Multi-dimensional monitoring and early warning system and method for displacement, axial force and water level in deep foundation pit

Abstract

The invention discloses a multi-dimensional monitoring and early warning system and method for displacement, axial force and water level in a deep foundation pit. The system comprises a data acquisition module, a data transmission module, a data processing module, a risk assessment module, an early warning issuing module and a user interaction module. The data acquisition module is composed of a displacement monitoring sub-module, an axial force monitoring sub-module and a water level monitoring sub-module and is used for respectively acquiring displacement, axial force and water level data in the foundation pit; the data transmission module is responsible for transmitting data to the data processing module; the data processing module cleans, fuses, analyzes and processes the data; the risk assessment module performs risk assessment on the processed data according to a preset standard; the early warning issuing module issues early warning information according to the risk assessment result; the user interaction module is used for displaying information and configuring system parameters. The method has the advantages that real-time, continuous and multi-dimensional monitoring of the construction process of the deep foundation pit is achieved, high-risk points can be accurately recognized, early warning can be conducted in time, and the construction safety of the foundation pit is guaranteed.

Description

Technical Field

[0001] This invention relates to the field of foundation pit engineering monitoring technology, specifically to a multi-dimensional monitoring and early warning system and method for displacement, axial force, and water level in deep foundation pits. Background Technology

[0002] With the advancement of urban construction, underground space has become a future direction for urban development. Rapid urban population growth is driving the rapid development of urban underground transportation networks. Since subway station foundation pit projects are mostly constructed in densely populated urban areas, they inevitably intertwine with urban underground spaces, buildings, subway tunnels, viaducts, and municipal pipelines. The exceptionally complex hydrogeological conditions further increase the difficulty of excavation and support for station foundation pits. Furthermore, foundation pit excavation faces a very high risk of support system instability, which can easily lead to major safety accidents. Therefore, how to effectively identify and manage major risk sources during foundation pit construction to ensure construction safety has become an urgent problem to be solved in current foundation pit engineering.

[0003] Traditional monitoring methods rely heavily on manual measurement, which has many drawbacks, such as inaccurate monitoring data, low monitoring frequency, inability to achieve real-time continuous monitoring, and difficulty in meeting the monitoring needs of deep foundation pit projects in complex environments. With the development of information technology, although some automated detection systems have appeared on the market, most of them have limited functions and data processing capabilities, and there is a lack of a system and method that can perform multi-dimensional and intelligent monitoring and early warning of multiple key indicators such as displacement, axial force, and water level in deep foundation pits. Summary of the Invention

[0004] The purpose of this invention is to provide a multi-dimensional monitoring and early warning system and method for displacement, axial force and water level in deep foundation pits, so as to overcome the shortcomings of the prior art.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows: On the one hand, a multi-dimensional monitoring and early warning system for displacement, axial force, and water level in deep foundation pits is provided, including a data acquisition module, a data transmission module, a data processing module, a risk assessment module, an early warning release module, and a user interaction module; The data acquisition module is connected to the data transmission module and is used to collect monitoring data of displacement, axial force and water level in the deep foundation pit, and transmit the collected monitoring data to the data transmission module. The data transmission module is connected to the data processing module and is used to transmit the monitoring data collected by the data acquisition module to the data processing module. The data processing module is connected to the risk assessment module and is used to clean, fuse, analyze and process the monitoring data transmitted from the data transmission module, and transmit the processed monitoring data to the risk assessment module. The risk assessment module is connected to the early warning release module and is used to conduct risk assessment on the processed monitoring data based on the preset risk level classification standard, obtain the risk assessment result, and transmit the risk assessment result to the early warning release module. The early warning release module is used to release early warning information to relevant personnel through SMS, email, and audible and visual alarms based on the risk assessment results. The user interaction module is connected to the data processing module, risk assessment module, and early warning release module, respectively, and is used to display monitoring data, risk assessment results, and early warning information, and supports users to configure system parameters.

[0006] Furthermore, the data acquisition module includes a displacement monitoring submodule, an axial force monitoring submodule, and a water level monitoring submodule. The displacement monitoring submodule is used to monitor and acquire the horizontal and vertical displacement information of the deep foundation pit retaining structure in real time. The axial force monitoring submodule is used to monitor and acquire the axial force on the deep foundation pit support structure in real time. The water level monitoring submodule is used to detect and acquire the changes in groundwater level inside and outside the deep foundation pit in real time.

[0007] Furthermore, the data transmission module is a 4G or 5G communication module or an NB-IoT narrowband Internet of Things module.

