Foundation pit deep soil body horizontal displacement monitoring system based on radio wave distance measurement

Abstract

The invention relates to the technical field of radio wave distance measurement, in particular to a foundation pit deep soil body horizontal displacement monitoring system based on radio wave distance measurement, which comprises a data acquisition module, a node strategy generation module, a data acquisition module, a displacement model construction module and a monitoring intelligent optimization module. According to the method, the sensor cluster is arranged through the sensor point distribution strategy, the foundation pit data are collected according to the sensor cluster, the displacement curve graph is generated according to the foundation pit data, the accurate displacement curve graph is finally obtained, the accuracy of the displacement curve graph is improved, and then the analysis precision of the foundation pit is improved.

Description

Technical Field

[0001] This invention relates to the field of radio wave ranging technology, and in particular to a system for monitoring the horizontal displacement of deep soil in foundation pits based on radio wave ranging. Background Technology

[0002] Accurate monitoring of soil displacement is crucial for ensuring construction safety and project quality during foundation pit construction. Traditional foundation pit soil displacement monitoring relies heavily on various sensors, which are pre-deployed to collect displacement data and generate displacement curves to analyze soil deformation trends. However, in actual construction, foundation pit geological conditions are complex and variable, and construction procedures are dynamically adjusted. Traditional sensor placement strategies are often fixed and cannot be flexibly adjusted according to real-time construction conditions. Consequently, the generated displacement curves fail to accurately reflect the actual deformation state of the soil, resulting in significantly reduced measurement accuracy and difficulty in accurately identifying potential safety hazards. This poses a significant risk to the safe construction of foundation pit projects. To address these issues, there is an urgent need for a monitoring system that can flexibly adjust sensor placement based on real-time construction conditions.

[0003] Chinese patent application CN110130421B discloses a method and system for monitoring foundation pit deformation. The method includes: setting up a reference sensor network and multiple monitoring sensor networks; each sensor node of the reference sensor network transmitting two radio waves of different frequencies to the first monitoring sensor network; other monitoring sensor networks sequentially transmitting their own measurement data and the received measurement data to the subsequent monitoring sensor networks, until the last monitoring sensor network transmits all measurement data to the reference sensor network; a server receives all measurement data transmitted by the reference sensor network and obtains the center point position of each monitoring sensor network based on all measurement data; the server calculates the deformation of the foundation pit based on the center point positions of each monitoring sensor network obtained multiple times. This invention monitors foundation pit deformation through a sensor network, eliminating the need for manual measurement and having low requirements for installation conditions, thus improving the convenience of measurement. However, this scheme cannot adjust the sensor placement strategy in real time according to the actual construction situation, resulting in low accuracy of displacement curve measurements due to an unreasonable sensor placement strategy. Summary of the Invention

[0004] To address this issue, the present invention provides a horizontal displacement monitoring system for deep soil in foundation pits based on radio wave ranging, which overcomes the problem in the prior art that the sensor placement strategy cannot be adjusted in real time according to the actual construction situation, resulting in low accuracy of displacement curve measurement due to unreasonable sensor placement strategy.

[0005] To achieve the above objectives, the present invention provides a system for monitoring the horizontal displacement of deep soil in foundation pits based on radio wave ranging, comprising: The data acquisition module is used to collect engineering data; The node strategy generation module is used to construct the node strategy model through the node strategy model construction method, and also to input engineering data into the node strategy model to obtain the sensor deployment strategy, and to deploy the sensor cluster according to the sensor deployment strategy. The data acquisition module is used to acquire foundation pit data through a sensor cluster, process the foundation pit data through data processing methods to obtain processed foundation pit data, and update the sensor deployment strategy based on the processed foundation pit data. The displacement model construction module is used to construct the displacement model using the displacement model construction method. It is also used to obtain the displacement curve based on the displacement model and the real-time field strength in the processed foundation pit data, render the displacement curve, and send the rendered displacement curve to the computer display terminal. The monitoring intelligent optimization module is used to optimize the displacement model based on the current water level and the previous water level in the processed foundation pit data, and to correct the optimization process of the displacement model based on the future rainfall in the processed foundation pit data. It is also used to calibrate the update process of the sensor placement strategy based on the current inclinometer temperature and the previous inclinometer temperature in the processed foundation pit data.

[0006] Furthermore, the node strategy generation module constructs a node strategy model using a node strategy model construction method to obtain the node strategy model. Engineering data is input into the node strategy model to obtain a set of sensor deployment strategies output by the node strategy model. This set of sensor deployment strategies includes a strategy order and the corresponding sensor deployment strategy content. The sensor deployment strategy content with the first strategy order is used as the sensor deployment strategy to arrange the sensor cluster. The node strategy model construction method includes: Step A1: Select a convolutional neural network model as the basic framework of the node policy model, and divide 70% of the policy generation sample dataset in the engineering data into a policy generation training set, 15% into a policy generation validation set, and 15% into a policy generation test set. Step A2: Train the convolutional neural network model using the training set generated according to the strategy, update the weights of the convolutional neural network model using the backpropagation algorithm, and validate the convolutional neural network model using the strategy generation validation set after each epoch to obtain the strategy generation validation loss value and the strategy generation validation accuracy. Step A3: When the policy generation verification loss value and the policy generation verification accuracy meet the preset verification conditions, the convolutional neural network model is tested according to the policy generation test set to obtain the policy generation test accuracy. When the policy generation test accuracy reaches the preset accuracy, the convolutional neural network model is output as a node policy model.

[0007] Furthermore, the data acquisition module also processes the foundation pit data using a data processing method to obtain processed foundation pit data. The data processing method includes: Step B1: Denoise the foundation pit data to obtain denoised foundation pit data; Step B2: Output the denoised foundation pit data as the processed foundation pit data.

[0008] Furthermore, the data acquisition module uses the positive dip angle in the processed foundation pit data... anti-duck angle and zero drift error For true inclination Perform calculations and set The true tilt angle Compared with the real-time tilt angle in the processed foundation pit data A comparison is performed, and the tilt angle measurement is judged based on the comparison results. The sensor placement strategy is then updated based on the judgment results, including: when = When the data acquisition module determines that the tilt angle measurement is valid, it does not update the sensor placement strategy. when < When the data acquisition module determines that the tilt angle measurement is invalid, it updates the sensor placement strategy to obtain the updated sensor placement strategy, sets the updated sensor placement strategy as the sensor placement strategy content of the next strategy order, and redeploys the sensor cluster according to the updated sensor placement strategy. when > When the data acquisition module determines that the tilt angle measurement is invalid, it updates the sensor placement strategy to obtain the updated sensor placement strategy, sets the updated sensor placement strategy as the next strategy order, and redeploys the sensor cluster according to the updated sensor placement strategy.

