Urban land subsidence monitoring and early warning method and device

Abstract

The present invention discloses a method for monitoring and early warning of urban ground subsidence, comprising the following steps: S1, deploying a split monitoring system, physically isolating an underground probe module from a ground relay station, wherein the underground probe module includes a trimodal sensing unit, an edge computing unit, a self-powered system, and a wireless communication module; S2, establishing a three-dimensional coordinate system for the monitoring area, calibrating the spatial position relationship of each sensor, and converting displacement meter data into vector displacement in the regional coordinate system; S3, calculating the subsidence risk probability based on the D-S evidence theory and generating an early warning level; S4, calculating the early warning level based on the basic threshold V base and seasonal correction factor k season Determine the final sedimentation rate threshold V th , triggering a graded early warning response based on cumulative settlement, displacement mutation, and inclination angle thresholds. This method offers higher monitoring precision, higher settlement identification accuracy, dynamic adjustment of thresholds, and the ability to implement graded early warnings. The present invention also discloses an urban land subsidence monitoring and early warning device.

Description

Technical Field

[0001] The present invention relates to the technical field of land subsidence monitoring, and in particular to a method and device for monitoring and early warning of urban land subsidence. Background Art

[0002] Traditional ground subsidence monitoring devices have defects such as isolated single-point sensor data, poor power supply continuity, and high false alarm rate; conventional equipment is mostly deployed in a centralized manner, which makes underground cables easy to damage and difficult to maintain, and the cost is high; existing early warning systems mostly use fixed thresholds and cannot adapt to the elastic deformation characteristics of the soil caused by seasonal groundwater level fluctuations. Summary of the Invention

[0003] In view of the shortcomings of the background technology, the technical problem to be solved by the present invention is to provide an urban land subsidence monitoring and early warning method with higher monitoring accuracy, higher settlement identification accuracy, dynamic adjustment of thresholds, and realization of graded early warning.

[0004] To this end, the present invention is achieved by adopting the following technical solutions:

[0005] The urban land subsidence monitoring and early warning method is characterized by comprising the following steps:

[0006] S1. Deploy a split monitoring system, physically isolate the underground probe modules from the ground relay station, drill holes at intervals on the ground in the monitoring area and install the underground probe modules, and install the ground relay station on the ground in the monitoring area;

[0007] The underground probe module includes a trimodal sensing unit, an edge computing unit, a self-powered system, and a wireless communication module. The trimodal sensing unit is composed of a distributed fiber grating sensor, a high-precision displacement meter, and a capacitive tilt sensor. The edge computing unit performs local preprocessing on the data of the trimodal sensing unit.

[0008] The ground relay station includes an energy supply system, a data relay module and an alarm device;

[0009] S2. Establish a three-dimensional coordinate system for the monitoring area, calibrate the spatial position relationship of each sensor, convert the displacement meter data into vector displacement in the regional coordinate system, and synchronize the time of all sensor data by using the built-in GPS taming clock module in the underground probe module;

[0010] S3. Normalize and unify the dimensions of the FBG wavelength and displacement meter voltage, extract the statistical and morphological characteristics of each sensor data, calculate the subsidence risk probability based on the DS evidence theory, and generate an early warning level;

[0011] S4, through the LSTM network model, takes environmental parameters, historical settlement data and engineering parameters as input, and outputs the seasonal correction factor k season; Based on the basic threshold V base and seasonal correction factor k season Determine the final sedimentation rate threshold V th , and trigger graded early warning responses based on cumulative settlement, displacement mutation and inclination threshold.

[0012] Furthermore, the edge computing unit in S1 adopts an STM32H743VIT6 microcontroller chip, which is connected to a Kalman filter. The Kalman filter fuses the optical fiber strain, displacement and inclination data to output the sedimentation rate, inclination angle and confidence index. The self-powered system includes a geothermal temperature difference generator and a lithium titanate battery. The wireless communication module includes a LoRa chip and an NB-IoT chip. The LoRa chip and the NB-IoT chip form dual-mode communication and automatically switch according to the signal strength. The distributed fiber grating sensor is arranged in a serpentine shape along the monitoring area, with a node spacing of 10m and a burial depth of 1.5-3m. The high-precision displacement meter is a magnetostrictive displacement meter, and the capacitive inclination sensor is a dual-axis MEMS inclinometer.

