Wellbore radar change attribution-based borehole grouting effect evaluation method and system

Abstract

This invention discloses a method and system for evaluating the effect of wellbore grouting based on change attribution, relating to the field of grouting quality evaluation technology. The method includes: acquiring pre-grouting and post-grouting data for the same detection well; extracting baseline defect characteristics from the pre-grouting data and establishing a pre-grouting defect template; constructing a non-grouting reference response based on the pre-grouting defect template and combining the response patterns of stable sections and adjacent background changes; extracting measured response characteristics from the post-grouting data; calculating the grouting action and residual defect amount using dual-channel quantization based on the measured response characteristics, the non-grouting reference response, and the pre-grouting defect template; identifying change attribution based on the grouting action and residual defect amount to obtain initial attribution results; and performing section-level continuity optimization based on the initial attribution results to obtain the grouting effect evaluation results.

Description

Technical Field

[0001] This invention relates to the field of grouting quality evaluation technology, specifically to a method and system for evaluating the effect of wellbore perforation grouting based on variation attribution using in-well radar. Background Technology

[0002] In the quality inspection of grouting for dam foundations and cutoff walls, the most difficult problem to solve is not whether there are cracks or defects on the borehole wall surface, but whether the medium within a certain range around the borehole has been truly transformed by the grouting. Borehole television can only observe the integrity of the borehole wall surface, the development of cracks, and locally visible defects. Its detection results are essentially limited to the surface layer of the borehole wall and cannot reliably determine whether fracture zones, voids, seepage channels, and areas of incomplete grouting within a range of approximately 1–2 m around the borehole have been effectively filled and sealed. For dam foundation leakage, localized failure of cutoff walls, and deep seepage problems, what truly determines the grouting quality and seepage prevention effect is often the state of the medium in the vicinity of this borehole.

[0003] Downhole radar can respond to anomalies in the surrounding medium, thus providing a technical basis for verifying the grouting effect within a certain range around the borehole. However, current applications of downhole radar in grouting quality evaluation are generally rather crude, mostly involving direct comparison of B-Scan depth profiles before and after grouting, observing changes in the intensity, range, or local structural changes of anomalies, and then making manual judgments based on experience. While this method can detect "image changes," it cannot further address the three most critical issues in engineering evaluation: first, whether the differences in images before and after grouting are caused by the actual effectiveness of grouting, or by differences in measurement conditions, background fluctuations, or local disturbances; second, whether, despite changes after grouting, a large number of original defect characteristics are still preserved; and third, whether a certain depth section should be classified as an effective grouting area, a partially effective area, or a residual leakage area.

[0004] In other words, the problem with existing methods is not that they "cannot compare before and after images," but that they cannot attribute changes to causes. They can only identify whether a change exists, but cannot determine what kind of state change it belongs to in an engineering sense. Therefore, they cannot directly translate differences in well radar images into conclusions about the effectiveness of perforation grouting, and it is even more difficult to provide a reliable basis for supplementary grouting and remediation.

[0005] Therefore, there is an urgent need in this field for a new method for evaluating the effect of well-drilled radar grouting, which can not only identify changes before and after grouting, but also further determine: how much the peripore medium has changed, how much of the original defects are still retained, and which depth sections still belong to key risk sections. Summary of the Invention

[0006] The technical problem this invention aims to solve is that existing borehole television methods cannot verify the grouting effect in the near-hole perimeter, and existing downhole radar methods mainly rely on direct comparison of before and after images, resulting in rough evaluation results and numerous misjudgments. The purpose of this invention is to provide a method and system for evaluating the effect of perforated grouting using downhole radar based on change attribution. Instead of simply judging whether the downhole radar response changes before and after grouting, it first constructs a reference response for the target depth window unit under conditions without grouting. Then, it calculates the deviation of the measured response after grouting from this reference response to quantify the degree of modification of the perforated medium state by grouting. Simultaneously, it calculates the degree of retention of the measured response after grouting relative to the pre-grouting defect template to quantify the residual degree of the original defect after grouting. Through joint analysis of these two types of quantitative results, the attribution and identification of changes before and after grouting are achieved. This invention enables the decomposition of changes in wellbore radar B-Scan depth profile data before and after grouting into changes in effective grouting, changes in residual defects, and changes in non-target disturbances. Ultimately, it outputs distribution maps of grouting effect sections, distribution maps of residual defect sections, suspected residual seepage channel sections, and suggested re-grouting sections. This transforms wellbore radar from an auxiliary means of "reading and judging" into a "quantifiable, zoned, and decision-making" tool for evaluating the quality of wellbore grouting.

[0007] This invention is achieved through the following technical solution:

[0008] In a first aspect, the present invention provides a method for evaluating the effect of wellbore radar peri-hole grouting based on variation attribution, the method comprising:

[0009] Data before and after grouting were obtained for the same detection well. The data before grouting was the B-Scan depth profile data of the radar in the well before grouting, and the data after grouting was the B-Scan depth profile data of the radar in the well after grouting.

[0010] Based on the pre-grouting data, baseline disease characteristics were extracted and a pre-grouting disease template was established; based on the pre-grouting disease template and combined with the response patterns of stable sections and the changes in adjacent background, a non-grouting reference response was constructed.

[0011] Based on the data after grouting, the measured response characteristics after grouting are extracted; and based on the measured response characteristics after grouting, the reference response before grouting, and the disease template before grouting, the grouting effect and the residual amount of disease are calculated using dual-channel quantization.

[0012] The changes in grouting volume and residual disease amount are identified by attribution to obtain initial attribution results; based on the initial attribution results, section-level continuous optimization is carried out to obtain grouting effect evaluation results.

[0013] Furthermore, the method also includes:

[0014] Data before and after grouting were standardized. The standardization process included:

[0015] The depth coordinates of the pre-grouting and post-grouting data are registered, the response amplitudes of the pre-grouting and post-grouting data are scaled, the background components of the pre-grouting and post-grouting data are suppressed, and the pre-grouting and post-grouting data are standardized to form a dual-temporal response pair before and after grouting.

[0016] Furthermore, based on the pre-grouting data, baseline disease characteristics were extracted and a pre-grouting disease template was established, including:

[0017] Based on the pre-grouting data, baseline disease features before grouting are extracted by depth window unit; the pre-grouting baseline disease features are the abnormal responses corresponding to the original diseases around the holes, including energy features, width features, morphological features and continuity features.

[0018] Based on the baseline defects characteristics before grouting, a baseline defect feature vector for the depth window unit before grouting is constructed;

[0019] Based on the combination relationship of each feature in the baseline disease feature vector, abnormal candidate segments are identified; and a pre-grouting disease template is established for the depth window unit within the abnormal candidate segments.

[0020] Furthermore, based on the pre-grouting defect template and combined with the response patterns of stable sections and the changes in adjacent background, a reference response without grouting is constructed, including:

[0021] Based on the response pattern of stable sections, the correction coefficients for baseline defects before grouting are calculated based on the set of stable sections.

[0022] Based on the correction coefficients and the pre-grouting defect template, a preliminary ungrouted reference response is constructed for the target depth window unit;

[0023] By combining the variation patterns of the adjacent background, the preliminary ungrouted reference response is locally modified to obtain the ungrouted reference response.

[0024] Furthermore, based on the post-grouting data, the measured response characteristics after grouting are extracted; and based on the measured response characteristics after grouting, the reference response before grouting, and the pre-grouting disease template, the grouting effect and residual disease amount are calculated using dual-channel quantization, including:

[0025] Based on the post-grouting data, the measured response characteristics after grouting are extracted from the depth window unit;

[0026] The first channel quantitative calculation: Based on the comparison between the measured response characteristics after grouting and the reference response without grouting, the deviation of each measured response characteristic after grouting is calculated; the deviation includes energy deviation, width deviation, morphological deviation and continuity deviation; and the deviations are weighted and combined to obtain the grouting effect of the target depth window unit; the grouting effect represents the degree of modification after grouting relative to the expected state without grouting.

[0027] The second channel quantitative calculation compares the measured response characteristics after grouting with the disease template before grouting, and calculates the retention amount of each measured response characteristic after grouting. The retention amount includes energy retention, width retention, morphology retention and continuity retention. The retention amounts are weighted and combined to obtain the disease residue of the target depth window unit. The disease residue represents the degree of retention of the original disease characteristics after grouting.

[0028] Furthermore, based on the grouting action amount and residual disease amount, the change attribution was identified to obtain initial attribution results, including:

[0029] Based on the relationship between grouting action and residual disease amount, an attribution analysis model is used to determine the type of change in the target depth window unit; the attribution analysis model includes:

[0030] When the grouting action amount is greater than or equal to the grouting action threshold, and the residual amount of disease is less than the residual amount of disease, it is determined to be an effective grouting zone;

[0031] When the amount of grouting is greater than or equal to the grouting threshold, and the amount of residual disease is also greater than or equal to the residual disease threshold, it is determined to be a partially effective area.

[0032] When the amount of grouting is less than the grouting threshold and the amount of residual disease is also less than the residual disease threshold, it is determined to be a non-target disturbance area or a normal background area.

[0033] When the amount of grouting is less than the grouting threshold and the amount of residual disease is greater than or equal to the residual disease threshold, it is determined to be a residual disease area or a suspected residual seepage channel area.

