Intelligent monitoring and early warning method and system for deep foundation pit

Abstract

The invention is suitable for the technical field of foundation pit monitoring, and relates to an intelligent monitoring and early warning method and system for a deep foundation pit, and the method comprises the steps: carrying out the real-time dynamic correction of a pre-constructed foundation pit construction whole process refined three-dimensional numerical model through the precise application of field monitoring data, and generating a highly accurate corrected numerical model; a data prediction result obtained based on the correction model not only contains various possible changes in the foundation pit construction process, but also accurately reflects the influence of construction conditions on the stability of the foundation pit, so that continuous and efficient monitoring and early warning of the state of the foundation pit can be realized, and tight combination and efficient cooperation of site construction and foundation pit monitoring are ensured; therefore, the intelligent monitoring and early warning method and system for the deep foundation pit provide reliable technical support and decision basis for safety assessment of the foundation pit, and the safety of foundation pit construction is greatly improved.

Description

Technical Field

[0001] The present application relates to the technical field of construction engineering foundation pit monitoring, and in particular to a deep foundation pit intelligent monitoring and early warning method and system. Background Art

[0002] As the process of urban construction in my country becomes more and more complete, the development of underground space has shown a trend of deepening, expanding and three-dimensionalization, and the role of foundation pit engineering in urban infrastructure construction has become increasingly important. During the excavation of the foundation pit, its stability, the safety of the support structure, and the impact on surrounding buildings and underground pipelines are all projects that need to be focused on. Therefore, it is necessary to strengthen the monitoring and measurement during the construction process, and promptly feedback a series of dynamic change information caused by the construction to the construction unit, so that it can adjust the construction parameters on site in a timely manner to avoid accidents that affect operational safety.

[0003] However, at present, foundation pit monitoring still mainly relies on manual inspection and measurement. This monitoring method is inefficient, greatly affected by the natural environment, and cannot guarantee monitoring at the required frequency and time. The monitoring cycle is unstable and there are great risks to personal safety. In addition, only a few deep foundation pit projects have realized automatic data collection and transmission. Some research institutions have developed monitoring data processing systems to classify and manage data and quickly query them. For example, the patent number CN116446473A provides an automatic construction monitoring and alarm system for super-large foundation pits. The system includes a construction monitoring center, a video monitoring module, and a data monitoring module. The safety of foundation pit construction is ensured through automatic monitoring and automatic alarm systems; the patent number CN117493815B provides a deep foundation pit monitoring method and system based on BIM. The stress condition of the deep foundation pit support structure is obtained by monitoring the support system and calculating the soil pressure theory, and the deep foundation pit support structure is accurately monitored. However, these monitoring systems are more of an analysis or prediction from the statistical perspective of monitoring data, and it is difficult to effectively and closely combine on-site construction with foundation pit monitoring to determine whether the foundation pit is safe. Summary of the invention

[0004] Based on this, it is necessary to provide a deep foundation pit intelligent monitoring and early warning method and system to address the problem that the existing foundation pit monitoring system is difficult to effectively and closely integrate on-site construction with foundation pit monitoring to determine whether the foundation pit is safe.

[0005] In a first aspect, the present application provides a deep foundation pit intelligent monitoring and early warning method. The method comprises:

[0006] Step S1, constructing a refined three-dimensional numerical model of the entire foundation pit construction process;

[0007] Step S2, obtaining on-site monitoring data;

[0008] Step S3, correcting the refined three-dimensional numerical model of the entire foundation pit construction process according to the on-site monitoring data to obtain a corrected numerical model;

[0009] Step S4, obtaining data prediction results according to the modified numerical model;

[0010] Step S5: monitor and warn the foundation pit based on the data prediction results.

[0011] In one embodiment, the deep foundation pit intelligent monitoring and early warning method further includes:

[0012] Step S61, constructing a deep foundation pit deformation prediction model based on LSTM;

[0013] Step S62, obtaining measured deformation data that changes with time during foundation pit construction;

[0014] Step S63, preprocessing the measured deformation data to obtain sample time series data;

[0015] Step S64, training the deep foundation pit deformation prediction model using the sample time series data to obtain a trained deep foundation pit deformation prediction model;

[0016] Step S65, predicting the recent deformation of the foundation pit according to the trained deep foundation pit deformation prediction model to obtain the recent development trend data of the foundation pit;

[0017] Step S66, monitoring and early warning of the foundation pit based on the recent development trend data of the foundation pit and the data prediction results.

[0018] In one embodiment, step S66 includes:

[0019] Step S660, obtaining the foundation pit parameter monitoring threshold, monitoring point alarm level and patrol inspection warning;

[0020] Step S661, generating a foundation pit early warning forecast based on the foundation pit recent development trend data, data prediction results, on-site monitoring data, foundation pit parameter monitoring thresholds, monitoring point alarm levels and patrol inspection warnings.

[0021] In one of the embodiments, the method for correcting the refined three-dimensional numerical model of the entire foundation pit construction process includes at least one of inversion of rock and soil mechanical parameters, correction of soil-structure interface contact parameters, correction of model boundary dimensions, and correction of soil and surrounding structure constitutive models and parameters.

[0022] In one embodiment, after step S4 and before step S5, the method further includes:

[0023] Step S41, comparing the data prediction results with the actual foundation pit monitoring data to verify the accuracy of the refined three-dimensional numerical model of the entire foundation pit construction process.

[0024] In the second aspect, the present application also provides a deep foundation pit intelligent monitoring and early warning system. The system includes a data acquisition and wireless remote transmission module, a data prediction module and a monitoring and early warning module, wherein the data acquisition and wireless remote transmission module is connected to the data prediction module, and the data prediction module is connected to the monitoring and early warning module;

[0025] The data acquisition and wireless remote transmission module is used to obtain the on-site monitoring data and send the on-site monitoring data to the data prediction module;

[0026] The data prediction module is used to construct a refined three-dimensional numerical model of the entire process of foundation pit construction; the refined three-dimensional numerical model of the entire process of foundation pit construction is corrected according to the on-site monitoring data to obtain a corrected numerical model; the data prediction results are obtained according to the corrected numerical model, and the data prediction results are sent to the monitoring and early warning module;

[0027] The monitoring and early warning module is used to monitor and warn the foundation pit based on data prediction results.

