Multi-parameter integrated detection method and device for frozen soil, terminal and storage medium

Abstract

The invention provides a multi-parameter integrated detection method and device for frozen soil, a terminal and a storage medium, and belongs to the technical field of frozen soil detection. The method comprises the following steps: acquiring soil temperature data of a to-be-detected area; wherein the soil temperature data comprises a three-dimensional space position, time and corresponding space-time temperature; inputting the soil temperature data into a pre-trained multi-parameter detection model to obtain the water content of soil, the ice content of the soil, the heat conductivity coefficient of the soil and the volume heat capacity of the soil in the to-be-detected area; wherein the multi-parameter detection model is obtained by training a PINN model based on a pre-constructed data set, and the training process of the multi-parameter detection model comprises the following steps: constructing a one-dimensional unsteady heat conduction equation according to soil temperature data of the frozen soil sample, constructing a loss function according to the one-dimensional unsteady heat conduction equation, and when the loss function reaches convergence, determining that the frozen soil sample is in a non-steady state; and training to obtain a multi-parameter detection model. The device can detect various parameters in frozen soil at the same time.

Description

Technical Field

[0001] The present invention relates to the technical field of frozen soil detection, and in particular to a multi-parameter integrated detection method, device, terminal and storage medium for frozen soil. Background Art

[0002] Ice-rich permafrost is widely distributed in high latitudes and high altitudes, and its physical and thermal parameters are of great significance to engineering construction, environmental monitoring and climate research. Accurately measuring parameters such as temperature, water content, ice content, thermal conductivity and volumetric heat capacity of ice-rich permafrost is crucial for predicting permafrost changes, assessing engineering stability and understanding climate change.

[0003] In the prior art, temperature measurement usually uses the method of burying thermocouples or thermistors, which can obtain soil temperature but cannot provide other physical parameters. Water content measurement uses time domain reflectometry (TDR) or frequency domain reflectometry (FDR) to infer soil moisture content by measuring the dielectric constant, but the accuracy decreases in frozen soil. Ice content measurement mostly uses methods such as sampling analysis or nuclear magnetic resonance, which are complicated and cannot be monitored in real time. The measurement of thermal conductivity and volumetric heat capacity uses steady-state or non-steady-state thermal conductivity methods, such as the hot plate method or the line heat source method, which requires complex experimental conditions and is difficult to apply in the field. The above methods are independent of each other and cannot achieve synchronous measurement of multiple parameters, and their application is limited in ice-rich frozen soil environments. Therefore, there is an urgent need for a method that can simultaneously detect multiple parameters in frozen soil. Summary of the invention

[0004] The embodiments of the present invention provide a method, device, terminal and storage medium for integrated multi-parameter detection of frozen soil, so as to solve the problem that multiple parameters in frozen soil cannot be detected simultaneously.

[0005] In a first aspect, an embodiment of the present invention provides a multi-parameter integrated detection method for frozen soil, comprising: Acquire soil temperature data of the area to be detected; wherein the soil temperature data includes three-dimensional spatial position, time and corresponding spatiotemporal temperature; The soil temperature data is input into a pre-trained multi-parameter detection model to obtain the water content, ice content, thermal conductivity and volumetric heat capacity of the soil in the area to be detected; wherein the multi-parameter detection model is obtained by training the PINN model based on a pre-constructed data set, and the data set includes the soil temperature data of the frozen soil sample, the water content, ice content, thermal conductivity and volumetric heat capacity of the frozen soil sample; The training process of the multi-parameter detection model includes: The soil temperature data of the frozen soil samples are taken as the input of the PINN model, and the water content, ice content, thermal conductivity and volumetric heat capacity of the frozen soil samples are taken as the output of the PINN model to train the PINN model; a one-dimensional non-steady-state heat conduction equation is constructed according to the soil temperature data of the frozen soil samples, and a loss function is constructed according to the one-dimensional non-steady-state heat conduction equation. When the loss function reaches convergence, a multi-parameter detection model is obtained through training.

[0006] In a possible implementation, the loss function includes a data loss function and a physical loss function; The loss function is constructed based on the one-dimensional unsteady heat conduction equation, including: Construct data loss function and physical loss function based on one-dimensional unsteady heat conduction equation; Construct the loss function based on the data loss function and the physical loss function.

