Intelligent monitoring method and system for deep foundation pit based on multi-source data fusion

Abstract

The application discloses a deep foundation pit intelligent monitoring method and system based on multi-source data fusion, relates to the technical field of construction monitoring, and comprises the following steps: step S1, collecting multi-physical field data of a foundation pit in real time through a distributed sensor network; step S2, performing a preprocessing operation on the multi-physical field data by using an edge computing node; step S3, transmitting the data after the preprocessing operation to a cloud platform through a multi-mode communication network; step S4, performing multi-source data fusion analysis on the data in the cloud platform, and calculating a comprehensive risk index by using an improved time-varying weight fusion algorithm; step S5, judging the comprehensive risk index, and realizing graded early warning; and step S6, executing corresponding automatic response measures according to the early warning level. The application realizes real-time collection of data such as foundation pit stability by embedding sensing elements, and realizes informatization and intelligentization of foundation pit monitoring through intelligent calculation and automatic alarm.

Description

Technical Field

[0001] This invention relates to the field of construction monitoring technology, specifically to a method and system for intelligent monitoring of deep foundation pits based on multi-source data fusion. Background Technology

[0002] With the rapid development of urban construction, deep foundation pit projects are showing a trend of becoming larger in scale, more complex in environment, and continuously increasing in depth. Traditional deep foundation pit monitoring technology mainly relies on manual periodic inspections and single-point automated monitoring, using professional equipment such as total stations, inclinometers, and levels to collect data periodically at preset monitoring points.

[0003] However, with increasingly stringent engineering requirements, traditional deep foundation pit monitoring also has relative limitations. In terms of data comprehensiveness, traditional methods often focus on monitoring single-type parameters, making it difficult to simultaneously perceive multiple physical fields such as the internal forces of the support structure, soil deformation, and groundwater changes. At the data analysis level, the lack of deep fusion and intelligent analysis capabilities for multi-source monitoring data makes it difficult to accurately assess the overall safety status of the system.

[0004] For example, Chinese patent CN115167212A discloses a dynamic construction control system and method for foundation pits based on a monitoring platform. The system includes an engineering visualization module, a wireless monitoring module for foundation pits, a data analysis module, a theoretical analysis interaction module, a dynamic construction control module, and an alarm module. Based on monitoring data, it compares the monitoring data with multiple sets of theoretical calculation results to select the optimal calculation condition and predict the changes in the foundation pit state in the next stage, thereby ensuring the safety of construction in the next stage. It has advantages such as convenient operation, low cost, and real-time analysis. This method considers the influence of geological conditions, groundwater, and construction procedures within the foundation pit site area, and can quantitatively predict the changing trends of each construction stage of the foundation pit, including but not limited to deformation of the support structure, changes in groundwater level, changes in support axial force, and the influence of the surrounding environment. This method can effectively reduce the risks of foundation pit construction and also provide assistance for adjusting measures, controlling costs, and conducting scientific research during the foundation pit construction process. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to address the shortcomings of the existing technology by providing a method and system for intelligent monitoring of deep foundation pits based on multi-source data fusion.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0007] The intelligent monitoring method for deep foundation pits based on multi-source data fusion includes the following steps:

[0008] Step S1: Collect multi-physics field data of the foundation pit in real time through a distributed sensor network;

[0009] Step S2: Perform preprocessing operations on the multiphysics data of the foundation pit based on edge computing nodes;

[0010] Step S3: Transmit the data after the preprocessing operation to the cloud platform through a multi-mode communication network;

[0011] Step S4: Perform multi-source data fusion analysis on the cloud platform and calculate the comprehensive risk index using an improved time-varying weighted fusion algorithm;

[0012] Step S5: Determine the comprehensive risk index to achieve tiered early warning;

[0013] Step S6: Implement corresponding automatic control measures according to the warning level.

