A coal rock fracture cross-interface identification method based on water pressure derivative and bimodal

Abstract

This invention discloses a method for identifying cross-interface fractures in coal and rock based on hydraulic pressure derivatives and bimodal data. It involves real-time acquisition of wellhead pressure and displacement data during fracturing operations to calculate the actual bottom-hole injection pressure. After filtering, denoising, and standardization preprocessing, the transient derivative curve of the pump injection pressure is calculated to amplify the weak mechanical response of the fracture cross-interface process. Based on the derivative curve, the "bimodal and trough" characteristic parameters of the corresponding fracture's "interface resistance-interface slip-cross-interface breakthrough" mechanical process are extracted. Automatic quantitative identification of hydraulic fracture cross-interface behavior is achieved through a preset joint threshold criterion. This invention requires no additional monitoring equipment and utilizes only conventional construction data to achieve real-time, high-precision, and low-cost identification of fracture cross-interface behavior. It effectively solves the technical problems of high misjudgment rate, poor real-time performance, and high cost associated with traditional methods, providing a scientific basis for real-time optimization and dynamic control of fracturing operation parameters.

Description

Technical Field

[0001] This invention belongs to the technical fields of safe coal mining, efficient extraction of unconventional natural gas (coalbed methane), and rock mass engineering monitoring. Specifically, it relates to a method for identifying coal-rock fracture interfaces based on water pressure derivative and bimodal peaks. Background Technology

[0002] In deep coal mining and coalbed methane (gas) extraction projects, hydraulic fracturing technology is a core technical means to weaken the hard roof to prevent rockbursts and improve the permeability of coal seams to enhance gas extraction efficiency. Coal-bearing strata generally exhibit a layered composite structure of "sandstone-coal seam-mudstone". Whether hydraulic fractures can effectively penetrate the lithological interface and enter the target coal seam directly determines the success or failure of fracturing for pressure relief or production enhancement. Therefore, accurate identification of the propagation behavior of hydraulic fractures across the coal-rock interface is a core prerequisite for parameter optimization and effect control during fracturing operations.

[0003] Currently, methods for identifying the propagation state and cross-interface behavior of hydraulic fractures in industrial settings are mainly divided into two categories: downhole microseismic / acoustic emission monitoring methods and conventional pressure curve empirical identification methods. Both methods have significant technical limitations, as detailed below: Firstly, microseismic / acoustic emission monitoring methods suffer from drawbacks such as high cost, poor anti-interference capability, and significant data lag. In the field, downhole microseismic or acoustic emission technologies are typically used to monitor the spatial distribution of fractures. However, the complex environment of deep well operations, severe noise interference from mechanical operations, and difficulty in extracting effective signals are all challenges. Furthermore, the deployment and equipment procurement costs of microseismic networks are high, the data processing flow is complex, and there is a significant lag, making it impossible to achieve real-time dynamic adjustments during the fracturing pumping process and difficult to guide real-time optimization of on-site operations. In contrast, the wellhead pressure curve collected by the surface pumping system during fracturing operations is the most convenient, real-time, and cost-effective basic monitoring data available on-site, and it is also the most suitable data source for real-time identification of fracture propagation behavior.

[0004] Secondly, traditional empirical identification methods based on hydraulic pressure curves have serious mechanistic blind spots and extremely high misjudgment rates. In traditional hydraulic fracturing operations, field engineers typically rely on linear elastic fracture mechanics experience, using the appearance of a single peak in pump injection pressure followed by a sudden, sharp drop as a sign of successful fracturing initiation and penetration of the formation. However, the inventors of this application have discovered that layered composite coal-rock assemblages exhibit significant elasto-plastic mismatch characteristics. When hydraulic fractures approach the interface of weak coal seams at a large dip angle, a series of nonlinear mechanical phenomena occur, causing traditional identification methods to completely fail. Specifically, this manifests in two typical misjudgment scenarios: The first type of misjudgment is "not compressed": The low strength characteristics of weak coal seams can easily lead to the blunting of the crack tip. The energy of fluid injection is dissipated by large-scale plastic deformation, forming a "plastic hindrance and friction lock-up" effect. At this time, the crack initiation pressure will rise abnormally, far exceeding the fracture pressure of normal rock. On-site, it is often misjudged as "the strata are dense and not compressed" and the pump is blindly and continuously blocked, which can easily cause engineering accidents such as damage to the surface high-pressure pump set or rupture of the borehole pipe.