[0028] In one embodiment, the data acquisition and wireless remote transmission module includes a wireless node data acquisition unit based on ZigBee technology, a ZigBee repeater, a DTU integrated intelligent gateway and a mobile communication base station; the wireless node data acquisition unit is connected to the ZigBee repeater, the ZigBee repeater is connected to the DTU integrated intelligent gateway, the DTU integrated intelligent gateway is connected to the mobile communication base station, and the mobile communication base station is connected to the data prediction module.

[0029] In one embodiment, the wireless node data acquisition unit includes a vibrating string sensor measurement circuit and a digital temperature sensor measurement circuit.

[0030] In one of the embodiments, the deep foundation pit intelligent monitoring and early warning system further includes a data analysis module, a data acquisition and wireless remote transmission module is connected to the data analysis module, and the data analysis module is connected to the data prediction module;

[0031] The data analysis module is used to perform data classification, preprocessing, benchmark stability analysis and adjustment processing on the field monitoring data, and send the analyzed and processed field monitoring data to the data prediction module.

[0032] In one of the embodiments, the deep foundation pit intelligent monitoring and early warning system also includes a sensor module, which includes a pore water pressure gauge, a vibrating wire axial force gauge, a vibrating wire rebar gauge, a fixed inclinometer, a level and a total station.

[0033] The above-mentioned deep foundation pit intelligent monitoring and early warning method and system, by using on-site monitoring data, performs real-time dynamic correction on the pre-constructed refined three-dimensional numerical model of the entire foundation pit construction process, thereby generating a highly accurate corrected numerical model; the data prediction results obtained based on this corrected model not only include various possible changes in the foundation pit construction process, but also accurately reflect the impact of construction conditions on the stability of the foundation pit, thereby enabling continuous and efficient monitoring and early warning of the foundation pit status, ensuring the close integration and efficient coordination of on-site construction and foundation pit monitoring; therefore, the deep foundation pit intelligent monitoring and early warning method and system provides reliable technical support and decision-making basis for the safety assessment of the foundation pit, greatly improving the safety of foundation pit construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 A schematic diagram of a process of a deep foundation pit intelligent monitoring and early warning method in one embodiment;

[0035] Figure 2 A schematic diagram of the structure of a deep foundation pit intelligent monitoring and early warning system in one embodiment;

[0036] Figure 3 A schematic diagram of the structure of a data acquisition and wireless remote transmission module in one embodiment;

[0037] Figure 4 is a schematic structural diagram of a sensor module in one embodiment;

[0038] Figure 5 A schematic diagram of a process for constructing a refined three-dimensional numerical model of the entire foundation pit construction process in one embodiment;

[0039] Figure 6 A schematic diagram of a process for constructing a deep foundation pit deformation prediction model based on LSTM in one embodiment;

[0040] Figure 7 The figure is a schematic diagram of the working principle of the monitoring and early warning module in one embodiment. DETAILED DESCRIPTION

[0041] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0042] In one embodiment, Figure 1 As shown, a deep foundation pit intelligent monitoring and early warning method is provided, which specifically includes the following steps:

[0043] Step S1, constructing a refined three-dimensional numerical model of the entire foundation pit construction process.

[0044] Among them, the refined three-dimensional numerical model of the whole process of foundation pit construction includes the main model of the foundation pit, the model of surrounding buildings and structures, and the model of underground pipeline tunnels. Specifically, the geological and hydrogeological survey data and the design drawings of the retaining structure can be obtained first; among them, based on the geological and hydrogeological survey data, the key information such as the geological structure, soil properties, and groundwater level of the construction area can be obtained, and based on the retaining structure design drawings, the boundary, depth, and support structure type of the foundation pit can be clarified. Then, based on the obtained geological and hydrogeological survey data and the retaining structure design drawings, the main model of the foundation pit is constructed. The model should be able to accurately reflect the size, shape, support structure, and interaction between the foundation pit and the soil. Then, the surrounding building model and the underground pipeline tunnel model are constructed. The surrounding building model can specifically include buildings, roads, bridges, etc. near the construction area to evaluate the impact of construction on them. The underground pipeline tunnel model is used to consider the direction, burial depth, material, etc. of various pipelines (such as water pipes, drainage pipes, power pipelines, etc.).

[0045] The whole process of foundation pit construction is simulated in the above model, including site leveling, layer-by-layer dewatering and excavation of foundation pit, support activation, support replacement and removal, main structure construction, etc. Specifically, according to the construction steps, the stress and strain states in the above model are gradually updated to simulate the real construction process.

[0046] Step S2, obtaining on-site monitoring data.

[0047] Among them, on-site monitoring data may include on-site real-time construction information and monitoring data. Real-time construction information may include weather conditions, construction progress, construction conditions (such as soil type, groundwater level, groundwater flow rate, etc.). Monitoring data may include road and surrounding surface settlement monitoring data, underground pipeline settlement monitoring data, surrounding building settlement and inclination monitoring data, retaining wall displacement monitoring data, column structure vertical displacement monitoring data, groundwater level and water pressure monitoring data and support axial force monitoring data. Specifically, on-site monitoring data can be obtained by installing sensors at key locations around and inside the foundation pit.

[0048] Furthermore, since the stress-deformation law of the foundation pit during the construction process can be obtained according to the simulation process of the above model, the potential dangerous points in the construction process can be identified according to the stress-deformation law, such as areas prone to collapse and locations where the supporting structure is prone to failure. In order to further improve the safety of the foundation pit during the construction process, the staff can intensively deploy sensors at potential dangerous points to monitor the deformation and stress state of the foundation pit in real time. Among them, the stress-deformation law can include the horizontal displacement of the side wall of the foundation pit, the bulge of the pit bottom, the axial force change of the supporting structure, etc.

[0049] Step S3, correcting the refined three-dimensional numerical model of the entire foundation pit construction process according to the on-site monitoring data to obtain a corrected numerical model.

[0050] The refined three-dimensional numerical model of the entire foundation pit construction process reflects the numerical calculation results and evolution laws of each monitoring parameter of the foundation pit. After the foundation pit is put into construction, the numerical model is corrected according to the real-time construction information and monitoring data on site. For example, when the weather condition during construction is rainy, the water level parameters of the corresponding construction stage in the numerical model can be adjusted according to the groundwater level and pore water pressure monitoring data of the current construction stage. In addition, other relevant parameters in the numerical model can also be adjusted, such as soil mechanics parameters, support structure stiffness, etc., to more accurately reflect the actual construction situation. The specific methods of model correction can include inversion of rock and soil mechanical parameters, soil-structure interface contact parameter correction, model boundary size correction, and soil and surrounding structure constitutive model and parameter correction. Then, the corrected numerical model is used to re-simulate and predict the entire process of foundation pit construction.

