monitoring nodes according to the distance, and obtain a monitoring node topology map; B2. Take each of the monitoring nodes as the target monitoring node in turn and execute: B201. Based on the monitoring node topology, select several neighboring monitoring nodes directly adjacent to the target monitoring node, and obtain the historical tilt angle data of the tilt sensor of each neighboring monitoring node within a preset time window; B202. For the tilt sensor of the target monitoring node, the Kalman filter algorithm is used to fuse the historical tilt angle data of the target monitoring node and each adjacent monitoring node, and the posture correction is performed on the tilt sensor of the target monitoring node to obtain the corrected tilt angle data of the tilt sensor of the target monitoring node as the original tilt angle data of the target monitoring node; during the posture correction process, the noise covariance matrix of the Kalman filter is dynamically adjusted according to the vibration frequency range of the construction equipment to reduce the influence of the vibration of the construction equipment on the posture correction.

4. A method for safety monitoring of a gas pipeline network construction site according to claim 3, characterized in that: Step B202 includes: Obtain the tilt angle data collected by the tilt sensors of the target monitoring node and each adjacent monitoring node within a preset time window, calculate the variance value of the tilt angle data of the target monitoring node and each of the adjacent monitoring nodes, and obtain the variance value of the tilt angle data of the tilt sensors of the target monitoring node and each of the adjacent monitoring nodes; According to the vibration frequency range of the construction equipment and the variance value of each tilt angle data, the weight ratio of the tilt sensors of the target monitoring node and each adjacent monitoring node affected by the vibration of the construction equipment is determined; Based on the weight proportions, dynamically adjust the process noise covariance matrix and the measurement noise covariance matrix of the Kalman filter to reduce the influence of the vibration of the construction equipment on the posture correction, and obtain an adjusted Kalman filter; The adjusted Kalman filter is used to fuse the historical tilt angle data of the target monitoring node and each adjacent monitoring node, and the tilt sensor of the target monitoring node is subjected to posture correction to obtain the corrected tilt angle data of the tilt sensor of the target monitoring node as the original tilt angle data of the target monitoring node.

5. A method for safety monitoring of a gas pipeline network construction site according to claim 1, characterized in that: Step A2 includes: A201. Obtaining the original tilt angle data within a preset time window, calculating the spectrum of the original tilt angle data, and determining the spectrum energy ratio corresponding to the vibration frequency range of the construction equipment; A202. According to the spectrum energy ratio, adjust the sliding window size of the sliding average filtering algorithm; when the spectrum energy ratio exceeds the preset threshold, reduce the sliding window size; when the spectrum energy ratio does not exceed the preset threshold, increase the sliding window size; A203. Use the adjusted sliding average filtering algorithm to filter the original tilt angle data to obtain filtered tilt angle data of each monitoring node.

6. A method for safety monitoring of a gas pipeline network construction site according to claim 1, characterized in that: Step A3 includes: A301. Obtain the filtered tilt angle data of adjacent monitoring nodes within a preset time window, and calculate the mean and standard deviation of the filtered tilt angle data of each adjacent monitoring node; A302. According to the mean and standard deviation of the filtered tilt angle data of each adjacent monitoring node, the confidence interval of the filtered tilt angle data of each adjacent monitoring node is calculated according to the formula: [mean - K*standard deviation, mean + K*standard deviation], where K is the confidence coefficient; A303. Determine whether the filtered tilt angle data of the adjacent monitoring nodes is within the corresponding confidence interval; A304. If there are adjacent monitoring nodes whose filtered tilt angle data are not within the corresponding confidence interval, the filtered tilt angle data that are not within the confidence interval are eliminated, and based on the remaining filtered tilt angle data, the tilt angle difference between the adjacent monitoring nodes is calculated to obtain the differential settlement value; A305. If the filtered tilt angle data of all adjacent monitoring nodes are within the corresponding confidence interval, the tilt angle difference between adjacent monitoring nodes is calculated directly based on the filtered tilt angle data of all adjacent monitoring nodes to obtain the differential settlement value.

