A method and system for characterizing the evolution of pore water state in a porous medium during infiltration

Abstract

The application discloses a kind of pore water state evolution characterization method and system in porous medium infiltration process, it is related to porous medium water migration nondestructive testing technical field, including the following steps: obtaining the same porous medium sample in multiple continuous infiltration stages T 2 relaxation spectrum, and the signal weighted logarithmic spectrum center and total signal amount of each T 2 relaxation spectrum;For each infiltration stage, the spectrum center migration factor of the stage is constructed based on the displacement amount of its signal weighted logarithmic spectrum center relative to the signal weighted logarithmic spectrum center of reference stage, the total signal scaling factor is constructed based on the ratio of its total signal amount and the total signal amount of reference stage, the spectrum center migration factor and total signal scaling factor of the same infiltration stage are combined to form the two-factor state coordinates of the stage;According to the order of infiltration stage, each two-factor state coordinates are connected, and the pore water state evolution trajectory is formed.The application effectively overcomes the technical defects that the stage state lacks comparability due to the inability to analyze multiple sources simultaneously.

Description

Technical Field

[0001] This invention relates to the field of non-destructive testing and data processing technology for moisture migration in porous media, and in particular to a method and system for characterizing the evolution of pore water state during the infiltration process of porous media. Background Technology

[0002] Rainfall, irrigation, seepage, or engineering water replenishment processes can cause water infiltration and redistribution within soil and rock masses, such as loess, clay, silt, sand, compacted soil, and other porous media. For unsaturated porous media, the infiltration process not only manifests as an increase in total water content but also involves the redistribution of water across different spatial locations and physical constraints. Therefore, macroscopic indicators such as cumulative water absorption, average water content, or wetting front location are insufficient to fully characterize the dynamic evolution of the internal pore water state.

[0003] To investigate the evolution of pore water state during infiltration into porous media, low-field nuclear magnetic resonance (NMR) technology can non-destructively obtain transverse relaxation time using the hydrogen proton relaxation response. T 2. Spectrum. Under fixed sample location, acquisition parameters, and inversion conditions, T 2. Spectroscopy can be used to characterize the NMR-detectable moisture response; the spectral distribution and morphology are influenced by factors such as the interaction between moisture and the solid surface, geometric constraints, local saturation, and pore connectivity. Existing... T 2. Spectral processing typically involves integrating to obtain the total spectral area or NMR-detectable water volume, extracting spectral characteristic parameters such as peak position, peak amplitude, or average relaxation time, or obtaining them based on empirical values, component intersections, and independent calibration. T 2. Cutoff value: Divide the spectrum into different relaxation time intervals and calculate the spectral area or relative proportion of each interval.

[0004] The methods described above typically characterize pore water conditions using either total spectral area or local spectral features, respectively. However, in dynamic infiltration processes, T The two spectra may simultaneously exhibit signal enhancement, changes in spectral shape and component ratios, and overall positional migration. Existing methods struggle to simultaneously determine changes in the detectable water response of NMR and the overall migration degree of the pore water relaxation environment. Therefore, when different infiltration stages... T When the 2-spectrum changes, existing methods struggle to determine whether the change primarily stems from an increase in water response, a redistribution of the relaxation environment, or a combination of both. This lack of a unified quantitative basis for comparing the state of pore water at different stages hinders the accurate identification of state evolution and stage transitions during infiltration. Summary of the Invention

[0005] Based on the deficiencies of the existing technology, the present invention provides a method and system for characterizing the evolution of pore water state during the infiltration process of porous media, which solves the existing problems.

[0006] The present invention adopts the following technical solution: In a first aspect, the present invention provides a method for characterizing the evolution of pore water state during the infiltration of porous media, comprising the following steps: Obtaining the same porous media sample at multiple successive infiltration stages T 2. Relaxation spectrum, and calculate each T 2. The signal weighted logarithmic spectrum of the relaxation spectrum and the total signal quantity; For each infiltration stage, a spectral center migration factor is constructed based on the displacement of its signal-weighted logarithmic spectral center relative to the signal-weighted logarithmic spectral center of the reference stage. A total signal scaling factor is constructed based on the ratio of its total signal quantity to the total signal quantity of the reference stage. The spectral center migration factor and the total signal scaling factor of the same infiltration stage are combined to form the two-factor state coordinates of that stage. The reference stage is a stage pre-selected from the plurality of consecutive infiltration stages. The spectral center migration factor is used to characterize the overall migration degree of the pore water relaxation environment, and the total signal scaling factor is used to characterize the relative change of the NMR-detectable water response. Connect the state coordinates of each dual factor in sequence according to the infiltration stage to form the pore water state evolution trajectory.

[0007] Preferably, the spectral center migration factor is as follows: ; In the formula, For the first n Infiltration stage spectral center migration factor For the first n The signal weighted logarithmic spectrum center during the infiltration phase, The weighted logarithmic spectrum center of the signal in the reference phase, For the first n Log-weighted representative relaxation time during the infiltration phase. The log-weighted representative relaxation time for the reference phase; The total signal scaling factor is specifically as follows: ; In the formula, This is the signal scaling factor. For the first n Total signal quantity during the infiltration phase This refers to the total semaphore during the reference phase.

[0008] Preferably, by T 2. After taking the natural logarithm of the transverse relaxation time of each discrete spectral point in the relaxation spectrum, a weighted average is performed using the corresponding spectral signal amplitude as the weight to obtain the weighted logarithmic spectral center of the signal. T The total signal quantity is obtained by summing the spectral signal amplitudes of each discrete spectral point in the relaxation spectrum.