[0014] Furthermore, step S1 specifically includes the following steps:

[0015] Step S1.1: Install vibrating wire sensors at the stress-bearing parts of the foundation pit support structure to monitor the internal forces of the support structure;

[0016] Step S1.2: Install a fiber optic strain sensor array along the depth of the foundation pit to monitor the horizontal displacement of the support structure;

[0017] Step S1.3: Install earth pressure cells in layers in the soil around the foundation pit to monitor active and passive earth pressure.

[0018] Step S1.4: Install water level sensors around the foundation pit to monitor changes in groundwater level;

[0019] Step S1.5: Install displacement sensors and tilt sensors on the walls of the foundation pit structure to monitor the settlement and tilt of the structure.

[0020] Furthermore, step S2 specifically includes the following steps:

[0021] Step S2.1: Perform temperature compensation and nonlinear correction on the frequency signal obtained from the vibrating wire sensor to obtain the internal force value of the support structure. The specific formula is as follows:

[0022]

[0023] in, This represents the measured force value, i.e., the internal force of the supporting structure. This represents the temperature-dependent sensitivity coefficient. Indicates the measured vibration frequency. Indicates the initial frequency related to temperature. This represents the coefficient of thermal expansion of the sensor material. This represents the difference between the current temperature and the reference temperature. Indicates the effective cross-sectional area of ​​the sensor. Indicates the elastic modulus of the sensor;

[0024] Step S2.2: Integrate the strain signal from the strain sensor array obtained from monitoring to reconstruct the horizontal displacement of the soil or support structure along the depth. The specific formula is as follows:

[0025]

[0026] in, This represents the horizontal displacement of the support structure at depth z. Indicates the initial rotation angle at the pile top. Indicates depth curvature at that point This represents the geometric nonlinearity correction factor. Indicates the inner integration variable. Indicates the outer integral variable;

[0027] Step S2.3: Perform temperature drift compensation and zero-point calibration on the original electrical signal of the earth pressure cell;

[0028] Step S2.4: Perform moving average filtering on the raw data from the water level sensor, displacement sensor, and tilt sensor.

[0029] Furthermore, in step S2.1, the specific formula for the temperature-related sensitivity coefficient is as follows:

[0030]

[0031] in, This represents the sensitivity coefficient at the reference temperature. and These represent the primary temperature coefficient and the secondary temperature coefficient, respectively.

[0032] The specific formula for the temperature-related initial frequency is as follows:

[0033]

[0034] in, This represents the initial frequency under no-load conditions. and These represent the primary temperature coefficient and the secondary temperature coefficient of the frequency, respectively.

[0035] Furthermore, step S3 specifically includes the following steps:

[0036] Step S3.1: Real-time assessment of communication link quality, monitoring indicators include signal strength, bit error rate, and transmission delay;

[0037] Step S3.2: Dynamically select the transmission path based on the communication link quality, and select the link with the highest communication link quality among 5G, NB-IoT and LoRaWAN for transmission;

[0038] Step S3.3: Start local data caching when communication is interrupted, and replenish the transmitted cached data after the connection is restored.

[0039] Furthermore, in step S4, the specific formula for the comprehensive risk index is as follows:

[0040]

[0041] in, This represents the comprehensive risk index at time t. This represents the time-varying weighting coefficient of the i-th monitoring parameter at time t. Indicates the total number of monitored parameters. This represents the single risk indicator of the i-th monitoring parameter at time t. Indicates the influence coefficient of the rate of change. This represents the j-th key parameter. This indicates the total number of key parameters, which include the internal forces of the support structure and the horizontal displacement of the support structure.

[0042] The specific formula for the time-varying weighting coefficient is as follows:

[0043]

[0044] in, This represents the baseline weight of the i-th monitoring parameter. Indicates the volatility sensitivity coefficient. Indicates the time adjustment factor. This represents the volatility contribution factor.

[0045] Furthermore, in step S5, the multi-level early warning specifically includes:

[0046] A Level 1 warning is triggered when the comprehensive risk index is greater than or equal to 0.3 and less than 0.6; a Level 2 warning is triggered when the comprehensive risk index is greater than or equal to 0.6 and less than 0.85; and a Level 3 warning is triggered when the comprehensive risk index is greater than or equal to 0.85.

