Saline-alkali frozen soil multi-field synchronous monitoring device and method based on weak reflection grating

Abstract

The invention discloses a saline-alkali frozen soil multi-field synchronous monitoring device and method based on a weak reflection grating. The device comprises a temperature, humidity, salt and ice sensing module, a strain sensing module, an interlocking connector, a weak reflection grating demodulation module, a voltage stabilization direct current heating module and a data analysis and processing module. During monitoring, the temperature of the saline-alkali frozen soil is measured in the whole time period based on pulse temperature change and water-ice phase change through weak reflection grating sensing; the humidity and the ice content of the saline-alkali frozen soil are calculated through the pulse heating temperature time history curve; calculating the salt content of the saline-alkali frozen soil based on the wavelength reflectivity; calculating layered displacement of the saline-alkali frozen soil according to the relationship between strain and displacement; and correcting a multi-field monitoring data error by using a neural network. According to the invention, synchronous measurement of temperature, moisture content, salt content, ice content and deformation parameters of the saline-alkali frozen soil is realized, the problem of difficulty in synchronous measurement of multi-field parameters caused by unstable properties of the saline-alkali frozen soil is solved, and the system has the advantages of distribution, real-time performance and continuity.

Description

Technical Field

[0001] The present invention relates to a multi-field parameter monitoring device and a monitoring method in the fields of geotechnical engineering, agriculture and optical fiber monitoring, and in particular to a multi-field synchronous dynamic monitoring device and a method for saline-alkali frozen soil based on a weak reflection grating. Background Art

[0002] Saline-alkali frozen soil is a special type of soil formed by freezing of water in the soil due to low temperature climate conditions in saline-alkali land. Its characteristics are related to basic physical and chemical parameters such as soil density and salt content. Among them, salt is harmful to plant growth, which will lead to a decrease in soil fertility and even make the soil lose its production capacity, seriously affecting agricultural development. Secondly, basic parameters such as soil temperature, moisture content, ice content and strain are important information in the theoretical research and engineering application of thermal-hydraulic-mechanical multi-field coupling in saline-alkali frozen soil areas. However, saline-alkali frozen soil is extremely sensitive to temperature. With climate change, freeze-thaw effects affect the migration of soil moisture and salt, and the accumulation of salt will lead to the deterioration of soil structure. Therefore, accurate measurement of these parameters is of great significance for theoretical research, engineering practice and agricultural protection of saline-alkali frozen soil.

[0003] The measurement methods of saline-alkali frozen soil temperature mainly include thermocouple method, thermal resistance method, infrared method and distributed optical fiber monitoring technology. Thermocouples are composed of two different metal materials. The temperature difference between the two metals will cause voltage changes, thereby measuring the temperature of frozen soil. However, its signal conditioning is complex, the accuracy is limited, it is susceptible to corrosion, and its anti-interference ability is poor. When the temperature of the thermal resistance sensor changes, its internal resistance changes, and the change in resistance value can be used to deduce the frozen soil temperature. However, it is greatly affected by environmental factors and has a greater impact on the temperature readings of bright or polished metal surfaces. The infrared radiation radiated by frozen soil can be used to measure its surface temperature, which is greatly affected by the environment and has a high price.

[0004] Distributed fiber optic monitoring technology (DFOS) uses optical fiber as a sensor for monitoring, and specifically demodulates temperature information through optical signals. Among them, weak reflection fiber grating (WFBG) is mainly based on the Bragg scattering effect of fiber grating. When the temperature changes, the fiber grating period changes, resulting in a corresponding change in the reflection wavelength. By measuring the wavelength change, the temperature change can be accurately monitored. WFBG has the advantages of high precision, high sensitivity, and distribution.

[0005] The measurement methods of moisture content and ice content of saline-alkali frozen soil mainly include dielectric spectroscopy and thermal pulse method. Dielectric spectroscopy includes time domain reflectometry (TDR) and frequency domain reflectometry (FDR), which indirectly reflects moisture content and ice content by measuring the dielectric constant of saline-alkali frozen soil. TDR and FDR are limited by field applications, and their applicability in saline-alkali frozen soil areas needs to be improved. The thermal pulse method uses the response characteristics of the soil to the thermal pulse to calculate its equivalent thermal conductivity to calculate the moisture content and ice content. Through active heating, based on the linear heat source model, direct current is passed through the sheath of the optical fiber to heat it, and its temperature response is measured. It has the advantages of distributed, anti-electromagnetic interference, and high sensitivity. However, this method has a long heating time, which has a large interference on the soil moisture and is prone to measurement errors.

