In-situ Spectral Observation Method for Sandstone Strata Based on Locally Washed Holes

Abstract

This application discloses a method for in-situ spectral observation of sandstone formations based on localized borehole washing: preliminary exploration and parameter setting; placement and positioning of detection equipment; localized borehole washing treatment of sandstone formations; borehole cleanliness detection and condition stabilization processing; high-precision in-situ spectral acquisition; spectral data preprocessing and interference correction; spectral feature extraction and formation property inversion; continuous multi-point detection and data aggregation. This application couples localized borehole washing with in-situ spectral observation, abandoning the traditional method of global borehole washing. The borehole washing parameters are precise and controllable with strong adaptability. A quantitative cleanliness assessment system ensures the reliability of the detection data. Spectral interference correction is used, resulting in higher data accuracy, high detection efficiency, and strong engineering practicality, achieving simultaneous and accurate detection of multiple parameters.

Description

Technical Field

[0001] This application belongs to the field of geological exploration and spectral detection technology, specifically involving an in-situ spectral observation method for sandstone strata based on localized well washing. Background Technology

[0002] Sandstone, as one of the most widely distributed rock types in sedimentary strata, is the core carrier of oil and gas reservoirs, groundwater occurrence, mineral storage, and surrounding rock for tunnels and underground engineering. Its physical properties, such as mineral composition, pore structure, cementation characteristics, water content, and integrity, directly determine reservoir productivity, formation stability, and engineering safety.

[0003] With its advantages of being non-destructive, rapid, high-precision, and capable of simultaneous detection of multiple parameters, spectral detection technology has become a core technical means for in-situ characterization of rock strata. Among them, borehole in-situ spectral observation technology can directly obtain data on the original occurrence state of deep underground strata. Compared with surface outcrop detection and core sampling laboratory detection, it effectively avoids detection errors caused by sampling disturbance, surface weathering, and stress release, and is widely used in geological exploration, energy development, geotechnical engineering monitoring and other fields.

[0004] Current mainstream in-situ borehole spectral observation techniques primarily rely on borehole probe-type spectral acquisition equipment, which is directly lowered to the target formation depth for spectral data acquisition. However, during the drilling process, the drill bit grinding generates a large amount of rock dust, and drilling mud continuously adheres to the borehole wall. Simultaneously, the sandstone formation itself has well-developed porosity and fractures, easily adsorbing mud impurities, suspended particles, and dust impurities, forming an unevenly thick impurity capping layer on the borehole wall surface. This capping layer completely obscures the original sandstone rock mass surface, altering the true spectral reflectance characteristics of the formation, leading to spectral peak shifts, baseline drift, and severe interference from extraneous peaks. This makes it impossible to accurately identify the core mineral components of sandstone, such as quartz, feldspar, and clay minerals, significantly reducing the accuracy of inversion parameters such as porosity, permeability, and cementation strength.

[0005] To address the interference from impurities on the borehole wall, current technologies generally employ a full-area flushing process, which involves thoroughly washing the entire borehole wall with high-pressure water and air jets before conducting spectral analysis. However, this process suffers from several unavoidable technical drawbacks in practical applications, particularly its poor suitability for sandstone formations. First, full-area flushing causes strong disturbance and damages the original structure of sandstone: Sandstone generally has the characteristics of well-developed pores, uneven cementation strength, and large differences in rock mass integrity. In particular, loose sandstone and weakly cemented sandstone will be eroded and destroyed by high-pressure full-area flushing, which will destroy the original pore structure of the borehole wall, expand the fine cracks, cause the surface sandstone to peel off and the rock mass to loosen, completely change the in-situ occurrence state of the strata, and cause the subsequent spectral detection data to fail to reflect the true physical properties of the strata, thus losing the core significance of in-situ detection.

[0006] Second, the accuracy of hole cleaning is uncontrollable, and the effect of impurity removal varies greatly: the whole-area hole cleaning cannot be used to clean a single observation point. There are problems such as over-rinsing in some areas and incomplete rinsing in the target observation area. The cleanliness of the hole wall is uneven, and residual impurities will continue to interfere with the spectral acquisition, resulting in poor repeatability and stability of the detection data.

[0007] Third, full-area flushing can cause sudden changes in formation water content: Sandstone formations are extremely sensitive to water. Full-area high-pressure flushing will significantly increase the water content of the borehole wall rock mass. The strong spectral absorption characteristics of water will mask the characteristic spectral information of sandstone minerals, resulting in a significant decrease in the accuracy of mineral identification and making it impossible to accurately distinguish the differences in physical properties between dry sandstone and water-bearing sandstone.

[0008] Fourth, the process is cumbersome and the detection efficiency is low: After the entire borehole is cleaned, it needs to be left to stand for a long time to drain and air dry, and spectral data can only be collected after the borehole wall condition has stabilized. The detection cycle is long, and it is impossible to achieve rapid in-situ detection, which makes it difficult to meet the engineering needs of batch exploration and real-time monitoring on site.

