Real-time synchronous testing device for soil freezing process and hyperspectral imaging

Abstract

The invention relates to a soil freezing process and hyperspectral imaging real-time synchronous testing device, and belongs to the technical field of frozen soil physical testing and hyperspectral imaging. The device comprises a freezing bin, a sample box, a hyperspectral imaging system and a light source system. The freezing bin is of a hollow structure with a groove, an observation window is formed in the top of the freezing bin, and temperature control modules are arranged at the two ends respectively. The sample box consists of a bottom groove plate and a top plate, and a sample area for filling a soil sample is formed between the bottom groove plate and the top plate; a plurality of thermocouple sensors are integrated in the sample box, one end of each thermocouple sensor is closed, and the other end of each thermocouple sensor is open; the hyperspectral imaging system comprises a hyperspectral camera and a data processing module; the light sources are symmetrically distributed on the two sides of the hyperspectral camera. Through structure integration and function coupling, high fusion of hyperspectral image acquisition and temperature synchronous monitoring in the soil freezing process is achieved, and the device has the advantages of being high in test precision, good in imaging quality, convenient to operate, low in cost and the like.

Description

Technical Field

[0001] The invention belongs to the technical field of frozen soil water and heat migration research, and in particular relates to a soil freezing process and hyperspectral imaging synchronous testing device. Background Art

[0002] The soil freezing process involves complex water and heat migration, ice crystal growth, unfrozen water retention and component spatial redistribution. These physical processes directly affect the stability of engineering foundations in frozen areas, the safety control of artificial freezing construction, the evolution of cold-region ecosystems, and agricultural frost damage assessment. Therefore, accurately monitoring the evolution of water status during soil freezing has important theoretical significance and engineering application value for studying the mechanism of water and heat migration in frozen soil.

[0003] At present, the measurement of soil moisture status during freezing mainly relies on soil moisture meters and X-ray CT scanning. Among them, although X-ray CT scanning can obtain microstructural images inside the soil, due to the limited difference in density between ice and water, it has obvious limitations in distinguishing ice lenses from unfrozen water. At the same time, CT equipment is expensive, bulky, and has high requirements for the operating environment.

[0004] In contrast, hyperspectral imaging technology has the advantages of non-contact, non-destructive, and multi-band parallel measurement, and has been successfully applied to soil texture analysis, organic matter content identification, and moisture content estimation. However, the integrated application of this technology in the process of soil freezing is still in its infancy, lacking a hyperspectral imaging system architecture and a dedicated synchronous testing device suitable for freezing environments.

[0005] Therefore, there is an urgent need to develop an integrated testing device with compact structure, easy operation, moderate cost, and the ability to achieve real-time synchronous acquisition of freezing process and hyperspectral imaging data, so as to improve the accuracy, efficiency and reliability of visualization research on the moisture status of frozen soil. Summary of the invention

[0006] In order to overcome the problems of insufficient accuracy in identifying moisture status during soil freezing, difficulty in synchronous observation, large size and high cost of the experimental system in the prior art, the present invention proposes a real-time synchronous test device for soil freezing process and hyperspectral imaging, which realizes the integration, automation and visualization of the freezing process and hyperspectral imaging, and is suitable for various application scenarios such as research on the mechanism of water and heat migration in frozen soil, identification of ice lens development process and analysis of freezing boundary evolution.