[0008] Furthermore, the specific process by which the data processing module processes the monitoring data is as follows: A1. Data Cleaning: First, check for missing values ​​in the monitoring data. For missing data, fill in the missing values ​​using the mean, median, or model-based prediction methods, depending on the characteristics and distribution of the data. Then, identify and process outliers in the monitoring data, using statistical methods or machine learning-based methods to detect outliers. For detected outliers, delete, correct, or mark them according to their causes. Next, standardize and unify the format of the monitoring data, converting monitoring data of different types and units into a unified format and standard. A2. Data Fusion: First, align the monitoring data of different types, such as displacement, axial force, and water level, according to the time series to ensure the consistency of each data in time; then, establish the correlation between different monitoring data based on the structural characteristics of the deep foundation pit and the location relationship of the monitoring points; finally, use a data fusion algorithm to fuse different types of monitoring data to obtain more accurate and comprehensive monitoring information. A3. Data Analysis and Processing: First, calculate the statistical characteristics of the monitoring data to describe the central tendency and dispersion of the monitoring data; then, process the monitoring data using time series analysis methods to predict the changing trend of the monitoring data; next, based on the mechanical properties of the deep foundation pit and the monitoring data, establish a mathematical model between the displacement, axial force, and water level monitoring data to analyze and predict the stability of the foundation pit.

[0009] Furthermore, the specific process by which the risk assessment module performs a risk assessment on the processed monitoring data is as follows: B1. Data Preprocessing and Feature Extraction: First, extract feature parameters related to risk assessment from the processed monitoring data; then, standardize the extracted feature parameters to eliminate the influence of different dimensions and make the parameters comparable. B2. Risk Indicator System Construction: First, based on the characteristics of deep foundation pit engineering and the needs of risk assessment, establish a risk indicator system that includes multiple aspects such as displacement, axial force, and water level; then, set corresponding threshold ranges for each risk indicator, and divide the risk indicators into different risk level ranges according to engineering experience and relevant specifications. B3. Selection and application of risk level assessment methods: First, the single indicator assessment method is used to assess the risk level of each risk indicator individually and determine its corresponding risk level; then, the comprehensive assessment method is used to consider the interrelationship and comprehensive impact between multiple risk indicators to assess the overall risk level of the deep foundation pit. B4. Determination and Update of Risk Level Assessment Results: First, based on the calculation results of the risk assessment method, determine the current risk level of the deep foundation pit and generate a risk assessment report. The risk assessment report includes the assessment results of each risk indicator, the overall risk level of the deep foundation pit, and an analysis of risk development trends. As the monitoring data is continuously updated, the risk level of the deep foundation pit is reassessed and updated regularly to promptly identify risk change trends and provide a basis for early warning decisions.

[0010] Furthermore, the displacement monitoring submodule includes a total station, an electronic level, and an automated displacement meter. The total station is used to measure the horizontal displacement changes of the deep foundation pit retaining structure on the plane. The electronic level is used to monitor the vertical displacement of the deep foundation pit retaining structure. The automated displacement meter is installed at key parts of the deep foundation pit retaining structure to acquire local displacement data in real time.

[0011] On the other hand, a multi-dimensional monitoring and early warning method for displacement, axial force, and water level in deep foundation pits is provided. It is implemented based on the aforementioned multi-dimensional monitoring and early warning system for displacement, axial force, and water level in deep foundation pits, and specifically includes the following steps: S1. Multi-dimensional data acquisition: The data acquisition module collects displacement, axial force, and water level data in the deep foundation pit and sends them to the data processing module through the data transmission module. S2. Data Processing and Fusion: The data processing module preprocesses and fuses the collected data to construct a multi-source monitoring data model of "displacement-axial force-water level". S3. Risk Assessment: Compare the fused data with preset standards and use the risk assessment model to calculate the risk level of the foundation pit. S4. Warning Issuance: When the risk level reaches or exceeds the warning threshold, the warning issuance module sends warning information to relevant personnel through multiple channels; S5. Feedback and Adjustment: Based on the early warning processing results and the actual situation on site, the system parameters and risk assessment model are adjusted.

[0012] Furthermore, in S2, a multi-source monitoring data model of "displacement-axial force-water level" is constructed. The specific steps are as follows: S201. The collected raw data is preprocessed, outliers are filtered by physical threshold and 3σ criterion, missing values ​​are filled by Kalman filtering and bidirectional LSTM, and the Z-score is used to standardize the data and synchronize it to a time series with a 15-minute interval. S202. Implement a three-level fusion strategy with spatiotemporal alignment, map the coordinates of monitoring points to the three-dimensional coordinate system of the foundation pit, use a sliding window to match data of different frequencies, estimate the system state through Kalman filtering, and fuse spatiotemporal correlations through Bayesian networks. S203. Construct a hybrid model that combines physics and data-driven approaches. Based on finite element software, establish a geotechnical mechanics model to reflect parameter coupling relationships. Use bidirectional LSTM and attention mechanism network to capture temporal dependencies, and use the output of the former as the prior knowledge of the latter. S204. The model's generalization ability is evaluated through k-fold cross-validation. Parameters are optimized based on MSE and MAE metrics, and an online learning mechanism is established to achieve dynamic updates. MSE (mean squared error) and MAE (mean absolute error) are metrics that measure the error between the model's predictions and the actual values. MSE is used to calculate the mean squared error, amplifying the weight of larger errors and being sensitive to outliers. MAE is used to calculate the mean absolute error, exhibiting strong robustness to outliers and good interpretability. The smaller the values ​​of both, the better the model performance. S205. Integrate the model into the BIM 3D visualization platform to intuitively display the parameter change status with color gradients. At the same time, dynamically adjust the risk warning threshold based on the prediction results to form a complete closed loop from data processing to risk warning.