[0007] In one possible implementation, the one-dimensional unsteady heat conduction equation is:

[0008] In the formula, T ( x,t )for x Location t The soil temperature at the time, ρ , c , k are the density, volumetric heat capacity and thermal conductivity of the soil, respectively. Q ( x , t ) is the heat source term; The loss function is:

[0009] In the formula, λ data is the weight coefficient, is the data loss function, λ physics is the weight coefficient, is the physical loss function; The data loss function is:

[0010] In the formula, To predict the temperature, T obs To observe the temperature, N is the number of data points; The physical loss function is:

[0011] In the formula, To predict the temperature,M is the number of sampling points, ρ , c , k are the density, volumetric heat capacity and thermal conductivity of the soil, respectively. Q is the heat source term.

[0012] In a possible implementation, obtaining soil temperature data of a to-be-detected area includes: The sensor is implanted into the frozen soil of the area to be detected; wherein the sensor includes three parallel probes, the middle probe can heat and measure the temperature, and the probes on both sides can measure the temperature; Make the middle probe apply pulses to the surrounding soil; The temperature data of the three probes that change with spatial position and time are collected at the same time to obtain soil temperature data.

[0013] In a possible implementation, before inputting soil temperature data into a pre-trained multi-parameter detection model, the following steps are further included: The soil temperature data of the area to be detected is filtered, the noise of the filtered soil temperature data is removed, and the noise-removed soil temperature data is input into a pre-trained multi-parameter detection model.

[0014] In a possible implementation, the soil temperature data of the area to be detected is filtered, including: Using a filter to filter the soil temperature data to obtain filtered soil temperature data; Remove noise from the filtered soil temperature data, including: The soil temperature data after filtering is subjected to noise removal by wavelet transform or median filtering to obtain the soil temperature data after noise removal.

[0015] In a possible implementation, the method further includes: Acquire a data set; the data set includes soil temperature data of the frozen soil sample, water content of the frozen soil sample, ice content of the frozen soil sample, thermal conductivity of the frozen soil sample, and volumetric heat capacity of the frozen soil sample; Use a preset proportion of data in the dataset as the training set and the remaining data as the test set; The PINN model is trained according to the training set, and the PINN model is tested according to the test set to obtain a multi-parameter detection model.

[0016] In a second aspect, an embodiment of the present invention provides a multi-parameter integrated detection device, comprising: A data acquisition module is used to acquire soil temperature data of the area to be detected; wherein the soil temperature data includes a three-dimensional spatial position, time and corresponding spatiotemporal temperature; A parameter detection module is used to input soil temperature data into a pre-trained multi-parameter detection model to obtain the water content of the soil in the area to be detected, the ice content of the soil, the thermal conductivity of the soil, and the volumetric heat capacity of the soil; wherein the multi-parameter detection model is obtained by training the PINN model based on a pre-constructed data set, and the data set includes soil temperature data of frozen soil samples, water content of frozen soil samples, ice content of frozen soil samples, thermal conductivity of frozen soil samples, and volumetric heat capacity of frozen soil samples; The training process of the multi-parameter detection model includes: The soil temperature data of the frozen soil samples are taken as the input of the PINN model, and the water content, ice content, thermal conductivity and volumetric heat capacity of the frozen soil samples are taken as the output of the PINN model to train the PINN model; a one-dimensional non-steady-state heat conduction equation is constructed according to the soil temperature data of the frozen soil samples, and a loss function is constructed according to the one-dimensional non-steady-state heat conduction equation. When the loss function reaches convergence, a multi-parameter detection model is obtained through training.

[0017] In a third aspect, an embodiment of the present invention provides a terminal, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the method described in the first aspect or any possible implementation of the first aspect are implemented.

[0018] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the steps of the method described in the first aspect or any possible implementation of the first aspect.

[0019] The embodiment of the present invention provides a multi-parameter integrated detection method, device, terminal and storage medium of frozen soil, by acquiring soil temperature data of the area to be detected; wherein the soil temperature data includes three-dimensional spatial position, time and corresponding spatiotemporal temperature, which is the basis for subsequent detection. The soil temperature data is input into a pre-trained multi-parameter detection model; wherein the multi-parameter detection model is obtained by training a PINN model based on a pre-constructed data set, and the data set includes soil temperature data of frozen soil samples, water content of frozen soil samples, ice content of frozen soil samples, thermal conductivity of frozen soil samples and volumetric heat capacity of frozen soil samples; the training process of the multi-parameter detection model includes: taking the soil temperature data of the frozen soil sample as the input of the PINN model, taking the water content of the frozen soil sample, ice content of the frozen soil sample, thermal conductivity of the frozen soil sample and volumetric heat capacity of the frozen soil sample as the output of the PINN model, and training the PINN model; constructing a one-dimensional non-steady-state heat conduction equation according to the soil temperature data of the frozen soil sample, constructing a loss function according to the one-dimensional non-steady-state heat conduction equation, evaluating the difference between the model prediction value and the true value according to the loss function, and optimizing the model parameters. When the loss function reaches convergence, a multi-parameter detection model is trained to facilitate the simultaneous detection of multiple parameters such as water content, ice content, thermal conductivity and volumetric heat capacity of frozen soil. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0021] Figure 1 It is a flowchart of the implementation of the multi-parameter integrated detection method of frozen soil provided by the embodiment of the present invention; Figure 2 It is a structural schematic diagram of a multi-parameter integrated detection device for frozen soil provided by an embodiment of the present invention; Figure 3 is a schematic diagram of a terminal provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0022] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present invention. However, it should be clear to those skilled in the art that the present invention may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present invention.