[0005] The second type is the misjudgment of "penetration": When high-pressure fluid breaks through the interface friction lock-up effect, the coal-rock interface will undergo catastrophic shear slip, causing the volume inside the fracture to expand instantaneously, the pumping pressure to drop sharply, and even generate transient negative pressure. Traditional methods are very likely to misjudge this "sharp pressure drop" as "the fracture has successfully penetrated the brittle layer". However, in reality, the fracture has not effectively penetrated into the target coal seam, but has been trapped at the coal-rock interface or entered a secondary pressure stagnation period, ultimately causing the fracturing transformation to completely fail to achieve the expected results.

[0006] Third, existing technologies lack in-depth analysis of the transient characteristics of hydraulic pressure curves and changes in the flexibility of fracture systems, making it impossible to establish a quantitative mapping relationship between these characteristics and mechanical mechanisms. During coal and rock fracturing, interface slippage and coal seam plastic trapping inevitably lead to dynamic abrupt changes in the "volume stiffness" of the fracture system, which are directly reflected in the temporal evolution of pressure. However, existing technologies only observe the absolute value of pressure and the magnitude of single peaks, failing to delve into the first derivative characteristics of pressure over time. Furthermore, they fail to establish a quantitative mapping relationship between the bimodal evolution pattern of "abnormal high pressure - transient drop - secondary rise" and the complex cross-interface dynamics at the coal-rock interface, thus hindering the accurate and quantitative identification of fracture cross-interface behavior.

[0007] To address the aforementioned shortcomings of existing technologies, there is an urgent need to develop a cross-interface identification method for hydraulic fractures in coal and rock that is highly real-time, accurate, cost-effective, and has a low misjudgment rate, so as to provide a scientific basis for real-time optimization and effect control of on-site fracturing operations. Summary of the Invention

[0008] To address the technical challenge of identifying whether hydraulic fractures penetrate the coal-rock interface in real-time, cost-effectively, and with high precision during existing hydraulic fracturing processes for coalbed methane / coal mine gas extraction, this paper proposes a method for identifying cross-interface coal-rock fractures based on water pressure derivatives and bimodal dynamics. By exploring the transient evolution patterns of pump injection pressure data and utilizing transient derivatives to amplify the weak mechanical response characteristics of fractures at the coal-rock interface, a quantitative mapping relationship between characteristic parameters and cross-interface mechanical behavior is established. This enables real-time and accurate identification of the cross-interface propagation behavior of hydraulic fractures, providing a scientific basis for real-time optimization of on-site fracturing construction parameters.

[0009] To achieve the above objectives, the technical solution adopted by this invention is: a method for identifying coal-rock fracture interfaces based on water pressure derivative and bimodal dynamics, comprising the following steps: S1. Real-time acquisition of construction data and conversion of bottom hole pressure: During fracturing operations or simulation experiments, real-time high-frequency acquisition of injection pressure data P at the wellhead is performed. surface The wellhead pressure is converted into the actual bottom hole injection pressure P(t) by taking into account the effects of wellbore friction and liquid column pressure, and the pumping displacement data Q(t) and pumping displacement data Q(t).

[0010] S2. Data Preprocessing: The bottom-hole injection pressure P(t) is filtered and denoised to eliminate high-frequency fluctuations and acquisition noise. Simultaneously, when the pump discharge rate fluctuates, the bottom-hole injection pressure is normalized to eliminate pressure fluctuations caused by non-geological factors, thus obtaining an effective pressure sequence P. smooth (t).