[0007] The purpose of the present invention and the technical problem to be solved are achieved by adopting the following technical solutions. The device described in the present invention includes a freezing chamber arranged in a temperature-controlled environment, a sample box for carrying soil samples, a hyperspectral imaging device for collecting spectral data, and a light source system for providing illumination; the freezing chamber is a hollow structure, and one end is provided with a slot for inserting the sample box, and the top is provided with an observation window for facilitating hyperspectral imaging, and the two ends of the freezing chamber are respectively provided with temperature control modules, wherein the end close to the slot is a high-temperature module, and the other end is a low-temperature module, so as to form a horizontal temperature gradient from the low-temperature end to the high-temperature end; the sample box includes a bottom slot plate, and a top plate covered on the bottom slot plate, and a filling area for filling soil is formed between the top plate and the bottom slot plate, and a number of thermocouple sensors are evenly distributed on the bottom slot plate for monitoring the temperature changes inside the soil during the freezing process, and the wires of the thermocouple sensors pass through the side walls of the sample box and are connected to the recorder. The sample box is closed at one end and open at the other end, and can be inserted into the freezing chamber through the slot; the hyperspectral imaging device is composed of a hyperspectral camera vertically arranged just above the observation window and a matching data processing module, and is used to obtain spectral image data of the soil surface during the freezing process; The light source system is symmetrically arranged on both sides of the hyperspectral camera, which can provide stable and uniform lighting conditions to meet the requirements of hyperspectral imaging.

[0008] Furthermore, the temperature control module includes aluminum water cooling blocks symmetrically arranged at the top and bottom of the freezing chamber, and the aluminum water cooling blocks are connected to the constant temperature water tank through heat conductive contact parts to achieve precise temperature control of the cold / hot ends.

[0009] Furthermore, the light source adopts a continuous spectrum broadband light source, and a high color temperature halogen lamp or a xenon lamp may be used.

[0010] Furthermore, the temperature control room is a low temperature room or a constant temperature box.

[0011] Furthermore, the sample box is made of a transparent material, which may be acrylic, organic glass, or PC plastic.

[0012] Furthermore, the light source host is arranged far away from the sample area and has a heat insulation and active heat dissipation structure to effectively avoid thermal interference.

[0013] Furthermore, the freezing chamber is made of a high thermal conductivity material, which may be pure copper, copper alloy, or copper-based composite material.

[0014] Furthermore, the light source adopts a multi-angle, diffuse reflection lighting design and acts on the imaging area in an oblique angle incident manner, thereby improving imaging uniformity and reducing stray light and reflection interference.

[0015] In summary, compared with the prior art, the device proposed in the present invention has the following effects: (1) Realize real-time synchronous observation of the freezing process This device combines hyperspectral imaging technology with a soil freezing test device, which can observe water migration, ice lens formation and unfrozen water distribution evolution during soil freezing in real time and synchronously. It breaks through the limitation of traditional methods that can only indirectly observe changes in soil freezing components, and greatly improves the experimental accuracy and information density.

[0016] (2) The device has a compact structure and strong applicability The sample box and freezing chamber adopt a modular design with a compact structure, easy to assemble, disassemble and maintain, and the overall device cost is lower than that of large imaging equipment such as X-ray CT, which is easy to promote and apply, and is suitable for the economic and operational needs of various types of laboratories or scientific research institutions.

[0017] (3) High data acquisition accuracy and strong system stability Combining a high-resolution hyperspectral camera, a precision temperature control module and a multi-point temperature sensing system, high consistency and synchronization between the physical state of soil freezing and spectral data can be achieved, improving data quality and experimental repeatability.

[0018] (4) Reduce labor costs and improve efficiency The system has an automatic synchronous data collection function, which avoids frequent manual sampling and monitoring, greatly reduces the operating burden, improves experimental efficiency and data collection rate, and is suitable for large-scale parameter tests and long-cycle freeze-thaw simulations.

[0019] (5) Effectively control external interference factors Special optimization has been carried out on the thermal interference of light sources and ambient light interference. Through the layout of long-distance diffuse reflection light sources, thermal insulation and heat dissipation measures and environmental control, the influence of the external environment on the test results during the experiment is effectively reduced, ensuring the accuracy and repeatability of the observed data and the credibility of the experimental results.

[0020] In summary, the soil freezing and hyperspectral imaging simultaneous testing device proposed in the present invention has significant innovative advantages in structural design, functional integration and experimental applicability, and can provide a high-resolution, high-efficiency and high-repeatability experimental platform for the study of frozen soil physical processes.