[0006] The main methods for measuring salt in saline-alkali frozen soil are conductivity measurement, saturation measurement and spectral measurement. The conductivity measurement method is based on the relationship between the concentration of ions in the solution and its conductivity. By measuring the conductivity of the solution, the salt content can be indirectly calculated. It is simple to operate, low cost, and suitable for rapid on-site measurement. However, this method is affected by the moisture content of frozen soil, and its accuracy is reduced. The saturation measurement method is to calculate the salt content by measuring the saturated mud conductivity of soil or water samples under certain conditions. This method can provide relatively accurate salt measurement results and is often used for soil salt determination, but it requires the extraction of frozen soil samples, destroying the original structure, and cannot achieve in-situ measurement. The spectral measurement method analyzes the composition and content of a substance based on its absorption characteristics of light of a specific wavelength. It has high sensitivity and selectivity, can be used to accurately determine the salt content, and is suitable for fields such as water quality analysis and environmental monitoring. However, in the field of saline-alkali frozen soil, it can only reflect the salt situation of the frozen soil surface, cannot provide vertical distribution information of the frozen soil profile, and is easily affected by the weather environment.

[0007] The main methods for measuring strain in saline-alkali frozen soil include borehole inclinometer, cone penetration test (CPT), distributed fiber optic monitoring technology (DFOS) and other methods. The borehole inclinometer is to install an inclinometer in the borehole to measure the inclination or displacement inside the soil. It is suitable for monitoring deep deformation of the soil. The operation is complicated and the installation cost is high. CPT uses a penetration device to measure the resistance of the soil to the probe, thereby evaluating the bearing capacity and deformation characteristics of the soil. CPT has a large disturbance to the soil layer and may cause damage or changes to the soil structure. DFOS uses distributed fiber optic sensors deployed along the soil to monitor temperature and other environmental parameters in real time, thereby predicting the deformation of the soil. DFOS can achieve long-distance, high spatial resolution, and distributed real-time monitoring. However, this method is easily affected by synchronous temperature measurement, resulting in measurement errors.

[0008] The above-mentioned traditional measurement methods for temperature, moisture content, salinity, ice content and strain of saline-alkali frozen soil all have certain limitations, and it is particularly difficult to conduct synchronous dynamic monitoring of multiple fields of saline-alkali frozen soil. Therefore, how to provide an effective monitoring device and method for multiple field parameters of saline-alkali frozen soil to achieve synchronous dynamic monitoring of temperature, moisture content, salinity, ice content and strain parameters in saline-alkali frozen soil areas, and provide strong support for frozen soil theory, engineering practice and agricultural development, has become a key technical problem that needs to be solved in saline-alkali frozen soil areas. Summary of the invention

[0009] Purpose of the invention: In view of the shortcomings of the existing monitoring technology, the present invention proposes a multi-field synchronous monitoring device and method of saline-alkali frozen soil based on weak reflection grating, which realizes the synchronous dynamic monitoring of temperature, moisture content, salinity, ice content and strain of saline-alkali frozen soil, solves the problem of difficulty in synchronous measurement of parameters caused by the unstable properties of saline-alkali frozen soil, and has the advantages of distribution, corrosion resistance, small disturbance, automation, real-time and stability.

[0010] Technical solution: The invention discloses a multi-field synchronous dynamic monitoring device for saline-alkali frozen soil based on weak reflection grating, comprising a temperature, humidity, salt and ice sensing module, a strain sensing module, an interlocking joint, a weak reflection grating demodulation module, a voltage-stabilized DC heating module and a data analysis and processing module;

[0011] The temperature, humidity, salt and ice sensing module consists of a temperature sensing unit, a humidity-ice content sensing unit, and a salt sensing unit; the temperature, humidity, salt and ice sensing modules are separated by thermal insulation materials;

[0012] A weak reflection grating is provided in the temperature sensing unit, and a heat insulation interface is connected between the temperature sensing unit and the humidity-ice content sensing unit;

[0013] The humidity-ice content sensing unit is embedded with a heating rod, and a weak reflection grating is used in the center of the heating rod to measure parameters; capillaries are provided on the side of the salt sensing unit.

[0014] The temperature, humidity, salt and ice sensing modules are fixed with interlocking connectors.

[0015] The temperature sensing unit is encapsulated by a high thermal conductivity material to isolate the strain.

[0016] The surface of the optical fiber in the salt sensing unit is coated with a weakening medium.

[0017] The temperature sensing unit provides temperature compensation for the strain sensing module.

[0018] The strain sensing module is packaged by a fully distributed sensing optical cable and has high strain sensitivity.

[0019] The side wall of the humidity-ice content sensing unit is provided with continuous holes to increase the contact area with the saline-alkali frozen soil.

[0020] The method for synchronous dynamic monitoring of saline-alkali frozen soil in multiple fields based on weak reflection grating of the present invention comprises the following steps:

[0021] (1) A hole is drilled on the surface of the saline-alkali frozen soil to be measured, and the temperature, humidity, salt and ice sensing modules are connected in series. Then, they are connected in parallel with the strain sensing module and vertically arranged in the saline-alkali frozen soil. Each sensing module is flexibly connected and configured to meet the measurement requirements under saline-alkali frozen soil environmental conditions.

[0022] (2) Samples of saline-alkali frozen soil at the same depth in each temperature, humidity, salt and ice sensing unit were collected, backfilled and compacted, and field temperature measurement and laboratory tests were conducted to determine the initial temperature, salinity, moisture content and ice content of the saline-alkali frozen soil at the corresponding depth.