[0011] S3. Calculation of transient pressure derivative: The effective pressure sequence after preprocessing is differentiated over time to obtain the transient derivative curve of the pump injection pressure. Δt is the sampling time interval, which is amplified by derivative calculation when the crack encounters the coal-rock interface during crack propagation.

[0012] S4. Bimodal Feature Extraction and Mechanical Mechanism Matching: Based on the zero-crossing point and positive-to-negative reversal characteristics of the transient derivative curve D(t), multiple feature points are extracted sequentially, namely the first peak value P. max1 Valley value P valley Second peak P max2 And complete the one-to-one matching of each feature point with the mechanical behavior of crack propagation.

[0013] S5. Quantitative discrimination of cross-interface behavior: Based on the extracted bimodal feature parameters, a joint threshold criterion matching the coal and rock geomechanical parameters of the target block is established. When the real-time monitored pressure sequence simultaneously meets all the joint threshold criteria, it is determined that the hydraulic fracture has successfully penetrated the coal and rock interface.

[0014] S6. Real-time optimization of construction parameters: Based on the judgment results of cross-interface behavior and combined with the fracturing project objectives, the fracturing construction parameters are adjusted in real time to achieve dynamic control of the fracturing process.

[0015] Furthermore, in step S1, the sampling frequency for data acquisition is not less than 1Hz, and preferably above 10Hz, to ensure effective capture of pressure transient characteristics. The conversion formula for the bottom-hole water injection pressure P(t) is: In the formula, ρ is the density of the fracturing fluid, g is the acceleration due to gravity, H is the vertical depth of the fracturing well, and P f(Q) represents the manifold and wellbore friction related to the pump injection rate, obtained through field calibration or fitting of historical fracturing data; when conducting laboratory experiments or shallow-hole conditions, the fluid column pressure ρgH and wellbore friction P are related. f The effect of (Q) is negligible, and at this time P(t) = P surface (t).

[0016] Furthermore, in step S2, the filtering and denoising process employs either a moving average filtering algorithm or a wavelet transform algorithm; the calculation formula for the moving average filtering algorithm is as follows: In the formula, P smooth (t) represents the effective pressure sequence, k is the sliding window radius, which is adaptively selected according to the sampling frequency, and Δt is the sampling time interval.

[0017] Furthermore, in step S3, the physical meaning of the transient derivative curve D(t) of the pumping pressure is as follows: D(t)>0 indicates that the fracture system is in a state of pressure buildup; D(t)<0 indicates that the energy is released due to fluid loss or rapid expansion of the fracture volume; the point where D(t)=0 and the curve reverses is the pressure extreme point (peak or valley).

[0018] Furthermore, in step S4, the first peak value P max1 Valley value P valley Second peak P max2 The extraction and mechanical mechanism matching are specifically as follows: S41. Extraction and matching of the first peak value (interfacial hindrance / cracking peak): When the transient derivative curve D(t) is first identified to turn from positive to negative, that is, when D(t1) = 0 and D'(t1) < 0 in the neighborhood of t1, the pressure value corresponding to time t1 is recorded as the first peak value P. max1 This peak corresponds to the point where, after the hydraulic fractures initiate and extend stably within the original rock, the fracture tip touches the interface between the coal seam and the roof and floor. Due to the resistance caused by the difference in geostress and the differences in rock mechanical properties (elastic modulus, fracture toughness), the fracture propagation is hindered, and the fluid energy in the wellbore and fracture system rapidly accumulates and reaches the first limit mechanical stage.

[0019] S42. Valley Value (Interface Open / Slide Zone) Extraction and Matching: After the first peak appears, when the transient derivative curve D(t) is detected to change from negative to positive, that is, when D(t2)=0 and D'(t2)>0 in the neighborhood of t2, the pressure value corresponding to time t2 is recorded as the valley value P. valley This valley corresponds to the mechanical stage in which, as fluid is continuously pumped in, the high-pressure fluid forces the fractures to undergo shear slip at the coal-rock interface, or the bedding / natural weak surfaces at the interface are opened, resulting in a sudden increase in the local micro-volume of the fracture system and a rapid release of fluid energy, manifested as a rapid drop in macroscopic pressure.