[0021] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention and to implement it according to the contents of the specification, and at the same time, to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the following specifically cites the preferred embodiments and describes them in detail with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a schematic diagram of a device for real-time synchronous testing of soil freezing process and hyperspectral imaging according to the present invention.

[0023] Figure 2The freezing process image of the soil sample observed by using the present invention.

[0024] Figure 3 The temperature distribution of the soil sample changes with time.

[0025] Figure 4 For different analytical fields ( Figure 2 Hyperspectral data of Class1-4).

[0026] Figure 5 This is the reflectivity map of soil in different temperature areas.

[0027] Figure 6 A visualization of the evolution of water status in frozen soil. DETAILED DESCRIPTION

[0028] The technical solution of the present invention is further described below in conjunction with the accompanying drawings and preferred embodiments.

[0029] See also Figure 1 A soil freezing process and hyperspectral imaging real-time synchronous testing device includes a freezing chamber 1, a sample box 2, a hyperspectral imaging device 12 and a light source system 11 in a temperature controlled environment. The temperature controlled environment can be selected from a low temperature chamber or a constant temperature box. At the same time, the whole system is arranged to operate in a dark or low illumination environment to avoid stray light interfering with the accuracy of hyperspectral imaging data.

[0030] The freezing chamber 1 is a hollow structure made of a metal material with high thermal conductivity, and pure copper, copper alloy or copper-based composite materials can be selected to ensure the uniformity of temperature distribution and response speed during soil freezing. A slotted structure is provided at one end of the freezing chamber 1 for inserting and fixing the sample box 2. An observation window 8 (i.e., imaging area) for observing the soil state is provided on the top of the freezing chamber 1. The observation window 8 is located directly above the filling area of ​​the sample box 2 and is aligned with the field of view of the hyperspectral camera to ensure imaging quality and data integrity. Temperature control modules 7 are provided at both ends of the freezing chamber 1, wherein the end close to the slot is a high-temperature module and the other end is a low-temperature module, which can form a unidirectional temperature gradient and realize the horizontal freezing process in an open system; The temperature control module 7 includes aluminum water-cooling blocks 10 symmetrically arranged at the top and bottom of the freezing chamber 1. The upper and lower aluminum water-cooling blocks 10 form a group. The two groups of aluminum water-cooling blocks 10 are respectively connected to the constant temperature water tanks through the heat-conducting contact parts 9. By setting a group of high-temperature water tanks and a group of low-temperature water tanks, the temperature at both ends of the freezing chamber can be accurately controlled to drive the soil freezing and moisture migration process, simulating the typical hydrothermal behavior in frozen soil.

[0031] The sample box 2 is made of a transparent material, has good spectral transmittance, and is suitable for hyperspectral imaging. The preparation material can be one of acrylic, organic glass, and PC plastic. The sample box 2 includes a bottom slot plate 4 and a top plate 3 covering the bottom slot plate 4. A filling area for filling soil is formed between the top plate 3 and the bottom slot plate 4. A plurality of thermocouple sensors 5 are evenly distributed on the bottom slot plate 4. The thermocouple sensors 5 can be arranged at a certain interval (such as 5 mm interval) to achieve real-time and precise temperature monitoring. The wires of the thermocouple sensors 5 pass through the side wall of the sample box 2 and are connected to the recorder. The sample box 2 is closed at one end and open at the other end. The open end can replenish water and provide space for soil frost heave. The sample box 2 can be inserted into the freezing chamber 1 from the slotted end of the freezing chamber 1. The insertion direction should ensure that its closed end corresponds to the low temperature end of the freezing chamber.