7. A method for monitoring safety at a gas pipeline network construction site according to claim 6, characterized in that: Step A304 includes: If there is an adjacent monitoring node whose filtered tilt angle data is not within the corresponding confidence interval, then based on a preset anomaly detection rule base, identify the abnormal type of the filtered tilt angle data that is not within the confidence interval; the abnormal type includes random noise type abnormality, sudden interference type abnormality and sensor failure type abnormality; For random noise type anomalies, the wavelet threshold filtering algorithm based on adaptive threshold is used to smooth the corresponding filtered tilt angle data to obtain the smoothed tilt angle data to replace the original filtered tilt angle data; For sudden interference anomalies, the linear interpolation method based on the adjacent time window data is used to interpolate and repair the corresponding filtered tilt angle data to obtain the interpolated and repaired tilt angle data to replace the original filtered tilt angle data; For sensor failure anomalies, the tilt angle data of the corresponding monitoring node is marked as invalid and removed from the subsequent differential settlement calculation, and the sensor failure warning information is sent to the monitoring center; Based on the processed filtered tilt angle data, the tilt angle differences between adjacent monitoring nodes are calculated to obtain differential settlement values.

8. A method for safety monitoring of a gas pipeline network construction site according to claim 1, characterized in that: Step A4 includes: A401. Obtain historical monitoring data within a preset period and extract a subset of historical settlement data that matches the geological conditions of the current construction area and the degree of aging of the pipeline network; A402. For the historical settlement data subset, a time series analysis method is used to predict the settlement change trend in the future time window to obtain the predicted settlement value; A403. Determine the geological parameters based on the geological survey report of the current construction area, and evaluate the structural strength and remaining life of the pipeline network in combination with the material, service life and repair record of the pipeline network to obtain the pipeline network risk assessment parameters; A404. Based on the predicted settlement value and the pipe network risk assessment parameters, the settlement risk level of the current construction area is assessed using a fuzzy logic reasoning method, and the settlement difference threshold is dynamically adjusted according to the settlement risk level; the higher the risk level, the smaller the settlement difference threshold; A405. Compare the differential settlement value with the dynamic settlement difference threshold. If the differential settlement value exceeds the dynamic settlement difference threshold, it is determined that there is a settlement risk, and an early warning message including the risk location, settlement risk level and recommended treatment measures is generated and sent to the monitoring center.

9. A gas pipeline network construction site safety monitoring device, used for monitoring the settlement of the old gas pipeline network trenchless transformation construction site, characterized in that: The device includes: A monitoring data acquisition module is used to acquire original tilt angle data collected by multiple monitoring nodes; the multiple monitoring nodes are spaced apart on the ground on both sides of the trenchless construction line of the gas pipeline network, and each monitoring node is equipped with at least two tilt sensors perpendicular to each other to sense the tilt angle of the ground in different directions to obtain the original tilt angle data; A filtering processing module, used for filtering the original tilt angle data using a sliding average filtering algorithm to obtain filtered tilt angle data of each monitoring node; A differential settlement calculation module is used to calculate the difference in inclination angles between adjacent monitoring nodes based on filtered inclination angle data of adjacent monitoring nodes to obtain a differential settlement value; The real-time early warning module is used to compare the differential settlement value with the dynamic settlement difference threshold to determine the settlement risk and issue early warning information in a timely manner; the dynamic settlement difference threshold is dynamically adjusted based on historical monitoring data, geological conditions and the degree of aging of the pipeline network.

10. A gas pipeline network construction site safety monitoring system, used for monitoring the settlement of the old gas pipeline network trenchless transformation construction site, characterized in that: The system includes a plurality of monitoring nodes, a mobile monitoring device, a cloud server and a monitoring center; the monitoring nodes are all connected to the mobile monitoring device in communication, and the mobile monitoring device and the monitoring center are both connected to the cloud server in communication; A plurality of the monitoring nodes are distributed at intervals on the ground on both sides of the trenchless construction line of the gas pipeline network, and each of the monitoring nodes is equipped with at least two tilt sensors perpendicular to each other to sense the tilt angle of the ground in different directions to obtain original tilt angle data; The mobile monitoring device is used to send the original tilt angle data collected by the plurality of monitoring nodes to the cloud server; The cloud server is used to execute: Obtaining original tilt angle data collected by multiple monitoring nodes; The original tilt angle data is filtered using a sliding average filtering algorithm to obtain filtered tilt angle data of each monitoring node; According to the filtered tilt angle data of adjacent monitoring nodes, the tilt angle difference between adjacent monitoring nodes is calculated to obtain the differential settlement value; The differential settlement value is compared with the dynamic settlement difference threshold to determine the settlement risk, and an early warning message is sent to the monitoring center in a timely manner; the dynamic settlement difference threshold is dynamically adjusted based on historical monitoring data, geological conditions and the degree of aging of the pipeline network.

Images