[0009] Preferably, the reference stage is the infiltration stage of the first infiltration.

[0010] Preferably, the T 2. Relaxation spectrum is retained T 2 spectra, retained T The acquisition of the spectrum specifically includes the following steps: Based on the stable inversion of each infiltration stage T 2. Public Scope: Identifying Each Infiltration Stage T 2. The relaxation spectrum represents the pore water response of the target porous medium and in T The target main spectrum distribution forms a continuous main body distribution on the two axes; If a non-target response exists outside the target main spectrum distribution, then a common boundary is selected to separate the target main spectrum distribution from the non-target response. T Two thresholds are used as cutoff boundaries to determine common... T 2. Scope of analysis; otherwise, based on commonly stable inversion. T 2. Public Scope as Common T 2. Scope of analysis; According to the common T 2. Scope of analysis: original samples at each infiltration stage T 2. The relaxation spectrum is truncated, and non-target responses are excluded to obtain the retention values ​​for each infiltration stage. T 2.

[0011] Preferred, original T 2. Relaxation spectrum was obtained by applying a CPMG pulse sequence to the porous medium sample.

[0012] Preferred options also include: Preservation of each infiltration stage T The two spectra are normalized according to their spectral areas, and the area-normalized average spectrum of all infiltration stages is calculated. exist T 2. Multi-component fitting of the area-normalized average spectrum within logarithmic space, using the intersection of adjacent fitted components to determine the operability that can be uniformly used throughout the entire infiltration stage. T Two boundary values ​​are used to form at least two different relaxation time intervals; Calculate the spectral area of ​​each infiltration stage in each relaxation time interval as a percentage of the area retained in that stage. T The proportion of the total signal quantity in the two spectra forms the operational relaxation domain contribution vector; The operational relaxation domain contribution vector is output together with the corresponding two-factor state coordinates to analyze the relative contributions of different relaxation environments during the spectral center migration process.

[0013] Preferred options also include: Obtain HSE2 sequence axial nuclear magnetic resonance signal profiles of porous media samples at at least one infiltration stage; The HSE2 axial signal profile is output in correspondence with the dual-factor state coordinates of the same infiltration stage to identify the spatial region of water response that causes changes in dual-factor state.

[0014] Preferred options also include: SE-SPI sequence was used to obtain the stratification of porous media samples at different spatial levels along the axial direction. T 2. Relaxation spectrum; For each spatial layer, calculate its local signal weighted logarithmic spectral center and local total signal quantity, and construct the local spectral center migration factor and local total signal scaling factor of the layer under each measurement state based on a pre-selected local reference state, forming a local two-factor state coordinate system. A spatiotemporal state matrix of pore water is constructed from local two-factor state coordinates at different spatial levels and under different measurement conditions. This matrix is ​​used to characterize the joint evolution of pore water state in the spatial and hydraulic dimensions.

[0015] Secondly, the present invention provides a system for characterizing the evolution of pore water state during the infiltration of porous media, comprising: The acquisition module is used to acquire data from the same porous media sample at multiple successive infiltration stages. T 2. Relaxation spectrum, and calculate each T 2. The signal weighted logarithmic spectrum of the relaxation spectrum and the total signal quantity; The calculation module is used to construct a spectral center migration factor for each infiltration stage based on the displacement of its signal-weighted logarithmic spectral center relative to the signal-weighted logarithmic spectral center of the reference stage, and to construct a total signal scaling factor based on the ratio of its total signal quantity to the total signal quantity of the reference stage. The spectral center migration factor and the total signal scaling factor of the same infiltration stage are combined to form a two-factor state coordinate for that stage. The reference stage is a stage pre-selected from the plurality of consecutive infiltration stages. The spectral center migration factor is used to characterize the overall migration degree of the pore water relaxation environment, and the total signal scaling factor is used to characterize the relative change of the NMR-detectable water response quantity. The characterization module is used to connect the state coordinates of each dual factor in chronological order of infiltration stages to form the pore water state evolution trajectory.