[0009] Fifth, existing spectral correction techniques have significant limitations: most existing techniques use pure data algorithms to correct spectral interference without taking into account the causes of impurities on the borehole wall, differences in borehole washing processes, and formation structural characteristics for targeted correction. They cannot effectively remove interference from trace impurities remaining after borehole washing or shallow structural disturbances, resulting in limited data correction accuracy.

[0010] Currently, existing technologies have been unable to simultaneously address the need for thorough removal of impurities from pore walls and complete preservation of the original formation structure. This has resulted in a long-standing difficulty in achieving breakthroughs in the accuracy of in-situ spectral detection of sandstone formations, hindering the technological development of fine geological exploration, evaluation of tight sandstone reservoirs, and precise assessment of surrounding rocks in underground engineering projects. Summary of the Invention

[0011] To address the problems of existing sandstone borehole spectral detection methods, such as borehole wall mud adhesion, rock powder coverage, fracture filling impurities interference, disturbance of the original formation structure by full-area borehole washing, low signal-to-noise ratio of spectral detection, poor accuracy of formation property parameter inversion, and inability to achieve in-situ true characterization, this application aims to provide a sandstone formation spectral in-situ observation method based on local borehole washing. This application couples local fixed-point borehole washing with in-situ spectral observation, abandons the traditional full-area borehole washing, and provides precise and controllable washing parameters with strong adaptability. It is equipped with a cleanliness quantification judgment system to ensure the reliability of detection data, adopts spectral interference correction, has higher data accuracy, high detection efficiency, and strong engineering practicality, and achieves simultaneous and accurate detection of multiple parameters.

[0012] The technical solution adopted in this application is: A method for in-situ spectroscopic observation of sandstone formations based on localized well washing includes the following steps: S1. Preliminary exploration and detection parameter setting: Conduct basic parameter surveys on boreholes in the sandstone strata to be tested, and obtain parameters such as borehole diameter, borehole depth, stratum lithology, cementation strength, pore development characteristics, and distribution of borehole wall impurity thickness; based on the survey results, preset local borehole washing parameters and spectral acquisition parameters, divide target observation points, and establish a layered and graded detection scheme; S2. Deployment and Positioning of Testing Equipment: Deploy the integrated testing equipment, which includes a local hole cleaning module, a spectrum acquisition module, an attitude calibration module, a high-definition imaging module, and a voltage-stabilized drying module, to the target observation depth of the borehole. Use the attitude calibration module to center and calibrate the equipment, lock the target observation point, and ensure that the hole cleaning area and the spectrum acquisition area completely overlap. S3. Localized Borehole Cleaning Treatment in Sandstone Formations: The localized borehole cleaning module is activated, and a low-pressure pulse cleaning process is used to perform targeted localized cleaning of a small area of ​​the borehole wall at the target observation point. This precisely removes the mud deposits, rock powder, and loose impurities from the observation area. During the cleaning process, the cleaning pressure, flow rate, and duration are monitored in real time, and the cleaning parameters are dynamically adjusted to avoid eroding and disturbing the original sandstone rock mass structure. S4. Hole Cleanliness Detection and Stabilization: After cleaning, high-definition imaging modules are used to acquire images of the borehole walls in the observation area. Combined with the preset sandstone borehole wall cleanliness grading standards, a quantitative cleanliness determination is completed. If the cleanliness meets the standards, the pressure-stabilizing air-drying module is activated to perform micro-drying, dehydration, and pressure stabilization treatment, restoring the rock mass in the observation area to its original stable state. If the cleanliness does not meet the standards, the hole cleaning parameters are fine-tuned for a second precise cleaning until the cleanliness meets the detection requirements. S5. In-situ high-precision spectral acquisition: After the formation state of the area to be observed stabilizes, the spectral acquisition module is activated. Adaptive parameter matching mode is used to acquire in-situ hyperspectral data of sandstone in the target area, covering the full-band spectral information of visible light, near-infrared and short-wave infrared. During the acquisition process, auxiliary parameters including equipment attitude, formation temperature, ambient humidity and drilling pressure are recorded simultaneously. S6. Spectral data preprocessing and interference correction: The acquired raw spectral data is preprocessed by noise reduction, baseline correction, and scattering correction. Combined with the characteristics of residual trace impurities in local washing holes and the characteristics of sandstone moisture interference, an adaptive interference correction algorithm is adopted to remove spectral interference caused by residual impurities, trace moisture, and equipment system errors, so as to obtain pure original sandstone spectral data. S7. Spectral Feature Extraction and Formation Property Inversion: Extract core feature parameters, including characteristic peak position, peak intensity, full width at half maximum (FWHM), and spectral reflectance, from the corrected pure spectral data. Combined with a pre-constructed sandstone spectral-physical property inversion model, identify physical property parameters, including sandstone mineral composition, porosity, permeability, cementation strength, water content, and rock mass integrity coefficient, to complete the in-situ state evaluation of the formation. S8. Multi-point continuous detection and data summary: After completing the single-point detection, move the integrated detection equipment to the next target observation point and repeat steps S2-S7 to complete the detection of the target area of ​​the entire borehole section. Summarize the spectral data and physical parameters of all points to form a comprehensive evaluation report of in-situ spectral detection of sandstone strata.