[0020] S43. Extraction and Matching of the Second Peak (Cross-interface Breakthrough Peak): After experiencing the trough phase, when the transient derivative curve D(t) is detected to turn from positive to negative again, i.e., satisfying D(t3)=0 and D'(t3)<0 in the neighborhood of t3, the pressure value corresponding to time t3 is recorded as the second peak value P. max2 This peak corresponds to the secondary accumulation of fluid energy reaching the fracturing pressure of the rock body on the other side of the coal-rock interface. The hydraulic fracture forcibly tears and breaks through the coal-rock interface, realizing the mechanical stage of cross-layer extension to adjacent rock / coal seams. After the breakthrough, the pressure usually drops and enters a stable extension period after the cross-interface.

[0021] Further, in step S5, the joint threshold criterion includes a time window criterion, an energy release drop criterion, and a secondary pressure buildup breakthrough criterion, specifically: Time window criterion: The bimodal characteristic occurs within a very short time after the crack encounters resistance; the time difference between the two peaks is defined. In the formula, T threshold The preset threshold time is calibrated based on the coal and rock geomechanical parameters of the target block and the on-site construction conditions.

[0022] Energy release drop criterion: The pressure drop caused by opening the interface must reach a preset threshold to define the first peak-to-valley pressure drop. In the formula, α is the preset first pressure drop ratio coefficient, and the first peak-to-valley pressure drop amplitude is defined as ΔP1=P1-P min The first pressure drop ratio coefficient α is defined as ΔP1 / P1. To accurately identify catastrophic shear slip and instantaneous expansion phenomena at the interface, this invention limits the preset value range of α to [0.2, 1), calibrated according to the coal and rock characteristics of the target block.

[0023] Criteria for a second pressure buildup breakthrough: The second peak must have substantial energy accumulation; the amplitude of the second pressure buildup is defined. In the formula, β is a preset secondary pressure-retaining ratio coefficient, defined as β = P2 / P1, β = P2 / P1. To confirm that the fracture has effectively invaded the weak coal seam, this invention limits the preset value range of β to [0.2, 0.85], calibrated according to the coal and rock characteristics of the target block.

[0024] Furthermore, in step S6, the specific rules for real-time optimization of the construction parameters are as follows: If the project objective is to "control the crack height" (to prevent cracks from penetrating the top and bottom slabs and causing accidents such as water intrusion and gas leakage), then when the first peak value is detected or the value is in the trough range, measures such as reducing the pump injection rate, adding temporary plugging agent, or stopping the pump should be taken immediately to inhibit the crack from continuing to expand and penetrate the interface.

[0025] If the engineering objective is to "penetrate the rock strata" (to achieve the transformation objectives such as increasing the permeability of the coal seam and weakening the roof), then after monitoring and confirming that the joint threshold criteria are met and the fracture has successfully crossed the interface, the pumping energy should be maintained or increased to ensure the effective extension of the fracture within the target stratum.

[0026] Compared with the prior art, the present invention has the following advantages: 1. Conventional pumping pressure curves show relatively gentle changes when fractures touch and penetrate the coal-rock interface, making it difficult for the naked eye or conventional algorithms to accurately capture the weak mechanical response signals. This invention introduces a transient derivative curve, transforming the subtle slope changes of the pressure curve into an explicit "double-peak, one-valley" characteristic, which greatly amplifies the mechanical response signal during the fracture-interface crossing process and significantly improves identification accuracy and signal sensitivity.