[0047] Furthermore, in step S6, the automatic control measures specifically include:

[0048] During a Level 1 early warning response, the monitoring frequency is increased to once every 30 seconds, and a text message is sent to notify on-site technical personnel.

[0049] During a Level II early warning response, the monitoring frequency is increased to once every 10 seconds, video surveillance linkage is activated, and a notification is sent to the on-site supervisor for verification.

[0050] During a Level 3 early warning response, the monitoring frequency is increased to once per second, automatically triggering audible and visual alarms, and construction is stopped if necessary.

[0051] The intelligent monitoring system for deep foundation pits based on multi-source data fusion, implemented according to any one of the aforementioned intelligent monitoring methods for deep foundation pits based on multi-source data fusion, includes:

[0052] Distributed sensor network, including vibrating wire sensors, fiber optic strain sensors, earth pressure cells, water level sensors and displacement sensors;

[0053] Edge computing nodes consist of edge computing nodes deployed on-site and are used for data preprocessing and local decision-making.

[0054] A multi-mode communication module to support 5G, NB-IoT and LoRaWAN communication;

[0055] A cloud service platform is used to provide data storage and risk index calculation services;

[0056] The remote monitoring terminal supports access via Web platform and mobile App, and is used to realize the release of early warning information and linkage control;

[0057] The early warning module is used to issue graded early warnings based on the comprehensive risk index and to execute corresponding automatic control measures.

[0058] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0059] 1. This invention achieves synchronous monitoring of the internal forces of the support structure, deep horizontal displacement, earth pressure, groundwater level and deformation of the surrounding environment by constructing a distributed monitoring network that includes multiple types of sensors such as vibrating wire sensors, fiber optic strain sensors, earth pressure cells, and water level gauges.

[0060] 2. This invention constructs a comprehensive risk index and achieves graded risk identification by setting threshold parameters of different levels, which not only ensures the timeliness of early warning but also effectively avoids false alarms.

[0061] 3. By constructing a time-varying weight algorithm, the system can automatically adjust the monitoring strategy according to the characteristics of the construction stage and changes in environmental conditions.

[0062] 4. The deployment of edge computing nodes in this invention enables local data preprocessing and intelligent decision-making, improves the real-time performance and reliability of the system, reduces the risks associated with network dependence, and can still maintain the ability to process and assess the quality of critical data locally even in the extreme case of network interruption.

[0063] 5. This invention effectively eliminates the influence of ambient temperature changes on the frequency-force conversion relationship by performing precise temperature compensation and nonlinear correction on the vibrating wire sensor; and by performing curvature integral calculation on the strain signal of the strain sensor array, it realizes the reconstruction of the horizontal displacement field of the soil or support structure along the depth direction, replacing the traditional monitoring method that must rely on manual operation. Attached Figure Description

[0064] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0065] Figure 1 This is a flowchart illustrating an embodiment of the present invention;

[0066] Figure 2 This is a system schematic diagram according to an embodiment of the present invention. Detailed Implementation

[0067] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0068] like Figure 1 As shown, the intelligent monitoring method for deep foundation pits based on multi-source data fusion includes the following steps:

[0069] Step S1: Collect multi-physics field data of the foundation pit in real time through a distributed sensor network;

[0070] Step S2: Perform preprocessing operations on the multiphysics data of the foundation pit based on edge computing nodes;

[0071] Step S3: Transmit the data after the preprocessing operation to the cloud platform through a multi-mode communication network;

[0072] Step S4: Perform multi-source data fusion analysis on the cloud platform and calculate the comprehensive risk index using an improved time-varying weighted fusion algorithm;

[0073] Step S5: Determine the comprehensive risk index to achieve tiered early warning;

[0074] Step S6: Implement corresponding automatic control measures according to the warning level.