[0034] Furthermore, based on the initial attribution results, segment-level continuity optimization is performed to obtain grouting effect evaluation results, including:

[0035] The initial attribution results of each depth window unit are subjected to segment-level continuity optimization, adjacent depth window units of the same type are merged, and isolated category segments are corrected or removed to form the grouting effect evaluation results.

[0036] The grouting effect evaluation results include the distribution map of the grouting effect around the hole, the distribution map of the residual disease section, the suspected residual seepage channel section, and the recommended supplementary grouting section.

[0037] Furthermore, the objective function for segment-level continuity optimization is:

[0038] ;

[0039] in, Represents depth window unit Assigned category The matching cost at that time, which is determined by the amount of grouting action. Residual amount of disease Determine the degree of consistency with the corresponding category; This is an indicator function; it takes a value of 1 when two adjacent depth window cells are of different categories, and a value of 0 otherwise. is the category jump penalty coefficient, used to control the smoothness of the category sequence; m is the total number of depth window units obtained along the depth direction of the detection hole; i is the index of the depth window unit, which takes values ​​of 1, 2, ..., m in the depth direction. Let i be the category of the i-th depth window unit. The category of the (i+1)th depth window unit.

[0040] Secondly, the present invention provides a well-drilled radar peri-hole grouting effect evaluation system based on change attribution, the system comprising:

[0041] The acquisition unit is used to acquire pre-grouting and post-grouting data for the same detection well. The pre-grouting data is the in-well radar B-Scan depth profile data before grouting, and the post-grouting data is the in-well radar B-Scan depth profile data after grouting.

[0042] The pre-grouting disease template establishment unit is used to extract the baseline disease characteristics before grouting and establish the pre-grouting disease template based on the pre-grouting data.

[0043] The ungrouted reference response construction unit is used to construct the ungrouted reference response based on the pre-grouting disease template and combined with the response law of the stable section and the change law of the adjacent background.

[0044] The dual-channel quantization calculation unit is used to extract the measured response characteristics after grouting based on the data after grouting; and to calculate the grouting effect and residual amount of disease based on the measured response characteristics after grouting, the reference response before grouting, and the disease template before grouting, using dual-channel quantization.

[0045] The attribution identification unit is used to identify the causes of changes based on the amount of grouting action and the amount of residual disease to obtain the initial attribution results.

[0046] The section continuity optimization unit is used to perform section-level continuity optimization based on the initial attribution results to obtain grouting effect evaluation results.

[0047] Furthermore, the dual-channel quantization calculation unit includes:

[0048] The measured response feature extraction subunit is used to extract the measured response features after grouting from the depth window unit based on the post-grouting data.

[0049] The first channel quantization calculation subunit is used to compare the measured response characteristics after grouting with the reference response before grouting, and calculate the deviation of each measured response characteristic after grouting. The deviation includes energy deviation, width deviation, morphological deviation and continuity deviation. The deviation is then weighted and combined to obtain the grouting effect of the target depth window unit. The grouting effect represents the degree of modification after grouting relative to the expected state before grouting.

[0050] The second channel quantization calculation sub-unit is used to compare the measured response characteristics after grouting with the disease template before grouting, and calculate the retention amount of each measured response characteristic after grouting. The retention amount includes energy retention, width retention, morphology retention and continuity retention. The retention amount is weighted and combined to obtain the disease residue amount of the target depth window unit. The disease residue amount represents the degree of preservation of the original disease characteristics after grouting.

[0051] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0052] 1. This invention relates to a method and system for evaluating the effect of peri-hole grouting using in-well radar based on change attribution. This invention proposes for the first time a "non-grouting reference response" construction mechanism. That is, instead of simply using the measured response before grouting for direct comparison, it constructs a reference response or reference interval for the target depth window unit under the condition of "no grouting but in the current re-measurement environment" based on a pre-grouting defect template and combined with the overall offset law of the stable section and the local background change law. This mechanism provides a clear normative benchmark for subsequent grouting effect evaluation, fundamentally differentiating it from ordinary direct difference methods between before and after images.

[0053] 2. This invention relates to a method and system for evaluating the effect of wellbore perforation grouting based on change attribution. The invention establishes a dual-channel quantitative evaluation model: one channel constructs the grouting action quantity based on the deviation of the measured response after grouting from the ungrouted reference response, used to evaluate whether grouting truly changed the state of the perforation medium; the other channel constructs the residual quantity of the defect based on the degree of retention of the measured response after grouting relative to the defect template before grouting, used to evaluate whether the original fracture zone, voids, or seepage channels still exist. This dual-channel model enables the evaluation of grouting effect to simultaneously assess both the degree of modification and the degree of residual defects, thus significantly improving the algorithm's engineering interpretability.

[0054] 3. This invention relates to a method and system for evaluating the effect of peri-hole grouting using downhole radar based on variation attribution. By jointly analyzing the grouting effect and residual defects, this invention attributes the variations in each depth window unit to different engineering categories, further forming section-level evaluation results. Thus, the variations in the downhole radar image are no longer merely differences, but are transformed into results with clear engineering implications, such as effective grouting sections, partially effective sections, residual defect sections, and suspected residual seepage channels. This variation attribution logic represents a substantial improvement over existing methods that rely solely on simple difference analysis.

[0055] 4. This invention, based on a variation attribution-based method and system for evaluating the effect of peri-hole grouting using in-well radar, proposes a continuous optimization output mechanism from depth window units to section-level results. This advances the invention from an image analysis algorithm to an engineering evaluation method. The output results are no longer just local feature differences or single-point identifications, but can directly provide engineering results such as effective grouting sections, residual defect sections, and suggested re-grouting sections, thereby truly meeting the practical needs of peri-hole grouting quality inspection for dam foundations and anti-seepage walls. Attached Figure Description

[0056] The accompanying drawings, which are included to provide a further understanding of embodiments of the invention and form part of this application, do not constitute a limitation thereof. In the drawings:

[0057] Figure 1 This is a flowchart of the well radar peri-hole grouting effect evaluation method based on change attribution of the present invention;

[0058] Figure 2 This is a schematic diagram of the depth window unit division and feature extraction of the present invention; Figure 2 This paper illustrates the process of extracting local response sequences from radar B-Scan depth profile data in wells before and after grouting, using the same depth window unit as the basic analysis object, and further extracting energy features, width features, morphological features and continuity features to form feature vectors before and after grouting.

[0059] Figure 3 This is a block diagram of the well-drilled radar peri-hole grouting effect evaluation system based on variation attribution, as per the present invention. Detailed Implementation

[0060] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments and accompanying drawings. The illustrative embodiments and descriptions of the present invention are only used to explain the present invention and are not intended to limit the present invention.

[0061] This invention addresses a key problem in the quality inspection of grouting for dam foundations and anti-seepage walls: borehole television can only reflect the integrity and visible defects of the borehole wall surface, and cannot reliably determine whether the medium within a range of about 1–2 m around the borehole has been effectively grouted; while existing in-well radar evaluation methods mostly only stop at the level of direct comparison of B-Scan images before and after grouting, which can only identify "whether there is a change", but cannot further determine whether such change is due to the effective effect of grouting, the retention of residual defects, or non-target disturbances, etc., and therefore cannot directly serve the evaluation of the grouting effect around the borehole and the decision-making of supplementary grouting.

[0062] To address the aforementioned issues, this invention presents a method and system for evaluating the effectiveness of borehole grouting using in-well radar based on variation attribution. This method uses B-Scan depth profile data from the same detection well before and after grouting as input. First, the dual-temporal data is standardized to ensure comparability in depth coordinates, response scale, and background conditions. Then, abnormal response features corresponding to the original defects are extracted from the pre-grouting baseline profile to establish a pre-grouting defect template. Based on this template and the response patterns of stable sections, a reference response for the target depth window unit under conditions without grouting is constructed. Subsequently, the deviation of the post-grouting measured response from the ungrouted reference response and the retention of the post-grouting measured response relative to the pre-grouting defect template are calculated, forming two quantitative evaluation lines: grouting effect and residual defect amount. Finally, based on these two quantitative results, the changes in each depth window unit are attributed and identified. After section-level continuity optimization, the effective grouting section, partially effective section, residual defect section, suspected residual seepage channel section, and suggested re-grouting section are output.

[0063] The core of this invention lies not in simply extracting differences before and after grouting, but in establishing an evaluation mechanism for attributing changes. Specifically, this invention does not directly use pre-grouting data as the sole reference, nor does it directly regard the difference before and after grouting as the grouting effect. Instead, it introduces an intermediate reference layer—the non-grouting reference response—so that subsequent evaluations are based on "what response should this depth section exhibit under the current retesting conditions if grouting were not performed." Simultaneously, it measures whether the original defects still remain after grouting by measuring the residual defect amount. With these two independent but complementary evaluation metrics, this invention can decompose changes in well-drilled radar images into information of different engineering attributes, thereby truly answering the three core engineering questions: "Was the grouting effective? Are the defects still present? Where is re-grouting needed?"

[0064] Furthermore, this invention employs depth window units instead of single depth sampling points as the basic analysis object. Within each depth window, multi-dimensional features such as energy, width, morphology, and continuity are extracted to improve the characterization of complex disease responses such as peri-hole fracture zones, voids, and seepage channels, and to enhance the algorithm's robustness to local noise and occasional fluctuations. Subsequently, by constructing a dual-channel evaluation system for grouting action and disease residue, combined with segment continuity optimization, the originally discrete, localized, and difficult-to-interpret image changes are transformed into segment-level evaluation results with clear engineering significance.