[0023] In order to make the purpose, technical solutions and advantages of the present invention more clear, specific embodiments will be described below in conjunction with the accompanying drawings.

[0024] See also Figure 1 , which shows a flowchart of the implementation of the multi-parameter integrated detection method of frozen soil provided by an embodiment of the present invention, which is described in detail as follows: Step 101: Acquire soil temperature data of the area to be detected; wherein the soil temperature data includes three-dimensional spatial position, time and corresponding spatiotemporal temperature.

[0025] In this embodiment, the three-dimensional spatial position is a coordinate point in a three-dimensional coordinate system (usually x, y, z axis), which helps to understand the distribution of soil temperature at different depths and horizontal positions, because the soil temperature may vary with depth and position. For example, in a frozen soil layer, the soil temperature close to the surface may be directly affected by changes in atmospheric temperature, while the temperature of the deep soil is relatively stable. By recording the three-dimensional spatial position, a spatial distribution map of the soil temperature can be more accurately depicted.

[0026] At the same time, soil temperature changes over time, especially when the seasons change, the day and night cycle, or is affected by external heat sources (such as solar radiation, human activities, etc.). Recording time information can capture these dynamic changes and analyze the temporal evolution of soil temperature. For example, during the day, the temperature of the soil surface may rise and fall as the sun rises and sets. By recording temperature data at different times, this daily variation characteristic can be studied.

[0027] Moreover, spatiotemporal temperature is a temperature data with integrated spatial and temporal dimensions, which is used to describe the change of soil temperature at different locations and time points. The soil temperature value measured at a specific three-dimensional spatial location and time point is the most direct temperature data, which is used to reflect the actual temperature state of the soil at that location and time. By combining the temperature value with the spatial location and time, a complete spatiotemporal evolution model of soil temperature can be constructed, providing detailed data support for subsequent parameter inversion and frozen soil characteristic analysis.

[0028] In a possible implementation, soil temperature data of the area to be detected is obtained by the following specific processing: a sensor is implanted into the frozen soil of the area to be detected; wherein the sensor includes three parallel probes, the middle probe can heat and measure temperature, and the probes on both sides can measure temperature; the middle probe applies pulses to the surrounding soil; and the temperature data of the three probes that change with spatial position and time are collected at the same time to obtain soil temperature data.

[0029] In this embodiment, the sensor is implanted in the frozen soil of the area to be detected to ensure that the probe is in close contact with the soil and reduce the contact thermal resistance so as to accurately measure the soil temperature. The middle probe can both heat the soil and measure the temperature. The heating function is used to apply transient heat pulses, and the temperature measurement function is used to record temperature changes. The probes on both sides are only used to measure temperature and record temperature changes in the surrounding soil. The sensor applies the same amount of energy to the soil as in the data calibration process during the previous training model, and collects the spatiotemporal evolution data of soil heat read by the probe. A transient heat pulse is applied to the surrounding soil through the middle probe to stimulate the transient thermal response of the soil. This transient thermal response can provide information on the thermophysical properties of the soil (such as thermal conductivity, volumetric heat capacity, etc.). The pulse duration is usually 10 to 20 seconds to avoid damage to the soil structure.

[0030] In addition, probes can be made of materials with different thermal conductivity to adapt to different temperature ranges and soil conditions. The heating method can be electric heating, laser heating or pulse heating to further improve the controllability of transient thermal response. By increasing the number of probes, a multi-point measurement array is formed to obtain more comprehensive spatial distribution information.