[0023] (3) Each sensing module component is connected to a weak reflection grating demodulation module, and the temperature, humidity, salt and ice sensing module is connected to a voltage-stabilized DC heating module. The temperature, humidity, salt and ice sensors are pulse-heated at a constant power and time. The weak reflection grating demodulation module records the wavelength data of the temperature along the burial depth direction at a set spatial resolution and acquisition frequency, and automatically converts it into temperature information output, thereby obtaining the temperature time history curve and temporal and spatial distribution changes of the in-situ saline-alkali frozen soil along the depth direction.

[0024] (4) weak reflection grating sensing data are collected to construct an artificial neural network (ANN). The number of nodes in the input layer of the artificial neural network corresponds to the central wavelength and reflectivity parameters output by the temperature sensing unit, and the number of nodes in the output layer corresponds to the initial temperature of the in-situ saline-alkali frozen soil measured on-site in step (2). The above parameters are forward propagated through each layer of the ANN, and the wavelength data is processed by the activation function and the temperature prediction value is output. The error between the output temperature and the actual temperature is quantified using the loss function. The weights of the parameters in the network are adjusted according to the prediction error by relying on the back propagation algorithm, and multiple iterations of training and optimization are performed until the sensor parameters are accurately mapped to the temperature, and finally the feedback is fed back to the weak reflection grating demodulation module to synchronously correct the output in-situ saline-alkali frozen soil temperature change.

[0025] (5) Select constant heating power and temperature data during heating time to calculate the thermal conductivity λ of saline-alkali frozen soil at each depth, and calculate the temperature change ΔT measured by the temperature sensing unit after heating. t , derive the equivalent thermal conductivity λ of the soil, the derivation process is:

[0026]

[0027]

[0028] Among them, t is the heating time, q is the heating power, λ is the equivalent thermal conductivity of saline-alkali frozen soil, and c is a constant related to soil diffusivity.

[0029] (6) Calibrate the in-situ saline-alkali frozen soil sample of step (1) and the thermal conductivity λ described in step (5) in an indoor test, and calculate the moisture content θ during the monitoring process of the saline-alkali frozen soil according to the calibration formula w .

[0030] (7) The in-situ saline-alkali frozen soil sample of step (1) and the wavelength change Δλ recorded by the salt sensing unit are calibrated in an indoor test, and the salt concentration C during the saline-alkali frozen soil monitoring process is calculated according to the calibration formula.

[0031] (8) According to the theoretical model of soil thermal conductivity λ = f(θ i ,θ w ), combined with the in-situ saline-alkali frozen soil moisture content θ w and equivalent thermal conductivity λ, calculate the total moisture content θ of the in-situ saline-alkali frozen soil t and ice content θ i ; The total moisture content θ t =θ w +θ i .

[0032] (9) The wavelength data of the strain sensing unit 7 along the buried depth direction is recorded by the spatial resolution and acquisition frequency set by the weak reflection grating demodulation module 2, and converted into strain information output to obtain the strain curve of the in-situ saline-alkali frozen soil along the depth direction.

[0033] In step (6), the moisture content indoor test calibration steps are:

[0034] (6.1) Prepare the soil samples taken from the monitoring site into several groups with known temperature T and moisture content θ at different depths. w samples;

[0035] (6.2) calculating the thermal conductivity λ of the soil at different depths according to the derivation formula of step (5) of claim 6;

[0036] (6.3) Establish the thermal conductivity λ and the known water content θ w The linear relationship is θ w =kλ+b, where k is the proportionality constant and b is the intercept determined during the calibration process.

[0037] In step (7), the steps of indoor test calibration of salinity are as follows:

[0038] (7.1) Prepare several groups of samples with known salt concentration C from the soil samples taken at the monitoring site;

[0039] (7.2) placing a salt sensing unit element of the same specification in a sample of known salt concentration in step (8.1) and measuring its wavelength change Δλ;

[0040] (7.3) A linear relationship between the wavelength change Δλ and the known salt concentration C is established, and the linear equation is C = kΔλ + b, where k is the proportionality constant obtained by linear regression analysis, and b is the intercept determined during the calibration process.

[0041] In step (8), the equivalent thermal conductivity λ obtained in step (6) and the equivalent thermal conductivity λ and ice content θ obtained in step (5) according to the calibration formula are deduced. i The functional relationship between them is used as the objective function to fit the data and calculate the corresponding ice content θ i , the derivation process is as follows:

[0042] The following ice content θ i Calculation formula

[0043]

[0044]

[0045] Among them, S r is the soil saturation coefficient, θ w is the soil moisture content of the in-situ saline-alkali frozen soil, n is the soil porosity, λ w is the thermal conductivity of water, λ i is the thermal conductivity of ice, λ s is the thermal conductivity of soil particles, k2 is the soil particle size distribution coefficient, x is the material parameter related to the thermal conductivity of soil particles, and η is the material coefficient related to the shape of soil particles.