[0013] Furthermore, in S3, the risk level of the foundation pit is calculated using a risk assessment model. The specific steps are as follows: S301. Calculate the comprehensive risk index of the foundation pit based on the risk assessment model. Among them, comprehensive risk indicators The specific calculation formula is as follows: ; in, The weight representing the displacement rate, The weight representing the rate of change of axial force, and ; Indicates displacement rate, Indicates the rate of change of axial force; The warning threshold representing the displacement rate; The warning threshold representing the rate of change of axial force; S302, Calculate the comprehensive risk index of the foundation pit The risk level of the foundation pit is determined by comparing it with the preset comprehensive risk index threshold range.

[0014] Compared with the prior art, the advantages of the present invention are as follows: it can realize real-time, continuous and multi-dimensional monitoring of deep foundation pit construction, accurately identify high-risk points and provide timely warnings, and ensure construction safety. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in this embodiment, the accompanying drawings used in the description of the embodiment will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This invention relates to a structural block diagram of a multi-dimensional monitoring and early warning system for displacement, axial force, and water level in deep foundation pits. Figure 2 This invention relates to a process flow diagram of a multi-dimensional monitoring and early warning method for displacement, axial force, and water level in deep foundation pits. Detailed Implementation

[0017] To make the technical means, creative features, objectives and effects of this invention easier to understand, the following specific embodiments further illustrate how this invention is implemented.

[0018] Example 1: See Figure 1 This paper presents a multi-dimensional monitoring and early warning system for displacement, axial force, and water level in deep foundation pits. The system includes a data acquisition module, a data transmission module, a data processing module, a risk assessment module, an early warning release module, and a user interaction module. The data acquisition module is connected to the data transmission module and is used to collect monitoring data on displacement, axial force and water level in the deep foundation pit, and transmit the collected monitoring data to the data transmission module. The data transmission module is connected to the data processing module and is used to transmit the monitoring data collected by the data acquisition module to the data processing module. The data processing module is connected to the risk assessment module and is used to clean, integrate, analyze and process the monitoring data transmitted from the data transmission module, and transmit the processed monitoring data to the risk assessment module. The risk assessment module is connected to the early warning release module. It is used to conduct risk assessment on the processed monitoring data based on the preset risk level classification standards, obtain the risk assessment results, and transmit the risk assessment results to the early warning release module. The early warning release module is used to release early warning information to relevant personnel through SMS, email, and audible and visual alarms based on the risk assessment results; The user interaction module is connected to the data processing module, risk assessment module, and early warning release module, respectively. It is used to display monitoring data, risk assessment results, and early warning information, and supports users in configuring system parameters.

[0019] Specifically, in Embodiment 1 of the present invention, the data acquisition module includes a displacement monitoring submodule, an axial force monitoring submodule, and a water level monitoring submodule. The displacement monitoring submodule is used to monitor and acquire the horizontal and vertical displacement information of the deep foundation pit retaining structure in real time. The axial force monitoring submodule is used to monitor and acquire the axial force on the deep foundation pit support structure in real time. The water level monitoring submodule is used to detect and acquire the changes in groundwater level inside and outside the deep foundation pit in real time.

[0020] More specifically, in Embodiment 1 of the present invention, the displacement monitoring submodule includes a total station, an electronic level, and an automated displacement meter. The total station is used to measure the horizontal displacement change of the deep foundation pit retaining structure on the plane, the electronic level is used to monitor the vertical displacement of the deep foundation pit retaining structure, and the automated displacement meter is installed at key parts of the deep foundation pit retaining structure to acquire local displacement data in real time.

[0021] In the construction monitoring system for deep foundation pit engineering, there are several key areas (i.e., key parts) that require close monitoring, including but not limited to: the external corners (such as corners of the foundation pit), internal corners, and special structural parts such as the connection between the support and retaining structure (such as the connection nodes between steel supports and underground continuous walls or piles); areas where buildings or underground pipelines (such as water supply and drainage, gas pipelines) are distributed around the foundation pit; areas where the excavation depth changes significantly (such as the retaining structure at the junction of deep and shallow foundation pits, elevator shafts, sump pits, etc., where local deepening occurs); areas with complex geological conditions and high uncertainty (such as retaining structures at soft soil layers, fault zones, or rock-soil interfaces, and retaining structures above confined aquifers); and parts that have shown abnormal signs such as deformation and cracking (such as those found in previous monitoring). In areas where the displacement rate is high or has exceeded the warning value, and where the retaining structure shows signs of cracks, leakage, or other defects, the following areas are particularly vulnerable: External corners (such as the corners of foundation pits) are prone to stress concentration and have a high risk of horizontal displacement; Internal corners may experience sudden local stress changes due to inconsistent soil deformation on both sides; Displacement changes at the junctions between the support and the retaining structure (such as the connection between steel supports and diaphragm walls or piles) directly reflect the stability of the support system; Areas with significant changes in excavation depth (such as the retaining structure at the junction of deep and shallow foundation pits, elevator shafts, and locally deepened areas like sump pits) are prone to uneven settlement; Areas with complex geological conditions and high uncertainty (such as retaining structures at soft soil strata, fault zones, or soil-rock interfaces, or retaining structures above confined aquifers) are prone to large displacements. In summary, due to the risks of displacement instability, sudden stress concentration changes, and uneven settlement in these areas, it is essential to strengthen the frequency and accuracy of monitoring. Real-time data collection and analysis, along with timely warnings of potential risks, can provide a solid guarantee for the safe construction and smooth progress of deep foundation pit projects.