[0009] Furthermore, the displacement model construction module constructs the displacement model using a displacement model construction method, which includes: Step C1: Divide the soil of the foundation pit to be tested into grid cells to obtain n soil cells, where n is the number of soil cells. Step C2: Construct the matrix equation for the attenuation coefficient β based on the measured field strength Hs, the length Li of the ray passing through the i-th soil unit, and the theoretical field strength H0, and set... The attenuation coefficient β is obtained by solving the matrix equation of the attenuation coefficient β using the SIRT algorithm. Step C3 outputs the matrix equation of the attenuation coefficient β and the SIRT algorithm as the displacement model.

[0010] Furthermore, the displacement model construction module inputs the real-time field strength from the processed foundation pit data as the measured field strength Hs into the displacement model, obtains the attenuation coefficient β output by the displacement model, converts all attenuation coefficients β output by the displacement model into a columnar section related to the unit soil, obtains the unit columnar section, and inputs the unit columnar section into the displacement curve generation model to obtain the displacement curve output by the displacement curve generation model and the displacement deviation r corresponding to the displacement curve. The displacement deviation r is compared with the preset displacement deviation r0, and the displacement deviation is judged according to the comparison result. Based on the judgment result, the sensor placement strategy is adjusted, wherein: When r≤r0, the displacement model construction module determines that the displacement deviation is small, does not adjust the sensor placement strategy, renders the displacement curve corresponding to the displacement deviation, obtains the rendered displacement curve, and sends the rendered displacement curve to the computer display terminal. When r > r0, the displacement model construction module determines that the displacement deviation is large, adjusts the sensor placement strategy, obtains the adjusted sensor placement strategy, sets the adjusted sensor placement strategy as the sensor placement strategy content of the next strategy order, and rearranges the sensor cluster according to the adjusted sensor placement strategy.

[0011] Furthermore, the monitoring intelligent optimization module includes: The water level monitoring and optimization unit is used to optimize the displacement model based on the current water level and the previous water level in the processed foundation pit data, and also to correct the optimization process of the displacement model based on the future rainfall in the processed foundation pit data. The temperature monitoring and optimization unit is used to calibrate the sensor placement strategy update process based on the current inclinometer temperature and the previous inclinometer temperature in the processed foundation pit data, and also to correct the sensor placement strategy update calibration process based on displacement deviation.

[0012] Furthermore, the water level monitoring and optimization unit compares the current water level P1 in the processed foundation pit data with the previous water level P2, judges the water level change based on the comparison result, and optimizes the displacement model based on the judgment result, wherein: When P1 > P2, the water level monitoring and optimization unit determines that the water level change is a rise, optimizes the displacement model, calculates the first water level difference ΔP1 based on the current water level P1 and the previous water level P2, sets ΔP1 = P1 - P2, inputs the first water level difference ΔP1 into the attenuation coefficient optimization model, obtains the first optimized attenuation coefficient β2 output by the attenuation coefficient optimization model, replaces the attenuation coefficient β with the first optimized attenuation coefficient β2, and reconstructs the displacement model based on the attenuation coefficient β. When P1=P2, the water level monitoring and optimization unit determines that the water level change is unchanged and does not optimize the displacement model. When P1 < P2, the water level monitoring and optimization unit determines that the water level change is a drop, optimizes the displacement model, calculates the second water level difference ΔP2 based on the current water level P1 and the previous water level P2, sets ΔP2 = P1 - P2, inputs the second water level difference ΔP2 into the attenuation coefficient optimization model, obtains the second optimized attenuation coefficient β3 output by the attenuation coefficient optimization model, replaces the attenuation coefficient β with the second optimized attenuation coefficient β3, and reconstructs the displacement model based on the attenuation coefficient β.

[0013] Furthermore, the water level monitoring and optimization unit compares the future rainfall W in the processed foundation pit data with the preset future rainfall W0, judges the future rainfall situation based on the comparison result, and corrects the optimization process of the displacement model based on the judgment result, wherein: When W≤W0, the water level monitoring optimization unit determines that the future rainfall is normal and does not correct the optimization process of the displacement model; When W > W0, the water level monitoring optimization unit determines that the future rainfall is abnormal and corrects the optimization process of the displacement model. The correction method is as follows: the process of optimizing the displacement model is replaced by doubling the number of grid units to be divided into in step C1, resulting in J units of soil. J is set to 2n. The third water level difference ΔP3 is calculated based on the current water level P1 and the previous water level P2. ΔP3 is set to P1 - P2. The third water level difference ΔP3 is input into the attenuation coefficient optimization model to obtain the third optimized attenuation coefficient β4 output by the attenuation coefficient optimization model. The attenuation coefficient β is replaced with the third optimized attenuation coefficient β4. The displacement model is then reconstructed based on the attenuation coefficient β and the J units of soil.

[0014] Furthermore, the temperature monitoring and optimization unit calculates the inclinometer temperature difference ΔT based on the current inclinometer temperature T0 and the previous inclinometer temperature T1 in the processed foundation pit data, setting ΔT = |T0 - T1|. It then compares the inclinometer temperature difference ΔT with the preset inclinometer temperature difference ΔT0, judges the inclinometer temperature change based on the comparison result, and calibrates the sensor placement strategy update process based on the judgment result. Wherein: When △T≤△T0, the temperature monitoring optimization unit determines that the temperature change of the inclinometer is a normal change and does not calibrate the update process of the sensor placement strategy. When ΔT > ΔT0, the temperature monitoring optimization unit determines that the temperature change of the inclinometer is abnormal, calibrates the sensor placement strategy update process, and adjusts the inclinometer based on the calibrated inclinometer coefficient α and the real-time inclinometer. For true inclination The calculation was re-performed to obtain the true tilt angle Qzj after calibration, and then set... = -α, α=wc×△T, wc is the calibration zero drift value, and the true tilt angle Replace with the calibrated true tilt angle Qzj, and set the true tilt angle... Re-tilt with real-time angle Perform a comparison; The temperature monitoring optimization unit compares the displacement deviation r with the preset displacement deviation r0, judges the displacement deviation based on the comparison result, and corrects the calibration process of updating the sensor placement strategy based on the judgment result, wherein: When r≤r0, the temperature monitoring optimization unit determines that the displacement deviation is small and does not correct the calibration process of updating the sensor placement strategy; When r > r0, the temperature monitoring optimization unit determines that the displacement deviation is large and corrects it during the calibration process of updating the sensor placement strategy. It corrects the preset inclinometer temperature difference ΔT0 according to the correction coefficient Bz, setting Bz = 0.75 + 0.20 × e -(r-r0) Let e ​​be the base of the natural logarithm. We obtain the corrected preset inclinometer temperature difference value △T0B. We set △T0B = △T0 × Bz, replace the preset inclinometer temperature difference value △T0 with the corrected preset inclinometer temperature difference value △T0B, and then compare the preset inclinometer temperature difference value △T0 with the inclinometer temperature difference value △T again.