[0013] In step S1, the method for conducting basic parameter surveys of the borehole in the sandstone stratum to be tested is as follows: a borehole imaging instrument is used to acquire images of the entire borehole wall to identify the cementation type (e.g., calcareous cementation, argillaceous cementation, siliceous cementation), porosity development level, and fracture distribution characteristics of sandstone at different depths; a borehole diameter meter is used to measure the actual borehole diameter and record the borehole deviation data; the thickness of impurities adhering to the borehole wall is quantified through image grayscale analysis, and the impurity thickness is divided into three levels: a thin adhering layer of 0-0.5 mm, a medium adhering layer of 0.5-1.5 mm, and a thick adhering layer of more than 1.5 mm.

[0014] In step S1, the preset local flushing parameters include flushing pressure, pulse frequency, flushing flow rate, flushing range, and flushing duration. The parameters are configured differently for sandstone formations with different impurity thicknesses and different cementation strengths: low pressure and low frequency parameters are used for thin-adhesion layers and weakly cemented sandstone, while medium pressure and high frequency parameters are used for thick-adhesion layers and strongly cemented sandstone. High-pressure continuous flushing is strictly prohibited throughout the process.

[0015] In the integrated testing equipment: a data transmission and control system is electrically connected to each module; the local hole cleaning module adopts a ring-shaped fixed-point spray structure, and the single cleaning area is a ring-shaped local area with a diameter of 3-5cm, which is completely matched with the spectral acquisition field of view, ensuring that the hole wall outside the cleaning area is free from scouring and disturbance throughout the process.

[0016] In step S3, the low-pressure pulse cleaning process is as follows: the cleaning pressure is controlled at 0.15-0.4MPa, the pulse frequency is 5-15Hz, the duration of a single pulse cleaning is 0.2-0.5s, and the interval duration is 0.3-0.8s; the cleaning medium is filtered pure water to avoid introducing new impurities that contaminate the hole wall; for loose sandstone, the cleaning pressure does not exceed 0.2MPa to avoid the surface rock mass peeling off.

[0017] In step S4, the cleanliness of the sandstone pore wall is classified into four levels according to the grading criteria: Level 1: No visible impurities on the pore wall, the original texture and pore structure of the sandstone are clearly visible, and the image grayscale uniformity is ≥95%; Level 2: There are trace amounts of point-like residual impurities, but no flaky or layered impurities covering the surface, and the image grayscale uniformity is 90%-95%; Level 3: There are localized flaky impurities, and the grayscale uniformity is 80%-90%; Level 4: There is a continuous impurity covering layer, and the grayscale uniformity is <80%. Only Level 1 and Level 2 clean areas meet the standards and can be used for spectral acquisition.

[0018] In step S4, the micro-drying, dehydration and pressure stabilization treatment is carried out by adopting a low temperature and low pressure drying mode, with a drying temperature of 30-40℃, a relative pressure of 0.08-0.12MPa, and a drying time of 20-60s. This quickly removes free water from the surface of the pore wall, retains the original bound water in the sandstone pores, avoids water from interfering with spectral detection, and does not change the original water content characteristics of the rock mass.

[0019] In step S5, the spectral acquisition module, in adaptive parameter matching mode, automatically matches the acquisition band, integration time, and number of scans according to the sandstone stratum type; the acquisition band covers the entire visible-near infrared band of 400-2500nm, the integration time is 10-50ms, and a single point is repeatedly scanned 20-50 times, and the average value is taken as the original spectral data to improve data stability.

[0020] In step S6, the adaptive interference correction algorithm includes four major units: baseline drift correction, particle scattering correction, trace impurity feature removal, and moisture absorption peak correction. By constructing an impurity spectral feature database, it accurately matches and removes the characteristic peaks of residual trace rock powder and mud impurities from the hole washing process. The moisture spectrum fitting algorithm corrects the spectral absorption interference of shallow free water, thereby restoring the original spectral curve of sandstone to the greatest extent.

[0021] In step S7, the sandstone spectral-physical property inversion model is a machine learning model that has been pre-calibrated using a large number of sandstone samples. It takes spectral characteristic parameters as input and laboratory measured porosity, permeability, cementation strength, and mineral content as output. After sample training and accuracy verification, the model inversion error is ≤3%, which meets the requirements of high-precision engineering testing.