[0027] 2. This invention achieves a precise one-to-one correspondence between the "double peak and one valley" characteristics of the transient derivative curve and the actual rock mechanical evolution process of fractures, namely "interface resistance - interface slippage opening - cross-interface breakthrough". It establishes a quantitative mapping relationship between characteristic parameters and mechanical mechanisms, which can effectively eliminate the interference of pressure fluctuations caused by factors such as sand shedding and local formation heterogeneity, avoid false positives and false negatives caused by traditional single pressure peak identification methods, and significantly reduce the identification misjudgment rate.

[0028] 3. Traditional microseismic monitoring technologies often suffer from severe data processing lag, typically requiring the completion of construction before a complete interpretation can be generated, making it impossible to guide real-time adjustments during the construction process. This invention utilizes only the ground high-frequency pump injection data from the fracturing truck / drilling system to complete real-time calculations, achieving millisecond- to second-level identification results output. It truly realizes closed-loop control of "fracturing, identification, and adjustment simultaneously," providing direct scientific basis for real-time optimization of on-site construction parameters.

[0029] 4. This invention relies solely on conventional engineering monitoring data (pressure and displacement) during the fracturing process, eliminating the need for additional expensive downhole monitoring instruments (such as optical fibers, microseismic detectors, etc.) and additional construction procedures, thus greatly reducing on-site monitoring costs. It is particularly suitable for routine applications in large-scale coalbed methane wells and underground gas extraction wells in coal mines, and can also be applied to monitoring fracture propagation behavior in indoor hydraulic fracturing physical simulation experiments. Attached Figure Description

[0030] Figure 1 This is the overall flowchart of the present invention.

[0031] Figure 2 This is a pumping pressure curve from a numerical simulation embodiment of the present invention.

[0032] Figure 3The figures shown are test results of the physical experiment embodiment of the present invention, wherein (a) is a real photograph of crack propagation and (b) is a comparison curve of the changes in pumping pressure and acoustic emission parameters over time. Detailed Implementation

[0033] The present invention will be further described below.

[0034] This invention discloses a method for identifying coal and rock fracture interfaces based on water pressure derivative and bimodal dynamics. The overall process is as follows: Figure 1 As shown, the core idea is to amplify the mechanical response of the crack interface by using the transient derivative of the pumping pressure, extract the "double peak and one valley" feature, and achieve accurate identification of the crack interface behavior by using a joint threshold criterion.

[0035] Example 1: Numerical simulation verification of the example This embodiment verifies the identification effect of the method of the present invention through numerical simulation of hydraulic fracturing of coal-rock composites. The specific implementation process is as follows: A numerical model of a two-layer coal-rock composite consisting of sandstone and coal seam was constructed. The model dimensions were 100mm × 100mm × 100mm, with the sandstone thickness at 75mm and the coal seam thickness at 25mm. The elastic modulus of the sandstone was set to 20 GPa and Poisson's ratio to 0.25, while the elastic modulus of the coal seam was set to 3 GPa and Poisson's ratio to 0.3. A true triaxial stress state was applied: vertical stress of 10 MPa, maximum horizontal principal stress of 6 MPa, and minimum horizontal principal stress of 4 MPa. A constant displacement of 0.001 m³ was used. 3 Hydraulic fracturing simulation was performed at a sampling frequency of 10Hz for 300 seconds. Data acquisition and conversion: Bottom-hole water injection pressure data P(t) and pump discharge data Q(t) were acquired in real time during the simulation process at a sampling frequency of 10Hz. In this embodiment, which is a numerical simulation, no wellhead-bottom-hole pressure conversion is required, and the bottom-hole pressure sequence is obtained directly.

[0036] S2. Data Preprocessing: A moving average filtering algorithm is used to denoise the pressure data, with a moving window radius of k=3, to obtain the smoothed effective pressure sequence P. smooth (t), in this embodiment, a constant displacement is used, and there is no need to perform displacement normalization.

[0037] S3. Transient Derivative Calculation: Differentiate the effective pressure sequence over time to obtain the transient derivative curve of the pumping pressure. .