[0065] Therefore, this invention does not provide a conventional method for analyzing differences before and after borehole radar testing, but rather a dual-temporal B-Scan variation attribution evaluation method for borehole grouting effects. This method can simultaneously evaluate the degree of borehole grouting modification, the degree of residual original defects, and the identification of re-grouting sections, thereby transforming borehole radar from an "auxiliary map-reading tool" into a "quantifiable, zone-based, and decision-making" tool for inspecting the quality of borehole grouting.

[0066] Example 1

[0067] like Figure 1 As shown, the present invention provides a method for evaluating the effect of wellbore radar peri-hole grouting based on variation attribution. This method includes:

[0068] S1, acquire the data before and after grouting of the same detection hole. The data before grouting is the radar B-Scan depth profile data in the well before grouting, and the data after grouting is the radar B-Scan depth profile data in the well after grouting.

[0069] Step S1 is used to establish the raw input data required for evaluating the effect of perforation grouting, and it is the starting point of the entire scheme. Since the evaluation objective of this invention is not to perform static interpretation of a single well radar image, but to perform attribution analysis on the changes in the state of the perforation medium before and after grouting, it is necessary to obtain well radar B-Scan depth profile data of the same detection well in two stages, before and after grouting, and ensure that the two sets of data have a one-to-one correspondence in terms of spatial range, measurement object, and acquisition method.

[0070] In this embodiment, before grouting, in-well radar measurements are first performed within the target detection borehole to obtain the pre-grouting B-Scan depth profile data. This data characterizes the original response of the borehole surrounding medium in the ungrouted state and serves as the basis for subsequent extraction of baseline defect responses, establishment of defect templates, and construction of ungrouted reference responses. Subsequently, after the corresponding grouting operation is completed and predetermined detection conditions are met, in-well radar measurements are performed again on the same detection borehole to obtain post-grouting B-Scan depth profile data. This data characterizes the actual response state of the borehole surrounding medium after grouting and is the direct object for subsequent calculations of grouting action and defect residue.

[0071] To ensure that subsequent calculations accurately reflect the grouting effect rather than differences in sampling conditions, data acquisition before and after grouting should adhere to the principles of "same borehole, same section, same system." "Same borehole" means both measurements are performed on the same detection borehole; "same section" means both measurements cover the same or substantially the same depth range; and "same system" means both measurements use the same or equivalent sampling interval, measurement step size, transmission and reception parameters, and recording methods as much as possible. Ensuring this consistency provides a reliable basis for subsequent dual-time-phase comparisons and reduces spurious differences caused by variations in measurement conditions.

[0072] During data acquisition, both pre- and post-grouting B-Scan depth profile data from the in-well radar were recorded with the depth direction as the primary axis. Each B-Scan depth profile is essentially a continuous response image of the detection borehole along the depth direction, reflecting the changes in the electromagnetic response of the surrounding medium at different depth locations. To facilitate subsequent algorithm analysis, in addition to acquiring the original B-Scan profile from the in-well radar, each measurement also simultaneously recorded basic information directly related to that data, including but not limited to: measurement time, detection borehole number, depth start position, depth end position, sampling interval, measurement step size, measurement parameter group, and the stage identifier to which the data group belongs. Among these, the stage identifiers are at least distinguished into two categories: pre-grouting baseline and post-grouting re-measurement.

[0073] In this invention, the radar B-Scan depth profile data in the well before grouting and the radar B-Scan depth profile data in the well after grouting are respectively denoted as:

[0074]

[0075] in, This refers to data prior to grouting, specifically the radar B-Scan depth profile data within the well before grouting. This represents the data after grouting, specifically the radar B-Scan depth profile data in the well after grouting. Together, these two data points constitute the dual-temporal input data pair for subsequent algorithm analysis.

[0076] If we further describe it from the perspective of data structure, then and Both can be viewed as response sequences or response matrices after discrete sampling along the depth direction. For any depth location, a corresponding response record should be found in both measurements before and after grouting to ensure the physical meaning of subsequent comparisons. In other words, the data output in this step is not two independent images, but a set of bi-temporal response data that can be compared segment by segment in a unified depth domain.

[0077] S2, unify the data before and after grouting to obtain unified data before and after grouting;

[0078] Step S2 is used to eliminate the differences introduced by non-target factors in the two well radar measurements before and after grouting, so that the two B-Scan depth profiles can be used in subsequent calculations on a unified basis. Since this invention requires the calculation of grouting effect and residual damage based on the difference in response before and after grouting, if the two sets of data are inconsistent in depth coordinates, sampling density, amplitude scale, or background composition, the resulting deviations and retentions will contain a large amount of components not caused by grouting, thus reducing the reliability of the evaluation results. Therefore, before feature extraction and quantification, it is necessary to first analyze the well radar B-Scan depth profile data before grouting. And grouting-in-well radar B-Scan depth profile data Standardize the process.

[0079] In this embodiment, the standardization process includes:

[0080] The depth coordinates of the pre-grouting and post-grouting data are registered, the response amplitudes of the pre-grouting and post-grouting data are scaled, the background components of the pre-grouting and post-grouting data are suppressed, and the pre-grouting and post-grouting data are standardized to form a dual-temporal response pair before and after grouting.

[0081] Specifically, firstly, the depth coordinates of the pre-grouting and post-grouting data are registered. Due to potential deviations in starting and ending depths, inconsistent depth sampling intervals, or slight drifts in recording points during field measurements, the same physical depth may not correspond to exactly the same sampling location in two profiles. Therefore, it is necessary to map the two sets of measurement data to the same depth coordinate sequence, ensuring a clear one-to-one correspondence between the pre-grouting and post-grouting data at each depth location. Specific methods may include depth axis interpolation, resampling, or depth correction to unify the two profiles to the same depth grid. Only after registration can the comparison of data from two grouting operations at the same depth have physical meaning.

[0082] Secondly, the response amplitudes of the data before and after grouting are standardized in scale. In-well radar measurements are affected not only by the medium itself but also by instrument status, coupling conditions, the state of the medium within the well, and changes in background energy levels. Therefore, even measurements targeting the same target section may show a shift in the overall response level. Directly comparing the original amplitudes can easily lead to misinterpreting the amplitude differences caused by systematic shifts as grouting effects. To avoid this problem, the responses before and after grouting need to be normalized to make them comparable on an overall scale. Normalization can be based on the statistical characteristics of the entire profile, the statistical characteristics of stable sections, or local background response characteristics. After this processing, the amplitude expressions of the two profiles before and after grouting are unified to the same scale, allowing subsequent change analysis to better reflect the state changes of the medium itself.

[0083] Secondly, background components in the pre- and post-grouting data are suppressed. The original radar profiles from the well typically contain background responses, low-frequency trends, or irrelevant interference components. These components do not directly correspond to peri-hole defects or grouting effects, but they can create spurious differences in two-phase comparisons. Therefore, in the standardization process, background components that coexist in the pre- and post-grouting profiles and have no direct correlation with the target anomalies need to be weakened or removed, allowing subsequent analysis to focus more on the anomaly sections themselves. After background suppression, target anomalies such as peri-hole fracture zones, voids, and seepage channels, as well as response changes caused by grouting, will be more prominent.

[0084] After depth registration, scale unification, and background suppression, the pre-grouting and post-grouting data are standardized. This means that both B-Scan profiles before and after grouting are represented as standardized response sequences in a unified depth coordinate system. For any depth location... Let the standardized responses before and after grouting be denoted as follows: and This results in a two-phase response before and after grouting:

[0085]

[0086] This dual-phase response pair is the direct input for subsequent window partitioning, feature extraction, construction of ungrouted references, and calculation of grouting action and residual disease amount.

[0087] In a preferred embodiment, to improve the characterization ability of subsequent analysis of local anomaly zones, after the unification process is completed, the standardized response sequence can be divided into depth windows according to a preset depth window width. That is, based on each center depth position... Create a corresponding analysis window around the center:

[0088]

[0089] And extract the local response sequences within the window range, denoted as follows: and The local response sequence of the depth window will serve as the basis for subsequent feature extraction and quantization calculations. By analyzing using a depth window instead of a single point, the algorithm's resistance to local noise and random fluctuations can be improved, resulting in more stable evaluation metrics.

[0090] S3. Based on the data before grouting, extract the baseline disease characteristics before grouting and establish a disease template before grouting;

[0091] Step S3 is used to extract the abnormal response corresponding to the original perforation defects from the in-well radar B-Scan depth profile data before grouting, and to establish a baseline defect template (i.e., the pre-grouting defect template) required for subsequent defect residue identification. Since the evaluation objective of this invention is not only to determine whether the perforation medium has changed after grouting, but also to further determine whether the original defects still remain after grouting, it is necessary to first clarify the response morphology and characteristic composition of the original defects in the in-well radar profile before grouting. In other words, the role of this step is not simply to find where there are anomalies before grouting, but to establish a standardized feature template that can represent the state of defects before grouting, providing a reference basis for subsequent defect residue calculation.