[0046] Working principle: During monitoring, the present invention uses weak reflection grating sensing, based on the principle of pulse temperature change and water-ice phase change, to measure the temperature of saline-alkali frozen soil throughout the whole period; uses the pulse heating temperature time curve to calculate the humidity and ice content of saline-alkali frozen soil; calculates the salt content of saline-alkali frozen soil based on wavelength reflectivity; calculates the stratified displacement of saline-alkali frozen soil based on the strain-displacement relationship; and uses neural networks to correct errors in multi-field monitoring data. The present invention realizes the synchronous measurement of the temperature, moisture content, salt content, ice content, and deformation parameters of saline-alkali frozen soil, solves the problem of difficulty in synchronously measuring multi-field parameters caused by the unstable nature of saline-alkali frozen soil, and has the advantages of distributed, durable, real-time, continuous, and multi-parameter monitoring.

[0047] The temperature, humidity, salt and ice sensing module of the present invention is composed of a temperature sensing unit, a humidity-ice content sensing unit and a salt sensing unit, and together with the strain sensing unit, dynamically observes the temperature, moisture content, salt, ice content and strain of saline-alkali frozen soil at different depths.

[0048] Among them, the temperature and strain of saline-alkali frozen soil are directly obtained by demodulating the wavelength information; combined with the moisture content data of in-situ saline-alkali frozen soil θ w, soil equivalent thermal conductivity λ and its theoretical model λ=f(θ i ,θ w ), and the total moisture content θ of the in-situ saline-alkali frozen soil is obtained t and ice content θ i , where the total moisture content of saline-alkali frozen soil is θ t =θ i +θ w ; According to the linear relationship between the laboratory calibration wavelength and the salinity of the in-situ saline-alkali frozen soil, the salinity data C of the in-situ saline-alkali frozen soil is obtained; finally, the synchronous dynamic measurement of multiple field parameters is realized, and their spatiotemporal distribution is obtained. The spatiotemporal distribution is reflected in that during the freezing period of the saline-alkali frozen soil, the temperature decreases, the water freezes into ice, the water content in the in-situ saline-alkali frozen soil decreases, the ice content increases, the water in the unfrozen area migrates to the frozen area, and the salt in the frozen area migrates to the unfrozen area. The soil body expands and deforms due to the freezing of water. The temperature rises during the freezing period, the water content increases after the ice crystals melt, the ice content decreases, the salt content increases, and the soil volume shrinks, resulting in the closure of soil cracks or the formation of new cracks. The present invention realizes the synchronous dynamic monitoring of multiple fields of saline-alkali frozen soil based on weak reflection grating sensing technology, which solves the problem that the multi-field parameters are difficult to monitor synchronously due to the unstable nature of saline-alkali frozen soil.

[0049] Beneficial effects: Compared with the prior art, the present invention has the following advantages:

[0050] (1) The present invention adopts weak reflection grating sensing technology to realize integrated monitoring of frozen soil temperature, moisture content, salinity, ice content, and strain, meeting the requirements of distributed, continuous, and multivariable monitoring, and significantly reducing the size and cost of the sensor.

[0051] (2) The present invention uses weak reflection grating sensing technology and its temperature, humidity, salt and ice sensing module to realize the identification of ice, water and salt components in saline-alkali frozen soil, solving the problem of being unable to directly obtain the in-situ parameters of saline-alkali frozen soil. It has the advantages of distribution, real-time, accuracy, automation and low disturbance.

[0052] (3) A hole structure is provided on the side of the temperature, humidity, salt and ice sensing module of the device of the present invention to increase the contact area between the frozen soil to be measured and the temperature, humidity, salt and ice sensing module, so that they are in full contact and are conducive to measurement.

[0053] (4) Adjacent temperature, humidity, salt and ice sensing modules of the present invention are connected by precise interlocking joints, which is convenient for on-site assembly and can meet different monitoring depths according to the number of temperature, humidity, salt and ice modules connected in series.

[0054] (5) The temperature compensation of the strain sensing module of the present invention is provided by the temperature sensing module packaged by a special process in the temperature, humidity, salt and ice sensing module, which has little disturbance to the soil and solves the problem of difficulty in accurately measuring strain due to unstable soil properties.

[0055] (6) By integrating multiple sensing technologies and combining real-time data correction and optimization methods, the present invention significantly improves the overall performance of the system and ensures the high reliability of the monitoring results.