[0015] Compared with existing technologies, the beneficial effects of this invention are as follows: The system collects engineering data through a data acquisition module to generate a sensor deployment strategy and installs a sensor cluster according to the strategy, thereby acquiring foundation pit data. The system also generates a deployment strategy through a node strategy generation module to install the sensor cluster, improving the accuracy of foundation pit data acquisition by the sensor cluster. Furthermore, the system acquires foundation pit data from the sensor cluster through a data acquisition module and processes the data to obtain processed foundation pit data for subsequent analysis, thereby drawing accurate displacement curves. Finally, the system constructs a displacement model through a displacement model building module. The system constructs and outputs displacement curves based on the processed foundation pit data. Furthermore, through a monitoring and intelligent optimization module, it optimizes the displacement model based on the current and previous water levels, eliminating interference from water level changes and making the model more closely reflect reality. The optimization process of correcting the displacement model based on future rainfall enhances the system's ability to predict environmental factors and improves the model's foresight. The system also uses the current and previous inclinometer temperatures to calibrate the sensor placement strategy update process, reducing the impact of temperature changes on sensor accuracy and ensuring the accuracy of the sensor placement strategy update. Simultaneously, the displacement deviation correction calibration process further eliminates various accumulated errors, ensuring the reliability of the sensor placement strategy update and thus guaranteeing the accuracy of the displacement curves. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the structure of the deep soil horizontal displacement monitoring system for foundation pits based on radio wave ranging in this embodiment; Figure 2 This is a flowchart illustrating the node strategy model construction method in this embodiment; Figure 3 This is a flowchart illustrating the displacement model construction method in this embodiment; Figure 4 This is a schematic diagram of the monitoring intelligent optimization module in this embodiment. Detailed Implementation

[0017] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0018] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0019] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0020] Please see Figure 1-4 As shown, the radio wave ranging-based deep soil horizontal displacement monitoring system for foundation pits includes: The data acquisition module is used to collect engineering data; The node strategy generation module is used to construct the node strategy model through the node strategy model construction method, and also to input engineering data into the node strategy model to obtain the sensor deployment strategy, and to deploy the sensor cluster according to the sensor deployment strategy. The node strategy generation module is connected to the data acquisition module. The data acquisition module is used to acquire foundation pit data through a sensor cluster, process the foundation pit data through a data processing method to obtain processed foundation pit data, and update the sensor deployment strategy based on the processed foundation pit data. The data acquisition module is connected to the node strategy generation module. The displacement model construction module is used to construct the displacement model using the displacement model construction method. It is also used to obtain the displacement curve based on the displacement model and the real-time field strength in the processed foundation pit data, render the displacement curve, and send the rendered displacement curve to the computer display terminal. The displacement model construction module is connected to the data acquisition module. The monitoring intelligent optimization module is used to optimize the displacement model based on the current water level and the previous water level in the processed foundation pit data. It is also used to correct the optimization process of the displacement model based on the future rainfall in the processed foundation pit data, and to calibrate the update process of the sensor placement strategy based on the current inclinometer temperature and the previous inclinometer temperature in the processed foundation pit data. The monitoring intelligent optimization module is connected to the displacement model construction module.

[0021] Specifically, the radio wave ranging-based deep soil horizontal displacement monitoring system for foundation pits is applied in a radio wave ranging terminal for foundation pits. The system deploys a sensor cluster using a sensor placement strategy, collects foundation pit data based on the sensor cluster, and generates a displacement curve from the foundation pit data. This results in a precise displacement curve for analysis by personnel. The system collects engineering data through a data acquisition module to generate the sensor placement strategy and install the sensor cluster according to the strategy, thereby acquiring foundation pit data. The system also uses a node strategy generation module to generate a placement strategy for installing the sensor cluster, improving the accuracy of the foundation pit data collected by the sensor cluster. Finally, the system uses a data acquisition module to acquire foundation pit data from the sensor cluster and processes the data to obtain processed foundation pit data. The system then analyzes the processed foundation pit data to generate accurate displacement curves. It also constructs a displacement model using a displacement model building module and outputs displacement curves based on the processed data. Furthermore, the system utilizes a monitoring and intelligent optimization module to optimize the displacement model based on the current and previous water levels, eliminating interference from water level changes and making the model more closely reflect reality. The optimization process adjusts the displacement model based on future rainfall, enhancing the system's ability to predict environmental factors and improving the model's foresight. The system uses the current and previous inclinometer temperatures to calibrate the sensor placement strategy update process, reducing the impact of temperature changes on sensor accuracy and ensuring the accuracy of the updated strategy. Simultaneously, displacement deviations are used to correct the calibration process of the sensor placement strategy update, further ensuring its reliability and thus guaranteeing the accuracy of the displacement curves.

[0022] Specifically, the data acquisition module collects engineering data, including engineering drawing data, project construction data, and strategy generation sample datasets. The engineering drawing data refers to the schematic diagrams of the construction project, and the project construction data refers to the specific parameters of the construction project, such as the coordinates of the construction scope, the construction area, and the depth of the foundation pit. The data acquisition module collects engineering data by having workers manually input the engineering drawings and project plans into the foundation pit radio wave ranging terminal. The strategy generation sample dataset refers to a learning dataset stored in the form of an engineering data-sensor placement strategy set, used to train a convolutional neural network model. The data acquisition module obtains the strategy generation sample dataset by collecting historical engineering data and sensor placement strategies.

[0023] Specifically, the node strategy generation module constructs a node strategy model using a node strategy model construction method to obtain the node strategy model. Engineering data is input into the node strategy model to obtain a set of sensor deployment strategies output by the node strategy model. This set of sensor deployment strategies includes a strategy order and the corresponding sensor deployment strategy content. The sensor deployment strategy content with the first strategy order is used as the sensor deployment strategy for the sensor cluster. The node strategy model construction method includes: Step A1: Select a convolutional neural network model as the basic framework of the node policy model, and divide 70% of the policy generation sample dataset in the engineering data into a policy generation training set, 15% into a policy generation validation set, and 15% into a policy generation test set. Step A2: Train the convolutional neural network model using the training set generated according to the strategy, update the weights of the convolutional neural network model using the backpropagation algorithm, and validate the convolutional neural network model using the strategy generation validation set after each epoch to obtain the strategy generation validation loss value and the strategy generation validation accuracy. Step A3: When the policy generation verification loss value and the policy generation verification accuracy meet the preset verification conditions, the convolutional neural network model is tested according to the policy generation test set to obtain the policy generation test accuracy. When the policy generation test accuracy reaches the preset accuracy, the convolutional neural network model is output as a node policy model.