[0051] Step S4, obtaining data prediction results according to the modified numerical model.

[0052] Specifically, the revised numerical model is used to predict the recent changes of key monitoring indicators to obtain data prediction results. Among them, the key monitoring indicators may include deformation speed, displacement, stress change, etc.

[0053] Step S5: monitor and warn the foundation pit based on the data prediction results.

[0054] Specifically, an overall construction monitoring plan can be generated based on the data prediction results. For example, the overall construction monitoring plan can include monitoring indicators and potential risk points that need to be focused on, as well as suggestions for adjusting and optimizing the layout and monitoring frequency of monitoring points to ensure continuous monitoring of key monitoring indicators. Early warning information can also be generated based on data prediction results. For example, a pre-set early warning threshold is obtained, and when the data prediction result exceeds the corresponding early warning threshold, a foundation pit early warning forecast is generated and sent to relevant personnel. The foundation pit early warning forecast can guide relevant personnel to take corresponding emergency measures to ensure the safe progress of construction.

[0055] The deep foundation pit intelligent monitoring and early warning method in this embodiment, by accurately using on-site monitoring data, performs real-time dynamic correction on the pre-constructed refined three-dimensional numerical model of the entire foundation pit construction process, thereby generating a highly accurate corrected numerical model; the data prediction results obtained based on this corrected model not only include various possible changes in the foundation pit construction process, but also accurately reflect the impact of construction conditions on the stability of the foundation pit, thereby enabling continuous and efficient monitoring and early warning of the foundation pit status, ensuring the close integration and efficient coordination of on-site construction and foundation pit monitoring; therefore, the deep foundation pit intelligent monitoring and early warning method and system provides reliable technical support and decision-making basis for the safety assessment of the foundation pit, greatly improving the safety of foundation pit construction.

[0056] In one embodiment, the deep foundation pit intelligent monitoring and early warning method further includes:

[0057] Step S61, constructing a deep foundation pit deformation prediction model based on LSTM.

[0058] Specifically, the structure of the deep foundation pit deformation prediction model based on LSTM is determined, including the input layer, hidden layer, output layer and network training optimization algorithm.

[0059] Step S62, obtaining measured deformation data that changes with time during the foundation pit construction process.

[0060] Among them, the measured deformation data is the actual data about the deformation of the foundation pit obtained by measuring equipment (such as total station, displacement meter, etc.) during the foundation pit construction process, such as horizontal displacement data, vertical settlement data, etc.

[0061] Step S63, preprocessing the measured deformation data to obtain sample time series data.

[0062] Among them, preprocessing can include data cleaning (removing outliers, processing missing values, etc.), data normalization or standardization, time series feature extraction, etc. The sample time series data obtained after preprocessing can be used as a data set for model training.

[0063] Step S64, using the sample time series data to train the deep foundation pit deformation prediction model to obtain a trained deep foundation pit deformation prediction model.

[0064] Specifically, the sample time series data is imported into the input layer, and the samples are trained until the training error of the LSTM network reaches the preset requirement and the number of iterations is completed, and the output layer data is obtained.

[0065] Step S65, predicting the recent deformation of the foundation pit according to the trained deep foundation pit deformation prediction model to obtain the recent development trend data of the foundation pit.

[0066] Among them, the recent development trend data of the foundation pit can reflect the future deformation trend of the foundation pit.

[0067] Step S66, monitoring and early warning of the foundation pit based on the recent development trend data of the foundation pit and the data prediction results.

[0068] Specifically, the foundation pit can be monitored in real time based on the data prediction results obtained in step S4 and the recent development trend data of the foundation pit obtained in step S65, and an early warning mechanism can be triggered according to set thresholds or conditions to ensure the safety of the foundation pit.

[0069] This embodiment constructs a deep foundation pit deformation prediction model based on LSTM, and uses the deep foundation pit deformation prediction model to predict the future deformation trend of the foundation pit, so as to achieve more comprehensive monitoring and early warning of the foundation pit and further improve the safety of the foundation pit.

[0070] In one embodiment, step S66 includes:

[0071] Step S660, obtaining the foundation pit parameter monitoring threshold, monitoring point alarm level and patrol inspection warning.

[0072] Specifically, first determine the foundation pit parameters that need to be monitored, which may include but are not limited to the deformation of the foundation pit, changes in groundwater levels, soil pressure, stress of the supporting structure, etc. Then set reasonable monitoring thresholds for each parameter based on specific engineering requirements and safety standards. Once the monitoring data exceeds these thresholds, it is necessary to trigger the corresponding monitoring point alarm level. Specifically, the alarm level can be set for each monitoring point based on the degree and duration of the on-site monitoring data exceeding the threshold. The alarm level should be able to reflect the current safety status and risk level of the foundation pit, and provide a basis for subsequent early warning measures. For example, it can be divided into a first-level warning (mild risk), a second-level warning (moderate risk) and a third-level warning (severe risk).

[0073] Step S661, generating a foundation pit early warning forecast based on the foundation pit recent development trend data, data prediction results, on-site monitoring data, foundation pit parameter monitoring thresholds, monitoring point alarm levels and patrol inspection warnings.

[0074] Specifically, the current status and development speed of the foundation pit are obtained based on the recent development trend data of the foundation pit; the development of each monitoring point in the future period is obtained based on the data prediction results; and the on-site monitoring data is compared with the corresponding foundation pit parameter monitoring threshold to determine which foundation pit parameters exceed the threshold; combined with the on-site inspection of the patrol personnel, the potential danger points are determined, such as the locations where the patrol personnel observe signs of cracks, settlement, etc. as potential danger points. Then, all the above information is combined to generate a foundation pit early warning forecast. The foundation pit early warning forecast can include the current safety status of the foundation pit, potential risks, development trends, recommended warning levels, and countermeasures that need to be taken.

[0075] The deep foundation pit intelligent monitoring and early warning method in this embodiment comprehensively considers the on-site monitoring data, the recent development trend data of the foundation pit, the data prediction results, the alarm level of the monitoring point, and the patrol inspection and early warning to form a comprehensive foundation pit safety assessment. This comprehensive assessment provides comprehensive decision-making support for managers, enabling them to formulate risk response strategies and prevention and control measures more scientifically.