[0038] S4. Bimodal Feature Extraction: Based on the zero-crossing point and positive / negative reversal features of the derivative curve, feature points are automatically extracted, such as... Figure 2 As shown: ① First peak value: When t1≈118s, the derivative D(t) first turns from positive to negative, and the first peak value P is extracted. max1=14.348MPa, corresponding to the interface hindrance stage where the crack tip touches the coal-rock interface and the expansion is hindered.

[0039] ② Valley value: When t2≈121s, the derivative D(t) changes from negative to positive, and the valley value P is extracted. valley =0.664 MPa corresponds to the interface opening stage where shear slip occurs at the coal-rock interface and fluid energy is released rapidly.

[0040] ③ Second peak: When t3≈160s, the derivative D(t) changes from positive to negative again, and the second peak P is extracted. max2 =8.399MPa, corresponding to the cross-interface breakthrough stage after the fluid accumulates twice and breaks through the coal-rock interface.

[0041] S5. Joint threshold discrimination: This embodiment presets a threshold T. threshold =50s, α=0.80, β=0.50, perform criterion verification: ① Time window criterion: ΔT = 160s - 118s = 42s ≤ 50s, which satisfies the criterion.

[0042] ② Criterion for energy release drop: ΔP drop =14.348-0.664=13.684MPa, α·P max1 =0.80×14.348=11.478MPa, 13.684MPa≥11.478MPa, which satisfies the criterion.

[0043] ③ Criterion for secondary pressure breakthrough: ΔP rise =8.399-0.664=7.735MPa, β·ΔP drop =0.50×13.684=6.842MPa, 7.735MPa≥6.842MPa, which satisfies the criterion.

[0044] Recognition results: All three criteria were met, and the system output a recognition signal in real time that "the hydraulic fracture has successfully crossed the coal-rock interface," which is completely consistent with the actual expansion state of the fracture penetrating the coal-rock interface in the numerical simulation results, verifying the accuracy of the method of the present invention.

[0045] Example 2: Indoor Physical Experiment Verification Example This embodiment verifies the identification effect of the method of the present invention under real working conditions through a physical simulation experiment of hydraulic fracturing of a coal-rock composite. The specific implementation process is as follows: Specimen preparation: A standard specimen of a three-layer coal-rock combination of sandstone-coal seam-sandstone was prepared. The specimen size was φ100mm×200mm, with a coal seam thickness of 50mm in the middle and sandstone thicknesses of 75mm on each side. A 10mm diameter and 120mm deep borehole was pre-drilled in the center of the specimen. The bottom of the borehole was located in the lower sandstone layer for fracturing fluid injection.

[0046] Experimental system setup: A true triaxial hydraulic fracturing experimental system was used, with a vertical ground stress of 25 MPa, a maximum horizontal principal stress of 20 MPa, and a minimum horizontal principal stress of 15 MPa. The hydraulic fracturing experiment was conducted with a constant flow rate of 10 mL / min for 900 s. Pump injection pressure, flow rate, and acoustic emission data were collected simultaneously, and the sampling frequency was set to 100 Hz.

[0047] S1. Data Acquisition and Conversion: Real-time acquisition of wellhead water injection pressure data P surface (t) and pump discharge data Q(t), this embodiment is an indoor short borehole experiment, neglecting the fluid column pressure and wellbore friction, bottom hole pressure P(t) = P surface (t).

[0048] S2. Data preprocessing: Wavelet transform filtering algorithm is used to denoise the high-frequency pressure data to eliminate mechanical vibration noise of the experimental system and obtain the effective pressure sequence Psmooth(t). In this embodiment, constant displacement is used, so displacement normalization is not required.

[0049] S3. Calculation of transient derivative: The transient derivative curve of pumping pressure D(t) is obtained by differentiating the effective pressure sequence over time.