[0031] Step 102: input the soil temperature data into a pre-trained multi-parameter detection model to obtain the water content of the soil in the area to be detected, the ice content of the soil, the thermal conductivity of the soil and the volumetric heat capacity of the soil; wherein the multi-parameter detection model is obtained by training the PINN model based on a pre-constructed data set, and the data set includes the soil temperature data of the frozen soil sample, the water content of the frozen soil sample, the ice content of the frozen soil sample, the thermal conductivity of the frozen soil sample and the volumetric heat capacity of the frozen soil sample; the training process of the multi-parameter detection model is as follows: the soil temperature data of the frozen soil sample is used as the input of the PINN model, and the water content of the frozen soil sample, the ice content of the frozen soil sample, the thermal conductivity of the frozen soil sample and the volumetric heat capacity of the frozen soil sample are used as the output of the PINN model to train the PINN model; construct a one-dimensional non-steady-state heat conduction equation according to the soil temperature data of the frozen soil sample, and construct a loss function according to the one-dimensional non-steady-state heat conduction equation. When the loss function reaches convergence, the multi-parameter detection model is obtained through training.

[0032] In this embodiment, by inputting soil temperature data into the PINN model, the water content of the soil in the area to be detected, the ice content of the soil, the thermal conductivity of the soil and the volumetric heat capacity of the soil can be obtained. These parameters are crucial for understanding the physical properties and thermal behavior of the soil.

[0033] In this embodiment, when constructing a data set, the sensor is vertically implanted into a pre-made frozen soil sample with a known ice content to ensure that the probe is in close contact with the soil and reduce the contact thermal resistance. A transient heat pulse is applied to the surrounding soil through the middle probe that can be heated to stimulate a thermal response. The time and power of the heating process are controllable to avoid damage to the soil structure. For soils with different soil properties, there is no fixed value for the value of the heat input per unit line heat source. The control principle is: the duration of the pulse heat source is 10~20s to prevent excessive power from causing frozen soil melting; after the energy is input, if the temperature rise characteristic curves of soils with different ice contents do not intersect each other, then the input power is suitable for measurement. When the input heat and power are determined during calibration, the measurement process must be carried out strictly in accordance with the heat input value and power. During data acquisition, the temperature changes of the three probes are recorded simultaneously to obtain a complete temperature rise characteristic curve. These data are obtained through calibration experiments under laboratory conditions, ensuring the accuracy and reliability of the data. The water content of the frozen soil sample, the ice content of the frozen soil sample, the thermal conductivity of the frozen soil sample and the volumetric heat capacity of the frozen soil sample are experimentally measured or known calibration values.

[0034] In this embodiment, a one-dimensional non-steady-state heat conduction equation is constructed based on the soil temperature data of the frozen soil sample. This equation describes the physical process of soil temperature changing with time and space. The loss function is used to measure the difference between the model prediction value and the actual value. By optimizing the loss function, the model prediction value is made as close to the actual value as possible. When the loss function reaches convergence, it means that the difference between the model prediction value and the actual value is small enough, and the model can accurately predict multiple physical parameters of the soil.

[0035] In one possible implementation, the loss function includes a data loss function and a physical loss function; the loss function is constructed according to the one-dimensional non-steady-state heat conduction equation, and the specific processing is as follows: the data loss function and the physical loss function are constructed according to the one-dimensional non-steady-state heat conduction equation; the loss function is constructed according to the data loss function and the physical loss function.

[0036] In this embodiment, the data loss function can measure the difference between the model prediction value and the actual observation value. The physical loss function can measure whether the model prediction value satisfies the residual of the physical equation (such as the one-dimensional non-steady-state heat conduction equation). The data loss function and the physical loss function are combined to form a comprehensive loss function, which is used for model training and optimization to comprehensively measure the prediction performance of the model.

[0037] In one possible implementation, the one-dimensional unsteady heat conduction equation is:

[0038] In the formula, T ( x,t )for xLocation t The soil temperature at the time, ρ , c , k are the density, volumetric heat capacity and thermal conductivity of the soil, respectively. Q ( x , t ) is the heat source term; The loss function is:

[0039] In the formula, λ data is the weight coefficient, is the data loss function, λ physics is the weight coefficient, is the physical loss function; The data loss function is:

[0040] In the formula, To predict the temperature, T obs To observe the temperature, N is the number of data points; The physical loss function is:

[0041] In the formula, To predict the temperature, M is the number of sampling points, ρ , c , k are the density, volumetric heat capacity and thermal conductivity of the soil, respectively. Q is the heat source term.