[0024] Specifically, the policy generation sample dataset refers to a learning dataset used to train a convolutional neural network model, stored in the form of engineering data-sensor deployment strategy sets. An epoch refers to the process of the convolutional neural network model completing one forward and backward propagation on the policy generation training set. The validation loss value refers to the loss value of the loss function in the convolutional neural network model when the policy generation validation set is input into the model for validation. The validation accuracy is the ratio of the number of policy generation results output by the model after inputting the validation set to the number of policy generation results in the validation set that match the output of the model to the total number of samples in the validation set. The preset validation condition is that the validation loss value does not decrease and the validation accuracy does not improve for eight consecutive epochs. The analytical test accuracy refers to the ratio of the data test results output by the model after inputting the policy generation test set to the output of the model to the number of samples in the validation set. The ratio of the number of data parsing results that are consistent with the total number of samples in the strategy generation test set is used. The preset accuracy rate refers to the preset value of the strategy generation accuracy rate for judging whether the convolutional neural network model has reached the output standard. For example, the preset accuracy rate can be set to 98%. The sensor placement strategy set refers to a collection of several sensor placement strategies. The strategy order refers to the order of the sensor placement strategies in the sensor placement strategy set. The sensor placement strategy content refers to the specific content of the sensor placement strategy, such as the specific placement points of the sensors and the number of sensors. The first strategy order refers to the sensor placement strategy content with the order of 1 in the sensor placement strategy set. This embodiment does not limit the way the order of the sensor placement strategies is set. Those skilled in the art can set it according to the actual situation. For example, the order of the sensor placement strategies can be set according to the cost. The one with the lowest cost is the first strategy order, the one with the second lowest cost is the second strategy order, and so on.

[0025] Specifically, the node strategy generation module inputs engineering data into the node strategy model to obtain the sensor deployment strategy output by the node strategy model, and then arranges the sensor cluster according to the sensor deployment strategy, thereby achieving the effect of rationally arranging the sensor cluster and accurately collecting foundation pit data.

[0026] Specifically, the data acquisition module acquires foundation pit data through a sensor cluster, which includes inclinometers, water level sensors, weather instruments, temperature sensors, and magnetic field sensors. The inclinometer measures the positive dip angle, negative dip angle, and real-time dip angle. The water level sensor collects data on the current and previous water levels. The weather instrument collects data on future rainfall. The temperature sensor collects data on the current and previous inclinometer temperatures. The magnetic field sensor collects data on the real-time field strength. The foundation pit data includes the positive dip angle, negative dip angle, real-time dip angle, current water level, previous water level, future rainfall, real-time field strength, current inclinometer temperature, and previous inclinometer temperature. The data acquisition module acquires the positive dip angle, negative dip angle, and real-time dip angle through the inclinometers in the sensor cluster. The positive dip angle refers to the inclinometer's forward measurement. The tilt angle obtained is as follows: the reverse tilt angle refers to the tilt angle measured after rotating the tilt meter 180 degrees; the real-time tilt angle refers to the tilt angle measured by the tilt meter in real time; the data acquisition module acquires the current water level and the previous water level through the water level sensor in the sensor cluster; the current water level refers to the water level value measured by the current water level sensor, and the previous water level refers to the water level value measured by the water level sensor in the last time; the data acquisition module acquires the future rainfall through the meteorological instrument; the future rainfall refers to the rainfall in the next day; the data acquisition module acquires the current tilt meter temperature and the previous tilt meter temperature through the temperature sensor; the current tilt meter temperature refers to the temperature of the tilt meter currently measured by the temperature sensor, and the previous tilt meter temperature refers to the temperature of the tilt meter measured by the temperature sensor in the last time; the data acquisition module acquires the real-time field strength through the magnetic field sensor; the real-time field strength refers to the electromagnetic intensity acquired in real time.

[0027] Specifically, the data acquisition module acquires the foundation pit data so that it can subsequently output the displacement curve of the foundation pit.

[0028] Specifically, the data acquisition module further processes the foundation pit data using a data processing method to obtain processed foundation pit data. The data processing method includes: Step B1: Denoise the foundation pit data to obtain denoised foundation pit data; Step B2: Output the denoised foundation pit data as the processed foundation pit data.

[0029] Specifically, this embodiment does not limit the specific implementation method for denoising the foundation pit data. Those skilled in the art can set it according to the actual situation, such as denoising the foundation pit data by using the Kalman filter method. The Kalman filter method refers to a real-time dynamic data optimization algorithm based on the state space model, which is used to eliminate interference data in the foundation pit data.

[0030] Specifically, the data acquisition module processes the foundation pit data using data processing methods to eliminate interfering data, maintain the validity of the foundation pit data, and thus improve the accuracy of the displacement curve.

[0031] Specifically, the data acquisition module uses the positive dip angle in the processed foundation pit data. anti-duck angle and zero drift error For true inclination Perform calculations and set The true tilt angle Compared with the real-time tilt angle in the processed foundation pit data A comparison is performed, and the tilt angle measurement is judged based on the comparison results. The sensor placement strategy is then updated based on the judgment results, including: when = When the data acquisition module determines that the tilt angle measurement is valid, it does not update the sensor placement strategy. when < When the data acquisition module determines that the tilt angle measurement is invalid, it updates the sensor placement strategy to obtain the updated sensor placement strategy, sets the updated sensor placement strategy as the sensor placement strategy content of the next strategy order, and redeploys the sensor cluster according to the updated sensor placement strategy. when > When the data acquisition module determines that the tilt angle measurement is invalid, it updates the sensor placement strategy to obtain the updated sensor placement strategy, sets the updated sensor placement strategy as the next strategy order, and redeploys the sensor cluster according to the updated sensor placement strategy.

[0032] Specifically, this embodiment does not limit the specific implementation method of sensor cluster arrangement. Those skilled in the art can set it according to the actual situation, such as arranging the sensor cluster according to the sensor placement points in the sensor placement strategy. The zero drift error refers to the undesirable offset of the sensor output value when the input is zero. This embodiment does not limit the method of obtaining the zero drift error. For example, it can be obtained through the laboratory benchmark calibration method. The laboratory benchmark calibration method refers to the scientific calibration method of obtaining zero drift by establishing an accurate mapping relationship between the sensor output value and the real physical quantity through high-precision equipment and a controlled environment. The tilt angle measurement situation refers to the measurement situation of the inclinometer judged based on the real tilt angle and the real-time tilt angle in the processed foundation pit data. The tilt angle measurement situation includes valid and invalid. The sensor placement strategy content of the next strategy sequence refers to the sensor placement strategy content of the next strategy sequence corresponding to the current sensor placement strategy content.