[0016] Compared with the prior art, the above-mentioned at least one technical solution adopted by the present invention can achieve the following beneficial effects: For existing low-field nuclear magnetic resonance T 2To address the technical challenge of lacking a unified quantitative basis for identifying the sources of pore water state changes at different infiltration stages in relaxation spectral analysis, this invention innovatively proposes a pore water state evolution analysis method based on dual-factor joint characterization. By extracting the signal-weighted logarithmic spectral center and total signal quantity from the T2 relaxation spectrum, a spectral center migration factor and a total signal scaling factor are constructed, respectively. This quantitatively decomposes the pore water state changes during infiltration into two independent and jointly analyzable dimensions: the overall migration of the relaxation environment and the change in NMR-detectable water response. The pore water state evolution trajectory is then formed by connecting the dual-factor state coordinates in the order of infiltration stages. Specifically, the spectral center migration factor characterizes the overall migration degree of the pore water relaxation environment relative to the reference stage, and the total signal scaling factor characterizes the relative change in NMR-detectable water response relative to the reference stage. The dual-factor state coordinates are connected in the order of infiltration stages to form the pore water state evolution trajectory, thereby achieving accurate identification and quantitative discrimination of the sources of state changes at infiltration stages. This effectively overcomes the technical deficiency of existing methods, which cannot simultaneously analyze multiple sources, leading to a lack of comparability of stage states. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of the overall flow of the measurement and quantitative characterization method of the present invention; Figure 2 For the present invention T 2. Schematic diagram of the overall relaxation spectrum transfer concept model; Figure 3 A schematic diagram of a phased one-dimensional infiltration and multi-sequence low-field nuclear magnetic resonance monitoring integrated experimental system; in, Figure 3 (a): Infiltration and boundary control system, Figure 3 (b): Phased mass measurement and wetting front observation, Figure 3 (c): Multi-sequence low-field nuclear magnetic resonance monitoring system; Figure 4 For the overall CPMG during the phased infiltration process T 2. Schematic diagram of the evolution of relaxation spectrum; Figure 5 For area-normalized average CPMG T 2. Operability of fitting relaxation spectrum with three Gaussian components T 2. Domain partitioning diagram; Figure 6For different infiltration stages, CPMG short T 2 domains, middle T 2 domains and length T A schematic diagram illustrating the changes in the relative signal contribution of the two domains; Figure 7 This is a schematic diagram of the HSE2 nuclear magnetic resonance signal profile of the sample along the axial direction during the staged infiltration process. Figure 8 SE-SPI stratification after infiltration is completed T 2. Schematic diagram of relaxation spectrum and spatial distribution of signal intensity; in, Figure 8 (a): SE-SPI layering T 2. Relaxation spectrum Figure 8 (b): Schematic diagram of spatial distribution of signal strength; Figure 9 SE-SPI stratification after sealing and standing for 48 hours following infiltration. T 2. Schematic diagram of relaxation spectrum and spatial distribution of signal intensity; in, Figure 9 (a): SE-SPI layering T 2. Relaxation spectrum Figure 9 (b): Schematic diagram of spatial distribution of signal strength; Figure 10 SE-SPI stratification under vacuum saturation after being sealed and left to stand for 48 hours T 2. Schematic diagram of relaxation spectrum and spatial distribution of signal intensity; in, Figure 10 (a): SE-SPI layering T 2. Relaxation spectrum Figure 10 (b): Schematic diagram of spatial distribution of signal strength. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] Based on existing shortcomings, this invention proposes a method... T A two-factor quantitative characterization method based on the overall statistical characteristics of the spectrum decomposes the changes in pore water state during infiltration into two distinct but jointly analyzable dimensions: the spectral core migration factor characterizes the overall migration degree of the pore water relaxation environment, and the total signal scaling factor characterizes the relative change in the NMR-detectable water response. Furthermore, it utilizes operational... T 2. Domain, Axial Signal Profile and LayersT 2. Spectrum analysis of the composition and spatial expansion of two-factor state changes.

[0021] Example 1: Based on the whole T Two-factor state characterization of the spectrum.

[0022] This embodiment uses the CPMG obtained during a phased infiltration process as a whole. T Using relaxation spectra as the core data source, this paper explains the implementation process of the two-factor characterization method for pore water state evolution, referring to... Figure 1 It includes the following steps: S1: Prepare porous media samples.

[0023] Select the porous media sample to be tested and establish the infiltration boundary. The porous media can be undisturbed soil, remolded soil, compacted soil, rock core, or other water-bearing porous materials that can be detected by low-field NMR.

[0024] In one specific embodiment, a undisturbed cylindrical sandy loess specimen with a diameter of 60 mm and a height of 120 mm is used. The sidewalls of the specimen are sealed with a waterproof material to inhibit radial moisture exchange, ensuring that the main direction of moisture migration is along the specimen's axial direction. Permeable components are provided at the upper and lower ends of the specimen, contacting the specimen's end faces. The upper component is used to uniformly apply infiltrated water, while the lower component supports the specimen and maintains the drainage boundary. Figure 3 As shown in (a).

[0025] S2: Infiltrate in stages.

[0026] The sample is infiltrated in stages. After each set water input is completed and the free water or accumulated water on the sample disappears, a measurement stage, i.e., the infiltration stage, is recorded. Sample mass, wetting front location, or other macroscopic hydraulic parameters can be recorded simultaneously as needed, but the two-factor state calculation itself is performed using low-field NMR. T 2. Spectrum is the input.

[0027] In this embodiment, a water supply method with a flow rate of 24 mL / min is used, with each water supply lasting 30 seconds, corresponding to a single input water volume of 12 mL. Each stage sequentially completes water supply, surface water disappearance, weighing, wetting front recording, and NMR measurement before proceeding to the next stage, as follows: Figure 3 As shown in (b). When the change in sample water volume becomes less significant after continued water supply, the infiltration stage ends; a termination verification stage can also be set, which is only used to confirm the termination state and does not participate in the calculation of subsequent stage evolution trajectories.

[0028] The same sample was repeatedly tested for NMR in all measurement stages, and the axial position of the sample in the magnet was kept consistent by the sample tube positioning mark.

[0029] In this embodiment, a low-field nuclear magnetic resonance (NMR) device with a proton resonance frequency of 12.77 MHz and a magnetic field strength of approximately 0.30 T is used. The overall CPMG sequence is used to obtain... T 2. Relaxation spectrum, such as Figure 3 As shown in (c). Example acquisition parameters are: receiving bandwidth 250kHz, echo time 0.220ms, waiting time 2000ms, echo count 8000, and scan count 16. The above parameters are only one feasible implementation method, and are intended to obtain repeatable overall results. T Other parameter combinations can be used under the condition of 2 spectra.

[0030] Maintain consistent sample geometry, NMR acquisition parameters, and inversion parameters throughout all measurement phases. First, determine the common, stable inversion parameters for each phase. T 2. Common scope. Jointly stable and invertible. T 2. The common scope refers to the common area where each stage has effective inversion results and is not affected by the truncation of the inversion time window boundary. T 2. Within this common range, the original data from each stage are superimposed and compared. T 2. Relaxation spectrum, which will be used to characterize the pore water response of the target porous medium. T The spectral distribution that forms a continuous main distribution on the two axes and participates in inter-stage comparisons is defined as the target main spectral distribution.