[0013] Furthermore, the energy supply system in S1 includes solar panels and lithium iron phosphate batteries, the data relay module includes a 4G / 5G communication module, a LoRa gateway for receiving data from the underground probe module and a storage card for caching data, and the alarm device includes an audible and visual alarm and a digital display screen.

[0014] Furthermore, the environmental parameters in S4 include real-time rainfall, groundwater level, and temperature from the municipal monitoring network; historical settlement data include settlement rate and soil moisture content over the same period of the past three years; and engineering parameters include allowable building settlement, tunnel convergence threshold, and foundation threshold V base According to the building safety regulations: such as subway tunnel V base =2mm / month, seasonal correction factor k season Output from LSTM network: rainy season k season =1.2, dry season k season =0.8, the final sedimentation rate threshold V th =V base × k season .

[0015] Furthermore, the graded warning response in S4 includes:

[0016] Level 1 warning: The trigger condition is sedimentation rate ≥ 1.2V th For three days, the response measures include marking the data yellow and transmitting it to the ground relay station. The ground relay station transmits the data to the cloud platform via the 4G / 5G communication module. The cloud platform pushes the data to the inspection app, and the sound and light alarm lights up yellow.

[0017] Level 2 warning: The trigger condition is that the cumulative settlement is ≥80% of the design value. The response measure is to start the drone re-test and the sound and light alarm will be constantly blue.

[0018] Level 3 warning: The triggering condition is an inclination angle ≥ 0.5° or a sudden displacement change ≥ 10mm. The response measures are an audible and visual alarm that sounds and a red light flashes, and the associated road traffic is automatically closed.

[0019] Furthermore, the underground probe module in S1 includes a stainless steel shell, which has a sensor cabin for placing a tri-modal sensing unit, a computing cabin for placing an edge computing unit, and a power cabin for placing a self-powered system and a wireless communication module, and the surface of the shell is provided with spiral guide fins.

[0020] After adopting the above technical solution, this method isolates the underground probe module from the ground relay station through a split architecture, and combines the multi-sensor data fusion and adaptive threshold algorithm of distributed fiber grating sensors, high-precision displacement meters and capacitive inclination sensors to break through the data silos, environmental fragility and high cost limitations of existing technologies, achieving higher monitoring accuracy, higher settlement identification accuracy, dynamic adjustment of thresholds, and implementation of graded early warning.

[0021] The present invention also provides an urban ground subsidence monitoring and early warning device, which is characterized in that it includes an underground probe module located underground in the monitoring area and a ground relay station located on the ground in the monitoring area. The underground probe module includes an outer shell and a three-modal sensing unit, an edge computing unit, a self-powered system and a wireless communication module located in the outer shell. The ground relay station includes a column inserted into the ground and a power supply system, a data relay module and an alarm device located on the column.

[0022] Furthermore, the trimodal sensing unit is composed of a distributed fiber grating sensor, a high-precision displacement meter and a capacitive inclination sensor. The edge computing unit includes a circuit board and an STM32H743VIT6 microcontroller chip provided on the circuit board. The circuit board is provided with a Kalman filter. The distributed fiber grating sensor, the high-precision displacement meter and the capacitive inclination sensor are connected to the STM32H743VIT6 microcontroller chip. The self-powered system includes a geothermal temperature difference generator and a lithium titanate battery connected to each other. The lithium titanate battery is connected to the circuit board. The wireless communication module includes a LoRa chip and an NB-IoT chip provided on the circuit board and connected to the STM32H743VIT6 microcontroller chip. The LoRa chip and the NB-IoT chip form dual-mode communication.

[0023] Furthermore, the energy supply system includes interconnected solar panels and lithium iron phosphate batteries, the data relay module includes a circuit board and a 4G / 5G communication module arranged on the circuit board, a LoRa gateway for receiving data from the underground probe module and a storage card for caching data, the alarm device includes an audible and visual alarm and a digital display connected to the circuit board, the audible and visual alarm and the digital display are powered by lithium iron phosphate batteries, the 4G / 5G communication module is connected to the cloud platform, and the cloud platform has a patrol APP for two-way communication.