[0022] The beneficial effects of this application are: This application couples localized, targeted borehole cleaning with in-situ spectral observation, abandoning the traditional full-area borehole cleaning: it precisely cleans only the small target area required for spectral acquisition, thoroughly removing mud and rock powder impurities from the observation area, while completely preserving the original sandstone structure, stress state, and porosity characteristics of the remaining borehole areas. This fundamentally solves the problems of formation disturbance, fracture expansion, and structural damage caused by full-area borehole cleaning, truly achieving undisturbed in-situ detection, with detection data perfectly matching the actual formation state. The borehole cleaning parameters of this application are precise and controllable, with extremely high adaptability: suitable for different cementation strengths and different impurity adhesion thicknesses. For sandstone formations, a differentiated pulsed well cleaning parameter system is established, employing a low-pressure pulsed gentle cleaning mode. This ensures thorough removal of impurities while preventing surface flaking damage in loose and weakly cemented sandstone. It is suitable for all sandstone formation types, including dense, loose, water-bearing, and cemented sandstone, making it widely applicable. The accompanying quantitative cleanliness assessment system ensures the reliability of the test data: a well wall cleanliness grading standard is established to achieve quantitative evaluation of the cleaning effect, eliminating subjective judgment errors. Spectral acquisition is only conducted after the cleanliness standard is met, avoiding spectral interference from residual impurities at the source and significantly improving spectral data. Signal-to-noise ratio, repeatability, and accuracy; This application employs spectral interference correction for higher data accuracy: Compared to traditional single data correction methods, this application combines local hole washing process characteristics, sandstone moisture sensitivity characteristics, and impurity residue characteristics to construct a multi-dimensional adaptive correction model. This model accurately eliminates interference from trace impurities, moisture, and systematic errors, effectively restoring the original spectral characteristics of sandstone. The accuracy of mineral identification and the precision of physical property parameter inversion are significantly improved, with the inversion error controlled within 3%. This application boasts high detection efficiency and strong engineering applicability: It eliminates the need for prolonged static drying after full-area hole washing, using local hole washing + rapid voltage stabilization. Air drying can quickly stabilize the formation, shortening the single-point detection cycle by more than 60%. It enables continuous and rapid multi-point detection across the entire borehole depth, adapting to the engineering needs of batch field exploration and real-time on-site monitoring. It is easy to operate, the equipment is portable, and the construction cost is low. This application achieves simultaneous and accurate detection of multiple parameters: through spectral acquisition and inversion models, multiple core physical property parameters such as sandstone mineral composition, porosity, permeability, cementation strength, water content, and rock mass integrity can be acquired simultaneously, achieving comprehensive evaluation of multiple indicators in a single process. Compared with traditional single-detection technologies, the detection dimensions are more comprehensive and the evaluation results are more scientific. Therefore, this application is applicable to in-situ spectral observation of sandstone formations with different burial depths, cementation strengths, and porosities, possessing high engineering application value. It is suitable for in-situ spectral characteristic detection, mineral composition identification, physical property parameter inversion, and formation condition evaluation of various sandstone rock masses, including oil and gas reservoir sandstone, hydraulic sandstone formations, and mining sandstone surrounding rocks. Attached Figure Description

[0023] Figure 1 This is the overall process flow diagram of this application. Detailed Implementation

[0024] The present invention will now be described in further detail and in complete detail with reference to specific embodiments. These embodiments are implemented based on the technical solutions of this application, providing detailed implementation steps and operating parameters; however, the scope of protection of this application is not limited to the following embodiments.

[0025] like Figure 1 The diagram shown is an overall process flow chart of this application. This method can be applied to a variety of situations. Two embodiments are given below.

[0026] Example 1: In-situ spectral observation of boreholes in loose sandstone formations This embodiment focuses on in-situ spectral observation of boreholes in shallow loose sandstone reservoirs of an oil and gas field. The borehole depth is 0-300m and the diameter is 110mm. The target formation is weakly cemented loose sandstone with high porosity, low rock mass cementation strength, and severe rock powder adhesion on the borehole wall. It is extremely susceptible to water erosion and disturbance, making traditional full-area borehole washing unsuitable.

[0027] The specific implementation steps are as follows: Step 1, Preliminary Exploration and Parameter Setting: A borehole imaging instrument was used to perform a full borehole scan of the 0-300m section to identify the target sandstone strata as argillaceous, weakly cemented, loose sandstone with well-developed pores and abundant fractures; the impurities on the borehole wall were mainly thin and medium-thick layers of rock powder mixed with mud, with an impurity thickness of 0.3-1.2mm; the local cleaning parameters were set as follows: cleaning pressure 0.18MPa, pulse frequency 8Hz, single pulse duration 0.3s, interval duration 0.5s, single cleaning area diameter 4cm, and total cleaning time per point 15s; the preset spectral acquisition band was 400-2500nm, integration time 20ms, and 30 scans per point.