[0031] When an isolated response exists outside the target main spectrum distribution and is clearly separated from it, and there is an identifiable low-signal interval or local valley between the two, the isolated response is marked as a non-target candidate response. Non-target candidate responses may include responses formed by signals from sample boundary components, auxiliary permeable components, instrument background, or other signals that do not participate in the characterization of the pore water state of the target porous medium. After confirming that the response is not used for characterizing the pore water state of the target porous medium by combining information from sample boundary configuration, auxiliary experiments, or other factors that can determine the signal source, it is excluded as a non-target response.

[0032] Based on the spectral distribution characteristics of all measurement stages, a common spectral distribution capable of stably separating the target main spectral distribution from non-target candidate responses is selected. T Two thresholds are used as cutoff boundaries, and the same cutoff criterion is applied in all measurement phases to determine common thresholds. T 2. Scope of Analysis. If there are no non-target responses that are significantly separated from the target main spectrum distribution, then a stable inversion method that can be achieved across all measurement stages will be used. T 2. The range is used as the analysis range.

[0033] Original T 2. Relaxation spectrum according to common T 2. The spectral distribution retained for subsequent parameter calculations after the analysis range is truncated and non-target responses are excluded according to a unified criterion is defined as the retained distribution. T2. Spectrum. Subsequent spectral center shift factor, total signal scaling factor, and operability. T Both domain ratios are based on reservation. T 2. Spectrum calculation.

[0034] In this embodiment, the original CPMG spectrum is in a longer period of time. T Region 2 exhibits isolated responses separate from the main continuous spectrum distribution, with a clear low-amplitude separation between them. Based on the boundary configuration of this embodiment and the consistent spectral separation characteristics across all measurement stages, the upper limit for common analysis is set at 100 ms, thereby preserving... T The two spectra are uniformly limited to T The range is 2 ≤ 100 ms. This 100 ms is only the upper limit of analysis determined in this embodiment under the tested material, boundary configuration, and NMR measurement conditions, and does not constitute an application to other porous media or other measurement conditions. T 2. Limitations on the scope of analysis, such as Figure 4 As shown, the shaded area to the right of the dashed line represents the boundary correlation length that separates from the target main spectrum distribution. T 2. Non-target responses are excluded in subsequent spectral parameter calculations.

[0035] S3: Calculate the spectral center shift factor and the total signal scaling factor.

[0036] For the n Retention during the measurement phase T 2. In unification T 2. Within the analysis interval, let the discrete spectral points be... ,in, For the first k A horizontal relaxation time point For the first n Measurement phase The amplitude of the spectral signal at that location. All values ​​are considered during calculation. Using the same time unit, this embodiment uniformly uses ms.

[0037] No. n Measurement phase T 2. Spectrum in logarithmic T The location of the weighted average signal in space is defined as the center of the weighted logarithmic spectrum of the signal. Calculate using the following formula: (1); Further Map back T 2 time scales, to obtain the first n Log-weighted representative relaxation time during the measurement phase : (2); in, and Use the same unit of time.

[0038] Calculate the weighted logarithmic spectrum center of the signal according to equation (1). According to equation (2), the log-weighted representative relaxation time is obtained. .

[0039] A reference measurement stage is pre-selected from multiple measurement stages to be compared, and remains unchanged during the comparison of stages within the same group. Let the weighted logarithmic spectrum center of the signal in the reference stage be... The log-weighted representative relaxation time is Then the first n Spectral center shift factor during measurement phase for:

[0040] (3); In this embodiment, the initial infiltration measurement phase is selected as the reference phase and denoted as the first measurement phase. , Spectral center migration factor Used to characterize the n Measurement phase retention T 2. The net migration of the spectrum relative to the reference stage in the logarithmic relaxation time space. When When the value is greater than 0, it indicates that the representative position of the entire spectrum is shifted towards a longer range. T 2-way migration; when When <0, it indicates that its representative position is shorter. T 2-way migration, such as Figure 2 As shown.

[0041] No. n Total signal quantity during the measurement phase Calculate according to the following formula: (4); Total semaphore in the reference phase Based on this, construct the first n Total signal scaling factor during measurement phase : (5); in, Used to characterize the change in the NMR-detectable moisture response relative to a reference state. When When this occurs, it indicates an increase in the total NMR response relative to the reference state. In this embodiment, .

[0042] The incremental spectral center migration factor between adjacent measurement stages can also be calculated. : (6); in, The characterization is by the first n -1 Measurement Phase to the n The newly added overall spectral shift during the measurement phase.

[0043] S4: Formation of the pore water state evolution trajectory.

[0044] The first n Spectral center shift factor during measurement phase and total signal scaling factor Composition of two-factor state coordinates : (7); The reference phase satisfies: In this embodiment, that is .

[0045] Each measurement stage Draw in phase sequence and Within a two-dimensional state space defined by coordinate axes, the pore water state evolution trajectory is formed by sequentially connecting the axes. Thus, the overall migration of the pore water relaxation environment and the changes in NMR-detectable water response during the infiltration process are extracted and jointly characterized within a unified state space.