[0033] Specifically, the data acquisition module judges the tilt angle measurement situation and updates the sensor placement strategy according to the judgment result. When the tilt angle measurement situation is invalid, the sensor placement strategy is adjusted in time to avoid the problem of invalid tilt angle measurement caused by unreasonable sensor placement strategy, which in turn leads to low accuracy of the final output displacement curve data.

[0034] Specifically, the displacement model construction module constructs the displacement model using a displacement model construction method, which includes: Step C1: Divide the soil of the foundation pit to be tested into grid cells to obtain n soil cells, where n is the number of soil cells. Step C2: Construct the matrix equation for the attenuation coefficient β based on the measured field strength Hs, the length Li of the ray passing through the i-th soil unit, and the theoretical field strength H0, and set... The attenuation coefficient β is obtained by solving the matrix equation of the attenuation coefficient β using the SIRT algorithm. Step C3 outputs the matrix equation of the attenuation coefficient β and the SIRT algorithm as the displacement model.

[0035] Specifically, i represents the order of the soil units, i ≤ n, and βi is the attenuation coefficient of the i-th soil unit. This embodiment does not limit the specific implementation method for dividing the soil of the foundation pit to be tested into grid units. Those skilled in the art can set it according to the actual situation, such as dividing the soil of the foundation pit to be tested into grid units by Delaunay triangulation. Delaunay triangulation refers to a mathematical method of connecting discrete point sets into a triangular grid. The length of the ray passing through the i-th soil unit refers to the travel distance of the radio wave through each grid unit in the foundation pit soil. This embodiment does not limit the method of obtaining the length of the ray passing through the i-th soil unit. Those skilled in the art can set it according to the actual situation, such as obtaining the length of the ray passing through the i-th soil unit by spatial scanning method. Spatial scanning method refers to the ray path length measurement based on discrete sampling. This computational technique is particularly suitable for processing complex unstructured meshes. The method discretizes continuous rays into dense sampling points, combines this with a spatial query algorithm to determine the mesh cell to which each point belongs, and finally calculates the length of the ray passing through the i-th soil cell. The theoretical field strength refers to the theoretical magnetic field strength. This embodiment does not limit the method of obtaining the theoretical field strength; for example, it can be obtained through expert experience. The measured field strength refers to the actual measured magnetic field strength. This embodiment does not limit the method of obtaining the measured field strength; those skilled in the art can set it according to actual conditions, such as obtaining the measured field strength through a field strength sensor. The attenuation coefficient refers to the field strength attenuation value when radio waves pass through the foundation pit soil. The SIRT algorithm refers to an iterative reconstruction algorithm used for tomographic imaging, which compares and corrects multiple rounds of projection data with the computational model to invert the distribution of physical parameters inside the medium.

[0036] Specifically, the displacement model construction module constructs the displacement model using a displacement model construction method so that a displacement curve can be generated subsequently.

[0037] Specifically, the displacement model construction module inputs the real-time field strength from the processed foundation pit data as the measured field strength Hs into the displacement model, obtains the attenuation coefficient β output by the displacement model, converts all attenuation coefficients β output by the displacement model into a columnar section related to the unit soil, obtains the unit columnar section, and inputs the unit columnar section into the displacement curve generation model, obtains the displacement curve output by the displacement curve generation model and the displacement deviation r corresponding to the displacement curve, compares the displacement deviation r with the preset displacement deviation r0, judges the displacement deviation based on the comparison result, and adjusts the sensor placement strategy based on the judgment result, wherein: When r≤r0, the displacement model construction module determines that the displacement deviation is small, does not adjust the sensor placement strategy, renders the displacement curve corresponding to the displacement deviation, obtains the rendered displacement curve, and sends the rendered displacement curve to the computer display terminal. When r > r0, the displacement model construction module determines that the displacement deviation is large, adjusts the sensor placement strategy, obtains the adjusted sensor placement strategy, sets the adjusted sensor placement strategy as the sensor placement strategy content of the next strategy order, and rearranges the sensor cluster according to the adjusted sensor placement strategy.

[0038] Specifically, this embodiment does not limit the specific implementation method for converting all attenuation coefficients output by the displacement model into a bar chart related to the unit soil. Those skilled in the art can set it according to actual conditions, such as using Matplotlib to convert all attenuation coefficients output by the displacement model into a bar chart related to the unit soil. Matplotlib refers to computer plotting software used to convert data into various static, dynamic, and interactive charts. The preset displacement deviation refers to a preset value used to judge the displacement deviation. This embodiment does not limit the specific value of the preset displacement deviation; those skilled in the art can set it according to actual conditions, such as setting r0=10%. The displacement deviation refers to the deviation of the foundation pit soil displacement judged based on the displacement deviation and the preset displacement deviation. The displacement deviation includes small deviation and large deviation. This embodiment does not limit the specific implementation method for rendering the displacement curve; those skilled in the art can set it according to actual conditions. The specific implementation of sending the rendered displacement curve to the computer display terminal can be customized according to actual needs. For example, Matplotlib can be used to render the displacement curve. This embodiment does not limit the specific implementation of sending the rendered displacement curve to the computer display terminal. For example, the rendered displacement curve can be sent to the computer display terminal via wireless signal transmission. The displacement curve generation model refers to a convolutional neural network model that takes a unit bar chart as input and outputs a displacement curve and the displacement deviation corresponding to the displacement curve. This embodiment does not limit the specific construction method of the displacement curve generation model. Those skilled in the art can customize it according to actual needs. For example, the convolutional neural network model can be trained using a curve generation dataset to obtain the displacement curve generation model. The curve generation dataset refers to the training data set used to construct the displacement curve generation model. The curve generation dataset includes historically acquired unit bar charts, displacement curves corresponding to the historically acquired unit bar charts, and displacement deviations corresponding to the displacement curves.

[0039] Specifically, the displacement model construction module inputs the real-time field strength from the processed foundation pit data into the displacement model as the measured field strength, obtains the set of ray-crossing unit soil lengths output by the displacement model, converts the set of ray-crossing unit soil lengths into a ray-crossing unit soil columnar section, and inputs the ray-crossing unit soil columnar section into the displacement curve generation model to obtain the displacement curve and displacement curve output by the displacement curve generation model. The module then judges the displacement deviation and adjusts the sensor placement strategy based on the judgment result to avoid excessive displacement deviation caused by improper sensor placement strategy. When the displacement deviation is large, the sensor placement strategy is adjusted in a timely manner.

[0040] Specifically, the monitoring intelligent optimization module includes: The water level monitoring and optimization unit is used to optimize the displacement model based on the current water level and the previous water level in the processed foundation pit data, and also to correct the optimization process of the displacement model based on the future rainfall in the processed foundation pit data. The temperature monitoring optimization unit is used to calibrate the sensor placement strategy update process based on the current inclinometer temperature and the previous inclinometer temperature in the processed foundation pit data. It is also used to correct the calibration process of the sensor placement strategy update based on the displacement deviation. The temperature monitoring optimization unit is connected to the water level monitoring optimization unit.