[0056] (7) The present invention is economical and safe, easy to operate, has high durability, high stability, high degree of automation, and is suitable for long-term monitoring in harsh environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 It is a schematic diagram of the overall structure of the multi-field synchronous monitoring device for saline-alkali frozen soil based on weak reflection grating of the present invention;

[0058] Figure 2 It is a schematic diagram of the temperature, humidity, salt and ice sensing module of the present invention;

[0059] Figure 3 is a schematic cross-sectional view of a temperature sensing unit of the present invention;

[0060] Figure 4 is a cross-sectional schematic diagram of a humidity-ice content sensing unit of the present invention;

[0061] Figure 5 is a schematic cross-sectional view of a salt sensing unit of the present invention;

[0062] Figure 6 is a schematic diagram of the precision interlocking joint of the present invention;

[0063] Figure 7 This is a calibration result diagram of the linear relationship between the thermal conductivity of frozen soil and the humidity of frozen soil in the present invention;

[0064] Figure 8 It is a distribution diagram of the moisture content-ice content-total moisture content of saline-alkali frozen soil along the depth in the present invention;

[0065] Fig. 9 It is a relationship diagram between strain and stratified displacement at different depths of frozen soil in the present invention. DETAILED DESCRIPTION

[0066] like Figure 1 As shown, the multi-field synchronous dynamic monitoring device for saline-alkali frozen soil based on weak reflection grating of the present invention includes a temperature, humidity, salt and ice sensing module, a strain sensing module 7, an interlocking joint 6, a weak reflection grating demodulation module 2, a voltage-stabilized DC heating module 9 and an analysis and processing data module 1. The temperature, humidity, salt and ice sensing module and the strain sensing module 7 are vertically buried in the in-situ saline-alkali frozen soil. The temperature sensing unit 3 and the strain sensing module 7 are connected to the weak reflection grating demodulation module 2 through an optical fiber lead 8. 10 is a temperature, humidity, salt and ice sensing optical fiber, 14 is a heat insulation interface between the humidity-ice content sensing unit and the salt sensing unit, 17 is a salt sensing unit interface, 18 is a snap connection, and 19 is a temperature sensing unit interface.

[0067] The temperature, humidity, salt and ice sensing module is composed of a temperature sensing unit 3, a humidity-ice content sensing unit 4, and a salt sensing unit 5. The temperature, humidity, salt and ice sensing module is separated by a thermal insulation material to prevent heat conduction and is fixed by an interlocking joint 6. The strain sensing module 7 is encapsulated by a fully distributed sensing optical cable, and then the soil temperature, moisture content, salt, ice content and strain at different depths are synchronously and dynamically monitored. In this embodiment, the interlocking joint 6 adopts a precision interlocking joint.

[0068] In this embodiment, a weak reflection grating 11 is provided in the temperature sensing unit 3, and is connected to the humidity-ice content sensing unit 4 by a heat insulation interface 12. The temperature sensing unit 3 is encapsulated by a high thermal conductivity material, is isolated from strain, is highly sensitive to temperature, and uses a heat insulation structure to connect each sensing module to avoid the influence of the heat conduction of the sensor itself on the soil temperature measurement, thereby ensuring the accuracy of the temperature data.

[0069] The humidity-ice content sensing unit 4 has a built-in heating rod 13, and a weak reflection grating is used in the center of the heating rod 13 for parameter measurement. The side wall of the unit is provided with continuous holes to increase the contact area with the saline-alkali frozen soil, and is connected to the voltage-stabilized DC heating module 9 for pulse heating.

[0070] Capillaries 15 are provided on the side of the salt sensing unit 5 so that the sensing grating inside the salt sensing unit contacts the frozen soil salt. The surface of the optical fiber in the salt sensing unit 5 is coated with a coating layer 16 of a salt-sensitive weakening material to enhance Raman scattering, so that the salt content measurement can also maintain high accuracy under low concentration conditions.

[0071] The contact area between the humidity-ice content sensing unit 4 and the saline-alkali frozen soil is increased, and the voltage-stabilized DC heating module 9 is connected to perform pulse heating to calculate the humidity and ice content.

[0072] The strain sensing module 7 is encapsulated by a high strain transfer sheath and has high strain sensitivity. The temperature compensation function of the strain sensing module 7 is provided by the temperature sensing unit 3 to correct the output error of the strain sensing unit caused by temperature changes to ensure measurement under different ambient temperatures.

[0073] The weak reflection grating demodulation module 2 provides data transmission demodulation and cloud storage functions for each sensing module, and performs data analysis 1 through the data analysis and processing module.

[0074] The multi-field synchronous dynamic monitoring device of saline-alkali frozen soil based on weak reflection grating of this embodiment is deployed in the transition zone from a desert and mountains to the Loess Plateau in northwest China. The temperature in this area is relatively low in winter, with an average low temperature of about -14°C in January and an extreme low temperature of -20°C. Affected by the topography and geological conditions, some cultivated land has become salinized and saline-alkali frozen soil has developed. The in-situ monitoring time of this embodiment is from January 1, 2023 to February 18, 2023.

[0075] The method for synchronous dynamic monitoring of saline-alkali frozen soil in multiple fields based on weak reflection grating of the present invention comprises the following steps:

[0076] (1) Drill a hole on the surface of the saline-alkali frozen soil to be tested, connect the temperature, humidity, salt and ice sensing modules in series, and vertically arrange them in the saline-alkali frozen soil together with the strain sensing module 7.