[0092] Step S3 specifically includes:

[0093] After S2 standardization, the radar B-Scan depth profile data in the well before grouting was obtained. It has been represented as a standardized response sequence in a unified depth coordinate system. Based on this sequence, a preset depth window unit is used. The basic analysis object is the local response sequence before grouting. Anomaly feature extraction is performed. A depth window, rather than a single depth sampling point, is used as the analysis object because the response of radar defects in wells typically manifests as a continuous anomaly band within a certain depth range, rather than an isolated single-point change. By analyzing the overall local response within the depth window, the structural characteristics of target defects such as fracture zones, voids, and seepage channels can be more stably characterized, and the impact of local noise and occasional disturbances on defect identification can be reduced.

[0094] like Figure 2 As shown, for each depth window unit From the local response sequence before grouting Extract at least the following four types of features:

[0095] One is the energy characteristic, which is used to characterize the overall strength of the anomalous response within the depth window;

[0096] The second is the width feature, which is used to characterize the expansion range of the effective anomaly response in the depth window;

[0097] Thirdly, morphological features are used to characterize the contour structure and waveform pattern of local anomalies in B-Scan;

[0098] Fourthly, there is the continuity feature, which is used to characterize the coherence of the anomalous response between the depth window and adjacent depth windows.

[0099] The above four types of features together constitute a comprehensive description of the original disease state of this depth window.

[0100] Among them, the depth window unit before grouting Energy characteristics It can be defined as:

[0101]

[0102] In the formula, Indicates the standardized response at depth before grouting. The value at this location. This feature reflects the total energy level of the abnormal response within the depth window and can effectively characterize the significance of the disease response.

[0103] Pre-grouting depth window unit Width features It can be defined as:

[0104]

[0105] In the formula, To preset an abnormal threshold, This indicates the number of depth sampling points that meet the conditions. This feature reflects the effective spread width of the local anomaly zone and can characterize the size of the area affected by the disease.

[0106] Pre-grouting depth window unit morphological characteristics It can be defined as the normalized morphological vector of its local response sequence:

[0107]

[0108] In the formula, This represents the norm of the local response sequence. This feature is used to characterize the structural outline and relative distribution of the abnormal band within the window, and is an important basis for subsequent analysis of whether the original lesion morphology still exists after grouting.

[0109] Pre-grouting depth window unit Continuity characteristics It can be defined as the average similarity between the window and its top and bottom adjacent windows:

[0110]

[0111] In the formula, The similarity function can be cosine similarity, correlation coefficient, or other functions that reflect the degree of local morphological similarity. This feature reflects the continuity and coherence of the anomalous response along the depth direction, and is of great significance for identifying continuous fracture zones or continuous seepage channels.

[0112] Based on the above characteristics, a depth window unit can be constructed before grouting. Baseline disease feature vector :

[0113]

[0114] For depth window cells located in clearly abnormal zones, their baseline defect feature vector is defined as the defect template for that cell:

[0115]

[0116] Right now:

[0117]

[0118] The disease template not only preserves the intensity information of local anomalies before grouting, but also retains the continuity characteristics of the anomaly's range, morphology, and depth. Therefore, it can express the original disease state from multiple dimensions. This is crucial because even if the response intensity of a certain section changes after subsequent grouting, the original disease may still be partially preserved in terms of morphology or continuity. If only a single image feature is used, it is often impossible to reliably identify this residual disease state.

[0119] During template creation, it is also necessary to distinguish between abnormal sections and background sections. For depth window units that do not belong to obvious disease anomalies, background templates or low-anomaly templates can be created to distinguish ordinary background changes from true disease changes in subsequent disease residual quantity calculation and change attribution identification. Preferably, abnormal candidate sections can be identified based on the combination relationship of energy features, width features, and continuity features in the pre-grouting feature vector, and then a disease template set can be created for the depth window units within the abnormal candidate sections, i.e., the pre-grouting disease template.

[0120] Furthermore, to maintain the structural integrity of the disease template within a section, this invention not only records the template features of individual depth window units but also retains their relative position and adjacency relationships within their respective abnormal sections. In other words, the template is not merely an isolated window feature but a local structural unit with a contextual relationship. The purpose of this is that when the strength of a certain depth window unit weakens after grouting, as long as its overall morphological relationship and positional characteristics within the section remain relatively consistent, subsequent algorithms can still identify it as part of the disease remnants, rather than misjudging it as the complete disappearance of the disease due to a change in a single feature.

[0121] S4, based on the pre-grouting defect template and combined with the response law of stable section and the change law of adjacent background, constructs the reference response before grouting; among them, the response law of stable section refers to the overall scale shift, morphological drift and continuity change law of the well radar response characteristics in a stable depth window unit that does not show obvious defect abnormalities before and after grouting and is less affected by grouting, as the detection conditions change between the two detections; the change law of adjacent background refers to the local response change trend of the non-defect background section around the target depth window unit in the two detections before and after grouting, which is used to characterize the non-target changes near the target location caused by local medium differences, changes in coupling conditions or background fluctuations.

[0122] Step S4 establishes a benchmark for calculating subsequent grouting action. Its core purpose is to answer the question: for a target depth window element, if that location was not subsequently subjected to grouting, what response characteristics should it exhibit under post-grouting re-measurement conditions? This invention argues that grouting effect evaluation cannot simply use the pre-grouting template as the sole control, because there may be non-target differences between the two measurements due to instrument status, coupling conditions, background levels, and local fluctuations. If these differences are not corrected, any subsequent "before-and-after changes" may contain components not caused by grouting, leading to overestimation or misjudgment of the grouting action. Therefore, this step does not directly replicate the pre-grouting defect template, but rather uses it as a basis to construct a "reference response that should appear in the current re-measurement environment under conditions where no grouting has occurred."

[0123] In this embodiment, step S4 specifically includes:

[0124] Based on the pre-grouting defect template established in S3 Based on this, a preliminary ungrouted reference response is constructed for each depth window cell. To do this, it is first necessary to identify offset information unrelated to grouting between measurements before and after grouting. This offset information is not directly estimated from the target disease area itself, but is preferably extracted from relatively stable depth window cells that are less affected by grouting and do not belong to major abnormal sections. Let these stable depth window cells form a set. The set can be determined by the following principles: the overall change in response before and after grouting is small, the template section is not significantly damaged before grouting, and it shows high consistency in the two profiles.

[0125] Set in the stable region The overall change patterns of characteristics before and after grouting are statistically analyzed to establish a non-target offset mapping between the pre-grouting and post-grouting re-measurement conditions. For energy characteristics, the energy correction coefficient for the stable section is defined as:

[0126]

[0127] in, This represents the median calculation. This coefficient reflects the overall energy scale shift after grouting relative to before grouting measurements in stable sections where grouting has not occurred. If This indicates that the overall energy level measured after grouting is lower than before grouting; if This indicates that the overall energy level measured after grouting is higher than before grouting.

[0128] Similarly, for width features and continuity features, the correction coefficients are defined as follows:

[0129]

[0130]

[0131] For morphological features, since they are essentially vectors or structural features, they are not suitable for simple scaling. Therefore, the average morphological drift term on the stable segment is defined as follows:

[0132]

[0133] in, This indicates the mean value calculation. This drift term reflects the baseline of morphological changes that are prevalent in the stable section between two measurements before and after grouting, and can be used to correct the pre-grouting morphological template of the target depth window unit.

[0134] After obtaining the above correction amount, for any target depth window unit Construct its ungrouted reference feature, i.e., the ungrouted reference response:

[0135]

[0136] in,

[0137]

[0138]

[0139]

[0140]

[0141] In other words, the ungrouted reference response is not the original characteristic before grouting, but rather a predicted expression of the pre-grouting characteristic under post-grouting retesting conditions. This is crucial because it imbues subsequent measurements with a clearer physical meaning: the post-grouting measured value is not directly compared to the pre-grouting value, but rather to "the characteristics that should have been present without grouting." Only within this comparative framework are the differences obtained in subsequent calculations more likely to represent the true grouting effect, rather than a measurement bias.

[0142] In a preferred embodiment, to further improve robustness, this step does not express the ungrouted reference as a single fixed value, but rather as a reference interval. That is, for any depth window unit... By combining the fluctuation range of stable sections, the level of local background fluctuations, or the statistical characteristics of repeated measurements, corresponding upper and lower limits for ungrouted references are established for various characteristics. For example, for energy characteristics, an energy reference interval can be constructed:

[0143]

[0144] Similarly, reference intervals for width, continuity, and morphological similarity can be constructed separately. The significance of using interval representation is that, in subsequent evaluations, it is no longer necessary to mechanically judge whether the measured value after grouting deviates from a single reference value, but rather to judge whether it significantly deviates from the reasonable fluctuation range under non-grouting conditions. This can reduce the impact of local noise and random measurement errors on the calculation of grouting action.

[0145] Furthermore, in a more refined implementation, this step can also introduce a local correction mechanism. That is, based on the correction coefficients for the globally stable section, and combined with the variation patterns of the background windows adjacent to the target depth window unit above and below, the correction mechanism is adjusted accordingly. Local corrections are made to make the ungrouted reference more closely resemble the local measurement environment at that depth. This avoids the oversmoothing problem that can result from relying solely on global correction, ensuring that the ungrouted reference has both overall consistency and retains local adaptability.

[0146] S5. Based on the data after grouting, extract the measured response characteristics after grouting; and based on the measured response characteristics after grouting, the reference response before grouting, and the disease template before grouting, calculate the grouting effect and the residual amount of disease using dual-channel quantization.