[0041] Specifically, the monitoring intelligent optimization module optimizes the displacement model and sensor placement strategy based on changes in water level and temperature through the water level monitoring optimization unit and the temperature monitoring optimization unit, thereby improving the accuracy of the displacement model and enabling the displacement model to output a displacement curve with high accuracy.

[0042] Specifically, the water level monitoring and optimization unit compares the current water level P1 in the processed foundation pit data with the previous water level P2, judges the water level change based on the comparison result, and optimizes the displacement model based on the judgment result, wherein: When P1 > P2, the water level monitoring and optimization unit determines that the water level change is a rise, optimizes the displacement model, calculates the first water level difference ΔP1 based on the current water level P1 and the previous water level P2, sets ΔP1 = P1 - P2, inputs the first water level difference ΔP1 into the attenuation coefficient optimization model, obtains the first optimized attenuation coefficient β2 output by the attenuation coefficient optimization model, replaces the attenuation coefficient β with the first optimized attenuation coefficient β2, and reconstructs the displacement model based on the attenuation coefficient β. When P1=P2, the water level monitoring and optimization unit determines that the water level change is unchanged and does not optimize the displacement model. When P1 < P2, the water level monitoring and optimization unit determines that the water level change is a drop, optimizes the displacement model, calculates the second water level difference ΔP2 based on the current water level P1 and the previous water level P2, sets ΔP2 = P1 - P2, inputs the second water level difference ΔP2 into the attenuation coefficient optimization model, obtains the second optimized attenuation coefficient β3 output by the attenuation coefficient optimization model, replaces the attenuation coefficient β with the second optimized attenuation coefficient β3, and reconstructs the displacement model based on the attenuation coefficient β.

[0043] Specifically, the water level change refers to the current water level change determined based on the current water level and the previous water level. The water level change includes rising water level, unchanged water level, and falling water level. The attenuation coefficient optimization model refers to a recurrent neural network model that takes the water level difference as input and the optimized attenuation coefficient as output. This embodiment does not limit the specific construction method of the attenuation coefficient optimization model. For example, the attenuation coefficient optimization model can be obtained by training the recurrent neural network model through an attenuation training dataset. The attenuation training dataset refers to the training data set for constructing the attenuation coefficient optimization model. The attenuation training dataset includes historically obtained water level differences and the optimized attenuation coefficients corresponding to the historically obtained water level differences.

[0044] Specifically, the water level monitoring and optimization unit judges the water level changes and optimizes the displacement model based on the judgment results. Since water level changes affect the absorption of water by the soil, which in turn affects the transmission of radio waves, in order to eliminate the impact of water level changes on radio wave transmission, when the water level changes are rising or falling, the displacement model is reconstructed based on the optimized attenuation coefficient. This reduces the probability of inaccurate displacement model output results caused by water level changes and improves the accuracy of the displacement model, so as to obtain a highly accurate displacement curve.

[0045] Specifically, the water level monitoring and optimization unit compares the future rainfall W in the processed foundation pit data with the preset future rainfall W0, judges the future rainfall situation based on the comparison result, and corrects the optimization process of the displacement model based on the judgment result, wherein: When W≤W0, the water level monitoring optimization unit determines that the future rainfall is normal and does not correct the optimization process of the displacement model; When W > W0, the water level monitoring optimization unit determines that the future rainfall is abnormal and corrects the optimization process of the displacement model. The correction method is as follows: the process of optimizing the displacement model is replaced by doubling the number of grid units to be divided into in step C1, resulting in J units of soil. J is set to 2n. The third water level difference ΔP3 is calculated based on the current water level P1 and the previous water level P2. ΔP3 is set to P1 - P2. The third water level difference ΔP3 is input into the attenuation coefficient optimization model to obtain the third optimized attenuation coefficient β4 output by the attenuation coefficient optimization model. The attenuation coefficient β is replaced with the third optimized attenuation coefficient β4. The displacement model is then reconstructed based on the attenuation coefficient β and the J units of soil.

[0046] Specifically, the preset future rainfall amount refers to a preset value for judging future rainfall. This embodiment does not limit the specific value of the preset future rainfall amount. Those skilled in the art can set it according to the actual situation, such as setting W0=30mm. The future rainfall situation refers to the rainfall situation of the next day based on the rainfall amount and the preset future rainfall amount. The future rainfall situation includes normal rainfall and abnormal rainfall.

[0047] Specifically, the water level monitoring and optimization unit judges future rainfall conditions and corrects the optimization process of the displacement model based on the judgment results. When the future rainfall is abnormal, the displacement model is reconstructed based on the optimized attenuation coefficient and the corrected unit soil to prevent the displacement curve output by the displacement model from being inaccurate due to rainfall, thereby improving the accuracy of the displacement model.

[0048] Specifically, the temperature monitoring and optimization unit calculates the inclinometer temperature difference ΔT based on the current inclinometer temperature T0 and the previous inclinometer temperature T1 in the processed foundation pit data, setting ΔT = |T0 - T1|. It then compares the inclinometer temperature difference ΔT with the preset inclinometer temperature difference ΔT0, judges the inclinometer temperature change based on the comparison result, and calibrates the sensor placement strategy update process based on the judgment result. When △T≤△T0, the temperature monitoring optimization unit determines that the temperature change of the inclinometer is a normal change and does not calibrate the update process of the sensor placement strategy. When ΔT > ΔT0, the temperature monitoring optimization unit determines that the temperature change of the inclinometer is abnormal, calibrates the sensor placement strategy update process, and adjusts the inclinometer based on the calibrated inclinometer coefficient α and the real-time inclinometer. For true inclination The calculation was re-performed to obtain the true tilt angle Qzj after calibration, and then set... = -α, α=wc×△T, wc is the calibration zero drift value, and the true tilt angle Replace with the calibrated true tilt angle Qzj, and set the true tilt angle... Re-tilt with real-time angle Compare them.

[0049] Specifically, in △T=|T0-T1|, || refers to the absolute value calculation symbol. The preset inclinometer temperature difference value refers to a preset value for judging the temperature change of the inclinometer. This embodiment does not limit the specific value of the preset inclinometer temperature difference value. Those skilled in the art can set it according to actual conditions. For example, to control the inclinometer error ≤0.005°, △T0=2℃ can be set. The inclinometer temperature change refers to the temperature change of the inclinometer judged based on the inclinometer temperature difference value and the preset inclinometer temperature difference value. The inclinometer temperature change includes positive... The zero drift value refers to the zero drift value of the inclinometer during the calibration process of updating the sensor placement strategy based on the temperature change of the inclinometer. This embodiment does not limit the specific value of the zero drift value. Those skilled in the art can set it according to the actual situation. For example, the zero drift value can be obtained through a constant temperature chamber calibration experiment. The constant temperature chamber calibration experiment refers to the method of calibrating and verifying the temperature of the constant temperature test chamber equipment through a standardized process. The change of the zero drift value of the inclinometer when the temperature changes can be obtained through this method, and the zero drift value can be obtained.