[0024] Furthermore, the shell is a stainless steel shell, which has a sensor cabin for placing a tri-modal sensing unit, a computing cabin for placing an edge computing unit, and a power cabin for placing a self-powered system and a wireless communication module, and the surface of the shell is provided with spiral guide fins. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The present invention has the following accompanying drawings:

[0026] Figure 1 This is a schematic block diagram of the circuit structure of the present invention. DETAILED DESCRIPTION

[0027] In order to further illustrate the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, the specific implementation methods, structures, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0028] Referring to the above-mentioned figures, the urban land subsidence monitoring and early warning method provided by the present invention is characterized by comprising the following steps:

[0029] S1. Deploy a split monitoring system, physically isolate the underground probe modules from the ground relay station, drill holes at intervals on the ground in the monitoring area and install the underground probe modules, and install the ground relay station on the ground in the monitoring area;

[0030] The underground probe module includes a trimodal sensing unit, an edge computing unit, a self-powered system, and a wireless communication module. The trimodal sensing unit is composed of a distributed fiber grating sensor, a high-precision displacement meter, and a capacitive tilt sensor. The edge computing unit performs local preprocessing on the data of the trimodal sensing unit.

[0031] The ground relay station includes an energy supply system, a data relay module and an alarm device;

[0032] S2. Establish a three-dimensional coordinate system for the monitoring area, calibrate the spatial position relationship of each sensor, convert the displacement meter data into vector displacement in the regional coordinate system, and synchronize the time of all sensor data by using the built-in GPS taming clock module in the underground probe module;

[0033] S3. Normalize and unify the dimensions of the FBG wavelength and displacement meter voltage, extract the statistical and morphological characteristics of each sensor data, calculate the subsidence risk probability based on the DS evidence theory, and generate an early warning level;

[0034] S4, through the LSTM network model, takes environmental parameters, historical settlement data and engineering parameters as input, and outputs the seasonal correction factor k season ; Based on the basic threshold V base and seasonal correction factor k season Determine the final sedimentation rate threshold V th , and trigger graded early warning responses based on cumulative settlement, displacement mutation and inclination threshold.

[0035] The edge computing unit in S1 uses an STM32H743VIT6 microcontroller chip, which is connected to a Kalman filter. The Kalman filter fuses the optical fiber strain, displacement and inclination data to output the sedimentation rate, inclination angle and confidence index. The self-powered system includes a geothermal temperature difference generator and a lithium titanate battery. The wireless communication module includes a LoRa chip and an NB-IoT chip. The LoRa chip and the NB-IoT chip form dual-mode communication and automatically switch according to the signal strength. The distributed fiber grating sensor is arranged in a serpentine shape along the monitoring area, with a node spacing of 10m and a burial depth of 1.5-3m. The high-precision displacement meter is a magnetostrictive displacement meter, and the capacitive inclination sensor is a dual-axis MEMS inclinometer.

[0036] The energy supply system in S1 includes solar panels and lithium iron phosphate batteries, the data relay module includes a 4G / 5G communication module, a LoRa gateway for receiving data from the underground probe module and a storage card for caching data, and the alarm device includes an audible and visual alarm and a digital display screen.

[0037] The environmental parameters in S4 include real-time rainfall, groundwater level, and temperature from the municipal monitoring network; historical settlement data include settlement rate and soil moisture content over the same period of the past three years; engineering parameters include building allowable settlement, tunnel convergence threshold, foundation threshold V base According to the building safety regulations: such as subway tunnel V base =2mm / month, seasonal correction factor k season Output from LSTM network: rainy season k season =1.2 / month, dry season k season =0.8, the final sedimentation rate threshold V th =V base ×k season .

[0038] The graded warning response in S4 includes:

[0039] Level 1 warning: The trigger condition is sedimentation rate ≥ 1.2V th For three days, the response measures include marking the data yellow and transmitting it to the ground relay station. The ground relay station transmits the data to the cloud platform via the 4G / 5G communication module. The cloud platform pushes the data to the inspection app, and the sound and light alarm lights up yellow.