[0022] Specifically, in Embodiment 1 of the present invention, the data transmission module can be a 4G communication module, a 5G communication module, or an NB-IoT narrowband Internet of Things module.

[0023] Specifically, in Embodiment 1 of this invention, the data processing module for deep foundation pit monitoring data processing mainly consists of three stages: data cleaning, data fusion, and data analysis. In the data cleaning stage, missing values ​​in the dataset are first identified, and then, based on the data distribution characteristics, methods such as mean imputation, median imputation, or model prediction are flexibly selected to complete the data. Then, outliers are accurately identified and properly handled, followed by standardization of various data formats. In the data fusion process, time synchronization of multi-source data is first achieved, then correlation analysis is used to uncover relationships between data, and finally, algorithms such as weighted averages are used to integrate data from different channels. In the data analysis stage, the focus is on calculating the statistical characteristics of the data, conducting trend analysis, and constructing mathematical models to ensure the accuracy and reliability of deep foundation pit monitoring data, providing strong data support for deep foundation pit stability analysis and prediction.

[0024] More specifically, in Embodiment 1 of the present invention, the specific process by which the data processing module processes the deep foundation pit monitoring data is as follows: A1. Data Cleaning First, check for missing values ​​in the monitoring data. For missing data, fill in the missing values ​​using the mean, median, or model-based prediction methods, depending on the characteristics and distribution of the data. For example, for missing values ​​in displacement monitoring data, if the data approximately follows a normal distribution, the mean can be used for filling; if the data distribution is uneven, the median can be used. Then, outliers in the monitoring data are identified and processed using statistical methods (such as the 3σ criterion) or machine learning-based methods (such as the isolated forest algorithm) to detect outliers. For detected outliers, they are deleted, corrected, or specially marked according to their causes. For example, if a certain axial force monitoring data is significantly outside the normal range and it is found to be an outlier caused by a sensor malfunction, it can be deleted or corrected to a reasonable value. Next, the monitoring data is formatted and standardized, converting monitoring data of different types and units into a unified format and standard; for example, the unit of water level monitoring data is unified to meters, and the unit of displacement monitoring data is unified to millimeters, etc. A2, Data Fusion First, the monitoring data of different types, such as displacement, axial force, and water level, are aligned according to time series to ensure the consistency of each data in time; for example, the displacement data, axial force data, and water level data collected once per hour are synchronized in time for subsequent analysis. Then, based on the structural characteristics of the deep foundation pit and the location relationship of the monitoring points, establish the correlation between different monitoring data; for example, analyze the relationship between the displacement change of a certain part of the foundation pit and the axial force change of the supporting structure of that part, as well as the influence of water level changes on displacement and axial force, etc. Next, data fusion algorithms (such as weighted average method, Kalman filter method, etc.) are used to fuse different types of monitoring data to obtain more accurate and comprehensive monitoring information. For example, displacement, axial force and water level data are fused by weighted average method, and different weights are assigned according to the importance and reliability of each data to obtain comprehensive monitoring indicators. A3. Data Analysis and Processing First, calculate the statistical characteristics of the monitoring data, such as mean, variance, and standard deviation, to describe the central tendency and dispersion of the monitoring data; for example, calculate the mean and standard deviation of displacement monitoring data to assess the range of variation and stability of the foundation pit displacement. Then, the monitoring data is processed using time series analysis methods (such as moving average method, exponential smoothing method, etc.) to predict the changing trend of the monitoring data; for example, predicting the changing trend of the foundation pit water level and discovering possible anomalies in advance. Next, based on the mechanical properties and monitoring data of the deep foundation pit, a mathematical model is established between the displacement, axial force, and water level monitoring data to analyze and predict the stability of the foundation pit. For example, a foundation pit stability model based on finite element analysis is established, and the model is corrected and verified by combining monitoring data to improve the accuracy and reliability of the model.