[0077] (2) Samples of saline-alkali frozen soil at the same depth in each temperature, humidity, salt and ice sensing unit were collected, backfilled and compacted, and field temperature measurement and laboratory tests were conducted to determine the initial temperature, salinity, moisture content and ice content of the saline-alkali frozen soil at the corresponding depth.

[0078] (3) Each sensing module component is connected to a weak reflection grating demodulation module, and the temperature, humidity, salt and ice sensing module is connected to a voltage-stabilized DC heating module. The temperature, humidity, salt and ice sensors are pulse-heated at a constant power and time. The weak reflection grating demodulation module records the wavelength data of the temperature along the burial depth direction at a set spatial resolution and acquisition frequency, and automatically converts it into temperature information output, thereby obtaining the temperature time history curve and temporal and spatial distribution changes of the in-situ saline-alkali frozen soil along the depth direction.

[0079] (4) weak reflection grating sensing data are collected to construct an artificial neural network (ANN), wherein the number of nodes in the input layer of the artificial neural network corresponds to the central wavelength and reflectivity parameters output by the temperature sensing unit, and the number of nodes in the output layer corresponds to the initial temperature of the in-situ saline-alkali frozen soil measured on site in step (2). The above parameters are forward propagated through each layer of the ANN, the wavelength data is processed by the activation function and the temperature prediction value is output. The error between the output temperature and the actual temperature is quantified using the loss function, and the weights of the parameters in the network are adjusted according to the prediction error by relying on the back propagation algorithm. Multiple iterations of training and optimization are performed until the sensor parameters are accurately mapped to the temperature, and finally the feedback is fed back to the weak reflection grating demodulation module to synchronously correct the output in-situ saline-alkali frozen soil temperature change.

[0080] (5) Select the temperature data during constant heating power and heating time, calculate the equivalent thermal conductivity λ of saline-alkali frozen soil at each depth, and calculate the temperature change ΔT measured by the temperature sensing unit after heating. t , derive the equivalent thermal conductivity λ of the soil:

[0081]

[0082] Among them, t is the heating time, q is the heating power, λ is the equivalent thermal conductivity of saline-alkali frozen soil, and c is a constant related to soil diffusivity.

[0083] (6) Calibrate the in-situ saline-alkali frozen soil sample described in step (1) and the equivalent thermal conductivity coefficient λ described in step (5) in an indoor test, and calculate the moisture content θ during the monitoring process of the saline-alkali frozen soil according to the calibration formulaw .

[0084] In step (6), the steps for calibrating the moisture content in the indoor test are as follows:

[0085] (6.1) Drill a hole at the monitoring site to a depth of L = 10 m, take 5 groups of saline-alkali frozen soil samples, and measure the moisture content of frozen soil at each depth using the drying method in the laboratory. w , the measurement results are: 10%, 20%, 30%, 40%, 50%;

[0086] (6.2) Calculate the equivalent thermal conductivity λ of the soil at different depths according to the derived formula in step (5), and the corresponding equivalent thermal conductivity is 0.5, 0.55, 0.6, 0.65, and 0.7;

[0087] (6.3) Establish the equivalent thermal conductivity λ and the known moisture content θ w The linear relationship obtained is: w =0.45λ+0.005, such as Figure 7 shown.

[0088] (7) The in-situ saline-alkali frozen soil sample of step (1) and the wavelength process data continuously recorded by the salt sensing unit are calibrated in an indoor test, and the salt concentration C during the saline-alkali frozen soil monitoring process is calculated according to the calibration formula.

[0089] In step (7), the steps of indoor test calibration of salinity are as follows:

[0090] (7.1) Prepare several groups of soil samples with known salt concentration C from the soil samples collected at the monitoring site, and soak them in solutions with salt contents of 0.5%, 1.0%, 1.5%, 2.0%, and 2.5% in sequence. After soaking in each solution with a salt content of 60 minutes, in order to avoid the influence of temperature on this experiment, this experiment was carried out in a constant temperature room with a room temperature of 25°C;

[0091] (7.2) placing a salt sensing unit element of the same specification in a sample of known salt concentration in step (7.1), and measuring the wavelength change Δλ in the corresponding concentration, which is 0.2, 0.25, 0.3, 0.35, and 0.4 in sequence;

[0092] (7.3) A linear relationship between the wavelength change Δλ and the known salt concentration C is established, and the linear equation is C = 2Δλ-0.5. Finally, the salt measurement formula for in-situ saline-alkali frozen soil is obtained.

[0093] (8) According to the theoretical model of soil thermal conductivity λ = f(θ i ,θ w ), combined with the in-situ saline-alkali frozen soil moisture content θ w and equivalent thermal conductivity λ, calculate the total moisture content θ of the in-situ saline-alkali frozen soilt and ice content θ i , the total moisture content θ t =θ w +θ i .

[0094] (9) The weak reflection grating demodulation module (2) converts the wavelength data of the strain sensing unit (7) along the buried depth direction into strain information output, thereby obtaining the strain curve of the in-situ saline-alkali frozen soil along the depth direction.