[0040] Level 2 warning: The trigger condition is that the cumulative settlement is ≥80% of the design value. The response measure is to start the drone re-test and the sound and light alarm will be constantly blue.

[0041] Level 3 warning: The triggering condition is an inclination angle ≥ 0.5° or a sudden displacement change ≥ 10mm. The response measures are an audible and visual alarm that sounds and a red light flashes, and the associated road traffic is automatically closed.

[0042] The underground probe module in S1 includes a stainless steel shell, which has a sensor cabin for placing a tri-modal sensing unit, a computing cabin for placing an edge computing unit, and a power cabin for placing a self-powered system and a wireless communication module. The surface of the shell is provided with spiral guide fins.

[0043] The present invention also provides an urban ground subsidence monitoring and early warning device, comprising an underground probe module arranged underground in the monitoring area and a ground relay station arranged on the ground in the monitoring area, wherein the underground probe module comprises a shell and a trimodal sensing unit, an edge computing unit, a self-powered system and a wireless communication module arranged in the shell, the ground relay station comprises a column inserted into the ground and an energy supply system, a data relay module and an alarm device arranged on the column, the trimodal sensing unit is composed of a distributed fiber grating sensor, a high-precision displacement meter and a capacitive tilt sensor, the edge computing unit comprises a circuit board and an STM32H743VIT6 microcontroller chip arranged on the circuit board, a Kalman filter connected to the STM32H743VIT6 microcontroller chip is provided on the circuit board, the distributed fiber grating sensor, the high-precision displacement meter and the capacitive tilt sensor are connected to the STM32H743VIT6 microcontroller chip, and the self-powered system comprises a geothermal temperature difference generator and a lithium titanate generator connected to each other. The battery, lithium titanate battery and circuit board are connected, the wireless communication module includes a LoRa chip and an NB-IoT chip arranged on the circuit board and connected to the STM32H743VIT6 microcontroller chip, the LoRa chip and the NB-IoT chip form dual-mode communication, the energy supply system includes mutually connected solar panels and lithium iron phosphate batteries, the data relay module includes a circuit board and a 4G / 5G communication module arranged on the circuit board, a LoRa gateway for receiving data from the underground probe module and a storage card for caching data, the 4G / 5G communication module is connected to the cloud platform, and the cloud platform has a two-way communication inspection APP, the alarm device includes an audible and visual alarm and a digital display connected to the circuit board, the audible and visual alarm and the digital display are powered by a lithium iron phosphate battery, the shell is a stainless steel shell, the stainless steel shell has a sensor cabin for accommodating a trimodal sensing unit, a computing cabin for accommodating an edge computing unit, a power cabin for accommodating a self-powered system and a wireless communication module, and the surface of the shell is provided with spiral guide fins.