[0042] In this embodiment, when training the PINN model, the network parameters and the physical parameters to be inverted (ρ, c, k) are first randomly initialized. In the forward propagation, the position x, time t and temperature T(x, t) are input to calculate the predicted water content, ice content, volumetric heat capacity and thermal conductivity. When calculating the loss, according to the loss function Calculate the total loss. Calculate the gradient of the loss to the network parameters and physical parameters in the back propagation. Use the optimization algorithm (Adam optimizer) to update the network parameters and physical parameters during parameter update. Repeat the above process until the loss function converges.

[0043] Specifically, the thermal conductivity of the soil as a whole is k The correlation between the thermal conductivity of each component (used to solve multiple parameters such as water content, ice content, thermal conductivity and volumetric heat capacity) is:

[0044] In the formula, k dry , k w , k i They are the thermal conductivity of dry soil, the thermal conductivity of water, and the thermal conductivity of ice; θ w , θ i are the volumetric water content and volumetric ice content of the soil, respectively.

[0045] The embodiment of the present invention obtains soil temperature data of the area to be detected; wherein the soil temperature data includes three-dimensional spatial position, time and corresponding spatiotemporal temperature, which is the basis for subsequent detection. The soil temperature data is input into a pre-trained multi-parameter detection model; wherein the multi-parameter detection model is obtained by training the PINN model based on a pre-constructed data set, and the data set includes soil temperature data of frozen soil samples, water content of frozen soil samples, ice content of frozen soil samples, thermal conductivity of frozen soil samples and volumetric heat capacity of frozen soil samples; the training process of the multi-parameter detection model includes: taking the soil temperature data of the frozen soil samples as the input of the PINN model, taking the water content of the frozen soil samples, ice content of the frozen soil samples, thermal conductivity of the frozen soil samples and volumetric heat capacity of the frozen soil samples as the output of the PINN model, and training the PINN model; constructing a one-dimensional non-steady-state heat conduction equation according to the soil temperature data of the frozen soil samples, constructing a loss function according to the one-dimensional non-steady-state heat conduction equation, evaluating the difference between the model prediction value and the true value according to the loss function, and optimizing the model parameters. When the loss function reaches convergence, a multi-parameter detection model is trained to facilitate the simultaneous detection of multiple parameters such as water content, ice content, thermal conductivity and volumetric heat capacity of frozen soil.

[0046] In a possible implementation, before inputting the soil temperature data into a pre-trained multi-parameter detection model, the following processing may be performed: filtering the soil temperature data of the area to be detected, removing noise from the filtered soil temperature data, and inputting the soil temperature data after noise removal into the pre-trained multi-parameter detection model.

[0047] In this embodiment, filtering can smooth the data, reduce high-frequency noise and unnecessary fluctuations in the data, and make the data more stable and reliable. Denoising can further clean up the data and remove random errors and interference that may affect the accuracy of model prediction.

[0048] In a possible implementation, the soil temperature data of the area to be detected is filtered, and the specific processing is as follows: the soil temperature data is filtered using a filter to obtain filtered soil temperature data; the filtered soil temperature data is de-noised, and the specific processing is as follows: the filtered soil temperature data is de-noised using a wavelet transform or a median filter to obtain de-noised soil temperature data.

[0049] In this embodiment, according to the sampling frequency and signal characteristics, a Butterworth filter or a Chebyshev filter is selected to filter the original temperature data to obtain a smooth temperature-time curve. The filter transfer function is:

[0050] f c is the cut-off frequency, n is the filter order.

[0051] Filtered temperature signal:

[0052] F and F -1 denote Fourier transform and inverse Fourier transform respectively, Represents the raw temperature signal.

[0053] In a possible implementation, the method can also be processed as follows: obtaining a data set; the data set includes soil temperature data of the frozen soil sample, the water content of the frozen soil sample, the ice content of the frozen soil sample, the thermal conductivity of the frozen soil sample and the volumetric heat capacity of the frozen soil sample; using a preset proportion of data in the data set as a training set, and using the remaining data as a test set; training the PINN model according to the training set, and testing the PINN model according to the test set to obtain a multi-parameter detection model.

[0054] In this embodiment, a preset proportion (e.g., 80%) of the data in the data set is used as a training set. The training set is used to train the model so that the model learns the patterns and relationships in the data. The remaining data (e.g., 20%) is used as a test set. The test set is used to verify the performance of the model and evaluate the performance of the model on unseen data. The performance of the model is verified by evaluating the test set. If the model performs well on the test set, it means that the model has good generalization ability and can be used for actual multi-parameter detection.