[0046] Two-factor state trajectories can be used to distinguish different types of staged responses: when adjacent stages mainly exhibit the following characteristics... Significantly increased When the change is small, it indicates that the new response is mainly reflected in the increase of NMR-detectable water response, while the overall migration of the pore water relaxation environment is limited; when the adjacent stages are along A significant increase in the direction of change indicates a substantial overall adjustment in the pore water relaxation environment; a significant increase in both directions indicates that an increase in water response and overall migration of the relaxation environment are occurring simultaneously. This can also be further analyzed... or The output is combined with the corresponding stage's water input, mass change, water content, saturation, wetting front location, or other external hydraulic indicators to identify whether external water volume changes are synchronized with the internal relaxation state migration.

[0047] In the sandy loess test of this embodiment, the log-weighted representative relaxation time gradually increased from approximately 1.123 ms in the first measurement stage to approximately 7.558 ms in the final measurement stage, and the corresponding spectral core migration factor increased from 0 to approximately 1.907, as shown in Table 1. The incremental spectral core migration factor is not the same in different stages, indicating that the overall spectral migration rate changes with the infiltration stage.

[0048] Table 1. Log-weighted representative relaxation time and spectral center migration parameters for each infiltration stage. Infiltration stage Log-weighted representative relaxation time (ms) Cumulative spectral center migration factor Incremental spectral center migration factor 1 1.123 0 - 2 1.497 0.287 0.287 3 2.065 0.609 0.322 4 2.707 0.880 0.271 5 3.406 1.110 0.230 6 3.938 1.255 0.145 7 4.552 1.400 0.145 8 5.590 1.605 0.205 9 6.497 1.755 0.150 10 6.922 1.819 0.064 11 7.189 1.856 0.037 12 7.558 1.907 0.051 Furthermore, most of the final spectral center migration was completed before the wetting front reached the bottom of the sample; while the sample continued to absorb water after the wetting front reached the bottom, the overall spectral center only underwent minor further migration. These results demonstrate that the two-factor state trajectory can identify staged pore water state changes that are difficult to distinguish using only the wetting front location or overall water volume index.

[0049] Example 2: Operability T 2. Establishment of the domain and analysis of the composition of the spectral center migration.

[0050] To further explain the source of the variation in the spectral center migration factor, all stages obtained in Example 1 were retained. T 2. Establish a unified operational relaxation time domain. This operational... T The 2-domain is used to analyze the relative contribution of different relaxation environments to the overall spectral shift, but is not a prerequisite for the two-factor state calculation.

[0051] First, the first n Retention during the measurement phase T 2. Spectrum normalized by spectral area: (8); in, For the first n Measurement phase The area-normalized spectral signal amplitude at that location.

[0052] For the total For each measurement stage involved in the analysis, the area-normalized average spectrum is calculated: (9); in, For the first indivual T The area-normalized average spectral amplitude at point 2. Since each stage is normalized according to its own total spectral area before averaging, the stage with higher moisture response will not occupy an excessively high weight in the average spectrum simply because the total signal is larger.

[0053] Then, the x-axis is transformed to Within this space, a multi-component fitting is performed on the area-normalized average spectrum. Multi-component fitting refers to representing the area-normalized average spectrum as a superposition of two or more relaxation response components with predetermined mathematical forms. Parameters such as the amplitude, center position, and distribution width of each component are determined through numerical fitting, so that the sum of the fitted components characterizes the main distribution features of the area-normalized average spectrum. The intersection points of adjacent fitted components are used to establish operational boundaries between different relaxation time environments.

[0054] In this embodiment, based on the distribution characteristics of the area-normalized average spectrum, three Gaussian functions are used for component fitting, i.e., three-Gaussian component fitting, such as... Figure 5 As shown, the intersection of adjacent Gaussian components is used to determine the short... T 2 domains, middle T 2 domains and length T The boundary value of the 2-domain is fitted by the following function: (10); in, , , The first The amplitude, center position, and width parameters of each fitted component. Each fitted component is used to establish an operational relaxation time partition for the same material and under the same measurement conditions. It is not required that each fitted component correspond one-to-one with a unique physical pore size category or a single moisture type.

[0055] The intersection point of adjacent fitted component curves is determined by... Spatial conversion to corresponding T The two values ​​serve as the operational standard uniformly adopted throughout all measurement stages. T Two boundary values ​​are used to divide short segments. T 2 domains, middle T 2 domains and length T Two fields, denoted as , and .

[0056] Then the first n Measure three operability T The relative signal contributions in the two domains are as follows: The intersection point of adjacent fitted component curves is determined by... Spatial conversion to corresponding T The two values ​​serve as the operational standard uniformly adopted throughout all measurement stages. T Two boundary values ​​are used to divide short segments. T 2 domains, middle T 2 domains and length T Two fields, denoted as , and . No. n Three operational aspects of the measurement phase T The relative signal contributions in the two domains are as follows:

[0057] (11); (12); (13); In the formula, , and They represent the first n Measurement phases: short, medium, and long T 2-domain semaphores are reserved for this stage T The proportion of the total signal in the two spectra, and satisfying , This forms the first n Operational relaxation region contribution vector during the measurement phase: (14); Relate the proportions to the two-factor state coordinates The combined output is used to analyze the relative contributions of different relaxation environments during the spectral center migration process.

[0058] Will With two-factor state coordinates The joint output is used to analyze the relative contributions of different relaxation environments during spectral center migration. Among them, Used for quantitative characterization retention T 2. Overall migration degree and total signal quantity change in the spectrum This is used to further explain the component contributions of different operational relaxation time domains to the overall migration.