[0075] Step S1 specifically includes the following steps:

[0076] Step S1.1: Install vibrating wire sensors at the stress-bearing parts of the foundation pit support structure to monitor the internal forces of the support structure;

[0077] Step S1.2: Fiber optic strain sensor arrays are deployed along the depth of the foundation pit to monitor the horizontal displacement of the support structure. The arrays are deployed at 0.5m intervals along the depth of the foundation pit to form a strain monitoring network.

[0078] Step S1.3: Install earth pressure cells in layers in the soil around the foundation pit to monitor active and passive earth pressure.

[0079] Step S1.4: Install water level sensors around the foundation pit to monitor changes in groundwater level;

[0080] Step S1.5: Install displacement sensors and tilt sensors on the walls of the foundation pit structure to monitor the settlement and tilt of the structure.

[0081] All sensors use a unified timestamp to ensure data timing consistency; the basic sampling frequency is set to once every 2 minutes, and the frequency for critical parts is increased to once every 30 seconds.

[0082] Step S2 specifically includes the following steps:

[0083] Step S2.1: Perform temperature compensation and nonlinear correction on the frequency signal obtained from the vibrating wire sensor to obtain the internal force value of the support structure. The specific formula is as follows:

[0084]

[0085] in, This represents the measured force value, i.e., the internal force of the supporting structure. This represents the temperature-dependent sensitivity coefficient. Indicates the measured vibration frequency. Indicates the initial frequency related to temperature. This indicates the coefficient of thermal expansion of the sensor material; for stainless steel, it is 1.2 × 10⁻⁵ / ℃. This represents the difference between the current temperature and the reference temperature. This represents the effective cross-sectional area of ​​the sensor, which is determined by the model; a typical value is 3.14 × 10⁻⁴. Indicates the elastic modulus of the sensor;

[0086] Step S2.2: Integrate the strain signal from the strain sensor array obtained from monitoring to reconstruct the horizontal displacement of the soil or support structure along the depth. The specific formula is as follows:

[0087]

[0088] in, This represents the horizontal displacement of the support structure at depth z. Indicates the initial rotation angle at the pile top. Indicates depth The curvature of the support structure, This represents the geometric nonlinearity correction factor. Indicates the inner integration variable. Indicates the outer integral variable;

[0089] The specific formula for the geometric nonlinearity correction factor is as follows:

[0090]

[0091] in, Indicates the height of the support piles;

[0092] depth The curvature of the support structure is the ratio of the difference between the microstrain on the tension side and the microstrain on the compression side to the distance of the sensor from the neutral axis;

[0093] Step S2.3: Perform temperature drift compensation and zero-point calibration on the original electrical signal of the earth pressure cell;

[0094] Step S2.4: Perform moving average filtering on the raw data from the water level sensor, displacement sensor, and tilt sensor.

[0095] Among them, the vibrating wire sensor monitors the internal forces of the support structure, such as the axial force of the support and the bending moment of the retaining pile. It senses the stress on the structure by measuring the change in the vibration frequency of the steel wire. There is a clear conversion relationship between its signal frequency and the stress, but it is significantly affected by temperature, so complex temperature compensation preprocessing is required.

[0096] Deep horizontal displacement reconstruction monitors the horizontal movement at different depths within the soil or support structure. It involves measuring the bending strain of a pre-embedded inclinometer tube or flexible inclinometer, and then reconstructing the displacement of the entire profile using mathematical methods such as integration.

[0097] In step S2.1, the specific formula for the temperature-related sensitivity coefficient is as follows:

[0098]

[0099] in, This represents the sensitivity coefficient at a reference temperature of 20°C, with a typical value of 0.024 kN / Hz². and These represent the primary and secondary temperature coefficients, respectively, determined through experiments in a temperature chamber ranging from -10℃ to +60℃. Typical values ​​are -3.5×10^-4 / ℃ and 2.1×10^-6 / ℃², respectively.