[0028] Step 2, Equipment Lowering and Positioning: Lower the integrated detection equipment at a constant speed to the target detection depth (three core observation points: 80m, 150m, and 220m) using a steel wire rope. Activate the attitude calibration module and use a three-axis tilt sensor to calibrate the equipment attitude in real time, controlling the equipment tilt angle to ≤0.1°. This completes the centering and positioning of the equipment, ensuring that the hole cleaning area and the spectral acquisition field of view are completely overlapped and locked.

[0029] Step 3, Localized Borehole Cleaning: The localized borehole cleaning module is activated, and a low-pressure pulse cleaning process is used to clean a 4cm diameter local area at each target point. Filtered pure water is used as the cleaning medium. During the cleaning process, the pressure sensor monitors the pressure fluctuation in real time and dynamically adjusts the flow rate to ensure that the cleaning pressure is stable at 0.18MPa. This thoroughly removes rock powder and mud impurities from the borehole wall in the observation area, without any scouring or disturbance to the surrounding borehole wall.

[0030] Step 4: Cleanliness Testing and State Stabilization: After cleaning, the high-definition imaging module was activated to acquire images of the borehole walls and perform grayscale uniformity analysis. The grayscale uniformity at the three observation points were 96.2%, 95.8%, and 97.1%, respectively, all meeting the Level 1 cleanliness standard, thus determining that the cleanliness met the requirements. Subsequently, the pressure-stabilized drying module was activated, setting the drying temperature to 35℃, the relative pressure to 0.1MPa, and the drying time to 30 seconds. This removed free moisture from the borehole wall surface while retaining the original pore bound water, allowing the formation state to stabilize rapidly.

[0031] Step 5: In-situ spectral data acquisition: After the formation condition stabilizes, start the spectral acquisition module, turn on shading and constant light source compensation, and acquire full-band spectral data according to preset parameters. Repeat the scan 30 times at a single point. The system automatically removes abnormal data and takes the average value as the original spectral data. Simultaneously record auxiliary parameters such as formation temperature, humidity, and borehole pressure at each point.

[0032] Step 6: Spectral Data Interference Correction: The original spectral data is preprocessed, including noise reduction, baseline correction, and scattering correction. A specific interference correction model for loose sandstone (i.e., an adaptive interference correction algorithm) is used to remove interference from trace amounts of residual rock powder peaks and correct for spectral absorption shifts in shallow free water, resulting in a clean, original spectral curve. After correction, the spectral curve has a stable baseline, no abnormal peaks, and clear, precisely positioned characteristic peaks of minerals such as quartz, feldspar, and kaolinite.

[0033] Step 7: Feature Extraction and Property Inversion: Extract core parameters such as spectral characteristic peak positions, peak intensities, and reflectance, and input them into a pre-trained spectroscopic-physical property inversion model for loose sandstone. This inversion yields parameters such as porosity, permeability, cementation strength, and mineral composition content at each point. Comparison with indoor core sample test data shows a porosity inversion error of 2.1%, a permeability inversion error of 2.5%, and a mineral identification accuracy of 98.7%, demonstrating significantly higher detection precision than traditional methods.

[0034] Step 8: Continuous multi-point detection: After completing the core point detection, the target point detection of the entire borehole section is completed in sequence. All data are summarized to generate an in-situ spectral detection and physical property evaluation report of the loose sandstone reservoir, accurately determining the porosity development characteristics and reservoir performance of the formation.

[0035] Example 2: In-situ spectral observation of boreholes in tight sandstone formations This embodiment focuses on in-situ spectral observation of a deep tight sandstone oil and gas reservoir. The borehole depth is 300-800m and the diameter is 90mm. The target formation is siliceous cemented tight sandstone with high rock integrity, strong cementation, and fine pores. There is a thick layer of mud impurities attached to the borehole wall, with an impurity thickness of 1.2-2.0mm. Traditional cleaning methods cannot completely remove impurities and cause serious spectral interference.

[0036] The specific implementation steps are as follows: Step 1: Preliminary Exploration and Parameter Setting: Through borehole imaging and borehole diameter detection, the target stratum was determined to be strongly cemented dense sandstone with a dense structure, few fractures, and strong resistance to erosion. The borehole wall impurities consisted of a thick layer of solidified drilling mud, making cleaning difficult. Differential borehole cleaning parameters were preset: cleaning pressure 0.35 MPa, pulse frequency 12 Hz, single pulse duration 0.4 s, interval duration 0.4 s, single cleaning area diameter 4 cm, total cleaning time per point 25 s; spectral acquisition integration time 30 ms, 40 scans per point.