[0059] In this embodiment, the centers of the three fitted components of the area-normalized average spectrum are located at approximately 1.5 ms, 5.3 ms, and 12.2 ms, respectively, and the intersection points of adjacent components correspond to... T The two cutoff values ​​are approximately 2.8ms and 9.4ms. Therefore, within the common range of this embodiment... T Within 2 ≤ 100ms, it is divided into:

[0060] short T 2 domains: T 2 < 2.8 ms; middle T 2 domains: 2.8 ms≤ T 2 < 9.4 ms; long T 2 domains: 9.4 ms≤ T 2 < 100 ms; In this embodiment, as Figure 6 As shown, the first measurement phase is short T Domain 2 response is dominant; as infiltration continues, the middle... T The proportion of domain 2 gradually increased and surpassed that of the short-term domain in the medium term. T 2 domains; in later stages, short, medium, and long. TThe relative contributions of the two domains are approximately 10.8%, 47.4%, and 41.8%.

[0061] The above short, medium and long T Domain 2 is only used as a reference range for the operational relaxation environment under the sandy loess and corresponding instrument configuration in this embodiment. Because T 2. It is also affected by factors such as surface relaxation characteristics, local saturation state, mineral composition, pore connectivity, and acquisition parameters, without using an independent method to establish... T When determining the quantitative calibration relationship between 2 and pore size, the above-mentioned boundary value is not used as the absolute pore size boundary applicable to other materials.

[0062] Example 3: HSE2 axial assisted positioning.

[0063] Based on the overall two-factor state characterization, the spatial location of the moisture response at different measurement stages can be further determined by using one-dimensional axial nuclear magnetic profiles.

[0064] HSE2 sequence testing was performed on the same specimen during at least part of the measurement phase. The HSE2 test obtained a one-dimensional NMR signal profile along the specimen height direction through axial spatial encoding. The signal at each axial position represents the combined NMR response within the corresponding spatial layer and specimen cross-section.

[0065] In this embodiment, the HSE2 receiving bandwidth is set to 20kHz, the waiting time is set to 2000ms, the number of scans is 16, and the same instrument parameters and sample axial position are maintained in each stage.

[0066] The first n HSE2 axial signal profile and corresponding two-factor state coordinates of the stage Establish a correspondence to help determine the spatial moisture response region that causes the overall two-factor state change. When When the total signal scaling factor increases but the spectral center migration factor changes only slightly, the HSE2 profile can be used to identify which axial regions of the sample the new response is mainly concentrated in, thereby distinguishing cases where the overall two-factor responses are similar but the spatial state response distributions are different.

[0067] The HSE2 axial signal is used for relative spatial comparison under the same sample and measurement settings. It is not required that the signal amplitude at a single location be directly equated to the absolute volumetric moisture content at that location.

[0068] In this embodiment, after the wetting front reached the bottom of the 12cm high sample, the overall water absorption continued to increase, and the HSE2 results showed that the newly added NMR response was mainly concentrated in the lower region of the sample, approximately 8-11cm high. Figure 7As shown, the spatial distribution of pore water response along the sample depth is used to characterize different infiltration stages, with depth measured from the top of the sample downwards. Meanwhile, the overall spectral center migration amplitude has significantly decreased. This result indicates that, through... When used in conjunction with the HSE2 profile, it can further identify the spatial localization of the new moisture response under the condition of "increased overall moisture response but limited overall relaxation environment migration".

[0069] Therefore, HSE2 is not used to replace two-factor state calculation, but rather as auxiliary information for its spatial positioning.

[0070] Example 4: SE-SPI hierarchical two-factor characterization and spatiotemporal state matrix of pore water.

[0071] For cases requiring further characterization of pore water states at different spatial strata, the SE-SPI sequence is used to obtain stratification data. T 2.

[0072] SE-SPI utilizes spatial encoding to obtain multi-echo attenuation signals at different axial positions of the sample, and performs separate analysis on the attenuation curves at different positions. T 2. Inversion, thereby obtaining the local values ​​of each spatial level. T 2. Relaxation spectrum.

[0073] In this embodiment, the SE-SPI receiving bandwidth is 250kHz, the echo time is 0.220ms, the waiting time is 2000ms, the number of echoes is 8000, and the number of scans is 16.

[0074] The sample is divided along the axial direction into The first layer. For the first layer Layer, First Local state T 2. Spectrum, let the discrete spectral points be... ,in, For the first Layer, First Hydraulic state at The amplitude of the local spectral signal at a given location. Following the principle that the overall spectrum is identical, the weighted logarithmic spectrum of the local signal is calculated. for:

[0075] (15); The local log-weighted representative relaxation time is: (16); For the comparison Q There are several hydraulic states. One hydraulic state is pre-selected as a local reference state, and the same reference state category is used in the local two-factor calculations of all layers. Let the first hydraulic state be... iThe weighted logarithmic spectral center, logarithmic weighted representative relaxation time, and local total signal quantity of the layer under the local reference state are respectively , and Then the first i Layer, number q The local spectral center migration factor of the hydraulic state is:

[0076] (17); No. i Layer, number q The local total signal quantity of the hydraulic state is: (18); The local total signal scaling factor is: (19); This allows us to construct local two-factor state coordinates: (20); The spatiotemporal state matrix of pore water is constructed from local two-factor state coordinates at different spatial levels and under different hydraulic states: (twenty one); in, I The number of spatial layers. Q The number of hydraulic states or measurement states being compared.

[0077] Each element in the matrix contains both local spectral migration information and local total signal quantity change information, thus forming a unified spatiotemporal representation of "spatial level-measurement state-two-factor state".

[0078] The hydraulic state can include at least two of the following: the infiltration end state, the state after sealing and settling, and the state of further saturation after reducing gas pressure, so as to identify the evolution of the internal moisture re-state after the cessation of external moisture input and the further water filling response.