[0100] The specific formula for the temperature-related initial frequency is as follows:

[0101]

[0102] in, This represents the initial frequency under no-load conditions, with a typical value of 1500Hz. and These represent the primary and secondary temperature coefficients of the frequency, respectively, with typical values ​​of -2.8 × 10^-4 / ℃ and 1.5 × 10^-6 / ℃².

[0103] Step S3 specifically includes the following steps:

[0104] Step S3.1: Real-time assessment of communication link quality, monitoring indicators include signal strength, bit error rate, and transmission delay;

[0105] Step S3.2: Dynamically select the transmission path based on the communication link quality, and select the link with the highest communication link quality among 5G, NB-IoT and LoRaWAN for transmission;

[0106] Step S3.3: Start local data caching when communication is interrupted, and replenish the transmitted cached data after the connection is restored.

[0107] The communication link quality assessment formula in step S3.2 is as follows:

[0108]

[0109] in, Indicates the quality of the communication link. Indicates signal strength. Indicates bit error rate. Indicates transmission delay;

[0110] The local data cache has a capacity of 72 hours of full-frequency data. After the connection is restored, the data is retransmitted in order of risk level, and a data integrity report during the network outage is automatically generated.

[0111] In step S4, the specific formula for the comprehensive risk index is as follows:

[0112]

[0113] in, This represents the comprehensive risk index at time t. This represents the time-varying weighting coefficient of the i-th monitoring parameter at time t. Indicates the total number of monitored parameters. This represents the single risk indicator of the i-th monitoring parameter at time t. This represents the influence coefficient of the rate of change, typically ranging from 0.1 to 0.3. This represents the j-th key parameter. This represents the total number of key parameters. Key parameters are those that are particularly sensitive to changes and are selected from all parameters. These key parameters include the internal forces of the support structure and the horizontal displacement of the support structure.

[0114] The specific formula for the individual risk indicator is as follows:

[0115]

[0116] in, and These represent the maximum and minimum allowed values ​​for the i-th parameter, respectively. This represents the measured value of the i-th parameter at time t. This represents the safety reference value of the i-th parameter. This represents the nonlinear amplification index, typically ranging from 1.5 to 2.0, used to amplify the risk of exceeding limits.

[0117] The specific formula for the time-varying weighting coefficient is as follows:

[0118]

[0119] in, This represents the baseline weight of the i-th monitoring parameter. This represents the fluctuation sensitivity coefficient, typically with a value of 0.15. Indicates the time adjustment factor. This represents the volatility contribution factor.

[0120] The specific weights for the benchmarks are: displacement 0.35, internal force 0.25, earth pressure 0.20, water level 0.15, and settlement and tilt 0.05.

[0121] The specific formulas for the time adjustment factor and the volatility contribution factor are as follows:

[0122]

[0123]

[0124] in, This indicates key construction phase dates, such as the start date of foundation pit excavation, the date of support removal, and the date of extreme weather events such as heavy rain. This represents the time decay constant, taken as 15 days for the excavation stage and 30 days for the support stage. This represents the standard deviation of the i-th parameter over a 24-hour sliding window at time t. This represents the mean of the sliding window for the i-th parameter. The calculation window is the same as the volatility calculation window, and is set to 24 hours.

[0125] The time adjustment factor automatically increases the importance of relevant monitoring parameters near the critical construction stage. When time t is closer to the critical time point and closer to 0, the value of A(t) is closer to 1. When the absolute value of time t is far away from the critical time point, the value of A(t) approaches 0.

[0126] The volatility contribution factor identifies and focuses on parameters that have recently exhibited unstable or abnormal behavior. The larger B(t) is, the more drastic the relative volatility of the parameter and the more unstable it is.

[0127] In step S5, the multi-level early warning specifically includes:

[0128] A Level 1 warning is triggered when the comprehensive risk index is greater than or equal to 0.3 and less than 0.6; a Level 2 warning is triggered when the comprehensive risk index is greater than or equal to 0.6 and less than 0.85; and a Level 3 warning is triggered when the comprehensive risk index is greater than or equal to 0.85.