[0044] The urban land subsidence monitoring and early warning method and device provided by the present invention uses a split architecture to isolate the underground probe module from the ground relay station. It combines multi-sensor data fusion and adaptive threshold algorithm of distributed fiber grating sensors, high-precision displacement meters and capacitive inclinometers, breaking through the data island, environmental vulnerability and high cost limitations of existing technologies. It has higher monitoring accuracy, higher settlement identification accuracy, dynamic adjustment of thresholds, and realizes graded early warning. The distributed fiber grating sensors are arranged in a serpentine pattern along the monitoring area, with a node spacing of 10m and a burial depth of 1.5-3m, avoiding the shallow backfill soil disturbance layer. The magnetostrictive displacement meter has a range of ±50mm, a resolution of 0.001mm, a sampling frequency of 10Hz, a dual-axis MEMS inclinometer range of ±30°, and a built-in temperature compensation chip (- The 40°C to 85°C (120°F to 180°F) sensor outputs an RS485 digital signal, and a built-in FIR filter can be used to suppress high-frequency vibration noise in typhoon conditions. The STM32H743VIT6 microcontroller chip performs wavelet threshold denoising on the optical fiber wavelength drift data (using the sym8 wavelet basis and a 5-layer decomposition), calculates the standard deviation and kurtosis coefficient of the displacement meter data, identifies sudden events, and fuses the optical fiber strain, displacement, and inclination data through a Kalman filter to output the sedimentation rate, inclination angle, and confidence index. The LSTM network model includes: 1. Input layer: 30-dimensional features (including time-series environmental data and sedimentation amount); 2. Output layer: Hidden layer: 2 LSTM layers (128 neurons per layer) with a dropout rate of 0.2; 3. Output layer: Sedimentation rate threshold V for the current season. th 4. Training data: 10,000 sets of historical monitoring data (covering extreme conditions such as typhoons and droughts). Using the temperature difference between underground probes and the ground surface (ΔT ≥ 10°C), Bi2Te3 thermoelectric materials generate electricity with an output power of ≥ 2W, which is used to charge and store lithium titanate batteries for self-powering. The lithium titanate batteries have a cycle life of >10,000 cycles and support low-temperature discharge at -30°C. The LoRa chip and NB-IoT chip form dual-mode communication, automatically switching based on signal strength to ensure data transmission success in complex terrain. LoRa mode has a transmission range of 3km (in urban areas) and 20km (in suburban areas) with a data rate of 5kbps. NB-IoT mode supports China Mobile / China Telecom Band 5 / Band 8, with an average monthly data consumption of <50MB. Solar panels collect solar energy to charge and store lithium iron phosphate batteries, which operate over a wide temperature range of -20°C to 60°C and have a cycle life of >3,000 cycles. A digital display screen displays the sedimentation rate, historical curves, and warning levels of each monitoring point in real time.

Claims

1. Urban land subsidence monitoring and early warning method, its characteristics are: The following steps are involved: S1. Deploy a split monitoring system, physically isolate the underground probe modules from the ground relay station, drill holes at intervals on the ground in the monitoring area and install the underground probe modules, and install the ground relay station on the ground in the monitoring area; The underground probe module includes a trimodal sensing unit, an edge computing unit, a self-powered system, and a wireless communication module. The trimodal sensing unit is composed of a distributed fiber grating sensor, a high-precision displacement meter, and a capacitive tilt sensor. The edge computing unit performs local preprocessing on the data of the trimodal sensing unit. The ground relay station includes an energy supply system, a data relay module and an alarm device; S2. Establish a three-dimensional coordinate system for the monitoring area, calibrate the spatial position relationship of each sensor, convert the displacement meter data into vector displacement in the regional coordinate system, and synchronize the time of all sensor data by using the built-in GPS taming clock module in the underground probe module; S3. Normalize and unify the dimensions of the FBG wavelength and displacement meter voltage, extract the statistical and morphological characteristics of each sensor data, calculate the subsidence risk probability based on the DS evidence theory, and generate an early warning level; S4, through the LSTM network model, takes environmental parameters, historical settlement data and engineering parameters as input, and outputs the seasonal correction factor k season ; Based on the basic threshold V base and seasonal correction factor k season Determine the final sedimentation rate threshold V th , and trigger graded early warning responses based on cumulative settlement, displacement mutation and inclination threshold.

2. The urban land subsidence monitoring and early warning method according to claim 1 is characterized by: The edge computing unit in S1 uses an STM32H743VIT6 microcontroller chip, which is connected to a Kalman filter. The Kalman filter fuses the optical fiber strain, displacement and inclination data to output the sedimentation rate, inclination angle and confidence index. The self-powered system includes a geothermal temperature difference generator and a lithium titanate battery. The wireless communication module includes a LoRa chip and an NB-IoT chip. The LoRa chip and the NB-IoT chip form dual-mode communication and automatically switch according to the signal strength. The distributed fiber grating sensor is arranged in a serpentine shape along the monitoring area, with a node spacing of 10m and a burial depth of 1.5-3m. The high-precision displacement meter is a magnetostrictive displacement meter, and the capacitive inclination sensor is a dual-axis MEMS inclinometer.

3. The urban land subsidence monitoring and early warning method according to claim 1 is characterized by: The energy supply system in S1 includes solar panels and lithium iron phosphate batteries, the data relay module includes a 4G / 5G communication module, a LoRa gateway for receiving data from the underground probe module and a storage card for caching data, and the alarm device includes an audible and visual alarm and a digital display screen.