[0095] In step (4), the neural network algorithm for correcting the parameter temperature difference output by the weak reflection grating demodulation module is an artificial neural network (ANN). The network calculates the prediction result through the forward propagation algorithm, and then uses the back propagation algorithm to adjust the weight according to the prediction error. These weights are continuously optimized through iterative training to minimize the loss function, thereby learning the mapping relationship from input to output and realizing the recognition and correction of complex data patterns.

[0096] In step (8), the equivalent thermal conductivity λ of step (6) and the equivalent thermal conductivity λ and moisture content θ of saline-alkali frozen soil are deduced according to the calibration formula in step (5). w 、Ice contentθ i The relationship between them is as follows:

[0097]

[0098] Among them, S r is the soil saturation coefficient, θ w is the soil moisture content of the in-situ saline-alkali frozen soil, n is the soil porosity, λ w is the thermal conductivity of water, λ i is the thermal conductivity of ice, λ s is the thermal conductivity of soil particles, k2 is the soil particle size distribution coefficient, x is the material parameter related to the thermal conductivity of soil particles, usually 0.75W / m℃, and η is the material coefficient related to the shape of soil particles, usually 1.2.

[0099] Among them, S r is the soil saturation coefficient, θ w is the soil moisture content of the in-situ saline-alkali frozen soil, θ i is the ice content of the in-situ saline-alkali frozen soil, n is the soil porosity, λ w is the thermal conductivity of water, λ i is the thermal conductivity of ice, λ s is the thermal conductivity of soil particles, k2, x and η are empirical coefficients related to soil particle size distribution and particle shape.

[0100] like Figure 8As shown in the figure, the freezing and thawing law of saline-alkali frozen soil is reflected by the moisture content-ice content of saline-alkali frozen soil. During the freezing period (January 1, 2023-January 23, 2023), the water in the soil freezes into ice, the ice content increases, the water content decreases, and the total water content increases; during the melting period (February 1, 2023-February 18, 2023), the ice crystals melt, the ice content decreases, the water content increases, the water migrates to the deep, and the total water content decreases.

[0101] The wavelength data of the strain sensing unit 7 is recorded by the weak reflection grating demodulation module 2 and automatically converted into strain information to obtain the strain time history curve, such as Fig. 9 As shown in the figure, the strain of saline-alkali frozen soil at different depths shows obvious stratification characteristics with depth changes. In the shallow layer (0-2m), the strain changes little and basically remains at a low level. The strain between 2-6m gradually increases, and the peak strain is about 350με. The strain in this depth segment is most significant. The strain between 6-8m gradually decreases and approaches the initial state. This is related to the mechanical properties and temperature changes of the saline-alkali frozen soil layer. Especially in the process of freezing and melting, the strain stratification phenomenon is obvious.

Claims

1. A multi-field synchronous monitoring device for saline-alkali frozen soil based on weak reflection grating, characterized in that: It comprises a temperature, humidity, salt and ice sensing module, a strain sensing module (7), an interlocking joint (6), a weak reflection grating demodulation module (2), a voltage-stabilized DC heating module (9) and a data analysis and processing module (1); The temperature, humidity, salt and ice sensing module is composed of a temperature sensing unit (3), a humidity-ice content sensing unit (4), and a salt sensing unit (5); the temperature, humidity, salt and ice sensing modules are separated by thermal insulation materials; A weak reflection grating (11) is provided in the temperature sensing unit (3), and a heat insulation interface (12) is connected between the temperature sensing unit (3) and the humidity-ice content sensing unit (4); The humidity-ice content sensing unit (4) has a built-in heating rod (13), and a weak reflection grating is used at the center of the heating rod to measure parameters; and capillary holes (15) are provided on the side of the salt sensing unit (5).

2. The multi-field synchronous monitoring device for saline-alkali frozen soil based on weak reflection grating according to claim 1 is characterized in that: The temperature, humidity, salt and ice sensing modules are fixed with interlocking joints (6).

3. The multi-field synchronous monitoring device for saline-alkali frozen soil based on weak reflection grating according to claim 1 is characterized in that: The temperature sensing unit (3) is encapsulated by a high thermal conductivity material to isolate strain.

4. The multi-field synchronous monitoring device for saline-alkali frozen soil based on weak reflection grating according to claim 1 is characterized in that: The surface of the optical fiber in the salt sensing unit (5) is coated with a weakening medium.

5. The multi-field synchronous monitoring device for saline-alkali frozen soil based on weak reflection grating according to claim 1 is characterized in that: The temperature sensing unit (3) provides temperature compensation for the strain sensing module (7).

6. The multi-field synchronous monitoring device for saline-alkali frozen soil based on weak reflection grating according to claim 1 is characterized in that: The strain sensing module (7) is packaged by a fully distributed sensing optical cable.

7. The multi-field synchronous monitoring device for saline-alkali frozen soil based on weak reflection grating according to claim 1 is characterized in that: The side wall of the humidity-ice content sensing unit is provided with continuous holes.