[0129] In step S6, the automatic control measures specifically include:

[0130] During a Level 1 early warning response, the monitoring frequency is increased to once every 30 seconds, and a text message is sent to notify on-site technical personnel.

[0131] During a Level II early warning response, the monitoring frequency is increased to once every 10 seconds, video surveillance linkage is activated, and a notification is sent to the on-site supervisor for verification.

[0132] During a Level 3 early warning response, the monitoring frequency is increased to once per second, automatically triggering audible and visual alarms, and construction is stopped if necessary.

[0133] like Figure 2 As shown, the intelligent monitoring system for deep foundation pits based on multi-source data fusion is implemented based on any one of the aforementioned intelligent monitoring methods for deep foundation pits based on multi-source data fusion, including:

[0134] Distributed sensor network, including vibrating wire sensors, fiber optic strain sensors, earth pressure cells, water level sensors and displacement sensors;

[0135] Edge computing nodes consist of edge computing nodes deployed on-site and are used for data preprocessing and local decision-making.

[0136] A multi-mode communication module to support 5G, NB-IoT and LoRaWAN communication;

[0137] A cloud service platform is used to provide data storage and risk index calculation services;

[0138] The remote monitoring terminal supports access via Web platform and mobile App, and is used to realize the release of early warning information and linkage control;

[0139] The early warning module is used to issue graded early warnings based on the comprehensive risk index and to execute corresponding automatic control measures.

[0140] The distributed sensor network adopts a multi-redundancy design, with primary and backup dual sensors deployed at key monitoring points. It improves measurement reliability through data fusion and cross-validation, and has automatic fault diagnosis and sensor health status monitoring functions.

[0141] The edge computing gateway adopts a modular design, supports automatic identification and parsing of multiple sensor protocols, has local data quality assessment and anomaly detection capabilities, and integrates a hardware encryption module to ensure data transmission security.

[0142] The examples described herein are merely preferred embodiments of the invention and are not intended to limit the concept and scope of the invention. Any modifications and improvements made by those skilled in the art to the technical solutions of the invention without departing from the design concept of the invention should fall within the protection scope of the invention.

Claims

1. A method for intelligent monitoring of deep foundation pits based on multi-source data fusion, characterized in that, Includes the following steps: Step S1: Collect multi-physics field data of the foundation pit in real time through a distributed sensor network; Step S2: Perform preprocessing operations on the multiphysics data of the foundation pit based on edge computing nodes; Step S3: Transmit the data after the preprocessing operation to the cloud platform through a multi-mode communication network; Step S4: Perform multi-source data fusion analysis on the cloud platform and calculate the comprehensive risk index using an improved time-varying weighted fusion algorithm; Step S5: Determine the comprehensive risk index to achieve tiered early warning; Step S6: Implement corresponding automatic control measures according to the warning level; Specifically, step S2 includes the following steps: Step S2.1: Perform temperature compensation and nonlinear correction on the frequency signal obtained from the vibrating wire sensor to obtain the internal force value of the support structure. The specific formula is as follows: ； in, This represents the measured force value, i.e., the internal force of the supporting structure. This represents the temperature-dependent sensitivity coefficient. Indicates the measured vibration frequency. Indicates the initial frequency related to temperature. This represents the coefficient of thermal expansion of the sensor material. This represents the difference between the current temperature and the reference temperature. Indicates the effective cross-sectional area of ​​the sensor. Indicates the elastic modulus of the sensor; Step S2.2: Integrate the strain signal from the strain sensor array obtained from monitoring to reconstruct the horizontal displacement of the soil or support structure along the depth. The specific formula is as follows: ； in, This represents the horizontal displacement of the support structure at depth z. Indicates the initial rotation angle at the pile top. Indicates depth curvature at that point This represents the geometric nonlinearity correction factor. Indicates the inner integration variable. Indicates the outer integral variable; Step S2.3: Perform temperature drift compensation and zero-point calibration on the original electrical signal of the earth pressure cell; Step S2.4: Perform moving average filtering on the raw data from the water level sensor, displacement sensor, and tilt sensor; In step S4, the specific formula for the comprehensive risk index is as follows: in, This represents the comprehensive risk index at time t. This represents the time-varying weighting coefficient of the i-th monitoring parameter at time t. Indicates the total number of monitored parameters. This represents the single risk indicator of the i-th monitoring parameter at time t. Indicates the influence coefficient of the rate of change. This represents the j-th key parameter. This indicates the total number of key parameters, which include the internal forces of the support structure and the horizontal displacement of the support structure. The specific formula for the time-varying weighting coefficient is as follows: ； in, This represents the baseline weight of the i-th monitoring parameter. Indicates the volatility sensitivity coefficient. Indicates the time adjustment factor. Indicates the volatility contribution factor; In step S5, the tiered early warning specifically includes: A Level 1 warning is triggered when the comprehensive risk index is greater than or equal to 0.3 and less than 0.6; a Level 2 warning is triggered when the comprehensive risk index is greater than or equal to 0.6 and less than 0.85; and a Level 3 warning is triggered when the comprehensive risk index is greater than or equal to 0.