[0147] In this embodiment, step S5 specifically includes:

[0148] Step S51: Based on the post-grouting data, extract the measured response features of the depth window unit after grouting.

[0149] Step S52, First channel quantization calculation: Based on the comparison between the measured response characteristics after grouting and the reference response before grouting, calculate the deviation of each measured response characteristic after grouting; the deviation includes energy deviation, width deviation, morphological deviation and continuity deviation; and weight the deviations to obtain the grouting effect of the target depth window unit; the grouting effect characterizes the degree of modification after grouting relative to the expected state before grouting;

[0150] Step S53, second channel quantification calculation: compare the measured response features after grouting with the disease template before grouting, and calculate the retention amount of each measured response feature after grouting; the retention amount includes energy retention amount, width retention amount, morphology retention amount and continuity retention amount, and weight the retention amount to obtain the disease residue amount of the target depth window unit; the disease residue amount represents the degree of retention of the original disease features after grouting.

[0151] Steps S51 and S52 are used to extract the measured response characteristics of the target depth window cell from the post-grouting data (i.e., the radar B-Scan depth profile data in the well after grouting), and on this basis, calculate the degree of deviation of the depth window cell from the ungrouted reference state, thereby forming a quantitative result characterizing the strength of the actual grouting effect. If S4 addresses what should have been like before grouting, then S51 and S52 address what actually happened after grouting, and how significant this change is. Therefore, this step is the key step in this invention that transitions from "reference construction" to "quantification of grouting effect".

[0152] Specifically, using the B-Scan depth profile data from the grouting well after S2 standardization. For input, such as Figure 2 As shown, for each depth window unit Extract the local response sequence after grouting within this window. And calculate the corresponding measured feature vector after grouting:

[0153]

[0154] in, This refers to the energy characteristics after grouting. This refers to the width characteristics after grouting. This refers to the morphological characteristics after grouting. This represents the continuity features after grouting. The extraction methods for the above features are consistent with the extraction methods for the pre-grouting disease template features in S3, to ensure that the pre-grouting, post-grouting, and ungrouted references are in the same feature expression system.

[0155] The energy characteristics of the depth window element after grouting are defined as follows:

[0156]

[0157] In the formula, Indicates the standardized response at depth after grouting. The value at this location. This feature is used to characterize the overall energy level of local abnormal responses after grouting.

[0158] The width feature of the depth window element after grouting is defined as follows:

[0159]

[0160] in, This is a preset anomaly threshold. This feature is used to characterize the variation in the extent of local anomalies after grouting.

[0161] The morphological characteristics of the depth window element after grouting are defined as follows:

[0162]

[0163] This feature is used to characterize the structural morphology and waveform distribution pattern of local anomalies within the window after grouting.

[0164] The continuity characteristic of the depth window element after grouting is defined as follows:

[0165]

[0166] This feature is used to characterize the degree of continuity of anomalies along the depth direction between the window and its neighboring windows after grouting.

[0167] Measured feature vectors after grouting Then, it was compared with the ungrouted reference response constructed in S4. Comparisons are made, and various deviations are extracted. These deviations reflect the degree of change in the actual state after grouting relative to the expected state before grouting. For any depth window unit... Its energy deviation Defined as:

[0168]

[0169] In the formula, To prevent tiny positive numbers with a denominator of zero, this quantity represents the relative deviation of the energy characteristics after grouting from the reference energy characteristics before grouting.

[0170] Width deviation Defined as:

[0171]

[0172] This quantity represents the degree of change in the area of ​​abnormal influence after grouting relative to the ungrouted reference area.

[0173] Morphological deviation Defined as:

[0174]

[0175] in, This is a similarity function. The more similar the grouted shape is to the ungrouted reference shape, the better. The smaller the value, the better; conversely, if the local shape after grouting deviates significantly from the shape it should have been without grouting, then... The larger.

[0176] Continuity deviation Defined as:

[0177]

[0178] This quantity represents the degree of change in the continuity of the anomaly in the depth direction after grouting relative to the ungrouted reference state.

[0179] After obtaining the above deviations, the deviations are weighted and combined to obtain the grouting amount of the target depth window element. :

[0180]

[0181] in, Let be the weighting coefficient, and satisfy:

[0182]

[0183] The weighting coefficients can be set according to the sensitivity of various features to grouting effects under different engineering scenarios. If the engineering object places greater emphasis on abnormal energy changes, the weighting coefficients can be increased. If more attention is paid to the morphological reconstruction or continuous changes of abnormal zones, the accuracy can be improved accordingly. or .

[0184] In a preferred embodiment, to avoid the amplification effect of local extreme values ​​on the grouting effect, various deviations can be normalized and constrained before being weighted and combined. A reference interval judgment mechanism can also be introduced: if the measured characteristics after grouting still fall within the ungrouted reference interval, the corresponding deviation is set to a low value or zero; if it significantly exceeds the reference interval, the deviation is calculated according to the degree of exceedance. This further enhances the sensitivity of the grouting effect to the actual grouting effect and reduces erroneous responses to normal measurement fluctuations.

[0185] The core result output in step S52 is the grouting amount for each depth window element. This quantity has clear physical and engineering meanings: the larger the value, the more obvious the deviation of the depth window unit from the expected state before grouting, and the more likely it is that grouting has substantially modified the surrounding medium; the smaller the value, the more likely it is that the state of the depth window unit after grouting is still close to the ungrouted reference, and the more likely it is that the grouting effect is not obvious or has not been effectively intervened.

[0186] Step S53 quantifies the degree to which the original disease characteristics are still retained after grouting, and is another core calculation line that distinguishes this invention from ordinary before-and-after difference evaluation methods. If the grouting effect in S5 answers "how much has changed compared to the ungrouted state after grouting", then S6 answers "how much of the original disease remains after grouting". In engineering evaluation, both questions are indispensable. Even if a section undergoes significant changes after grouting, it does not mean that the original fracture zone, void, or seepage channel has been truly eliminated; on the contrary, in many cases, although grouting has intervened and caused a response shift, the original disease characteristics may still be retained to a high degree. Such sections are often the partially effective area, the residual disease area, or the suspected residual seepage channel area. Therefore, the role of this step is to transform "whether the original disease still exists" into a directly calculable quantitative result.

[0187] Specifically, using the pre-grouting defect template established in S3 For reference, the measured feature vector after grouting extracted from S51 is used. For comparison objects, for any depth window unit The retention rate of the measured state after grouting relative to the original state before grouting is calculated. This retention rate is not used to characterize the magnitude of change, but rather to characterize whether the original disease characteristics are still preserved. Therefore, a higher retention rate indicates that the unit remains closer to the original disease template after grouting, and the original disease is more likely not to have been effectively eliminated; a lower retention rate indicates that the similarity between the unit and the original disease state after grouting is lower, and the original disease is more likely to have been significantly weakened or reconstructed.

[0188] For any depth window unit First, the energy retention is calculated. If the abnormal energy of a unit remains close before and after grouting, it indicates that the original abnormal strength has been largely retained; conversely, if the abnormal energy decreases significantly after grouting, it indicates that the original disease has been weakened in terms of strength. Therefore, the energy retention is defined. for:

[0189]

[0190] in, To prevent tiny positive numbers with a denominator of zero, the closer this value is to 1, the closer the abnormal energy after grouting is to the energy of the disease before grouting, and the more likely the original disease intensity is to be preserved; the closer its value is to 0, the more the abnormal energy after grouting has deviated significantly from the original disease energy.

[0191] Secondly, calculate the width retention amount. If the effective expansion range of the abnormal zone after grouting is still similar to that of the damaged template before grouting, it indicates that the original abnormal influence range may still exist; if the abnormal width shrinks significantly or the structure is reconstructed, it indicates that the damaged area has been affected by the grouting. Therefore, define the width retention amount. for:

[0192]

[0193] This quantity reflects the degree to which the abnormal width after grouting is retained relative to the width of the template before grouting.

[0194] Next, calculate the morphological retention amount. Compared to strength and width, morphological features are more revealing of whether the structural outline and waveform pattern of local anomalies have been continued. In many cases, although the intensity of the anomaly changes after grouting, if the outline and morphology of the local anomaly are still highly similar to those before grouting, it indicates that the original defect may still exist structurally. Therefore, a morphological retention amount is defined. for:

[0195]

[0196] in, The similarity function can be cosine similarity, correlation coefficient, or other functions suitable for local morphological matching. The larger this value, the closer the local abnormal morphology after grouting is to the morphology before grouting, and the more obvious the original lesion is structurally preserved.

[0197] Furthermore, the continuity retention amount is calculated. Continuity characteristics reflect the continuity of anomalies along the depth direction, and are particularly important for determining whether fracture zones and seepage channels remain coherent. If, after grouting, the similarity relationship between a certain depth window element and its neighboring windows remains essentially the same as before grouting, it indicates that the element's position and connectivity within the depth segment structure are still preserved. Therefore, the continuity retention amount is defined. for:

[0198]

[0199] This quantity is used to characterize the degree to which the original disease features along the depth direction are preserved after grouting.

[0200] After obtaining the above four types of retention, they are weighted and combined to construct the disease residue of the target depth window unit, that is, the disease retention rate. :

[0201]

[0202] in, Let be the weighting coefficients corresponding to each retention amount, and satisfy the following:

[0203]

[0204] The weights can be set according to the importance of various disease characteristics in different engineering scenarios. For example, in scenarios where the continuity of seepage channels is of greater importance, the weight of continuity retention can be appropriately increased; in scenarios where the stability of voids and local broken zones is of greater importance, the weight of morphological retention can be increased.