[0055] In other possible ways, one or more stages in the PINN model and the sensor may adopt other ways. This application mainly uses the PINN model and sensor processing as an example for explanation.

[0056] The present invention uses a three-probe sensor coupled at close range to measure transient thermal response. The series of parameters and measurement processes involved are used to achieve multi-parameter integrated perception of ice-rich frozen soil, including high-precision synchronous measurement of physical parameters such as temperature, water content, ice content, thermal conductivity and volumetric heat capacity. By precisely controlling the heating process and data processing, the accuracy and reliability of the measurement can be significantly improved.

[0057] This invention breaks through the limitation of traditional methods that can only measure a single parameter, and greatly improves the measurement efficiency. The transient thermal response and advanced data processing algorithms significantly improve the accuracy of the measurement. The sensor has a simple structure and is easy to deploy on site, suitable for various complex ice-rich frozen soil environments. For the first time, the close-coupled three-probe technology is combined with machine learning to achieve integrated perception of multiple parameters of ice-rich frozen soil.

[0058] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0059] The following is an embodiment of the device of the present invention. For details not described in detail, reference may be made to the corresponding method embodiments described above.

[0060] Figure 2 The structure diagram of the multi-parameter integrated detection device for frozen soil provided by the embodiment of the present invention is shown. For the convenience of description, only the part related to the embodiment of the present invention is shown, which is described in detail as follows: like Figure 2 As shown, the multi-parameter integrated detection device 2 for frozen soil includes: The data acquisition module 21 is used to acquire soil temperature data of the area to be detected; wherein the soil temperature data includes a three-dimensional spatial position, time and corresponding spatiotemporal temperature; The parameter detection module 22 is used to input the soil temperature data into a pre-trained multi-parameter detection model to obtain the water content of the soil in the area to be detected, the ice content of the soil, the thermal conductivity of the soil, and the volumetric heat capacity of the soil; wherein the multi-parameter detection model is obtained by training the PINN model based on a pre-constructed data set, and the data set includes the soil temperature data of the frozen soil sample, the water content of the frozen soil sample, the ice content of the frozen soil sample, the thermal conductivity of the frozen soil sample, and the volumetric heat capacity of the frozen soil sample; The training process of the multi-parameter detection model includes: The soil temperature data of the frozen soil samples are taken as the input of the PINN model, and the water content, ice content, thermal conductivity and volumetric heat capacity of the frozen soil samples are taken as the output of the PINN model to train the PINN model; a one-dimensional non-steady-state heat conduction equation is constructed according to the soil temperature data of the frozen soil samples, and a loss function is constructed according to the one-dimensional non-steady-state heat conduction equation. When the loss function reaches convergence, a multi-parameter detection model is obtained through training.

[0061] In a possible implementation, the loss function includes a data loss function and a physical loss function, and the parameter detection module 22 can be used to: Construct data loss function and physical loss function based on one-dimensional unsteady heat conduction equation; Construct the loss function based on the data loss function and the physical loss function.

[0062] In a possible implementation, the data acquisition module 21 may be used to: The sensor is implanted into the frozen soil of the area to be detected; wherein the sensor includes three parallel probes, the middle probe can heat and measure the temperature, and the probes on both sides can measure the temperature; Make the middle probe apply pulses to the surrounding soil; The temperature data of the three probes that change with spatial position and time are collected at the same time to obtain soil temperature data.

[0063] In a possible implementation, the data acquisition module 21 may be used to: The soil temperature data of the area to be detected is filtered, the noise of the filtered soil temperature data is removed, and the noise-removed soil temperature data is input into a pre-trained multi-parameter detection model.

[0064] In a possible implementation, the data acquisition module 21 may be used to: Using a filter to filter the soil temperature data to obtain filtered soil temperature data; Remove noise from the filtered soil temperature data, including: The soil temperature data after filtering is subjected to noise removal by wavelet transform or median filtering to obtain the soil temperature data after noise removal.

[0065] In a possible implementation, the parameter detection module 22 may be used to: Acquire a data set; the data set includes soil temperature data of the frozen soil sample, water content of the frozen soil sample, ice content of the frozen soil sample, thermal conductivity of the frozen soil sample, and volumetric heat capacity of the frozen soil sample; Use a preset proportion of data in the dataset as the training set and the remaining data as the test set; The PINN model is trained according to the training set, and the PINN model is tested according to the test set to obtain a multi-parameter detection model.