4. The urban land subsidence monitoring and early warning method according to claim 1 is characterized by: The environmental parameters in S4 include real-time rainfall, groundwater level, and temperature from the municipal monitoring network; historical settlement data include settlement rate and soil moisture content over the same period of the past three years; engineering parameters include building allowable settlement, tunnel convergence threshold, foundation threshold V base According to the building safety regulations: such as subway tunnel V base =2mm / month, seasonal correction factor k season Output from LSTM network: rainy season k season =1.2, dry season k season =0.8, the final sedimentation rate threshold V th =V base ×k season .

5. The urban land subsidence monitoring and early warning method according to claim 4 is characterized by: The graded warning response in S4 includes: Level 1 warning: The trigger condition is sedimentation rate ≥ 1.2V th For three days, the response measures include marking the data yellow and transmitting it to the ground relay station. The ground relay station transmits the data to the cloud platform via the 4G / 5G communication module. The cloud platform pushes the data to the inspection app, and the sound and light alarm lights up yellow. Level 2 warning: The trigger condition is that the cumulative settlement is ≥80% of the design value. The response measure is to start the drone re-test and the sound and light alarm will be constantly blue. Level 3 warning: The triggering condition is an inclination angle ≥ 0.5° or a sudden displacement change ≥ 10mm. The response measures are an audible and visual alarm that sounds and a red light flashes, and the associated road traffic is automatically closed.

6. The urban land subsidence monitoring and early warning method according to claim 1 is characterized by: The underground probe module in S1 includes a stainless steel shell, which has a sensor cabin for placing a tri-modal sensing unit, a computing cabin for placing an edge computing unit, and a power cabin for placing a self-powered system and a wireless communication module. The surface of the shell is provided with spiral guide fins.

7. A device for implementing the urban land subsidence monitoring and early warning method according to any one of claims 1 to 6, characterized in that: It includes an underground probe module installed underground in the monitoring area and a ground relay station installed on the ground in the monitoring area. The underground probe module includes an outer shell and a three-modal sensing unit, an edge computing unit, a self-powered system and a wireless communication module installed in the outer shell. The ground relay station includes a column inserted into the ground and a power supply system, a data relay module and an alarm device installed on the column.

8. The urban land subsidence monitoring and early warning device according to claim 7 is characterized by: The trimodal sensing unit is composed of a distributed fiber grating sensor, a high-precision displacement meter and a capacitive inclinometer. The edge computing unit includes a circuit board and an STM32H743VIT6 microcontroller chip provided on the circuit board. The circuit board is provided with a Kalman filter. The distributed fiber grating sensor, the high-precision displacement meter and the capacitive inclinometer are connected to the STM32H743VIT6 microcontroller chip. The high-precision displacement meter is a magnetostrictive displacement meter, and the capacitive inclinometer is a dual-axis MEMS inclinometer. The self-powered system includes a geothermal temperature difference generator and a lithium titanate battery connected to each other. The lithium titanate battery is connected to the circuit board. The wireless communication module includes a LoRa chip and an NB-IoT chip provided on the circuit board and connected to the STM32H743VIT6 microcontroller chip. The LoRa chip and the NB-IoT chip form dual-mode communication.

9. The urban land subsidence monitoring and early warning device according to claim 7 is characterized by: The energy supply system includes interconnected solar panels and lithium iron phosphate batteries. The data relay module includes a circuit board and a 4G / 5G communication module arranged on the circuit board, a LoRa gateway for receiving data from the underground probe module and a storage card for caching data. The alarm device includes an audible and visual alarm and a digital display connected to the circuit board. The audible and visual alarm and the digital display are powered by lithium iron phosphate batteries. The 4G / 5G communication module is connected to the cloud platform, and the cloud platform has a patrol APP for two-way communication.

10. The urban land subsidence monitoring and early warning device according to claim 7 is characterized by: The shell is a stainless steel shell, which has a sensor cabin for placing a tri-modal sensing unit, a computing cabin for placing an edge computing unit, and a power cabin for placing a self-powered system and a wireless communication module. Spiral guide fins are provided on the surface of the shell.

Images