[0205] In a preferred embodiment, to avoid misleading the residual amount of disease due to abnormal fluctuations in a single feature, each retention amount can be restricted or smoothed. For example, when a single feature is significantly abnormal due to local noise, it can be constrained in conjunction with the remaining retention amounts; alternatively, a neighborhood consistency condition can be introduced to jointly correct the retention trend of multiple consecutive depth windows, thereby improving the stability and reliability of disease retention.

[0206] It is particularly important to note the amount of residual disease. Significance and Grouting Amount They are completely different. Grouting effect measures how much has changed after grouting relative to the ungrouted state, while residual disease measures how much remains after grouting relative to the original disease state. The two are not interchangeable; it is precisely because they characterize the state of the same depth window unit from the two directions of modification degree and retention degree, respectively, that this invention can be further elevated from simple change detection to change attribution evaluation.

[0207] S6, based on the grouting action amount and the residual amount of disease, the change attribution judgment is carried out to obtain the initial attribution result;

[0208] Step S6 is used to apply the grouting action amount calculated above. and residual amount of disease Transforming the results into evaluations with clear engineering implications is a crucial step in this invention, moving beyond simply quantifying changes to interpreting them. If S52 and S53 answer the questions of "how much has changed after grouting" and "how much of the original defect remains," then S6 addresses whether the changes exhibited by this depth window unit constitute effective grouting, partial effectiveness, residual defect, or non-target disturbance. In other words, this step no longer focuses on comparing the magnitude of a single indicator but uses a combined dual-indicator approach to attribute the changes in the images before and after grouting to engineering causes.

[0209] In traditional grouting effect evaluation, a significant difference between before and after images is usually interpreted as indicating effective grouting. However, the same difference may correspond to completely different engineering conditions. For example, a significant deviation in the response after grouting at a certain depth section from the pre-grouting state may mean that the grouting has effectively modified the surrounding medium, or it may simply mean that the grout has entered but the original defects are still largely preserved. Conversely, insignificant changes in a section may indicate that the grouting is essentially ineffective, or it may simply mean that there is only minor local background disturbance. Because simple differences lack explanatory power, this invention establishes a set of change attribution rules by combining the amount of grouting action and the amount of residual defects, enabling the same "change" to be classified into different engineering categories.

[0210] Specifically, for any depth window unit The amount of grouting action has been obtained in the aforementioned steps. and residual amount of disease .in, This is used to characterize the degree of deviation of the depth window from the ungrouted reference state after grouting; This depth window is used to characterize the degree to which the original disease state before grouting is preserved after grouting. To transform these two factors into directly interpretable engineering conclusions, step S6 first sets a grouting action threshold. Threshold for distinguishing between residual disease amount The threshold can be determined based on the specific engineering scenario, historical sample statistics, expert experience calibration results, or preset evaluation level requirements.

[0211] Based on this, for any depth window unit The initial identification is carried out according to the following attribution identification model.

[0212] (1) When the grouting action amount Greater than or equal to the grouting action threshold And the residual amount of disease Less than the residual disease threshold If the depth window unit has significantly deviated from the ungrouted state after grouting and retains fewer original disease characteristics, it is considered an effective grouting area.

[0213] The typical characteristics of this type of section are: grouting has significantly altered the state of the surrounding medium, and the abnormal features of the original fractured zone, voids, or seepage channels have been significantly weakened or essentially disappeared. Therefore, it can be considered that grouting has achieved a good modification effect in this section. The criteria for this determination can be expressed as:

[0214]

[0215] (2) When the grouting action volume Greater than or equal to the grouting action threshold And the residual amount of disease It is also greater than or equal to the residual disease threshold. If the depth window unit has undergone significant changes after grouting compared to the ungrouted state, but the original disease characteristics are still largely preserved, it is determined to be a partially effective area.

[0216] The typical situation for this type of section is as follows: grouting has been introduced, and the local response has been altered, but the original defects have not been completely eliminated. Residual fracture zones, incompletely filled voids, or seepage channels whose connectivity has not been completely severed may still exist. The criteria for this can be expressed as:

[0217]

[0218] (3) When the grouting action volume Less than the grouting action threshold And the residual amount of disease It is also less than the residual disease threshold. This indicates that the depth window unit neither shows obvious deviation from the grouting process nor retains obvious original disease characteristics. Such sections are then classified as non-target disturbance areas or ordinary background areas.

[0219] These sections typically do not constitute effective grouting zones or areas of significant residual defects, and can be considered low-concern areas in subsequent engineering evaluations. The criteria for this distinction can be expressed as:

[0220]

[0221] (4) When the grouting action amount Less than the grouting action threshold And the residual amount of disease Greater than or equal to the residual disease threshold If the depth window unit does not deviate significantly from the ungrouted state after grouting, and the original disease characteristics are still obvious, it is determined to be a residual disease area. Under specific engineering conditions, it can also be further identified as a suspected residual seepage channel area.

[0222] These sections are often key risk areas where grouting is insufficient, grouting is difficult to penetrate, or the damaged structure remains connected. They are also the focus of subsequent re-grouting and remedial treatment. The criteria for this determination can be expressed as:

[0223]

[0224] By combining the above two indicators for discrimination, this invention transforms image changes that could only express "difference before and after" into classification results with clear engineering attributes, thereby achieving a leap from change detection to change attribution. For ease of algorithmic expression, the identification result of each depth window unit can be recorded as a category label. ,Right now:

[0225]

[0226] in, This represents an attribution discriminant function based on the amount of grouting applied and the amount of residual disease. The function outputs category labels. It includes at least the following categories: effective grouting zone, partially effective zone, residual disease zone, suspected residual seepage channel zone, and non-target disturbance zone.

[0227] In a preferred embodiment, this step can also introduce a more refined classification mechanism. For example, based on the amount of grouting, the effective grouting area can be further subdivided into high-efficiency grouting areas and general effective grouting areas; similarly, based on the amount of residual disease, the residual disease area can be further subdivided into medium-residual areas and strong-residual areas. For suspected residual seepage channels, a higher priority risk identification can be applied based on continuity characteristics or the degree of segment connectivity. While this level of refinement does not change the dual-index attribution logic of the present invention, it enhances the hierarchical nature of the results and its engineering guidance value.

[0228] Furthermore, in a more robust implementation, this step can also employ a soft-discrimination method instead of a single hard threshold segmentation. That is, the grouting action amount... and residual amount of disease The input is fed into a preset discrimination function, which outputs the membership degree or confidence score for each category. The final category is then determined based on the principle of maximum membership. The advantage of this approach is that when a depth window cell happens to be near the threshold, it can avoid abrupt category changes due to small fluctuations, thus improving the continuity and stability of the identification results.

[0229] S7. Based on the initial attribution results, perform section-level continuity optimization to obtain grouting effect evaluation results. The grouting effect evaluation results include the distribution map of the grouting effect section around the hole, the distribution map of the residual disease section, the suspected residual seepage channel section, and the recommended supplementary grouting section.

[0230] Step S7 further integrates the initial attribution results of a single depth window unit obtained in S6 into a section-level evaluation result with engineering usability, ultimately forming the output result of the peri-hole grouting effect. Since the initial attribution results in S6 are obtained based on depth window units, although they can distinguish different engineering states such as effective grouting, partially effective grouting, residual defects, and suspected residual seepage channels, in practical applications, the category of a single depth window unit may still be affected by local noise, fluctuations in individual anomalies, or unstable boundary positions, manifesting as scattered category jumps, short isolated anomalies, or local discontinuities. Directly outputting this result as an engineering conclusion can easily lead to fragmented interpretation, which is detrimental to subsequent re-grouting and restoration decisions. Therefore, it is necessary to introduce depth continuity constraints based on the unit-level attribution results to integrate and optimize adjacent results, transforming the evaluation result from discrete units into continuous sections.

[0231] Specifically, let each depth window unit output by S6... The category label is The category labels include at least the effective grouting area, partially effective area, residual disease area, suspected residual seepage channel area, and non-target disturbance area. For the category sequence arranged along the depth direction... First, the consistency and continuity of category among adjacent depth window units are analyzed. If multiple adjacent depth window units have the same category and are continuously distributed in depth, they are considered as components of the same engineering segment. If the category of a certain depth window unit is significantly inconsistent with the adjacent units above and below, and its segment length is too short and lacks neighborhood support, it is preferentially determined to be a locally unstable result, and it is smoothed or merged into the adjacent main segment.

[0232] To achieve the above integration process, this invention introduces deep continuity optimization into the initial category sequence. Preferably, the segment optimization process is represented as an optimization problem with category jump constraints. Let the optimized category sequence be... Then the following objective function can be constructed:

[0233] ;

[0234] in, Represents depth window unit Assigned category The matching cost at that time, which is determined by the amount of grouting action. Residual amount of disease Determine the degree of consistency with the corresponding category; This is an indicator function; it takes a value of 1 when two adjacent depth window cells are of different categories, and a value of 0 otherwise. is the category jump penalty coefficient, used to control the smoothness of the category sequence; m is the total number of depth window units obtained along the depth direction of the detection hole; i is the index of the depth window unit, which takes values ​​of 1, 2, ..., m in the depth direction. Let i be the category of the i-th depth window unit. The category of the (i+1)th depth window unit.