[0032] The hyperspectral imaging device 12 includes a hyperspectral camera vertically mounted just above the observation window 8, and a data processing module for image acquisition and analysis. The hyperspectral camera is vertically mounted just above the observation window 8 and at a certain distance from the observation window 8 to ensure data acquisition stability and imaging clarity. The light source system 11 is symmetrically arranged on both sides of the hyperspectral camera, and adopts a continuous spectrum broadband light source, preferably a high color temperature halogen lamp, a xenon lamp, etc., with good spectral coverage capability. The light source adopts a multi-angle, diffuse reflection lighting structure, and acts on the sample surface at an oblique angle to ensure uniform lighting without local overexposure or shadow interference; in order to control the thermal interference of the light source on the frozen state of the soil sample, the light source host is arranged away from the sample box, and is equipped with a heat insulation structure and an active heat dissipation device to avoid local thawing or interference with the freezing process due to thermal radiation.

[0033] Through the highly integrated design of the above components, the present invention enables the synchronous operation of the temperature monitoring system and the hyperspectral acquisition system. During the freezing process, the internal temperature change data of the soil and the surface hyperspectral reflectance information can be collected in real time, and the key physical states such as the formation of ice lenses and the changes in the distribution of unfrozen water during the freezing process can be visualized, dynamically monitored and accurately identified, providing an experimental support platform with high temporal and spatial resolution for the study of the water and heat migration mechanism of frozen soil.

[0034] A method for using a soil freezing process and hyperspectral imaging real-time synchronous testing device comprises the following steps: S1, sample preparation.

[0035] Select representative fine-grained soil (such as a certain type of silty clay, etc.), screen and homogenize it, and adjust its water content to a near-saturated state. Fill the prepared soil sample evenly in a transparent sample box 2, and control the density and size of the sample to meet the standardized experimental requirements.

[0036] S2, boundary setting and water supply control.

[0037] The sample box 2 is placed horizontally, and a proper amount of distilled water is added to the open end of the sample box 2 to provide the sample with the water supply required during the freezing process.

[0038] S3, specimen insertion and temperature control arrangement.

[0039] Insert the sample box 2 horizontally into the device from the side of the freezing chamber 1 where the slot is provided, and ensure that the closed end of the sample box corresponds to the low temperature end of the freezing chamber. Control the temperature control modules 7 at both ends of the freezing chamber 1 respectively, adjust the temperature difference between the high temperature end and the low temperature end, and form a horizontal unidirectional freezing. In order to ensure the stability of the test environment, heat preservation materials (such as polystyrene foam boards) are arranged around the freezing device to maintain the uniformity of the ambient temperature and the temperature control accuracy.

[0040] S4, hyperspectral imaging acquisition.

[0041] The sample surface is imaged and observed by a hyperspectral camera installed above the device. The imaging band covers the visible light to near-infrared region (such as 400-1000 nm), with moderate spectral resolution (such as 5 nm) and sub-millimeter spatial resolution. The imaging cycle is set to collect images every 0.5-1 hour, and the image data is transmitted to the data processing module in real time for spectral feature extraction and moisture state analysis.

[0042] S5, synchronous monitoring of the internal temperature of the sample.

[0043] The temperature changes inside the soil during the freezing process are recorded in real time using thermocouple sensors 5 arranged at multiple locations in the sample box 2. The temperature data sampling frequency is set to once every minute, and is collected synchronously with the hyperspectral image data to facilitate subsequent freezing front identification and moisture state matching analysis.

[0044] During the use of the device described in the present invention, the collected hyperspectral data and temperature monitoring data need to be synchronously processed and deeply analyzed to achieve quantitative identification and visual display of different water states (such as unfrozen water, ice lenses, etc.) during the soil freezing process; (1) Hyperspectral data processing and characteristic band extraction First, the collected raw hyperspectral image data was preprocessed, including noise reduction, band clipping, smoothing and outlier removal, to ensure the integrity and reliability of the data. Then, principal component analysis (PCA) and correlation coefficient analysis were used to reduce the dimension and screen the sensitivity of the full-band data, and identify the key characteristic bands that are most sensitive to changes in soil freezing status. Experiments show that some bands in the near-infrared range (such as 700–900 nm) have a strong response to moisture status and are suitable as the main spectral parameters for frozen soil identification.