8. A method for synchronously monitoring multi-fields of saline-alkali frozen soil based on weak reflection grating, characterized in that: The monitoring method is implemented by using the multi-field synchronous monitoring device for saline-alkali frozen soil based on weak reflection grating as claimed in claim 1, and the monitoring method comprises the following steps: (1) drilling a hole on the surface of the saline-alkali frozen soil to be tested, connecting the temperature, humidity, salt and ice sensing modules in series, and then connecting them in parallel with the strain sensing module (7) and vertically arranging them in the saline-alkali frozen soil; (2) Take samples of saline-alkali frozen soil at the same depth of each temperature, humidity, salt and ice sensing unit, and measure the initial temperature, salinity, moisture content and ice content of the saline-alkali frozen soil at the corresponding depth; (3) connecting each sensing module component to a weak reflection grating demodulation module (2), connecting the temperature, humidity, salt and ice sensing module to a voltage-stabilized DC heating module (9), pulse heating the temperature, humidity, salt and ice sensor with constant power and time, and converting the wavelength data of the temperature along the buried depth direction into a temperature output by the weak reflection grating demodulation module, thereby obtaining a temperature time course curve of the in-situ saline-alkali frozen soil along the depth direction and the temporal and spatial distribution changes; (4) weak reflection grating sensing data are collected to construct an artificial neural network, wherein the number of nodes in the input layer of the artificial neural network corresponds to the central wavelength and reflectivity parameters output by the temperature sensing unit, and the number of nodes in the output layer corresponds to the initial temperature of the in-situ saline-alkali frozen soil measured on site in step (2). The parameters are forward propagated through each layer of the ANN, the wavelength data is processed by an activation function and a temperature prediction value is output. The error between the output temperature and the actual temperature is quantified using a loss function, and the weights of the parameters in the network are adjusted according to the prediction error. Multiple iterations of training and optimization are performed until the sensor parameters are mapped to the temperature, and finally the parameters are fed back to the weak reflection grating demodulation module to synchronously correct the output in-situ saline-alkali frozen soil temperature change. (5) Select constant heating power and temperature data during heating time to calculate the thermal conductivity λ of saline-alkali frozen soil at each depth, and calculate the temperature change ΔT measured by the temperature sensing unit after heating. t , derive the equivalent thermal conductivity λ of the soil: Among them, t is the heating time, q is the heating power, λ is the equivalent thermal conductivity of saline-alkali frozen soil, and c is a constant related to soil diffusivity; (6) Calibrate the in-situ saline-alkali frozen soil sample of step (1) and the thermal conductivity λ described in step (5) in an indoor test, and calculate the moisture content θ during the monitoring process of the saline-alkali frozen soil according to the calibration formula. w ; (7) performing indoor test calibration on the in-situ saline-alkali frozen soil sample of step (1) and the wavelength change Δλ recorded by the salt sensing unit, and calculating the salt concentration C during the saline-alkali frozen soil monitoring process according to the calibration formula; (8) According to the in-situ saline-alkali frozen soil moisture content θ w and equivalent thermal conductivity λ, calculate the total moisture content θ of the in-situ saline-alkali frozen soil t and ice content θ i ; Among them, S r is the soil saturation coefficient, θ w is the soil moisture content of the in-situ saline-alkali frozen soil, n is the soil porosity, λ w is the thermal conductivity of water, λ i is the thermal conductivity of ice, λ s is the thermal conductivity of soil particles, k2 is the soil particle size distribution coefficient, x is the material parameter related to the thermal conductivity of soil particles, and η is the material coefficient related to the shape of soil particles; The total moisture content θ t =θ w +θ i ; (9) The weak reflection grating demodulation module (2) converts the wavelength data of the strain sensing unit (7) along the buried depth direction into strain information output, thereby obtaining the strain curve of the in-situ saline-alkali frozen soil along the depth direction.

9. The method for synchronously monitoring multi-fields of saline-alkali frozen soil based on weak reflection grating according to claim 8 is characterized in that: In step (6), the moisture content indoor test calibration steps are: (6.1) Prepare the soil samples into several groups with known temperature T and moisture content θ at different depths. w samples; (6.2) Calculate the thermal conductivity λ of the soil at different depths according to the derived formula in step (5); (6.3) Establish the thermal conductivity λ and the known water content θ w The linear relationship is θ w =kλ+b, where k is the proportional constant and b is the intercept determined during the calibration process.

10. The method for synchronously monitoring multi-fields of saline-alkali frozen soil based on weak reflection grating according to claim 8 is characterized in that: In step (7), the indoor test calibration steps are: (7.1) Prepare several groups of samples with known salt concentration C from the soil samples taken at the monitoring site; (7.2) placing a salt sensing unit element of the same specification in a sample with a known salt concentration in step (8.1), and measuring the wavelength change Δλ of the sample; (7.3) A linear relationship between the wavelength change Δλ and the known salt concentration C is established, and the linear equation is C=kΔλ+b, where k is the proportionality constant and b is the intercept determined during the calibration process.

Images