[0076] In one embodiment, the method for correcting the refined three-dimensional numerical model of the entire foundation pit construction process includes at least one of inversion of rock and soil mechanical parameters, correction of soil-structure interface contact parameters, model boundary size correction, and soil and surrounding structure constitutive model and parameter correction.

[0077] Among them, the inversion correction method of geotechnical mechanical parameters includes obtaining on-site monitoring data and correcting the geotechnical mechanical parameters in combination with the inversion algorithm to improve the accuracy of the model. Specifically, the physical and mechanical properties of the rock and soil around the foundation pit can be obtained based on the on-site geological survey data, and a small number of samples can be learned and trained based on the support vector machine (SVM) or other machine learning algorithms to establish a mapping relationship between the geotechnical mechanical parameters and the on-site monitoring data. Using the established mapping relationship, the geotechnical mechanical parameters in the refined three-dimensional numerical model of the entire foundation pit construction process are inverted to make it closer to the actual engineering conditions.

[0078] The soil-structure interface contact parameter correction method improves the model's ability to simulate interface behavior by correcting the soil-structure interface contact parameters. Specifically, the mechanical properties of the contact surface are determined by analyzing the interaction between the soil and the structure, and then appropriate contact surface units are set in the refined three-dimensional numerical model of the entire foundation pit construction process, and appropriate contact parameters such as friction coefficient and adhesion are assigned. Then, the changes in stress and strain of the contact surface are obtained through simulation analysis, and the contact parameters are iteratively corrected until the simulation results are consistent with the actual engineering situation.

[0079] The model boundary size correction method includes reasonably adjusting the model boundary size according to the actual construction situation and monitoring data to ensure the accuracy and reliability of the simulation results. Specifically, first obtain the spatial range required for simulation analysis; based on the spatial range, establish appropriate boundary conditions, such as fixed boundaries, free boundaries, etc. in the refined three-dimensional numerical model of the entire foundation pit construction process; then obtain the influence of boundary effects on the deformation and force of the foundation pit through simulation analysis, and adjust the boundary size appropriately to reduce the influence of boundary effects. Among them, the spatial range required for simulation analysis can be determined according to the actual situation of the foundation pit project. For example, the spatial range of the simulation analysis of the foundation pit project can be determined according to the actual excavation depth, shape and surrounding environment of the foundation pit. The simulation range should extend at least to the stable stratum below the bottom of the foundation pit in the direction of the foundation pit excavation depth to ensure that the simulated bottom boundary conditions are stable. In the plane direction, the simulation range can cover the four sides of the foundation pit and extend outward, and the extension distance can be several times the excavation depth of the foundation pit to ensure that the potential impact of the foundation pit excavation on the surrounding soil and structure is taken into account. At the same time, it is also necessary to ensure that important buildings, roads and other structures around the foundation pit are also included in the simulation range.

[0080] The method for modifying the constitutive model and parameters of soil and retaining structures includes modifying the constitutive model and parameters of soil and retaining structures according to on-site monitoring data and material mechanical properties analysis, so as to improve the model's simulation ability of material behavior during construction. Specifically, the constitutive model and parameters of the mechanical properties of soil and retaining structures are first obtained, and then the selected constitutive model and parameters are applied to the refined three-dimensional numerical model of the entire foundation pit construction process for simulation analysis. The constitutive model and parameters are iteratively modified according to the comparison between the simulation results and the actual engineering conditions until the simulation results meet the engineering requirements. Among them, the constitutive model of the mechanical properties of soil and retaining structures can include elastic models, elastoplastic models, etc., and the relevant parameters in the constitutive model can include elastic modulus, Poisson's ratio, yield stress, etc.

[0081] Specifically, the above correction method may be selected and implemented by comprehensively considering factors such as the actual situation of the construction site, the accuracy and reliability of the on-site monitoring data, and the applicability and precision of the numerical simulation method.

[0082] The deep foundation pit intelligent monitoring and early warning method in this embodiment corrects the refined three-dimensional numerical model of the entire foundation pit construction process by using at least one of the following methods: inversion of rock and soil mechanical parameters, correction of soil-structure interface contact parameters, correction of model boundary dimensions, and correction of soil and surrounding structure constitutive models and parameters. This can achieve close integration and efficient coordination between on-site construction and foundation pit monitoring, further improve the real-time and reliability of foundation pit monitoring and early warning, and greatly enhance the safety of foundation pit construction.

[0083] In one embodiment, after step S4 and before step S5, the deep foundation pit intelligent monitoring and early warning method further includes:

[0084] Step S41, comparing the data prediction results with the actual foundation pit monitoring data to verify the accuracy of the refined three-dimensional numerical model of the entire foundation pit construction process.

[0085] Among them, the actual foundation pit monitoring data is the real-time measured foundation pit monitoring data, which may include key indicators such as displacement, stress, and strain. Specifically, first obtain the output data of the refined three-dimensional numerical model of the entire foundation pit construction process and the actual foundation pit monitoring data, compare the output data of the three-dimensional numerical model with the actual foundation pit monitoring data in terms of numerical size, change trend, etc., and obtain the residual of each monitoring point (that is, the difference between the measured value and the simulated value), and evaluate the size and distribution of the residual. If the residual is small and evenly distributed, the model accuracy is high. In addition, the overall fitting error of the refined three-dimensional numerical model of the entire foundation pit construction process can also be calculated, such as the root mean square error or the mean absolute error. The smaller the error value, the more accurate the model.

[0086] If the refined three-dimensional numerical model of the entire process of foundation pit construction is not accurate, the parameters of the refined three-dimensional numerical model of the entire process of foundation pit construction can be adjusted, such as adjusting the mechanical parameters of the rock and soil, contact parameters, boundary dimensions, etc.; if the accuracy of the model is still not high after the parameter adjustment, further consideration can be given to optimizing the model results or using more advanced numerical analysis methods. After parameter adjustment or model optimization, re-conduct numerical simulation and compare and verify with real-time measured data. Treat the entire verification process as a cycle, continuously collect real-time data, and continuously correct and optimize the numerical model to improve its accuracy.

[0087] The deep foundation pit intelligent monitoring and early warning method in this embodiment evaluates the accuracy of the refined three-dimensional numerical model of the entire foundation pit construction process based on the actual foundation pit monitoring data, and further optimizes the model when the accuracy of the model is low. This can improve the reliability of the refined three-dimensional numerical model of the entire foundation pit construction process, thereby improving the reliability of the data prediction results, and providing reliable data support for intelligent monitoring and early warning of the foundation pit.