[0235] By minimizing this objective function, unreasonable high-frequency category jumps can be suppressed while maintaining the engineering significance of the unit-level identification results, so that similar depth window units can form continuous segments as much as possible.

[0236] In a simplified implementation, instead of explicitly constructing an optimization function, a regular continuous integration method can be used. Specifically, this may include the following processing: directly merging adjacent depth window units of the same category; merging isolated category segments whose length is less than a preset minimum segment length based on their upper and lower neighboring main categories; and considering short, dissimilar segments sandwiched between two large segments of the same category as boundary transition zones and merged into adjacent main segments if their grouting volume and residual disease volume do not exhibit significant independence. Through this regular integration, the discrete unit results can also be transformed into a more stable segment representation.

[0237] After segment-level continuity optimization, the final set of grouting effect segments can be formed. For any continuous segment... The starting depth, ending depth, segment category, and comprehensive evaluation value of the segment are given. The segment category may include: effective grouting segment, partially effective segment, residual disease segment, suspected residual seepage channel segment, and non-target disturbance segment. The comprehensive evaluation value of the segment can be further given by the average grouting action, average residual disease amount, maximum value, or weighted statistical results of each depth window unit within the segment, to enhance the expressive power of the segment-level results. For example, for the first... Section The comprehensive value of its grouting effect It can be defined as:

[0238]

[0239] Its disease residual value It can be defined as:

[0240]

[0241] in, Indicates a section The number of depth window units included. The comprehensive value of a section can be used to further determine whether it is a high-efficiency grouting section, a generally effective section, a medium-residual section, or a strong-residual section.

[0242] After the sectional results are generated, this step further outputs the grouting effect evaluation results. The evaluation results include at least the following: First, an output of a grouting effect sectional distribution map along the depth direction, used to visually display the depth location distribution of effective grouting areas, partially effective areas, and residual disease areas; second, an output of a disease residual sectional distribution map, used to characterize the main locations where the original disease remains after grouting; third, an output of suspected residual seepage channel sections, used to identify high-risk sections that require subsequent close monitoring; and fourth, an output of recommended re-grouting sections. The recommended re-grouting sections are preferably selected based on sections with "low grouting effect and high residual disease amount," that is, continuous sections with insufficient grouting effect but still significantly existing original disease are prioritized for re-grouting recommendations.

[0243] Furthermore, in a preferred embodiment, a re-irrigation priority can be given for the recommended re-irrigation sections. The priority can be based on the comprehensive residual disease value of the section. The criteria for determining the priority of irrigation are comprehensively considered, including continuous length, continuity characteristics, and segment category. For example, a suspected residual seepage channel segment that simultaneously meets the criteria of "high residual disease content, long segment length, and high continuity" can be designated as a high-priority irrigation segment; a portion of the effective segment with moderate residual disease content and relatively short segment length can be designated as a medium-priority irrigation segment.

[0244] This invention has the following innovative features:

[0245] 1. Innovation Point 1: A reference response construction mechanism for ungrouted areas was proposed.

[0246] Current methods for evaluating the effectiveness of radar grouting in wells typically use pre-grouting images as the sole reference, then compare them with post-grouting images. This approach assumes that the measured values ​​before grouting are the absolute reference for the subsequent non-grouting state, ignoring non-target differences introduced by background fluctuations, changes in coupling conditions, and system offsets between the two time-phase measurements. Consequently, it is easy to misjudge non-grouting factors as grouting effects.

[0247] This invention proposes for the first time a "non-grouting reference response" construction mechanism. Instead of simply using the measured response before grouting for comparison, it constructs a reference response or reference interval for the target depth window unit under the condition of "no grouting but in the current re-measurement environment" based on a pre-grouting defect template and considering the overall offset pattern of stable sections and local background changes. This mechanism provides a clear normative benchmark for subsequent grouting effect evaluation, fundamentally differentiating it from ordinary direct difference methods between before and after images.

[0248] 2. Innovation Point Two: A dual-channel quantitative evaluation model for grouting effect and residual disease amount was proposed.

[0249] Most existing methods only judge the magnitude of the difference before and after grouting, essentially answering only whether a change has occurred, but not what the change means. What truly needs to be distinguished in engineering is whether, although a certain section has changed, the grouting has effectively improved the medium, or whether the original defects have been largely preserved despite the intervention of grouting.

[0250] To address this, this invention establishes a dual-channel quantitative evaluation model: one channel constructs the grouting action quantity based on the deviation of the measured response after grouting from the ungrouted reference response, used to evaluate whether grouting truly altered the state of the perforated medium; the other channel constructs the residual quantity of the defect based on the degree of retention of the measured response after grouting relative to the defect template before grouting, used to evaluate whether the original fracture zone, voids, or seepage channels still exist. This dual-channel model enables the evaluation of grouting effects to simultaneously assess both the degree of modification and the degree of residual defects, thereby significantly improving the algorithm's engineering interpretability.

[0251] 3. Innovation Point Three: It proposes an evaluation logic based on change attribution rather than change detection.

[0252] The core of traditional radar grouting evaluation methods in wells is to detect differences, while the core of this invention is to interpret those differences. In other words, this invention does not stop at the superficial fact that images differ before and after grouting, but further determines which engineering attribute this difference belongs to: whether it is a change in effective grouting, a change in partially effective grouting, a change in residual defects, or a change in non-target disturbances.

[0253] Specifically, this invention, through joint analysis of grouting action and residual defects, attributes changes in each depth window unit to different engineering categories, further forming section-level evaluation results. Thus, changes in the wellbore radar image are no longer merely differences, but are transformed into results with clear engineering implications, such as effective grouting sections, partially effective sections, residual defect sections, and suspected residual seepage channels. This change attribution logic represents a substantial improvement over existing methods that rely solely on simple difference analysis.

[0254] 4. Innovation Point Four: A continuous optimization output mechanism from depth window unit to segment-level results is proposed.

[0255] Existing image comparison results typically present as scattered outliers or localized difference zones, making it difficult to directly translate them into grouting effect sections usable in engineering. Therefore, this invention uses depth window units as the basic analysis object. After feature extraction and change attribution are completed at the unit level, depth continuity constraints are introduced to segment and integrate adjacent similar results, eliminating short, isolated outliers and correcting boundary fluctuations. The final output is a continuous, stable, and directly usable segment-level result for engineering decision-making.

[0256] The innovative significance of this mechanism lies in its advancement from image analysis algorithms to engineering evaluation methods. The output is no longer merely a matter of local feature differences or single-point identification, but can directly provide engineering results such as effective grouting sections, residual defect sections, and recommended re-grouting areas, thereby truly meeting the practical needs of grouting quality inspection around dam foundations and anti-seepage walls.

[0257] 5. Innovation Point Five: A framework for evaluating the grouting effect within 1–2 m around the borehole was established.

[0258] This invention is not applied to general underground anomaly detection, but specifically addresses the engineering challenge of evaluating the effectiveness of grouting in the perimeter area, which is not covered by borehole television. To achieve this goal, this invention forms a complete algorithm chain: dual-temporal data acquisition, unified processing, baseline defect template establishment, construction of ungrouted reference response, calculation of grouting action, calculation of residual defect amount, attribution of changes, and section-level continuity optimization. This algorithm framework is not simply pieced together from existing in-well radar interpretation methods, but is designed around the core questions of "whether the perimeter grouting effect is truly effective, whether the original defects still remain, and which sections need re-grouting," thus possessing clear specialization, systematicity, and engineering relevance.

[0259] Example 2

[0260] like Figure 3 As shown, the difference between this embodiment and Embodiment 1 is that this embodiment provides a well-drilled radar peri-hole grouting effect evaluation system based on variation attribution. This system corresponds one-to-one with the well-drilled radar peri-hole grouting effect evaluation method based on variation attribution in Embodiment 1. The system includes:

[0261] The acquisition unit is used to acquire pre-grouting and post-grouting data for the same detection well. The pre-grouting data is the in-well radar B-Scan depth profile data before grouting, and the post-grouting data is the in-well radar B-Scan depth profile data after grouting.

[0262] The pre-grouting disease template establishment unit is used to extract the baseline disease characteristics before grouting and establish the pre-grouting disease template based on the pre-grouting data.

[0263] The ungrouted reference response construction unit is used to construct the ungrouted reference response based on the pre-grouting disease template and combined with the response law of the stable section and the change law of the adjacent background.

[0264] The dual-channel quantization calculation unit is used to extract the measured response characteristics after grouting based on the data after grouting; and to calculate the grouting effect and residual amount of disease based on the measured response characteristics after grouting, the reference response before grouting, and the disease template before grouting, using dual-channel quantization.

[0265] The attribution identification unit is used to identify the causes of changes based on the amount of grouting action and the amount of residual disease to obtain the initial attribution results.

[0266] The section continuity optimization unit is used to perform section-level continuity optimization based on the initial attribution results to obtain grouting effect evaluation results.

[0267] As a further implementation, the dual-channel quantization calculation unit includes:

[0268] The measured response feature extraction subunit is used to extract the measured response features after grouting from the depth window unit based on the post-grouting data.