[0066] The embodiment of the present invention obtains soil temperature data of the area to be detected; wherein the soil temperature data includes three-dimensional spatial position, time and corresponding spatiotemporal temperature, which is the basis for subsequent detection. The soil temperature data is input into a pre-trained multi-parameter detection model; wherein the multi-parameter detection model is obtained by training the PINN model based on a pre-constructed data set, and the data set includes soil temperature data of frozen soil samples, water content of frozen soil samples, ice content of frozen soil samples, thermal conductivity of frozen soil samples and volumetric heat capacity of frozen soil samples; the training process of the multi-parameter detection model includes: taking the soil temperature data of the frozen soil samples as the input of the PINN model, taking the water content of the frozen soil samples, ice content of the frozen soil samples, thermal conductivity of the frozen soil samples and volumetric heat capacity of the frozen soil samples as the output of the PINN model, and training the PINN model; constructing a one-dimensional non-steady-state heat conduction equation according to the soil temperature data of the frozen soil samples, constructing a loss function according to the one-dimensional non-steady-state heat conduction equation, evaluating the difference between the model prediction value and the true value according to the loss function, and optimizing the model parameters. When the loss function reaches convergence, a multi-parameter detection model is trained to facilitate the simultaneous detection of multiple parameters such as water content, ice content, thermal conductivity and volumetric heat capacity of frozen soil.

[0067] Figure 3 is a schematic diagram of a terminal provided by an embodiment of the present invention. Figure 3 As shown, the terminal 3 of this embodiment includes: a processor 30, a memory 31, and a computer program 32 stored in the memory 31 and executable on the processor 30. When the processor 30 executes the computer program 32, the steps in the above-mentioned multi-parameter integrated detection method of frozen soil are implemented, for example Figure 1 Alternatively, when the processor 30 executes the computer program 32, the functions of the modules in the above-mentioned device embodiments are realized, for example Figure 2 The functions of each module are shown.

[0068] Exemplarily, the computer program 32 may be divided into one or more modules, which are stored in the memory 31 and executed by the processor 30 to implement the present invention. The one or more modules may be a series of computer program instruction segments capable of implementing specific functions, which are used to describe the execution process of the computer program 32 in the terminal 3. For example, the computer program 32 may be divided into Figure 2 The modules shown.

[0069] The terminal 3 may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The terminal 3 may include, but is not limited to, a processor 30 and a memory 31. Those skilled in the art will appreciate that Figure 3 It is only an example of terminal 3 and does not constitute a limitation on terminal 3. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the terminal may also include input and output devices, network access devices, buses, etc.

[0070] The processor 30 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0071] The memory 31 may be an internal storage unit of the terminal 3, such as a hard disk or memory of the terminal 3. The memory 31 may also be an external storage device of the terminal 3, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the terminal 3. Further, the memory 31 may also include both an internal storage unit and an external storage device of the terminal 3. The memory 31 is used to store the computer program and other programs and data required by the terminal. The memory 31 may also be used to temporarily store data that has been output or is to be output.

[0072] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0073] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0074] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0075] In the embodiments provided by the present invention, it should be understood that the disclosed devices / terminals and methods can be implemented in other ways. For example, the device / terminal embodiments described above are only schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0076] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0077] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0078] If the integrated module is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned multi-parameter integrated detection method embodiments of each frozen soil can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device that can carry the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practices in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practices, computer-readable media does not include electrical carrier signals and telecommunication signals.

[0079] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A multi-parameter integrated detection method for frozen soil, characterized in that: include: Acquire soil temperature data of the area to be detected; wherein the soil temperature data includes three-dimensional spatial position, time and corresponding spatiotemporal temperature; The soil temperature data is input into a pre-trained multi-parameter detection model to obtain the water content of the soil in the area to be detected, the ice content of the soil, the thermal conductivity of the soil, and the volumetric heat capacity of the soil; wherein the multi-parameter detection model is obtained by training the PINN model based on a pre-constructed data set, and the data set includes the soil temperature data of the frozen soil sample, the water content of the frozen soil sample, the ice content of the frozen soil sample, the thermal conductivity of the frozen soil sample, and the volumetric heat capacity of the frozen soil sample; The training process of the multi-parameter detection model includes: The soil temperature data of the frozen soil sample is used as the input of the PINN model, and the water content of the frozen soil sample, the ice content of the frozen soil sample, the thermal conductivity of the frozen soil sample and the volumetric heat capacity of the frozen soil sample are used as the output of the PINN model to train the PINN model; a one-dimensional non-steady-state heat conduction equation is constructed according to the soil temperature data of the frozen soil sample, and a loss function is constructed according to the one-dimensional non-steady-state heat conduction equation; when the loss function reaches convergence, the multi-parameter detection model is obtained through training.