[0037] Step 2, Equipment lowering, alignment and attitude calibration: The operation process is the same as in Example 1, ensuring that the equipment is vertically centered and that the acquisition field of view and the hole washing area are precisely aligned.

[0038] Step 3, Localized Targeted Hole Cleaning: The medium-pressure high-frequency pulse cleaning process is used to clean the target area in a targeted manner, specifically flushing away thick layers of solidified mud impurities. The pressure and flow rate are monitored in real time during the cleaning process to avoid excessive local pressure causing micro-cracks in the hole wall. While accurately removing surface impurities, the original dense structure of the dense sandstone is completely preserved.

[0039] Step 4, Cleanliness Testing and Stabilization: After cleaning, the grayscale uniformity of the borehole wall image is 94.5%, meeting the Class II cleanliness standard; start low-temperature air drying treatment, air drying time is 45 seconds, to thoroughly remove residual moisture and trace dust on the surface and ensure the stability of the formation.

[0040] Step 5, In-situ spectral data acquisition: The operation procedure is the same as in Example 1.

[0041] Step 6: Spectral data interference correction: Collect in-situ spectral data across the entire band, and use a dense sandstone-specific interference correction model (i.e., adaptive interference correction algorithm) to remove interference from thick impurities and trace moisture in deep strata. After correction, the resolution of spectral characteristic peaks is significantly improved, and trace mineral components such as quartz, calcite, and illite in dense sandstone can be accurately distinguished.

[0042] Step 7, Feature Extraction and Physical Property Inversion: The formation porosity, permeability, and cementation tightness parameters are calculated through the inversion model to accurately identify the physical property heterogeneity of tight sandstone reservoirs. The data has a high degree of agreement with the core measured data, and the inversion error is ≤2.8%, which can meet the needs of fine evaluation of deep tight reservoirs.

[0043] Step 8, Continuous multi-point detection: The operation procedure is the same as in Example 1.

[0044] In conclusion: This application couples localized fixed-point borehole washing with in-situ spectral observation, abandoning the traditional full-area borehole washing: only a small target area required for spectral acquisition is precisely cleaned to thoroughly remove mud and rock powder impurities from the observation area, while completely preserving the original sandstone structure, stress state, and porosity characteristics of the rest of the borehole area. This fundamentally solves the problems of formation disturbance, fracture expansion, and structural damage caused by full-area borehole washing, truly achieving undisturbed in-situ detection, and the detection data fully reflects the actual occurrence state of the formation.

[0045] The hole-washing parameters of this application are precise and controllable, and highly adaptable: a differentiated pulse hole-washing parameter system is established for sandstone formations with different cementation strengths and different impurity adhesion thicknesses. A low-pressure pulse gentle cleaning mode is adopted to ensure that impurities are thoroughly removed while preventing surface peeling damage to loose sandstone and weakly cemented sandstone. It is suitable for all sandstone formation types, including dense sandstone, loose sandstone, water-bearing sandstone, and cemented sandstone, and has a wide range of applications.

[0046] The accompanying cleanliness quantification judgment system of this application ensures the reliability of test data: it establishes a grading standard for the cleanliness of the well wall, realizes the quantitative evaluation of the cleaning effect, eliminates subjective judgment errors, and conducts spectral acquisition only after the cleanliness meets the standard, thereby avoiding spectral interference from residual impurities from the source and significantly improving the signal-to-noise ratio, repeatability and accuracy of spectral data.

[0047] This application employs spectral interference correction, resulting in higher data accuracy: Compared to traditional single data correction methods, this application combines local hole washing process characteristics, sandstone moisture sensitivity characteristics, and impurity residue characteristics to construct a multi-dimensional adaptive correction model. This model accurately eliminates interference caused by trace impurities, moisture, and systematic errors, effectively restoring the original spectral characteristics of sandstone. The accuracy of mineral identification and the precision of physical property parameter inversion are significantly improved, with the inversion error controlled within 3%.

[0048] This application boasts high testing efficiency and strong engineering practicality: it eliminates the need for prolonged static drying after full-area borehole washing; localized borehole washing combined with rapid pressure stabilization and drying can quickly stabilize the formation, shortening the single-point testing cycle by more than 60%. It enables continuous and rapid testing of multiple points across the entire borehole depth, meeting the engineering needs of batch field exploration and real-time on-site monitoring. It is easy to operate, the equipment is portable, and the construction cost is low.

[0049] This application achieves simultaneous and accurate detection of multiple parameters: through spectral acquisition and inversion models, multiple core physical property parameters such as sandstone mineral composition, porosity, permeability, cementation strength, water content, and rock mass integrity can be obtained simultaneously, realizing comprehensive evaluation of multiple indicators in a single process. Compared with traditional single detection technology, the detection dimensions are more comprehensive and the evaluation results are more scientific.