[0025] Specifically, in Embodiment 1 of this invention, the risk assessment module mainly focuses on preprocessing, indicator construction, assessment methods, and result updates when conducting risk assessments on the processed monitoring data. That is, firstly, characteristic parameters such as the displacement rate and axial force change rate of the deep foundation pit are extracted and standardized; then, a risk indicator system is constructed based on engineering requirements and thresholds are set; next, a single or comprehensive assessment method is used to determine the risk level; finally, a report is generated based on the calculation results, and the data is periodically reassessed as the monitoring data is updated to provide support for early warning decisions.

[0026] More specifically, in Embodiment 1 of the present invention, the specific process by which the risk assessment module performs a risk assessment on the processed monitoring data is as follows: B1. Data Preprocessing and Feature Extraction First, extract characteristic parameters related to risk assessment from the processed monitoring data, such as displacement rate, axial force change rate, and water level fluctuation amplitude. For example, calculate the change in displacement monitoring data per unit time to obtain the displacement rate; calculate the change rate of axial force monitoring data to reflect the change in the stress state of the supporting structure. Then, the extracted feature parameters are standardized to eliminate the influence of different dimensions and make the parameters comparable. For example, the Z-score standardization method is used to convert feature parameters such as displacement rate and axial force change rate into standard normal distribution data with a mean of 0 and a standard deviation of 1. B2. Construction of Risk Indicator System First, based on the characteristics of deep foundation pit engineering and the needs of risk assessment, a risk indicator system including multiple aspects such as displacement, axial force, and water level should be established. For example, risk indicators for displacement may include cumulative displacement, displacement rate, and displacement gradient; risk indicators for axial force may include axial force change rate and axial force exceeding limit; and risk indicators for water level may include water level change range and distance between water level and bottom of foundation pit. Then, a corresponding threshold range is set for each risk indicator. Based on engineering experience and relevant specifications, the risk indicators are divided into different risk level ranges. For example, for the displacement rate indicator, it is set that when the displacement rate is less than 0.5 mm / d, it is a low risk level; when the displacement rate is between 0.5 mm / d and 1.0 mm / d, it is a medium risk level; and when the displacement rate is greater than 1.0 mm / d, it is a high risk level. B3. Selection and Application of Risk Level Assessment Methods First, a single-indicator assessment method is used to assess the risk level of each risk indicator individually and determine its corresponding risk level. For example, for the displacement rate indicator of a certain monitoring point, its value is used to determine whether it belongs to a low-risk, medium-risk, or high-risk level. Then, a comprehensive assessment method is adopted to evaluate the overall risk level of the deep foundation pit by considering the interrelationships and comprehensive impacts among multiple risk indicators. For example, the Analytic Hierarchy Process (AHP) can be used to determine the weight of each risk indicator, and then the comprehensive risk score can be calculated by weighted summation. Finally, the overall risk level of the deep foundation pit is determined based on the comprehensive risk score. B4. Determination and Update of Risk Level Assessment Results First, based on the calculation results of the risk assessment method, the current risk level of the deep foundation pit is determined, and a risk assessment report is generated. The risk assessment report includes the assessment results of each risk indicator, the overall risk level of the deep foundation pit, and an analysis of the risk development trend. Then, as the monitoring data is continuously updated, the risk level of the deep foundation pit is reassessed and updated regularly to promptly identify risk change trends and provide a basis for early warning decisions. For example, the risk level of the deep foundation pit is reassessed once a day. When an abnormal trend of a certain risk indicator is found, the assessment frequency is increased and the risk changes are closely monitored.

[0027] Specifically, in Embodiment 1 of this invention, the early warning release module relies on a multi-dimensional collaborative mechanism to release early warning information through multi-mechanism early warning triggering, early warning grading, early warning classification, multi-channel early warning release, and closed-loop processing. That is, when monitored data indicators or trends reach a pre-set threshold, the early warning mechanism is immediately activated; then, based on the severity of the risk, four levels of early warning—blue, yellow, orange, and red—are established, and further subdivided according to the monitoring indicator category; next, early warning information is quickly pushed to the recipient through diverse channels such as SMS and email; after confirmation by the recipient, each responsible entity responds according to the early warning level, and the response results are promptly fed back, forming a closed-loop process of "early warning—response—feedback". Furthermore, the system records the entire early warning process log in real time, mines potential patterns through data analysis and statistics, supports historical early warning information retrospective query, and provides strong support for the optimization and upgrading of the early warning system and the improvement of management processes.

[0028] More specifically, in Embodiment 1 of this invention, the multi-mechanism early warning triggering includes threshold triggering and trend analysis triggering. Threshold triggering refers to automatically triggering an early warning process when a certain monitoring indicator or comprehensive risk level reaches or exceeds a preset early warning threshold, based on the risk level results output by the risk assessment module. For example, when the displacement rate exceeds 1.0 mm / d or the comprehensive risk level is assessed as "high risk," the system immediately initiates the early warning procedure. Trend analysis triggering refers to the system issuing early warnings based on the changing trends of monitoring data, in addition to threshold triggering. For example, when the displacement amount does not reach the early warning threshold but shows an accelerating growth trend for three consecutive days, the system will issue an "abnormal trend" warning to remind management personnel to pay attention to potential risks.