[0045] (2) Temperature-spectrum model construction and identification algorithm Based on the reflectivity of the characteristic bands selected above and the local temperature data collected by the thermocouples, a temperature-spectrum coupling model is constructed. The model can use a multivariate linear regression method, or introduce a machine learning algorithm such as a support vector machine (SVM) or a random forest (RF) for fitting modeling to obtain the functional relationship between temperature change and reflectivity change, and realize automatic recognition and classification of different freezing states.

[0046] (3) Visual coding and image generation of unfrozen water state Based on the established model, the predicted water state and the corresponding spectral reflectance characteristics are color-coded to generate a spatial distribution image of the sample surface. By analyzing images at different time periods, the spatiotemporal evolution characteristics of phenomena such as ice lens formation and unfrozen water migration during the freezing process can be intuitively displayed.

[0047] (4) Experimental verification and image analysis The typical freezing experiment duration of this device is 12-24 hours, and the number and duration of freeze-thaw cycles can be set according to research needs. During the experiment, hyperspectral images and temperature data are collected regularly. At the same time, after the freezing process is completed, the sample is sampled, dried and weighed in sections along the freezing direction to obtain the actual moisture content distribution for reverse verification.

[0048] See also Figure 2 , which is an image of the soil freezing state collected 6 hours after the start of the experiment. In the image, the black crack-like area advancing along the freezing direction can be clearly observed, which is determined to be an ice lens. After the freezing surface advances to about 27 mm in the middle of the sample, the rate gradually slows down, and ice lenses begin to form near the freezing surface, and a larger ice lens layer is formed on the upper part in the 4th hour.

[0049] See also Figure 3 , showing that the trajectory of the 0℃ isotherm during the freezing process is highly consistent with the location of the ice lens, further verifying the accuracy of ice lens identification in hyperspectral images. The temperature change trend measured by the thermocouple data is highly consistent with the final water content determination results.

[0050] See also Figure 4 By extracting the hyperspectral reflectance curves of different spatial regions (Class1~Class4), it was found that the reflectance in the 400–600nm band showed an upward trend, and the reflectance difference between frozen soil and unfrozen soil in the 750–850nm range was obvious, about several percentage points, which can be used as a basis for discrimination.

[0051] See also Figure 5, showing the change in average reflectivity in different temperature ranges. The results show that the reflectivity did not change significantly before the temperature dropped to -0.9℃, but in the range of -0.9℃ to -1.2℃, the reflectivity increased significantly and then stabilized, indicating that this range may be a transition zone for the rapid transformation of unfrozen water into ice crystals. Combined with the characteristic absorption behavior of water molecules in the near-infrared band, this phenomenon further verifies the close correlation between reflectivity and the water state in the soil.

[0052] See also Figure 6 , the figure shows a visualization image of the frozen soil moisture state generated based on the hyperspectral-temperature coupling model. By color-coding the predicted moisture state, the spatiotemporal evolution of ice lenses and unfrozen water during the freezing process is intuitively displayed. This figure verifies the dynamic recognition and visualization analysis capabilities of this device in different moisture states in frozen soil, providing image support for water and heat migration research.

[0053] In other embodiments of the present invention, the soil freezing and hyperspectral imaging synchronous testing device provided by the present invention is also suitable for carrying out extended experiments under various parameter conditions, for example: Different soil types: such as sand, loam, silty clay, etc., to study the effects of different particle gradations and structures on ice lens formation and water migration; Different initial moisture content: By setting different moisture contents (such as 20%, 30%, 50%, saturation, etc.), the effect of moisture content on freezing rate, unfrozen water distribution and spectral response is analyzed; Different salt content and types: Different concentrations and types of salts (such as NaCl, CaCl 2 etc.), studying the regulatory effects of salt on freezing temperature, ice crystal growth and spectral reflectance characteristics; Different temperature gradients: The temperature difference between the two ends of the freezing device can be adjusted (such as -2℃&2℃, -10℃&5℃, etc.) to explore the changes in freezing rate and ice lens formation mechanism under different thermal conditions; Freeze-thaw cycle experiment: The sample is repeatedly cycled in a freeze-thaw environment for several times to observe the structural evolution process and the changing trend of the spectral response with the number of cycles, and further expanded to the study of freeze-thaw damage mechanism and durability.