[0050] Therefore, this application is applicable to in-situ spectral observation of sandstone formations with different burial depths, different cementation strengths, and different porosities. It has high engineering application value and is suitable for in-situ spectral characteristic detection, mineral component identification, physical property parameter inversion, and formation condition evaluation of various sandstone rock bodies, such as oil and gas reservoir sandstone, hydraulic sandstone formations, and mining sandstone surrounding rocks.

[0051] The above description is merely a preferred embodiment of this application and is not intended to limit the scope of patent protection of this application. Any equivalent transformations, process modifications, parameter optimizations, scenario adaptations made within the scope of the description and drawings of this application, or the direct or indirect application of the technical solution of this application to other fields of radioactive mineral exploration, shall similarly fall within the protection scope of this invention.

Claims

1. A method for in-situ spectral observation of sandstone strata based on localized well washing, characterized in that, Includes the following steps: S1. Preliminary exploration and detection parameter setting: Conduct basic parameter surveys on boreholes in the sandstone strata to be tested, and obtain parameters such as borehole diameter, borehole depth, stratum lithology, cementation strength, pore development characteristics, and distribution of borehole wall impurity thickness; based on the survey results, preset local borehole washing parameters and spectral acquisition parameters, divide target observation points, and establish a layered and graded detection scheme; S2. Deployment and Positioning of Testing Equipment: Deploy the integrated testing equipment, which includes a local hole cleaning module, a spectrum acquisition module, an attitude calibration module, a high-definition imaging module, and a voltage-stabilized drying module, to the target observation depth of the borehole. Use the attitude calibration module to center and calibrate the equipment, lock the target observation point, and ensure that the hole cleaning area and the spectrum acquisition area completely overlap. S3. Localized Borehole Cleaning Treatment in Sandstone Formations: The localized borehole cleaning module is activated, and a low-pressure pulse cleaning process is used to perform targeted localized cleaning of a small area of ​​the borehole wall at the target observation point. This precisely removes the mud deposits, rock powder, and loose impurities from the observation area. During the cleaning process, the cleaning pressure, flow rate, and duration are monitored in real time, and the cleaning parameters are dynamically adjusted to avoid eroding and disturbing the original sandstone rock mass structure. S4. Hole Cleanliness Detection and Stabilization: After cleaning, high-definition imaging modules are used to acquire images of the borehole walls in the observation area. Combined with the preset sandstone borehole wall cleanliness grading standards, a quantitative cleanliness determination is completed. If the cleanliness meets the standards, the pressure-stabilizing air-drying module is activated to perform micro-drying, dehydration, and pressure stabilization treatment, restoring the rock mass in the observation area to its original stable state. If the cleanliness does not meet the standards, the hole cleaning parameters are fine-tuned for a second precise cleaning until the cleanliness meets the detection requirements. S5. In-situ high-precision spectral acquisition: After the formation state of the area to be observed stabilizes, the spectral acquisition module is activated. Adaptive parameter matching mode is used to acquire in-situ hyperspectral data of sandstone in the target area, covering the full-band spectral information of visible light, near-infrared and short-wave infrared. During the acquisition process, auxiliary parameters including equipment attitude, formation temperature, ambient humidity and drilling pressure are recorded simultaneously. S6. Spectral data preprocessing and interference correction: The acquired raw spectral data is preprocessed by noise reduction, baseline correction, and scattering correction. Combined with the characteristics of residual trace impurities in local washing holes and the characteristics of sandstone moisture interference, an adaptive interference correction algorithm is adopted to remove spectral interference caused by residual impurities, trace moisture, and equipment system errors, so as to obtain pure original sandstone spectral data. S7. Spectral Feature Extraction and Formation Property Inversion: Extract core feature parameters, including characteristic peak position, peak intensity, full width at half maximum (FWHM), and spectral reflectance, from the corrected pure spectral data. Combined with a pre-constructed sandstone spectral-physical property inversion model, identify physical property parameters, including sandstone mineral composition, porosity, permeability, cementation strength, water content, and rock mass integrity coefficient, to complete the in-situ state evaluation of the formation. S8. Multi-point continuous detection and data summary: After completing the single-point detection, move the integrated detection equipment to the next target observation point and repeat steps S2-S7 to complete the detection of the target area of ​​the entire borehole section. Summarize the spectral data and physical parameters of all points to form a comprehensive evaluation report of in-situ spectral detection of sandstone strata.