[0029] More specifically, in Embodiment 1 of the present invention, the early warning classification refers to mapping the risk level in the risk assessment result to the corresponding early warning level, such as: low risk: blue warning (advice level); medium risk: yellow warning (warning level); high risk: orange warning (serious level); extremely high risk: red warning (emergency level).

[0030] More specifically, in Embodiment 1 of the present invention, the early warning classification refers to classifying early warnings into different types such as displacement early warning, axial force early warning, water level early warning, and comprehensive early warning based on the type of monitoring indicator triggered by the early warning, so that managers can quickly locate the risk source.

[0031] More specifically, in Embodiment 1 of this invention, multi-channel early warning dissemination includes SMS warnings, email warnings, audible and visual alarms, and APP push notifications. SMS warnings refer to sending warning SMS messages to pre-set management personnel's mobile phone numbers via an SMS platform to ensure timely notification in emergencies. Email warnings refer to sending detailed warning reports to management personnel's email addresses via an internet platform, including monitoring data charts, trend analysis, risk assessment results, and other information to facilitate in-depth analysis by management personnel. Audible and visual alarms refer to setting up audible and visual alarms at the monitoring center or on-site; when a high-level warning (such as an orange or red warning) is triggered, a piercing alarm sound and flashing lights are emitted to remind on-site personnel to take immediate action. APP push notifications refer to simultaneously pushing warning information to management personnel's mobile APP, supporting real-time viewing of detailed data and historical records.

[0032] More specifically, in Embodiment 1 of this invention, closed-loop processing refers to the process after an early warning is issued. First, it tracks whether management personnel have viewed or confirmed the warning information, for example, through SMS receipts, email reading status, or APP operation records. Then, based on different warning levels, corresponding response procedures and responsible personnel are set. For example, for a blue warning: the on-duty personnel view and record the information without special handling; for a yellow warning: the on-site manager must view the data and submit preliminary handling opinions within 1 hour; for an orange warning: the project manager must organize a meeting within 30 minutes to formulate countermeasures; for a red warning: the emergency plan must be activated immediately, personnel must be evacuated, and relevant departments must be notified. Finally, after taking countermeasures, each responsible person enters the processing results into the system, forming a closed loop for early warning processing.

[0033] Example 2: See Figure 2 This paper provides a multi-dimensional monitoring and early warning method for displacement, axial force, and water level in deep foundation pits. It is based on the multi-dimensional monitoring and early warning system for displacement, axial force, and water level in deep foundation pits described in Example 1 above, and specifically includes the following steps: S1. Multi-dimensional data acquisition: The data acquisition module collects displacement, axial force, and water level data in the deep foundation pit and sends them to the data processing module through the data transmission module. S2. Data Processing and Fusion: The data processing module preprocesses and fuses the collected data to construct a multi-source monitoring data model of "displacement-axial force-water level". S3. Risk Assessment: Compare the fused data with preset standards and use the risk assessment model to calculate the risk level of the foundation pit. S4. Warning Issuance: When the risk level reaches or exceeds the warning threshold, the warning issuance module sends warning information to relevant personnel through multiple channels; S5. Feedback and Adjustment: Based on the early warning processing results and the actual situation on site, the system parameters and risk assessment model are adjusted.

[0034] Specifically, in step S2 of embodiment 2 of the present invention, a multi-source monitoring data model of "displacement-axial force-water level" is constructed, and the specific steps are as follows: S201. The collected raw data is preprocessed, outliers are filtered by physical threshold and 3σ criterion, missing values ​​are filled by Kalman filtering and bidirectional LSTM, and the Z-score is used to standardize the data and synchronize it to a time series with a 15-minute interval. S202. Implement a three-level fusion strategy with spatiotemporal alignment, map the coordinates of monitoring points to the three-dimensional coordinate system of the foundation pit, use a sliding window to match data of different frequencies, estimate the system state through Kalman filtering, and fuse spatiotemporal correlations through Bayesian networks. S203. Construct a hybrid model that combines physics and data-driven approaches. Based on finite element software, establish a geotechnical mechanics model to reflect parameter coupling relationships. Use bidirectional LSTM and attention mechanism network to capture temporal dependencies, and use the output of the former as the prior knowledge of the latter. S204. The model's generalization ability is evaluated through k-fold cross-validation. Parameters are optimized based on MSE and MAE metrics, and an online learning mechanism is established for dynamic updates. MSE (mean squared error) and MAE (mean absolute error) are metrics that measure the error between the model's predictions and the actual values. MSE is used to calculate the mean squared error, amplifying the weight of larger errors and being sensitive to outliers. MAE is used to calculate the mean absolute error, exhibiting strong robustness to outliers and good interpretability. The smaller the values ​​of both MSE and MAE, the better the model performance. S205. Integrate the model into the BIM 3D visualization platform to intuitively display the parameter change status with color gradients. At the same time, dynamically adjust the risk warning threshold based on the prediction results to form a complete closed loop from data processing to risk warning.