[0088] In one embodiment, Figure 2 As shown, a deep foundation pit intelligent monitoring and early warning system is provided, including a data acquisition and wireless remote transmission module 200, a data prediction module 300 and a monitoring and early warning module 400, wherein the data acquisition and wireless remote transmission module 200 is connected to the data prediction module 300, and the data prediction module 300 is connected to the monitoring and early warning module 400.

[0089] The data acquisition and wireless remote transmission module 200 is used to obtain the on-site monitoring data and send the on-site monitoring data to the data prediction module 300. The on-site monitoring data may include road and surrounding surface settlement monitoring data, underground pipeline settlement monitoring data, surrounding building settlement and tilt monitoring data, retaining wall displacement monitoring data, column structure vertical displacement monitoring data, groundwater level and water pressure monitoring data, and support axial force monitoring data.

[0090] The data prediction module 300 is used to construct a refined three-dimensional numerical model of the entire process of foundation pit construction, and to modify the refined three-dimensional numerical model of the entire process of foundation pit construction according to the on-site monitoring data to obtain a modified numerical model, and to obtain data prediction results according to the modified numerical model, and to send the data prediction results to the monitoring and early warning module 400. The specific steps of model construction and model modification can refer to the above-mentioned method embodiment, which will not be repeated here. Furthermore, the data prediction module 300 is also used to compare the data prediction results with the actual foundation pit monitoring data to verify the accuracy of the refined three-dimensional numerical model of the entire process of foundation pit construction.

[0091] In addition, the data prediction module 300 is also used to construct a deep foundation pit deformation prediction model based on LSTM; obtain the measured deformation data that changes with time during the foundation pit construction process; obtain sample time series data after preprocessing the measured deformation data; use the sample time series data to train the deep foundation pit deformation prediction model to obtain the trained deep foundation pit deformation prediction model; predict the recent deformation of the foundation pit based on the trained deep foundation pit deformation prediction model to obtain the recent development trend data of the foundation pit.

[0092] The monitoring and early warning module 400 is used to monitor and warn the foundation pit based on the data prediction results. The specific steps may include: obtaining the monitoring threshold of foundation pit parameters, the alarm level of monitoring points, and the early warning of patrol inspections; generating a foundation pit early warning forecast based on the recent development trend data of the foundation pit, data prediction results, on-site monitoring data, the monitoring threshold of foundation pit parameters, the alarm level of monitoring points, and the early warning of patrol inspections.

[0093] The deep foundation pit intelligent monitoring and early warning system in this embodiment, by accurately using on-site monitoring data, performs real-time dynamic correction on the pre-built refined three-dimensional numerical model of the entire foundation pit construction process, thereby generating a highly accurate corrected numerical model; the data prediction results obtained based on this corrected model not only include various possible changes in the foundation pit construction process, but also accurately reflect the impact of construction conditions on the stability of the foundation pit, thereby enabling continuous and efficient monitoring and early warning of the foundation pit status, ensuring the close integration and efficient coordination of on-site construction and foundation pit monitoring; therefore, the deep foundation pit intelligent monitoring and early warning method and system provides reliable technical support and decision-making basis for the safety assessment of the foundation pit, greatly improving the safety of foundation pit construction.

[0094] In one embodiment, Figure 3 As shown, the data acquisition and wireless remote transmission module 200 includes a wireless node data acquisition unit, a ZigBee repeater, a DTU integrated intelligent gateway and a mobile communication base station based on ZigBee technology; the wireless node data acquisition unit is connected to the ZigBee repeater, the ZigBee repeater is connected to the DTU integrated intelligent gateway, the DTU integrated intelligent gateway is connected to the mobile communication base station, and the mobile communication base station is connected to the data prediction module 300.

[0095] Among them, the wireless node data acquisition unit based on ZigBee technology includes ZigBee nodes and sensors, which can be understood as including ZigBee nodes and sensors related thereto, which can be integrated together or independent devices. The sensor can receive data acquisition instructions at a specified time and frequency, and collect field monitoring data in real time according to the data acquisition instructions, and then the ZigBee node sends the field monitoring data collected by the sensor to the ZigBee repeater. The ZigBee repeater plays a bridge role in the data acquisition and wireless remote transmission module 200. It is responsible for receiving the field monitoring data from the ZigBee node and forwarding it to the DTU integrated intelligent gateway. Since ZigBee technology has the characteristics of low power consumption, high reliability and self-organizing network, the ZigBee repeater can ensure the stability and security of data during transmission. The DTU integrated intelligent gateway integrates multiple functions such as data processing, storage and forwarding. The DTU integrated intelligent gateway receives data from the ZigBee repeater and performs necessary processing on it, such as data compression, encryption, etc. The processed data is then sent to the mobile communication base station through the communication module inside the gateway. The mobile communication base station then sends the processed on-site monitoring data to the data prediction module 300 .

[0096] The deep foundation pit intelligent monitoring and early warning system in this embodiment transmits on-site monitoring data by adopting wireless communication technology based on ZigBee technology, which can ensure the stability and security of data during transmission.

[0097] In one embodiment, the wireless node data acquisition unit includes a vibrating string sensor measurement circuit and a digital temperature sensor measurement circuit. Among them, the vibrating string sensor measurement circuit is mainly used for real-time monitoring and early warning of deformation of underground structures in deep foundation pit monitoring. When the excavation of the foundation pit or changes in the surrounding environment cause changes in the force of the underground structure, these changes will act on the vibrating string and change its vibration frequency. The measurement circuit of the vibrating string sensor converts the change in the vibration frequency of the string into a change in an electrical signal, and sends it to the data prediction module 300 through the ZigBee repeater, the DTU integrated intelligent gateway and the mobile communication base station. The digital temperature sensor measurement circuit is mainly used for real-time monitoring and early warning of temperature changes inside the foundation pit or the surrounding environment in deep foundation pit monitoring. The digital temperature sensor senses the ambient temperature through its internal thermistor and converts the temperature signal into a digital signal output.

[0098] The wireless node data acquisition unit in this embodiment includes a vibrating string sensor measurement circuit and a digital temperature sensor measurement circuit. By monitoring the deformation of the underground structure and the changes in ambient temperature, it can provide important data support for the safe construction of the foundation pit, thereby realizing comprehensive, real-time and accurate monitoring and early warning of the foundation pit.