2. The in-situ spectroscopic observation method for sandstone strata based on localized well washing as described in claim 1, characterized in that, In step S1, the method for conducting basic parameter surveys of the borehole in the sandstone formation to be tested is as follows: a borehole imaging instrument is used to acquire images of the entire borehole wall to identify the cementation type, porosity development level, and fracture distribution characteristics of sandstone at different depths; The actual borehole diameter was measured using a borehole diameter tester, and the borehole deviation data was recorded. The thickness of impurities adhering to the borehole wall was quantified by image grayscale analysis, and the impurity thickness was divided into three levels: thin adhesion layer of 0-0.5mm, medium adhesion layer of 0.5-1.5mm, and thick adhesion layer of more than 1.5mm.

3. The in-situ spectroscopic observation method for sandstone strata based on localized well washing as described in claim 1, characterized in that: In step S1, the preset local flushing parameters include flushing pressure, pulse frequency, flushing flow rate, flushing range, and flushing duration. The parameters are configured differently for sandstone formations with different impurity thicknesses and different cementation strengths: low pressure and low frequency parameters are used for thin-adhesion layers and weakly cemented sandstone, while medium pressure and high frequency parameters are used for thick-adhesion layers and strongly cemented sandstone. High-pressure continuous flushing is strictly prohibited throughout the process.

4. The in-situ spectroscopic observation method for sandstone strata based on localized well washing as described in claim 1, characterized in that, In the integrated testing equipment: a data transmission and control system is electrically connected to each module; the local hole cleaning module adopts a ring-shaped fixed-point spray structure, and the single cleaning area is a ring-shaped local area with a diameter of 3-5cm, which is completely matched with the spectral acquisition field of view, ensuring that the hole wall outside the cleaning area is free from scouring and disturbance throughout the process.

5. The in-situ spectroscopic observation method for sandstone strata based on localized well washing as described in claim 1, characterized in that, In step S3, the low-pressure pulse cleaning process is as follows: the cleaning pressure is controlled at 0.15-0.4MPa, the pulse frequency is 5-15Hz, the duration of a single pulse cleaning is 0.2-0.5s, and the interval duration is 0.3-0.8s; the cleaning medium is filtered pure water to avoid introducing new impurities that contaminate the hole wall; for loose sandstone, the cleaning pressure does not exceed 0.2MPa to avoid the surface rock mass peeling off.

6. The in-situ spectroscopic observation method for sandstone strata based on localized well washing as described in claim 1, characterized in that: In step S4, the cleanliness of the sandstone pore wall is classified into four levels according to the grading criteria: Level 1: No visible impurities on the pore wall, the original texture and pore structure of the sandstone are clearly visible, and the image grayscale uniformity is ≥95%; Level 2: There are trace amounts of point-like residual impurities, but no flaky or layered impurities covering the surface, and the image grayscale uniformity is 90%-95%; Level 3: There are localized flaky impurities, and the grayscale uniformity is 80%-90%; Level 4: There is a continuous impurity covering layer, and the grayscale uniformity is <80%. Only Level 1 and Level 2 clean areas meet the standards and can be used for spectral acquisition.

7. The in-situ spectroscopic observation method for sandstone strata based on localized well washing as described in claim 1, characterized in that, In step S4, the micro-drying, dehydration and pressure stabilization treatment is carried out by adopting a low temperature and low pressure drying mode, with a drying temperature of 30-40℃, a relative pressure of 0.08-0.12MPa, and a drying time of 20-60s. This quickly removes free water from the surface of the pore wall, retains the original bound water in the sandstone pores, avoids water from interfering with spectral detection, and does not change the original water content characteristics of the rock mass.

8. The in-situ spectroscopic observation method for sandstone strata based on localized well washing as described in claim 1, characterized in that, In step S5, the spectral acquisition module, in adaptive parameter matching mode, automatically matches the acquisition band, integration time, and number of scans according to the sandstone stratum type; the acquisition band covers the entire visible-near infrared band of 400-2500nm, the integration time is 10-50ms, and a single point is repeatedly scanned 20-50 times, and the average value is taken as the original spectral data to improve data stability.

9. The in-situ spectroscopic observation method for sandstone strata based on localized well washing as described in claim 1, characterized in that: In step S6, the adaptive interference correction algorithm includes four major units: baseline drift correction, particle scattering correction, trace impurity feature removal, and moisture absorption peak correction. By constructing an impurity spectral feature database, it accurately matches and removes the characteristic peaks of residual trace rock powder and mud impurities from the hole washing process. The moisture spectrum fitting algorithm corrects the spectral absorption interference of shallow free water, thereby restoring the original spectral curve of sandstone to the greatest extent.

10. The in-situ spectroscopic observation method for sandstone strata based on localized well washing as described in claim 1, characterized in that: In step S7, the sandstone spectral-physical property inversion model is a machine learning model that has been pre-calibrated using a large number of sandstone samples. It takes spectral characteristic parameters as input and laboratory measured porosity, permeability, cementation strength, and mineral content as output. After sample training and accuracy verification, the model inversion error is ≤3%, which meets the requirements of high-precision engineering testing.

Images