[0035] Specifically, in step S3 of embodiment 2 of the present invention, the risk level of the foundation pit is calculated using a risk assessment model, and the specific steps are as follows: S301. Calculate the comprehensive risk index of the foundation pit based on the risk assessment model. Among them, comprehensive risk indicators The specific calculation formula is as follows: ; in, The weight representing the displacement rate, The weight representing the rate of change of axial force, and ; Indicates displacement rate, Indicates the rate of change of axial force; The warning threshold representing the displacement rate; The warning threshold representing the rate of change of axial force; S302, Calculate the comprehensive risk index of the foundation pit The risk level of the foundation pit is determined by comparing it with the preset comprehensive risk index threshold range.

[0036] In practical applications, we can set different comprehensive risk indicator threshold ranges to correspond to different risk levels according to actual needs; for example: when When <0.6, it is considered a low-risk level; when 0.6 ≤ When <0.8, it is classified as a medium-risk level; when 0.8 ≤ A value less than 1.0 indicates a high-risk level. A value of ≥1.0 indicates an extremely high risk level.

[0037] The above formula allows us to calculate a comprehensive risk index based on monitored data such as displacement rate and axial force change rate. By comparing this calculated comprehensive risk index with a pre-defined comprehensive risk threshold range, we can determine the risk level of the foundation pit. Of course, actual risk assessment models may consider more monitoring indicators and complex factors, requiring further optimization and adjustments based on specific engineering conditions and requirements.

[0038] Finally, it should be noted that the above description is only an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A multi-dimensional monitoring and early warning system for displacement, axial force, and water level in deep foundation pits, characterized in that, It includes a data acquisition module, a data transmission module, a data processing module, a risk assessment module, an early warning release module, and a user interaction module; The data acquisition module is connected to the data transmission module and is used to collect monitoring data of displacement, axial force and water level in the deep foundation pit, and transmit the collected monitoring data to the data transmission module. The data transmission module is connected to the data processing module and is used to transmit the monitoring data collected by the data acquisition module to the data processing module. The data processing module is connected to the risk assessment module and is used to clean, fuse, analyze and process the monitoring data transmitted from the data transmission module, and transmit the processed monitoring data to the risk assessment module. The risk assessment module is connected to the early warning release module and is used to conduct risk assessment on the processed monitoring data based on the preset risk level classification standard, obtain the risk assessment result, and transmit the risk assessment result to the early warning release module. The early warning release module is used to release early warning information to relevant personnel through SMS, email, and audible and visual alarms based on the risk assessment results. The user interaction module is connected to the data processing module, risk assessment module, and early warning release module, respectively, and is used to display monitoring data, risk assessment results, and early warning information, and supports users to configure system parameters.

2. The multi-dimensional monitoring and early warning system for displacement, axial force, and water level in deep foundation pits according to claim 1, characterized in that, The data acquisition module includes a displacement monitoring submodule, an axial force monitoring submodule, and a water level monitoring submodule. The displacement monitoring submodule is used to monitor and acquire the horizontal and vertical displacement information of the deep foundation pit retaining structure in real time. The axial force monitoring submodule is used to monitor and acquire the axial force on the deep foundation pit support structure in real time. The water level monitoring submodule is used to detect and acquire the changes in groundwater level inside and outside the deep foundation pit in real time.

3. The multi-dimensional monitoring and early warning system for displacement, axial force, and water level in deep foundation pits according to claim 1, characterized in that, The data transmission module is a 4G or 5G communication module or an NB-IoT narrowband Internet of Things module.

4. The multi-dimensional monitoring and early warning system for displacement, axial force, and water level in deep foundation pits according to claim 1, characterized in that, The specific process by which the data processing module processes the monitoring data is as follows: A1. Data Cleaning: First, check for missing values ​​in the monitoring data. For missing data, fill in the missing values ​​using the mean, median, or model-based prediction methods, depending on the characteristics and distribution of the data. Then, identify and process outliers in the monitoring data, using statistical methods or machine learning-based methods to detect outliers. For detected outliers, delete, correct, or mark them according to their causes. Next, standardize and unify the format of the monitoring data, converting monitoring data of different types and units into a unified format and standard. A2. Data Fusion: First, align the monitoring data of different types, such as displacement, axial force, and water level, according to the time series to ensure the consistency of each data in time; then, establish the correlation between different monitoring data based on the structural characteristics of the deep foundation pit and the location relationship of the monitoring points; finally, use a data fusion algorithm to fuse different types of monitoring data to obtain more accurate and comprehensive monitoring information. A3. Data Analysis and Processing: First, calculate the statistical characteristics of the monitoring data to describe the central tendency and dispersion of the monitoring data; Then, the monitoring data is processed using time series analysis to predict the trend of the monitoring data. Next, based on the mechanical properties of the deep foundation pit and the monitoring data, a mathematical model is established between the displacement, axial force, and water level monitoring data to analyze and predict the stability of the foundation pit.