[0050] Specifically, the temperature monitoring and optimization unit judges the temperature change of the inclinometer and calibrates the sensor placement strategy update process based on the judgment result. When the temperature change of the inclinometer is abnormal, the true inclinometer is recalculated based on the calibrated inclinometer coefficient and the real-time inclinometer to obtain the true inclinometer with abnormal temperature change, thereby improving the accuracy of the true inclinometer and thus improving the accuracy of the displacement curve.

[0051] Specifically, the temperature monitoring optimization unit compares the displacement deviation r with the preset displacement deviation r0, judges the displacement deviation based on the comparison result, and corrects the calibration process of updating the sensor placement strategy based on the judgment result, wherein: When r≤r0, the temperature monitoring optimization unit determines that the displacement deviation is small and does not correct the calibration process of updating the sensor placement strategy; When r > r0, the temperature monitoring optimization unit determines that the displacement deviation is large and corrects it during the calibration process of updating the sensor placement strategy. It corrects the preset inclinometer temperature difference ΔT0 according to the correction coefficient Bz, setting Bz = 0.75 + 0.20 × e -(r-r0)Let e ​​be the base of the natural logarithm. We obtain the corrected preset inclinometer temperature difference value △T0B. We set △T0B = △T0 × Bz, replace the preset inclinometer temperature difference value △T0 with the corrected preset inclinometer temperature difference value △T0B, and then compare the preset inclinometer temperature difference value △T0 with the inclinometer temperature difference value △T again.

[0052] Specifically, the temperature monitoring optimization unit judges the displacement deviation and corrects the sensor placement strategy update calibration process based on the judgment result. When the displacement deviation is large, a correction coefficient is set that decreases from 0.95 to infinitely close to 0.75 as the displacement deviation increases. This reduces the preset inclinometer temperature difference value, making it easier to judge the inclinometer temperature change as an abnormal change when the displacement deviation is large. This allows for the calibration of the true inclinometer and improves the accuracy of the displacement curve.

[0053] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A system for monitoring the horizontal displacement of deep soil in foundation pits based on radio wave ranging, characterized in that, include: The data acquisition module is used to collect engineering data; The node strategy generation module is used to construct the node strategy model through the node strategy model construction method, and also to input engineering data into the node strategy model to obtain the sensor deployment strategy, and to deploy the sensor cluster according to the sensor deployment strategy. The data acquisition module is used to acquire foundation pit data through a sensor cluster, process the foundation pit data through data processing methods to obtain processed foundation pit data, and update the sensor deployment strategy based on the processed foundation pit data. The displacement model construction module is used to construct the displacement model using the displacement model construction method. It is also used to obtain the displacement curve based on the displacement model and the real-time field strength in the processed foundation pit data, render the displacement curve, and send the rendered displacement curve to the computer display terminal. The monitoring intelligent optimization module is used to optimize the displacement model based on the current water level and the previous water level in the processed foundation pit data, and to correct the optimization process of the displacement model based on the future rainfall in the processed foundation pit data. It is also used to calibrate the update process of the sensor placement strategy based on the current inclinometer temperature and the previous inclinometer temperature in the processed foundation pit data.

2. The deep soil horizontal displacement monitoring system for foundation pits based on radio wave ranging according to claim 1, characterized in that, The node strategy generation module constructs a node strategy model using a node strategy model construction method. Engineering data is input into the node strategy model to obtain a set of sensor deployment strategies output by the model. This set includes a strategy order and the corresponding sensor deployment strategy content. The sensor deployment strategy content with the first strategy order is used as the sensor deployment strategy for the sensor cluster. The node strategy model construction method includes: Step A1: Select a convolutional neural network model as the basic framework of the node policy model, and divide 70% of the policy generation sample dataset in the engineering data into a policy generation training set, 15% into a policy generation validation set, and 15% into a policy generation test set. Step A2: Train the convolutional neural network model using the training set generated according to the strategy, update the weights of the convolutional neural network model using the backpropagation algorithm, and validate the convolutional neural network model using the strategy generation validation set after each epoch to obtain the strategy generation validation loss value and the strategy generation validation accuracy. Step A3: When the policy generation verification loss value and the policy generation verification accuracy meet the preset verification conditions, the convolutional neural network model is tested according to the policy generation test set to obtain the policy generation test accuracy. When the policy generation test accuracy reaches the preset accuracy, the convolutional neural network model is output as a node policy model.

3. The deep soil horizontal displacement monitoring system for foundation pits based on radio wave ranging according to claim 2, characterized in that, The data processing method includes: Step B1: Denoise the foundation pit data to obtain denoised foundation pit data; Step B2: Output the denoised foundation pit data as the processed foundation pit data.

4. The deep soil horizontal displacement monitoring system for foundation pits based on radio wave ranging according to claim 3, characterized in that, The data acquisition module uses the positive inclination angle in the processed foundation pit data. anti-duck angle and zero drift error For true inclination Perform calculations and set To achieve the true tilt angle Compared with the real-time tilt angle in the processed foundation pit data A comparison is performed, and the tilt angle measurement is judged based on the comparison results. The sensor placement strategy is then updated based on the judgment results, including: when = When the data acquisition module determines that the tilt angle measurement is valid, it does not update the sensor placement strategy. when < When the data acquisition module determines that the tilt angle measurement is invalid, it updates the sensor placement strategy to obtain the updated sensor placement strategy, sets the updated sensor placement strategy as the sensor placement strategy content of the next strategy order, and redeploys the sensor cluster according to the updated sensor placement strategy. when > When the data acquisition module determines that the tilt angle measurement is invalid, it updates the sensor placement strategy to obtain the updated sensor placement strategy, sets the updated sensor placement strategy as the next strategy order, and redeploys the sensor cluster according to the updated sensor placement strategy.

5. The deep soil horizontal displacement monitoring system for foundation pits based on radio wave ranging according to claim 4, characterized in that, The displacement model construction module constructs the displacement model using a displacement model construction method, which includes: Step C1: Divide the soil of the foundation pit to be tested into grid cells to obtain n soil cells, where n is the number of soil cells. Step C2: Construct the matrix equation for the attenuation coefficient β based on the measured field strength Hs, the length Li of the ray passing through the i-th soil unit, and the theoretical field strength H0, and set... The attenuation coefficient β is obtained by solving the matrix equation of the attenuation coefficient β using the SIRT algorithm. Step C3 outputs the matrix equation of the attenuation coefficient β and the SIRT algorithm as the displacement model.