[0079] In this embodiment, the 12cm high sample was divided into 7 continuous layers along the axial direction, and SE-SPI measurements were performed in three states: the first state was immediately after the staged infiltration ended; the second state was after external moisture input was stopped and the sample was sealed and left to stand for 48 hours; the third state was after the gas pressure was reduced and further saturated, as shown below. Figures 8-10 As shown.

[0080] To calculate the local total signal scaling factor between different states, SE-SPI preferably uses consistent spatial coding, receiver gain, pulse parameters, and inversion parameters for each state, ensuring inter-stage comparability of local spectral signal quantities at the same layer. If only intra-state normalized local spectra are available, local spectral center migration factors and inter-layer relative contributions can be used for auxiliary comparisons, rather than directly interpreting intra-state normalization ratios as absolute local water volume changes.

[0081] By comparing local areas at the same stratum under different conditions T 2. Spectrum Using #imgpt118#, we can determine whether there is still an internal moisture state adjustment after infiltration ends, and further identify whether there is an additional moisture response after reducing gas pressure.

[0082] In this embodiment, no new external moisture input occurred during the 48-hour sealed and static period, but the local NMR responses at different layers still changed, indicating that the end of infiltration was not the final state where the internal moisture state remained completely unchanged; after further saturation, a longer period of time... T The enhanced response within the range of 2 indicates that a relaxation environment that could further generate NMR water response still existed.

[0083] Thus, SE-SPI expands the global two-factor state coordinates #imgpt119# into local two-factor state coordinates #imgpt120# under different spatial layers and different hydraulic states, and further forms the pore water spatiotemporal state matrix.

[0084] S6: Pore water state evolution characterization system.

[0085] Based on the same concept, this embodiment provides a pore water state evolution characterization system during porous media infiltration, including an acquisition module, a calculation module, and a characterization module.

[0086] The acquisition module is used to acquire data from the same porous media sample at multiple successive infiltration stages. T 2. Relaxation spectrum, and calculate each T 2. The signal weighted logarithmic spectrum of the relaxation spectrum and the total signal quantity; The calculation module is used to construct the spectral center migration factor for each infiltration stage based on the displacement of its signal weighted logarithmic spectral center relative to the signal weighted logarithmic spectral center of the reference stage, and to construct the total signal scaling factor based on the ratio of its total signal to the total signal of the reference stage. The spectral center migration factor and the total signal scaling factor of the same infiltration stage are combined to form the two-factor state coordinates of that stage. The reference stage is a stage pre-selected from multiple consecutive infiltration stages. The spectral center migration factor is used to characterize the overall migration degree of the pore water relaxation environment, and the total signal scaling factor is used to characterize the relative change of the NMR detectable water response. The characterization module is used to connect the state coordinates of each dual factor in chronological order of infiltration stages to form the pore water state evolution trajectory.

[0087] This invention expands the capability for non-destructive testing of porous media infiltration processes.

[0088] This invention introduces multi-sequence low-field nuclear magnetic resonance into the testing of porous media infiltration processes, enabling continuous or phased acquisition of overall data from the same sample. T 2. Spectrum, height-direction moisture signal and stratification T 2. Distribution reduces the uncertainty caused by destructive sampling.

[0089] This invention can simultaneously characterize macroscopic infiltration response and pore water occurrence state.

[0090] This invention can not only obtain water absorption, moisture content, saturation, and wetting front position, but also identify pore water at different temperatures. T The distribution ratio in the interval 2 is used to realize the correlation analysis between macroscopic hydraulic response and pore water occurrence state.

[0091] This invention can identify the continued water filling and internal redistribution after the wetting front has penetrated.

[0092] This invention utilizes HSE2 height signal profiles and SE-SPI layering. T The distribution of moisture can determine whether there is still a redistribution of moisture in the already moistened area and continued water filling of the pore space after the wetting front reaches the bottom of the sample, thus avoiding misjudging the penetration of the wetting front as complete saturation or internal equilibrium.

[0093] This invention provides a quantitative characterization index for pore water spectral migration.

[0094] This invention proposes a logarithmic-based approach. T 2. Spectral center offset descriptor, which can quantify the pore water occurrence state towards a longer range. T 2. The degree of migration of pore domains with weaker constraints or higher mobility provides a unified indicator for comparison between different infiltration stages, different samples, or different porous media.

[0095] This invention has a wide range of applications.

[0096] This invention can be used not only for testing the infiltration process of loess, but also extended to testing and evaluation of pore water migration, changes in pore water occurrence state, and redistribution after infiltration in silt, clay, sand, compacted soil, core samples, and other water-bearing porous media.

[0097] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the invention.

[0098] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for characterizing the evolution of pore water state during the infiltration of porous media, characterized in that, Includes the following steps: Obtaining the same porous media sample at multiple successive infiltration stages T 2. Relaxation spectrum, and calculate each T 2. The signal weighted logarithmic spectrum of the relaxation spectrum and the total signal quantity; For each infiltration stage, a spectral center migration factor is constructed based on the displacement of its signal-weighted logarithmic spectral center relative to the signal-weighted logarithmic spectral center of the reference stage. A total signal scaling factor is constructed based on the ratio of its total signal quantity to the total signal quantity of the reference stage. The spectral center migration factor and the total signal scaling factor of the same infiltration stage are combined to form the two-factor state coordinates of that stage. The reference stage is a stage pre-selected from the plurality of consecutive infiltration stages. The spectral center migration factor is used to characterize the overall migration degree of the pore water relaxation environment, and the total signal scaling factor is used to characterize the relative change of the NMR-detectable water response. Connect the state coordinates of each dual factor in sequence according to the infiltration stage to form the pore water state evolution trajectory.