5. The multi-dimensional monitoring and early warning system for displacement, axial force, and water level in deep foundation pits according to claim 1, characterized in that, The specific process by which the risk assessment module performs a risk assessment on the processed monitoring data is as follows: B1. Data Preprocessing and Feature Extraction: First, extract feature parameters related to risk assessment from the processed monitoring data; then, standardize the extracted feature parameters to eliminate the influence of different dimensions and make the parameters comparable. B2. Risk Indicator System Construction: First, based on the characteristics of deep foundation pit engineering and the needs of risk assessment, establish a risk indicator system that includes multiple aspects such as displacement, axial force, and water level; then, set corresponding threshold ranges for each risk indicator, and divide the risk indicators into different risk level ranges according to engineering experience and relevant specifications. B3. Selection and application of risk level assessment methods: First, the single indicator assessment method is adopted to assess the risk level of each risk indicator separately and determine its corresponding risk level. Then, a comprehensive assessment method is adopted to evaluate the overall risk level of the deep foundation pit by considering the interrelationship and comprehensive impact among multiple risk indicators. B4. Determination and Update of Risk Level Assessment Results: First, based on the calculation results of the risk assessment method, determine the current risk level of the deep foundation pit and generate a risk assessment report. The risk assessment report includes the assessment results of each risk indicator, the overall risk level of the deep foundation pit, and an analysis of the risk development trend. With the continuous updating of monitoring data, the risk level of deep foundation pits is regularly reassessed and updated to promptly identify risk trends and provide a basis for early warning decisions.

6. The multi-dimensional monitoring and early warning system for displacement, axial force, and water level in deep foundation pits according to claim 2, characterized in that, The displacement monitoring submodule includes a total station, an electronic level, and an automated displacement meter. The total station is used to measure the horizontal displacement of the deep foundation pit retaining structure on the plane. The electronic level is used to monitor the vertical displacement of the deep foundation pit retaining structure. The automated displacement meter is installed at key parts of the deep foundation pit retaining structure to acquire local displacement data in real time.

7. A multi-dimensional monitoring and early warning method for displacement, axial force, and water level in deep foundation pits, characterized in that, Includes the following steps: S1. Multi-dimensional data acquisition: The data acquisition module collects displacement, axial force, and water level data in the deep foundation pit and sends them to the data processing module through the data transmission module. S2. Data Processing and Fusion: The data processing module preprocesses and fuses the collected data to construct a multi-source monitoring data model of "displacement-axial force-water level"; S3. Risk Assessment: Compare the fused data with the preset standards and use the risk assessment model to calculate the risk level of the foundation pit. S4. Warning Issuance: When the risk level reaches or exceeds the warning threshold, the warning issuance module sends warning information to relevant personnel through multiple channels; S5. Feedback and Adjustment: Based on the early warning processing results and the actual situation on site, the system parameters and risk assessment model are adjusted.

8. The multi-dimensional monitoring and early warning method for displacement, axial force, and water level in deep foundation pits according to claim 7, characterized in that, In S2, a multi-source monitoring data model of "displacement-axial force-water level" is constructed. The specific steps are as follows: S201. The collected raw data is preprocessed, outliers are filtered by physical threshold and 3σ criterion, missing values ​​are filled by Kalman filtering and bidirectional LSTM, and the Z-score is used to standardize the data and synchronize it to a time series with a 15-minute interval. S202. Implement a three-level fusion strategy with spatiotemporal alignment, map the coordinates of monitoring points to the three-dimensional coordinate system of the foundation pit, use a sliding window to match data of different frequencies, estimate the system state through Kalman filtering, and fuse spatiotemporal correlations through Bayesian networks. S203. Construct a hybrid model that combines physics and data-driven approaches. Based on finite element software, establish a geotechnical mechanics model to reflect parameter coupling relationships. Use bidirectional LSTM and attention mechanism networks to capture temporal dependencies, and use the output of the former as the prior knowledge of the latter. S204. The generalization ability of the model is evaluated through k-fold cross-validation, the parameters are optimized based on the MSE and MAE indicators, and an online learning mechanism is established to achieve dynamic updates. S205. Integrate the model into the BIM 3D visualization platform to intuitively display the parameter change status with color gradients. At the same time, dynamically adjust the risk warning threshold based on the prediction results to form a complete closed loop from data processing to risk warning.

9. The multi-dimensional monitoring and early warning method for displacement, axial force, and water level in deep foundation pits according to claim 7, characterized in that, In S3, the risk level of the foundation pit is calculated using a risk assessment model. The specific steps are as follows: S301. Calculate the comprehensive risk index of the foundation pit based on the risk assessment model. Among them, comprehensive risk indicators The specific calculation formula is as follows: ； in, The weight representing the displacement rate, The weight representing the rate of change of axial force, and ; Indicates displacement rate, Indicates the rate of change of axial force; The warning threshold representing the displacement rate; The warning threshold representing the rate of change of axial force; S302, Calculate the comprehensive risk index of the foundation pit The risk level of the foundation pit is determined by comparing it with the preset comprehensive risk index threshold range.

Images