[0269] The first channel quantization calculation subunit is used to compare the measured response characteristics after grouting with the reference response before grouting, and calculate the deviation of each measured response characteristic after grouting. The deviation includes energy deviation, width deviation, morphological deviation and continuity deviation. The deviation is then weighted and combined to obtain the grouting effect of the target depth window unit. The grouting effect represents the degree of modification after grouting relative to the expected state before grouting.

[0270] The second channel quantization calculation sub-unit is used to compare the measured response characteristics after grouting with the disease template before grouting, and calculate the retention amount of each measured response characteristic after grouting. The retention amount includes energy retention, width retention, morphology retention and continuity retention. The retention amount is weighted and combined to obtain the disease residue amount of the target depth window unit. The disease residue amount represents the degree of preservation of the original disease characteristics after grouting.

[0271] The execution process of each unit can be carried out according to the process flow of the well radar peri-hole grouting effect evaluation method based on change attribution in Example 1, and will not be described in detail in this example.

[0272] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0273] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0274] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0275] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0276] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for evaluating the effect of wellbore radar peri-hole grouting based on variation attribution, characterized in that, The method includes: Data before and after grouting of the same detection well are obtained. The data before grouting is the B-Scan depth profile data of the radar in the well before grouting, and the data after grouting is the B-Scan depth profile data of the radar in the well after grouting. Based on the pre-grouting data, baseline disease characteristics were extracted and a pre-grouting disease template was established; based on the pre-grouting disease template, and combined with the response patterns of stable sections and the changes in adjacent background, a non-grouting reference response was constructed. Based on the post-grouting data, the measured response characteristics after grouting are extracted; and based on the measured response characteristics after grouting, the ungrouted reference response, and the pre-grouting disease template, the grouting effect and disease residue are calculated using dual-channel quantization. Based on the grouting action amount and the residual amount of disease, the change attribution judgment is performed to obtain the initial attribution result; based on the initial attribution result, section-level continuous optimization is performed to obtain the grouting effect evaluation result.

2. The method for evaluating the effect of wellbore radar peri-hole grouting based on variation attribution as described in claim 1, characterized in that, The method also includes: The data before and after grouting are standardized, and the standardization process includes: The depth coordinates of the pre-grouting and post-grouting data are registered, the response amplitudes of the pre-grouting and post-grouting data are scaled, the background components of the pre-grouting and post-grouting data are suppressed, and the pre-grouting and post-grouting data are standardized to form a dual-temporal response pair before and after grouting.

3. The method for evaluating the effect of wellbore radar peri-hole grouting based on variation attribution as described in claim 1, characterized in that, Based on the pre-grouting data, baseline disease characteristics were extracted and a pre-grouting disease template was established, including: Based on the pre-grouting data, baseline disease features before grouting are extracted by depth window unit; the pre-grouting baseline disease features are the abnormal responses corresponding to the original diseases around the holes, including energy features, width features, morphological features and continuity features. Based on the baseline defect characteristics before grouting, a baseline defect feature vector for the depth window unit before grouting is constructed. Based on the combination relationship of each feature in the baseline disease feature vector, abnormal candidate segments are identified; and a pre-grouting disease template is established for the depth window unit within the abnormal candidate segments.

4. The method for evaluating the effect of wellbore radar peri-hole grouting based on variation attribution as described in claim 1, characterized in that, Based on the pre-grouting defect template, and combined with the response patterns of stable sections and the changes in adjacent background, a reference response without grouting is constructed, including: Based on the response pattern of stable sections, the correction coefficients for baseline defects before grouting are calculated based on the set of stable sections. Based on the correction coefficient and the pre-grouting defect template, a preliminary ungrouted reference response is constructed for the target depth window unit; The preliminary ungrouted reference response is locally modified based on the variation pattern of the adjacent background to obtain the ungrouted reference response.

5. The method for evaluating the effect of wellbore radar peri-hole grouting based on variation attribution as described in claim 1, characterized in that, Based on the post-grouting data, extract the measured response characteristics after grouting; Based on the measured response characteristics after grouting, the reference response before grouting, and the disease template before grouting, the grouting effect and residual disease amount are calculated using dual-channel quantization, including: Based on the post-grouting data, the measured response features after grouting are extracted from the depth window unit; First-channel quantization calculation: Based on the comparison between the measured response characteristics after grouting and the reference response before grouting, the deviation of each measured response characteristic after grouting is calculated; the deviation includes energy deviation, width deviation, morphological deviation, and continuity deviation; and the deviations are weighted and combined to obtain the grouting effect of the target depth window unit; the grouting effect characterizes the degree of modification after grouting relative to the expected state before grouting; The second channel quantization calculation: compare the measured response features after grouting with the disease template before grouting, and calculate the retention amount of each measured response feature after grouting; the retention amount includes energy retention amount, width retention amount, morphology retention amount and continuity retention amount, and perform weighted combination of each retention amount to obtain the disease residue amount of the target depth window unit; the disease residue amount characterizes the degree of retention of the original disease features after grouting.

6. The method for evaluating the effect of wellbore radar peri-hole grouting based on variation attribution as described in claim 1, characterized in that, Based on the grouting effect and the residual amount of disease, a change attribution judgment is performed to obtain initial attribution results, including: Based on the relationship between the grouting effect and the residual amount of disease, an attribution is made to the change type of the target depth window unit based on an attribution identification model; the attribution identification model includes: When the grouting action amount is greater than or equal to the grouting action threshold, and the residual amount of disease is less than the residual amount of disease, it is determined to be an effective grouting zone; When the grouting action amount is greater than or equal to the grouting action threshold, and the residual amount of disease is also greater than or equal to the residual amount of disease, it is determined to be a partially effective area; When the grouting action amount is less than the grouting action threshold and the residual amount of disease is also less than the residual amount of disease, it is determined to be a non-target disturbance area or a normal background area. When the grouting action amount is less than the grouting action threshold, and the residual amount of disease is greater than or equal to the residual amount of disease threshold, it is determined to be a residual disease area or a suspected residual seepage channel area.

7. The method for evaluating the effect of wellbore radar peri-hole grouting based on variation attribution as described in claim 1, characterized in that, Based on the initial attribution results, segment-level continuity optimization is performed to obtain grouting effect evaluation results, including: The initial attribution results of each depth window unit are subjected to segment-level continuity optimization, adjacent depth window units of the same type are merged, and isolated category segments are corrected or removed to form the grouting effect evaluation results. The grouting effect evaluation results include a distribution map of the grouting effect around the hole, a distribution map of the remaining diseased sections, suspected residual seepage channels, and recommended supplementary grouting sections.

8. The method for evaluating the effect of wellbore radar peri-hole grouting based on variation attribution as described in claim 7, characterized in that, The objective function for the segment-level continuity optimization is: ； in, Represents depth window unit Assigned category The matching cost at that time, which is determined by the amount of grouting action. Residual amount of disease Determine the degree of consistency with the corresponding category; This is an indicator function; it takes a value of 1 when two adjacent depth window cells are of different categories, and a value of 0 otherwise. is the category jump penalty coefficient, used to control the smoothness of the category sequence; m is the total number of depth window units obtained along the depth direction of the detection hole; i is the index of the depth window unit, which takes values ​​of 1, 2, ..., m in the depth direction. Let i be the category of the i-th depth window unit. The category of the (i+1)th depth window unit.

9. A well-drilled radar peri-hole grouting effect evaluation system based on variation attribution, characterized in that, The system includes: The acquisition unit is used to acquire pre-grouting and post-grouting data for the same detection hole. The pre-grouting data is the in-well radar B-Scan depth profile data before grouting, and the post-grouting data is the in-well radar B-Scan depth profile data after grouting. The pre-grouting disease template establishment unit is used to extract the pre-grouting baseline disease characteristics and establish a pre-grouting disease template based on the pre-grouting data. The ungrouted reference response construction unit is used to construct an ungrouted reference response based on the pre-grouting disease template and in combination with the response law of the stable section and the change law of the adjacent background. The dual-channel quantization calculation unit is used to extract the measured response features after grouting based on the grouting data; and to calculate the grouting effect and residual amount of disease based on the measured response features after grouting, the ungrouted reference response, and the disease template before grouting, using dual-channel quantization. The attribution identification unit is used to perform change attribution identification based on the grouting amount and the residual amount of disease to obtain the initial attribution result; The section continuity optimization unit is used to perform section-level continuity optimization based on the initial attribution results to obtain grouting effect evaluation results.

10. The well-drilled radar peri-hole grouting effect evaluation system based on variation attribution according to claim 9, characterized in that, The dual-channel quantization calculation unit includes: The measured response feature extraction subunit is used to extract the measured response features after grouting from the depth window unit based on the grouting post-grouting data. The first channel quantization calculation subunit is used to compare the measured response characteristics after grouting with the reference response before grouting, and calculate the deviation of each measured response characteristic after grouting; the deviation includes energy deviation, width deviation, morphological deviation and continuity deviation; and the deviations are weighted and combined to obtain the grouting effect of the target depth window unit; the grouting effect represents the degree of modification after grouting relative to the expected state before grouting; The second channel quantization calculation subunit is used to compare the measured response features after grouting with the disease template before grouting, and calculate the retention amount of each measured response feature after grouting; the retention amount includes energy retention amount, width retention amount, morphology retention amount and continuity retention amount, and the retention amount is weighted and combined to obtain the disease residue amount of the target depth window unit; the disease residue amount represents the degree of retention of the original disease features after grouting.

Images