[0099] In one embodiment, the deep foundation pit intelligent monitoring and early warning system further includes a data analysis module, the data acquisition and wireless remote transmission module 200 is connected to the data analysis module, and the data analysis module is connected to the data prediction module 300. The data analysis module is used to perform data classification, preprocessing, benchmark stability analysis and adjustment processing on the field monitoring data, and send the analyzed and processed field monitoring data to the data prediction module 300.

[0100] This embodiment can improve data quality and enhance data reliability by performing data classification, preprocessing, benchmark stability analysis and adjustment processing on the collected field monitoring data.

[0101] In one embodiment, the deep foundation pit intelligent monitoring and early warning system also includes a sensor module. After constructing a refined three-dimensional numerical model of the entire foundation pit construction process, the stress and deformation law can be obtained, and an overall construction monitoring plan can be formed by encrypting and deploying sensors at dangerous points. The sensor module is used to monitor the surface settlement of roads and surrounding areas, the settlement of underground pipelines, the settlement and inclination of surrounding buildings, the displacement of retaining walls, the vertical displacement of column structures, the groundwater level and water pressure, and the support axial force. After each sensor that needs to be measured receives the collection command sent by the cloud platform at the specified time and frequency, real-time data measurement is performed and sent to the data collection and wireless remote transmission module 200. Among them, if Figure 4 As shown, the sensor modules may include a pore water pressure gauge, a vibrating wire axial force gauge, a vibrating wire rebar gauge, a fixed inclinometer, a level and a total station.

[0102] The sensor modules in this embodiment include a pore water pressure gauge, a vibrating wire axial force gauge, a vibrating wire rebar gauge, a fixed inclinometer, a level and a total station, which can perform more comprehensive monitoring of the foundation pit.

[0103] In order to better understand the above-mentioned deep foundation pit intelligent monitoring and early warning method and system, a more detailed embodiment is provided below for explanation.

[0104] In one embodiment, Figure 3As shown, the deep foundation pit intelligent monitoring and early warning system includes a monitoring and early warning module (not shown in the figure), a cloud platform, a mobile communication base station, a DTU integrated intelligent gateway, a ZigBee repeater, and a ZigBee node and sensor. Among them, the cloud platform is a data display, data storage center and a client interface, which can include the data prediction module 300 and the data analysis module in the above embodiment; the monitoring and early warning module 400 can be integrated into the cloud platform or can be a separate module. The DTU integrated intelligent gateway is a multifunctional acquisition module with a built-in embedded processor, and a built-in DTU (Data Transfer Unit) module, which can send the data transmitted by the ZigBee repeater to the cloud platform; the ZigBee node, ZigBee repeater, the DTU integrated intelligent gateway and the mobile communication base station can constitute the data acquisition and wireless remote transmission module 200 in the above embodiment.

[0105] The deep foundation pit intelligent monitoring and early warning method specifically includes the following steps:

[0106] S1. Based on geological and hydrogeological survey data and retaining structure design drawings, a refined three-dimensional numerical model of the entire foundation pit construction process is established on the cloud platform to obtain the stress-deformation law. By analyzing the stress-deformation law, dangerous points are obtained, and sensors are densely deployed at dangerous points to form an overall construction monitoring plan. For example, when the refined three-dimensional numerical model of the entire foundation pit construction process simulates the construction steps of layer-by-layer precipitation excavation, the middle and lower parts of the foundation pit ground-connected wall have a large displacement of 25mm and a maximum stress of 1.8MPa. Therefore, when the actual construction reaches the layer-by-layer precipitation excavation stage, a reminder message can be generated. The reminder message may include paying attention to speeding up the monitoring frequency of fixed inclinometers, such as increasing the monitoring frequency from 1 time / 1d (day) to 3 times / d (day), so as to better reflect the state changes of the foundation pit retaining structure during this construction stage.

[0107] S2. Arrange various types of sensors at the foundation pit construction site according to the overall construction monitoring plan of step S1. After the construction starts, the cloud platform sends a data acquisition instruction to the DTU integrated intelligent gateway through the mobile communication base station. The DTU integrated intelligent gateway sends the data acquisition instruction to the ZigBee node through the ZigBee repeater. The ZigBee node then sends the data acquisition instruction to the sensor. The sensor performs real-time data measurement according to the preset time and frequency to obtain the on-site monitoring data. The on-site monitoring data is then transmitted to the cloud platform for storage and analysis through the ZigBee node, ZigBee repeater, DTU integrated intelligent gateway and mobile communication base station in turn. Among them, the ZigBee node integrates the measurement circuit for the vibrating string sensor and the measurement circuit for the digital temperature sensor, and is expanded to 4 outputs through a switch.

[0108] S3. The cloud platform screens, processes and stores the on-site monitoring data; the processing steps specifically include data classification, preprocessing, benchmark point stability analysis and adjustment processing.

[0109] S4. The cloud platform is used to construct a refined three-dimensional numerical model of the entire foundation pit construction process and a deep foundation pit deformation prediction model based on LSTM, and obtain data prediction results based on the refined three-dimensional numerical model of the entire foundation pit construction process, and obtain the recent development trend data of the foundation pit based on the LSTM-based deep foundation pit deformation prediction model; monitor and warn the foundation pit based on the recent development trend data of the foundation pit and the data prediction results.

[0110] S4.1, if Figure 5 As shown, the data prediction results obtained based on the refined three-dimensional numerical model of the entire foundation pit construction process specifically include the following steps: (1) Construct a refined three-dimensional numerical model of the entire foundation pit construction process based on geological and hydrogeological survey data, retaining structure design drawings and surrounding environmental conditions. The entire foundation pit construction process specifically includes layer-by-layer dewatering excavation of the foundation pit, activation of support, replacement and removal of support, main structure construction, until the end of construction. Through this model, the numerical calculation results and evolution laws of various monitoring parameters of the foundation pit can be obtained. (2) After the foundation pit is put into construction, the numerical model is corrected according to real-time construction information and monitoring data to obtain a corrected numerical model. The correction method includes inversion of rock and soil mechanical parameters, correction of soil-structure interface contact parameters, correction of model boundary size, and correction of soil and retaining structure constitutive model and parameters. The numerical calculation results of the corrected numerical model are compared with the real-time measured foundation pit monitoring data to verify the accuracy of the refined three-dimensional numerical model. The specific verification steps can refer to the above embodiment. (3) The corrected numerical model is used to predict the recent changes of key monitoring indicators. The refined three-dimensional numerical model of the entire foundation pit construction process in this embodiment includes a Mohr-Coulomb constitutive model, which can be used to simulate the state changes of a certain rock and soil layer during the construction process. Take the correction of the Mohr-Coulomb constitutive model based on monitoring data and the inversion correction method of the mechanical parameters of the rock and soil as an example: After preliminary simulation, it was found that the displacement of the ground-anchored wall simulated by the Mohr-Coulomb constitutive model deviated by 5.2 mm from the actual displacement of the ground-anchored wall. In order to more accurately predict the displacement of the ground-anchored wall, it is necessary to correct the physical and mechanical parameters in the Mohr-Coulomb constitutive model. The details are as follows:

[0111] First, the initial values ​​of physical and mechanical parameters are obtained, including elastic modulus (E) 20MPa, Poisson's ratio (ν) 0.36, friction angle (φ) 16.12° and cohesion (c) 36.4kPa. Then the parameter sensitivity analysis is carried out, specifically including changing the elastic modulus (E) by -15%, -10%, -5%, 5%, 10%, and 15%, respectively, keeping other parameters unchanged, simulating and calculating the sum of the Euclidean distances between the measured value and the simulated value of the lateral displacement of the ground-connected wall from 5m depth to 20m depth (denoted as ΔE). Changing the Poisson's ratio (ν) by -15%, -10%, -5%, 5%, 10%, and 15%, respectively, keeping other parameters unchanged, simulating and calculating the sum of the Euclidean distances (denoted as Δν). Changing the friction angle (φ) by -15%, -10%, -5%, 5%, 10%, and 15%, respectively, keeping other parameters unchanged, simulating and calculating the sum of the Euclidean distances (denoted as Δφ). The cohesion (c) was varied by -15%, -10%, -5%, 5%, 10%, and 15% respectively, while other parameters were kept unchanged. The simulation was performed and the sum of the Euclidean distances (denoted as Δc) was calculated.

[0112] Then perform parameter correction, specifically including, based on the results of the above sensitivity analysis, selecting the physical and mechanical parameters that have the greatest impact on the sum of Euclidean distances for priority correction. Assume that the change in elastic modulus (E) has the greatest impact on the sum of Euclidean distances. According to the result of ΔE, find the rate of change of elastic modulus (E) that minimizes the sum of Euclidean distances. Assuming that when the elastic modulus (E) increases by 5%, ΔE reaches the minimum value, the elastic modulus (E) can be corrected to 20MPa*(1+5%)=21MPa, keeping other parameters unchanged and re-simulating. If there is still a deviation between the corrected simulation results and the actual displacement of the ground-connected wall, repeat the above steps to correct other physical and mechanical parameters until the simulation results are close enough to the actual displacement.

[0113] In addition, due to the complex and changeable actual construction conditions, the parameters in the model need to be corrected according to the real-time construction information to ensure the accuracy of the prediction. This embodiment will illustrate how to correct the water level and pore pressure parameters in the numerical model according to the real-time construction information after encountering rainy weather, and re-simulate and analyze. Specifically including real-time construction information acquisition: through the water level gauge and pore water pressure gauge sensor in the observation well, real-time monitoring of the changes in groundwater level and pore water pressure, recording the height of the groundwater level rise (such as 1.6m in this embodiment) and the pore pressure increase value monitored by the pore water pressure gauge sensor. Model parameter correction: According to the height of the groundwater level rise (1.6m), modify the water level parameters of the corresponding construction stage in the numerical model to ensure that the water level in the model matches the actual observation data; according to the pore pressure increase value monitored by the pore water pressure gauge sensor, modify the pore pressure parameters of the corresponding position in the model. Resubmit numerical simulation: Use the corrected numerical model to re-calculate the fluid-solid coupling after precipitation, and the simulation result is that the maximum displacement of the ground-connected wall increases by 2.86mm.

[0114] S4.2, such as Figure 6 As shown in the figure, the specific steps for obtaining the recent development trend data of the foundation pit according to the LSTM-based deep foundation pit deformation prediction model are as follows: (1) Determine the framework of the deep foundation pit deformation prediction model based on LSTM, that is, determine the input layer, hidden layer, output layer and network training optimization algorithm selection; (2) Organize and preprocess the measured deformation data that changes with time during the foundation pit construction process to obtain sample time series data; (3) Import the sample time series data into the input layer and train the samples until the LSTM network training error meets the requirements and the number of iterations is completed, and obtain the output layer data; (4) Use the trained LSTM model to predict the recent deformation of the foundation pit and obtain the recent development trend data of the foundation pit.

[0115] S5, the monitoring and early warning module 400 issues an early warning based on the recent development trend data and data prediction results of the foundation pit obtained in step S4, the alarm level of the monitoring point, and the patrol inspection and early warning, and maintains tracking and supervision until the alarm is eliminated. Figure 7As shown in FIG, the specific steps include: (1) determining the foundation pit parameters to be monitored, and determining the foundation pit parameter monitoring threshold and warning level according to the project requirements and safety standards; for example, the warning level can be divided into three types: yellow alarm at the monitoring point, orange alarm at the monitoring point, and red alarm at the monitoring point. Among them, the yellow alarm at the monitoring point corresponds to the cumulative value of the actual change of each monitoring indicator of the foundation pit and the rate value "double control" indicator exceeding 70% of the corresponding foundation pit parameter monitoring threshold, or one of the double control indicators reaches 80% of the foundation pit parameter monitoring threshold. The orange alarm at the monitoring point corresponds to the cumulative value and the rate value "double control" indicator exceeding 80% of the corresponding foundation pit parameter monitoring threshold, or one of the double control indicators exceeds the foundation pit parameter monitoring threshold. The red alarm at the monitoring point corresponds to the cumulative value and the rate value "double control" indicator exceeding the foundation pit parameter monitoring threshold, or the measured change rate increases sharply, and the project shows signs of instability. In addition, the predicted values ​​of each foundation pit parameter can be obtained based on the refined three-dimensional numerical model of the entire foundation pit construction process and the deep foundation pit deformation prediction model based on LSTM. When it is predicted that the monitoring indicator at a certain construction stage may exceed 70% of the corresponding foundation pit parameter monitoring threshold, a yellow warning is issued. When the predicted value reaches 80% of the monitoring threshold of the foundation pit parameters, an orange warning is issued. Once the predicted value exceeds the monitoring threshold of the foundation pit parameters, a red warning is immediately issued. Furthermore, after the yellow warning is issued, the monitoring frequency can be increased to 2 times / d. After the orange warning is issued, the monitoring frequency can be increased to 3 times / d. In the case of a red warning, it is judged that the foundation pit is in an unacceptable risk state at this time, and work is immediately stopped and emergency rescue is carried out. (2) A comprehensive warning is issued based on the recent results of data prediction, the alarm level of the monitoring point, and the patrol inspection warning, and uploaded to the cloud platform for foundation pit early warning forecast; (3) Increase the monitoring frequency and organize units to survey the site, and each unit discusses and puts forward suggestions and control measures; (4) After the warning is issued, if effective measures are taken on site, the monitoring data for several consecutive days shows a convergence trend, and the alarm situation is effectively controlled or lifted, apply for the alarm to be eliminated.