[0054] The above is only a preferred embodiment of the present invention. Any technician familiar with the profession, without departing from the scope of the technical solution of the present invention, makes any simple modification, equivalent change and modification to the above embodiment based on the technical essence of the present invention, any equivalent replacement, functional adjustment or structural optimization, all of which still fall within the scope of the technical solution of the present invention.

Claims

1. A soil freezing process and hyperspectral imaging real-time synchronous testing device, characterized in that: It comprises a freezing chamber (1), a sample box (2), a hyperspectral imaging device (12) and a light source system (11) arranged in a temperature-controlled environment; The freezing chamber (1) is a hollow structure, and one end is provided with a slot for inserting a sample box (2), the top of the freezing chamber (1) is provided with an observation window (8), and both ends are provided with temperature control modules (7), wherein the end close to the slot is a high-temperature module, and the other end is a low-temperature module, so as to form a unidirectional temperature gradient; The sample box (2) comprises a bottom slot plate (4) and a top plate (3) covered thereon, a filling area for filling soil samples is formed between the top plate (3) and the bottom slot plate (4), a plurality of thermocouple sensors (5) are evenly distributed on the bottom slot plate (4), and the wires of the thermocouple sensors (5) are connected to a recorder after passing through the side wall of the sample box (2). One end of the sample box (2) is closed and the other end is open, and the sample box (2) can be positioned and inserted into the freezing chamber (1) from the slot; The hyperspectral imaging device (12) comprises a hyperspectral camera and a data processing module. The hyperspectral camera is vertically mounted directly above the observation window (8). The light source system (11) is symmetrically arranged on both sides of the hyperspectral camera to provide uniform illumination for the imaging area.

2. The soil freezing process and hyperspectral imaging real-time synchronous testing device according to claim 1, characterized in that: The temperature control module (7) comprises aluminum water cooling blocks (10) symmetrically arranged at the top and bottom of the freezing chamber (1); the aluminum water cooling blocks (10) are connected to the constant temperature water tank via a heat-conducting contact piece (9).

3. The soil freezing process and hyperspectral imaging real-time synchronous testing device according to claim 1, characterized in that: The light source system (11) adopts a broadband light source with continuous spectrum characteristics, and the light source is one of a high color temperature halogen lamp and a xenon lamp.

4. The device for real-time synchronous testing of soil freezing process and hyperspectral imaging according to claim 1, characterized in that: The temperature control environment is one of a low temperature chamber and a constant temperature box.

5. The device for real-time synchronous testing of soil freezing process and hyperspectral imaging according to claim 1, characterized in that: The sample box (2) is made of a transparent material, and can be made of one of acrylic, organic glass, and PC plastic.

6. The soil freezing process and hyperspectral imaging real-time synchronous testing device according to claim 1, characterized in that: The light source system (11) is connected to a light source host disposed away from the sample area via a wire, and the light source host is provided with a heat insulation structure and an active heat dissipation unit.

7. The soil freezing process and hyperspectral imaging real-time synchronous testing device according to claim 1, characterized in that: The freezing chamber (1) is made of a high thermal conductivity material, which may be pure copper, a copper alloy, or a copper-based composite material.

8. The device for real-time synchronous testing of soil freezing process and hyperspectral imaging according to claim 1, characterized in that: The light source system (11) adopts a multi-angle, diffuse reflection lighting design, and is incident on the observation window (8) at an oblique angle.

Images