6. The deep soil horizontal displacement monitoring system for foundation pits based on radio wave ranging according to claim 5, characterized in that, The displacement model construction module inputs the real-time field strength from the processed foundation pit data as the measured field strength Hs into the displacement model, obtains the attenuation coefficient β output by the displacement model, converts all attenuation coefficients β output by the displacement model into a columnar section related to the unit soil, obtains the unit columnar section, and inputs the unit columnar section into the displacement curve generation model, obtains the displacement curve output by the displacement curve generation model and the displacement deviation r corresponding to the displacement curve, compares the displacement deviation r with the preset displacement deviation r0, judges the displacement deviation based on the comparison result, and adjusts the sensor placement strategy based on the judgment result, wherein: When r≤r0, the displacement model construction module determines that the displacement deviation is small, does not adjust the sensor placement strategy, renders the displacement curve corresponding to the displacement deviation, obtains the rendered displacement curve, and sends the rendered displacement curve to the computer display terminal. When r > r0, the displacement model construction module determines that the displacement deviation is large, adjusts the sensor placement strategy, obtains the adjusted sensor placement strategy, sets the adjusted sensor placement strategy as the sensor placement strategy content of the next strategy order, and rearranges the sensor cluster according to the adjusted sensor placement strategy.

7. The deep soil horizontal displacement monitoring system for foundation pits based on radio wave ranging according to claim 6, characterized in that, The monitoring intelligent optimization module includes: The water level monitoring and optimization unit is used to optimize the displacement model based on the current water level and the previous water level in the processed foundation pit data, and also to correct the optimization process of the displacement model based on the future rainfall in the processed foundation pit data. The temperature monitoring and optimization unit is used to calibrate the sensor placement strategy update process based on the current inclinometer temperature and the previous inclinometer temperature in the processed foundation pit data, and also to correct the sensor placement strategy update calibration process based on displacement deviation.

8. The deep soil horizontal displacement monitoring system for foundation pits based on radio wave ranging according to claim 7, characterized in that, The water level monitoring and optimization unit compares the current water level P1 in the processed foundation pit data with the previous water level P2, judges the water level change based on the comparison result, and optimizes the displacement model based on the judgment result, wherein: When P1 > P2, the water level monitoring and optimization unit determines that the water level change is a rise, optimizes the displacement model, calculates the first water level difference ΔP1 based on the current water level P1 and the previous water level P2, sets ΔP1 = P1 - P2, inputs the first water level difference ΔP1 into the attenuation coefficient optimization model, obtains the first optimized attenuation coefficient β2 output by the attenuation coefficient optimization model, replaces the attenuation coefficient β with the first optimized attenuation coefficient β2, and reconstructs the displacement model based on the attenuation coefficient β. When P1=P2, the water level monitoring and optimization unit determines that the water level change is unchanged and does not optimize the displacement model. When P1 < P2, the water level monitoring and optimization unit determines that the water level change is a drop, optimizes the displacement model, calculates the second water level difference ΔP2 based on the current water level P1 and the previous water level P2, sets ΔP2 = P1 - P2, inputs the second water level difference ΔP2 into the attenuation coefficient optimization model, obtains the second optimized attenuation coefficient β3 output by the attenuation coefficient optimization model, replaces the attenuation coefficient β with the second optimized attenuation coefficient β3, and reconstructs the displacement model based on the attenuation coefficient β.

9. The deep soil horizontal displacement monitoring system for foundation pits based on radio wave ranging according to claim 8, characterized in that, The water level monitoring and optimization unit compares the future rainfall W in the processed foundation pit data with the preset future rainfall W0, judges the future rainfall situation based on the comparison result, and corrects the optimization process of the displacement model based on the judgment result, wherein: When W≤W0, the water level monitoring optimization unit determines that the future rainfall is normal and does not correct the optimization process of the displacement model; When W > W0, the water level monitoring optimization unit determines that the future rainfall is abnormal and corrects the optimization process of the displacement model. The correction method is as follows: the process of optimizing the displacement model is replaced by doubling the number of grid units to be divided into in step C1, resulting in J units of soil. J is set to 2n. The third water level difference ΔP3 is calculated based on the current water level P1 and the previous water level P2. ΔP3 is set to P1 - P2. The third water level difference ΔP3 is input into the attenuation coefficient optimization model to obtain the third optimized attenuation coefficient β4 output by the attenuation coefficient optimization model. The attenuation coefficient β is replaced with the third optimized attenuation coefficient β4. The displacement model is then reconstructed based on the attenuation coefficient β and the J units of soil.

10. The deep soil horizontal displacement monitoring system for foundation pits based on radio wave ranging according to claim 9, characterized in that, The temperature monitoring and optimization unit calculates the inclinometer temperature difference ΔT based on the current inclinometer temperature T0 and the previous inclinometer temperature T1 in the processed foundation pit data, setting ΔT = |T0 - T1|. It then compares the inclinometer temperature difference ΔT with the preset inclinometer temperature difference ΔT0, judges the inclinometer temperature change based on the comparison result, and calibrates the sensor placement strategy update process based on the judgment result. When △T≤△T0, the temperature monitoring optimization unit determines that the temperature change of the inclinometer is a normal change and does not calibrate the update process of the sensor placement strategy. When ΔT > ΔT0, the temperature monitoring optimization unit determines that the temperature change of the inclinometer is abnormal, calibrates the sensor placement strategy update process, and adjusts the inclinometer based on the calibrated inclinometer coefficient α and the real-time inclinometer. For true inclination The calculation was re-performed to obtain the true tilt angle Qzj after calibration, and then set... = -α, α=wc×△T, wc is the calibration zero drift value, and the true tilt angle Replace with the calibrated true tilt angle Qzj, and set the true tilt angle... Re-tilt with real-time angle Perform a comparison; The temperature monitoring optimization unit compares the displacement deviation r with the preset displacement deviation r0, judges the displacement deviation based on the comparison result, and corrects the calibration process of updating the sensor placement strategy based on the judgment result, wherein: When r≤r0, the temperature monitoring optimization unit determines that the displacement deviation is small and does not correct the calibration process of updating the sensor placement strategy; When r > r0, the temperature monitoring optimization unit determines that the displacement deviation is large and corrects it during the calibration process of updating the sensor placement strategy. It corrects the preset inclinometer temperature difference ΔT0 according to the correction coefficient Bz, setting Bz = 0.75 + 0.20 × e -(r-r0) Let e ​​be the base of the natural logarithm. We obtain the corrected preset inclinometer temperature difference value △T0B. We set △T0B = △T0 × Bz, replace the preset inclinometer temperature difference value △T0 with the corrected preset inclinometer temperature difference value △T0B, and then compare the preset inclinometer temperature difference value △T0 with the inclinometer temperature difference value △T again.

Images