2. The multi-parameter integrated detection method for frozen soil according to claim 1 is characterized in that: The loss function includes a data loss function and a physical loss function; The step of constructing a loss function according to the one-dimensional non-steady-state heat conduction equation comprises: Constructing the data loss function and the physical loss function according to the one-dimensional non-steady-state heat conduction equation; A loss function is constructed according to the data loss function and the physical loss function.

3. The multi-parameter integrated detection method for frozen soil according to claim 2 is characterized in that: The one-dimensional unsteady heat conduction equation is: In the formula, T ( x,t )for x Location t The soil temperature at the time, ρ , c , k are the density, volumetric heat capacity and thermal conductivity of the soil, respectively. Q ( x , t ) is the heat source term; The loss function is: In the formula, λ data is the weight coefficient, is the data loss function, λ physics is the weight coefficient, is the physical loss function; The data loss function is: In the formula, To predict the temperature, T obs To observe the temperature, N is the number of data points; The physical loss function is: In the formula, To predict the temperature, M is the number of sampling points, ρ , c , k are the density, volumetric heat capacity and thermal conductivity of the soil, respectively. Q is the heat source term.

4. The multi-parameter integrated detection method for frozen soil according to claim 1 is characterized in that: The step of obtaining soil temperature data of the area to be detected includes: The sensor is implanted into the frozen soil of the area to be detected; wherein the sensor includes three probes arranged in parallel, the middle probe can heat and measure the temperature, and the probes on both sides can measure the temperature; Make the middle probe apply pulses to the surrounding soil; The temperature data of the three probes that change with spatial position and time are collected simultaneously to obtain the soil temperature data.

5. The multi-parameter integrated detection method for frozen soil according to claim 1 is characterized in that: Before inputting the soil temperature data into the pre-trained multi-parameter detection model, the method further includes: The soil temperature data of the area to be detected is filtered, the noise of the filtered soil temperature data is removed, and the soil temperature data after the noise is removed is input into a pre-trained multi-parameter detection model.

6. The multi-parameter integrated detection method for frozen soil according to claim 5 is characterized in that: The filtering process of the soil temperature data of the area to be detected includes: Using a filter to filter the soil temperature data to obtain the filtered soil temperature data; The step of removing noise from the soil temperature data after filtering comprises: The soil temperature data after filtering is subjected to noise removal processing by using wavelet transform or median filtering to obtain the soil temperature data after noise removal.

7. The multi-parameter integrated detection method for frozen soil according to claim 1, characterized in that: The method further comprises: Acquire a data set; the data set includes soil temperature data of the frozen soil sample, water content of the frozen soil sample, ice content of the frozen soil sample, thermal conductivity of the frozen soil sample, and volumetric heat capacity of the frozen soil sample; Using a preset proportion of data in the data set as a training set, and using the remaining data as a test set; The PINN model is trained according to the training set, and the PINN model is tested according to the test set to obtain the multi-parameter detection model.

8. A multi-parameter integrated detection device for frozen soil, characterized in that: include: A data acquisition module, used to acquire soil temperature data of the area to be detected; wherein the soil temperature data includes a three-dimensional spatial position, time and corresponding spatiotemporal temperature; A parameter detection module, used for inputting the soil temperature data into a pre-trained multi-parameter detection model to obtain the water content of the soil in the area to be detected, the ice content of the soil, the thermal conductivity of the soil, and the volumetric heat capacity of the soil; wherein the multi-parameter detection model is obtained by training the PINN model based on a pre-constructed data set, and the data set includes the soil temperature data of the frozen soil sample, the water content of the frozen soil sample, the ice content of the frozen soil sample, the thermal conductivity of the frozen soil sample, and the volumetric heat capacity of the frozen soil sample; The training process of the multi-parameter detection model includes: The soil temperature data of the frozen soil sample is used as the input of the PINN model, and the water content of the frozen soil sample, the ice content of the frozen soil sample, the thermal conductivity of the frozen soil sample and the volumetric heat capacity of the frozen soil sample are used as the output of the PINN model to train the PINN model; a one-dimensional non-steady-state heat conduction equation is constructed according to the soil temperature data of the frozen soil sample, and a loss function is constructed according to the one-dimensional non-steady-state heat conduction equation; when the loss function reaches convergence, the multi-parameter detection model is obtained through training.

9. A terminal comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method as claimed in any one of claims 1 to 7 are implemented.

Images