[0116] The deep foundation pit intelligent monitoring and early warning method and system in this embodiment builds an automated and intelligent real-time monitoring system by relying on the Internet of Things and cloud platforms. The system not only realizes the automatic collection and efficient transmission of data, but also accurately classifies massive data through the data analysis module, which is convenient for users to quickly query. At the same time, the system closely combines the on-site construction of the foundation pit with real-time monitoring, uses three-dimensional numerical models to carefully plan the construction monitoring plan, accurately locates potential danger points and encrypts the sensor layout. In addition, combined with construction conditions and geological environment, data prediction modules and monitoring and early warning modules have been developed based on three-dimensional numerical models and LSTM models. These innovative technologies have greatly promoted the widespread application of intelligent, dynamic and integrated monitoring technologies in deep foundation pit projects.

[0117] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0118] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A deep foundation pit intelligent monitoring and early warning method, characterized in that: The method comprises: Step S1, constructing a refined three-dimensional numerical model of the entire foundation pit construction process; Step S2, obtaining on-site monitoring data; Step S3, correcting the refined three-dimensional numerical model of the entire foundation pit construction process according to the on-site monitoring data to obtain a corrected numerical model; Step S4, obtaining a data prediction result according to the modified numerical model; Step S5: monitoring and early warning of the foundation pit based on the data prediction results.

2. The deep foundation pit intelligent monitoring and early warning method according to claim 1 is characterized in that: The method further comprises: Step S61, constructing a deep foundation pit deformation prediction model based on LSTM; Step S62, obtaining measured deformation data that changes with time during foundation pit construction; Step S63, preprocessing the measured deformation data to obtain sample time series data; Step S64, using the sample time series data to train the deep foundation pit deformation prediction model to obtain a trained deep foundation pit deformation prediction model; Step S65, predicting the recent deformation of the foundation pit according to the trained deep foundation pit deformation prediction model to obtain the recent development trend data of the foundation pit; Step S66: monitoring and issuing early warnings to the foundation pit based on the recent development trend data of the foundation pit and the data prediction results.

3. The deep foundation pit intelligent monitoring and early warning method according to claim 2 is characterized in that: The step S66 comprises: Step S660, obtaining the foundation pit parameter monitoring threshold, monitoring point alarm level and patrol inspection warning; Step S661, generating a foundation pit early warning forecast based on the foundation pit recent development trend data, the data prediction results, the on-site monitoring data, the foundation pit parameter monitoring threshold, the monitoring point alarm level and the patrol inspection warning.

4. The deep foundation pit intelligent monitoring and early warning method according to claim 1 is characterized in that: The method for correcting the refined three-dimensional numerical model of the entire foundation pit construction process includes at least one of rock and soil mechanical parameter inversion, soil-structure interface contact parameter correction, model boundary size correction, and soil and surrounding structure constitutive model and parameter correction.

5. The deep foundation pit intelligent monitoring and early warning method according to claim 1 is characterized in that: After step S4 and before step S5, the method further includes: Step S41, comparing the data prediction results with the actual foundation pit monitoring data to verify the accuracy of the refined three-dimensional numerical model of the entire foundation pit construction process.

6. A deep foundation pit intelligent monitoring and early warning system, characterized in that: It includes a data acquisition and wireless remote transmission module, a data prediction module and a monitoring and early warning module, wherein the data acquisition and wireless remote transmission module is connected to the data prediction module, and the data prediction module is connected to the monitoring and early warning module; The data acquisition and wireless remote transmission module is used to obtain on-site monitoring data and send the on-site monitoring data to the data prediction module; The data prediction module is used to construct a refined three-dimensional numerical model of the entire process of foundation pit construction; the refined three-dimensional numerical model of the entire process of foundation pit construction is corrected according to the on-site monitoring data to obtain a corrected numerical model; a data prediction result is obtained according to the corrected numerical model, and the data prediction result is sent to the monitoring and early warning module; The monitoring and early warning module is used to monitor and warn the foundation pit based on the data prediction results.

7. The deep foundation pit intelligent monitoring and early warning system according to claim 6 is characterized in that: The data acquisition and wireless remote transmission module includes a wireless node data acquisition unit based on ZigBee technology, a ZigBee repeater, a DTU integrated intelligent gateway and a mobile communication base station; the wireless node data acquisition unit is connected to the ZigBee repeater, the ZigBee repeater is connected to the DTU integrated intelligent gateway, the DTU integrated intelligent gateway is connected to the mobile communication base station, and the mobile communication base station is connected to the data prediction module.

8. The deep foundation pit intelligent monitoring and early warning system according to claim 7 is characterized in that: The wireless node data acquisition unit includes a vibrating string sensor measurement circuit and a digital temperature sensor measurement circuit.

9. The deep foundation pit intelligent monitoring and early warning system according to claim 6 is characterized in that: It also includes a data analysis module, the data acquisition and wireless remote transmission module is connected to the data analysis module, and the data analysis module is connected to the data prediction module; The data analysis module is used to perform data classification, preprocessing, benchmark stability analysis and adjustment processing on the field monitoring data, and send the analyzed and processed field monitoring data to the data prediction module.

10. The deep foundation pit intelligent monitoring and early warning system according to claim 6, characterized in that: The device also comprises a sensor module, which comprises a pore water pressure gauge, a vibrating wire axial force gauge, a vibrating wire rebar gauge, a fixed inclinometer, a level and a total station.

Images