2. The method for characterizing the evolution of pore water state during porous media infiltration as described in claim 1, characterized in that, The specific spectral center migration factor is as follows: ； In the formula, For the first n Infiltration stage spectral center migration factor For the first n The signal weighted logarithmic spectrum center during the infiltration phase, The weighted logarithmic spectrum center of the signal in the reference phase, For the first n Log-weighted representative relaxation time during the infiltration phase. The log-weighted representative relaxation time for the reference phase; The total signal scaling factor is specifically as follows: ； In the formula, This is the signal scaling factor. For the first n Total signal quantity during the infiltration phase This refers to the total semaphore during the reference phase.

3. The method for characterizing the evolution of pore water state during porous media infiltration as described in claim 1, characterized in that, By T 2. After taking the natural logarithm of the transverse relaxation time of each discrete spectral point in the relaxation spectrum, a weighted average is performed using the corresponding spectral signal amplitude as the weight to obtain the weighted logarithmic spectral center of the signal. T The total signal quantity is obtained by summing the spectral signal amplitudes of each discrete spectral point in the relaxation spectrum.

4. The method for characterizing the evolution of pore water state during porous media infiltration as described in claim 1, characterized in that, The reference stage is the initial infiltration stage.

5. The method for characterizing the evolution of pore water state during porous media infiltration as described in claim 1, characterized in that, The T 2. Relaxation spectrum is retained T 2 spectra, retained T The acquisition of the spectrum specifically includes the following steps: Based on the stable inversion of each infiltration stage T 2. Public Scope: Identifying Each Infiltration Stage T 2. The relaxation spectrum represents the pore water response of the target porous medium and in T The target main spectrum distribution forms a continuous main body distribution on the two axes; If a non-target response exists outside the target main spectrum distribution, then a common boundary is selected to separate the target main spectrum distribution from the non-target response. T Two thresholds are used as cutoff boundaries to determine common... T 2. Scope of analysis; otherwise, based on commonly stable inversion. T 2. Public Scope as Common T 2. Scope of analysis; According to the common T 2. Scope of analysis: original samples at each infiltration stage T 2. The relaxation spectrum is truncated, and non-target responses are excluded to obtain the retention values ​​for each infiltration stage. T 2.

6. The method for characterizing the evolution of pore water state during porous media infiltration as described in claim 5, the original... T 2. Relaxation spectrum was obtained by applying a CPMG pulse sequence to the porous medium sample.

7. The method for characterizing the evolution of pore water state during porous media infiltration as described in claim 4, characterized in that, Also includes: Preservation of each infiltration stage T The two spectra are normalized according to their spectral areas, and the area-normalized average spectrum of all infiltration stages is calculated. exist T 2. Multi-component fitting of the area-normalized average spectrum within logarithmic space, using the intersection of adjacent fitted components to determine the operability that can be uniformly used throughout the entire infiltration stage. T Two boundary values ​​are used to form at least two different relaxation time intervals; Calculate the spectral area of ​​each infiltration stage in each relaxation time interval as a percentage of the area retained in that stage. T The proportion of the total signal quantity in the two spectra forms the operational relaxation domain contribution vector; The operational relaxation domain contribution vector is output together with the corresponding two-factor state coordinates to analyze the relative contributions of different relaxation environments during the spectral center migration process.

8. The method for characterizing the evolution of pore water state during porous media infiltration as described in claim 1, characterized in that, Also includes: Obtain HSE2 sequence axial nuclear magnetic resonance signal profiles of porous media samples at at least one infiltration stage; The HSE2 axial signal profile is output in correspondence with the dual-factor state coordinates of the same infiltration stage to identify the spatial region of water response that causes changes in dual-factor state.

9. The method for characterizing the evolution of pore water state during porous media infiltration as described in claim 1, characterized in that, Also includes: SE-SPI sequence was used to obtain the stratification of porous media samples at different spatial levels along the axial direction. T 2. Relaxation spectrum; For each spatial layer, calculate its local signal weighted logarithmic spectral center and local total signal quantity, and construct the local spectral center migration factor and local total signal scaling factor of the layer under each measurement state based on a pre-selected local reference state, forming a local two-factor state coordinate system. A spatiotemporal state matrix of pore water is constructed from local two-factor state coordinates at different spatial levels and under different measurement conditions. This matrix is ​​used to characterize the joint evolution of pore water state in the spatial and hydraulic dimensions.

10. A system for characterizing the evolution of pore water state during the infiltration of porous media, characterized in that, include: The acquisition module is used to acquire data from the same porous media sample at multiple successive infiltration stages. T 2. Relaxation spectrum, and calculate each T 2. The signal weighted logarithmic spectrum of the relaxation spectrum and the total signal quantity; The calculation module is used to construct a spectral center migration factor for each infiltration stage based on the displacement of its signal-weighted logarithmic spectral center relative to the signal-weighted logarithmic spectral center of the reference stage, and to construct a total signal scaling factor based on the ratio of its total signal quantity to the total signal quantity of the reference stage. The spectral center migration factor and the total signal scaling factor of the same infiltration stage are combined to form a two-factor state coordinate for that stage. The reference stage is a stage pre-selected from the plurality of consecutive infiltration stages. The spectral center migration factor is used to characterize the overall migration degree of the pore water relaxation environment, and the total signal scaling factor is used to characterize the relative change of the NMR-detectable water response quantity. The characterization module is used to connect the state coordinates of each dual factor in chronological order of infiltration stages to form the pore water state evolution trajectory.

Images