85. In step S6, the automatic control measures specifically include: During a Level 1 early warning response, the monitoring frequency is increased to once every 30 seconds, and a text message is sent to notify on-site technical personnel. During a Level II early warning response, the monitoring frequency is increased to once every 10 seconds, video surveillance linkage is activated, and a notification is sent to the on-site supervisor for verification. During a Level 3 early warning response, the monitoring frequency is increased to once per second, automatically triggering audible and visual alarms, and construction is stopped if necessary.

2. The method according to claim 1, characterized in that, Step S1 specifically includes the following steps: Step S1.1: Install vibrating wire sensors at the stress-bearing parts of the foundation pit support structure to monitor the internal forces of the support structure; Step S1.2: Install a fiber optic strain sensor array along the depth of the foundation pit to monitor the horizontal displacement of the support structure; Step S1.3: Install earth pressure cells in layers in the soil around the foundation pit to monitor active and passive earth pressure. Step S1.4: Install water level sensors around the foundation pit to monitor changes in groundwater level; Step S1.5: Install displacement sensors and tilt sensors on the walls of the foundation pit structure to monitor the settlement and tilt of the structure.

3. The method according to claim 2, characterized in that, In step S2.1, the specific formula for the temperature-related sensitivity coefficient is as follows: ； in, This represents the sensitivity coefficient at the reference temperature. and These represent the primary temperature coefficient and the secondary temperature coefficient, respectively. The specific formula for the temperature-related initial frequency is as follows: ； in, This represents the initial frequency under no-load conditions. and These represent the primary temperature coefficient and the secondary temperature coefficient of the frequency, respectively.

4. The method according to claim 3, characterized in that, Step S3 specifically includes the following steps: Step S3.1: Real-time assessment of communication link quality, monitoring indicators include signal strength, bit error rate, and transmission delay; Step S3.2: Dynamically select the transmission path based on the communication link quality, and select the link with the highest communication link quality among 5G, NB-IoT and LoRaWAN for transmission; Step S3.3: Start local data caching when communication is interrupted, and replenish the transmitted cached data after the connection is restored.

5. A deep foundation pit intelligent monitoring system based on multi-source data fusion, implemented based on the deep foundation pit intelligent monitoring method based on multi-source data fusion as described in any one of claims 1-4, characterized in that, include: Distributed sensor network, including vibrating wire sensors, fiber optic strain sensors, earth pressure cells, water level sensors and displacement sensors; Edge computing nodes consist of edge computing nodes deployed in the field, used for data preprocessing and local decision-making; A multi-mode communication module to support 5G, NB-IoT and LoRaWAN communication; A cloud service platform is used to provide data storage and risk index calculation services; The remote monitoring terminal supports access via Web platform and mobile App, and is used to realize the release of early warning information and linkage control; The early warning module is used to issue graded early warnings based on the comprehensive risk index and to execute corresponding automatic control measures.

Images