[0050] S4. Bimodal Feature Extraction and Criterion Verification: Based on the zero-crossing point and positive / negative reversal features of the derivative curve, feature points are automatically extracted, such as... Figure 3 As shown: ① First peak capture: When t1≈50s, the system captures the first zero crossing of the derivative D(t) and its change from positive to negative, and automatically extracts the first peak P. max1 =35.65MPa, indicating that the fracture initiated around the shaft and encountered resistance at the high-stress coal-rock interface, reaching the first energy limit; this embodiment presets α=0.20, β=0.30, T threshold =120s.

[0051] ② Valley value capture and drop criterion verification: When t2≈80s, the system captures the derivative D(t) changing from negative to positive and extracts the valley value P. valley =25.80MPa; Calculate the first peak-to-valley pressure drop ΔP drop =35.65-25.80=9.85MPa, α·P max1 =0.20×35.65=7.13MPa, 9.85MPa>7.13MPa, the pressure drop criterion is met, and it is determined that the coal-rock interface has undergone substantial displacement and opening.

[0052] ③ Second peak capture and secondary pressure criterion verification: When t3≈140s, the system captures the derivative D(t) crossing zero again and turning from positive to negative, and extracts the second peak P.max2 =28.30MPa; Calculate the secondary pressure amplitude ΔP rise =28.30-25.80=2.50MPa, β·ΔP drop =0.20×9.85=1.97MPa, 2.50MPa>1.97MPa, the secondary pressure criterion is met.

[0053] ④ Time window criterion verification: ΔT=140s-50s=90s≤120s, the time window criterion is satisfied.

[0054] Identification Results and Verification: All three joint threshold criteria were met, and the system output a real-time identification signal of "hydraulic fracture has successfully crossed the coal-rock interface" at t3=140s. After the experiment, the specimen was cross-sectionally observed, and the fracture had successfully penetrated the interface between the underlying sandstone and the coal seam and entered the coal seam, which was completely consistent with the identification results of the method of this invention. At the same time, acoustic emission monitoring data showed that after 140s, acoustic emission events were concentrated inside the coal seam, further verifying that the fracture had completed cross-interface expansion, proving the high accuracy and reliability of the method of this invention.

[0055] Example 3: On-site Engineering Application Example This example demonstrates a field application of hydraulic fracturing in a gas extraction borehole at a coal mine. The specific implementation process is as follows: Project Overview: The target borehole is a cross-layer gas drainage borehole in the underground coal mine floor, with a vertical depth of 650m. The target coal seam is No. 3 coal seam, with a thickness of 6.2m. The roof of the coal seam is mudstone, and the floor is fine sandstone. The project objective is to use hydraulic fracturing to create fractures that penetrate the floor sandstone and enter the coal seam, thereby increasing the permeability of the coal seam and improving the gas drainage efficiency.

[0056] S1. Data Acquisition: The monitoring system built into the fracturing truck is used to collect the wellhead water injection pressure P in real time at a sampling frequency of 10Hz. surface (t) and pumping displacement Q(t), fracturing fluid density ρ=1020kg / m³ 3 The acceleration due to gravity is g = 9.8 m / s². 2 With a vertical depth H=650m, the wellbore friction P was obtained through on-site calibration. f (Q) = 0.08Q 2 Q is in meters. 3 / min.

[0057] S2. Bottom-of-well pressure conversion: The bottom-of-well injection pressure P(t) is calculated in real time according to the formula P = P surface (t)+1020×9.8×650×10 -6 -0.08Q 2 =P surface (t)+6.50-0.08Q 2.

[0058] S3. Data Preprocessing: The bottom hole pressure data is denoised using a moving average filter with a moving window radius of k=5. At the same time, the pressure data is normalized to eliminate pressure interference caused by pump discharge fluctuations, thus obtaining an effective pressure sequence.

[0059] S4. Transient Derivative Calculation and Feature Extraction: Calculate the pressure transient derivative curve in real time and extract the first peak value P sequentially. max1 =42.8MPa (t1=185s), valley value P valley =31.2MPa (t2=210s), second peak value P max2 =36.5MPa (t3=260s).

[0060] S5. Joint threshold determination: The preset threshold T for this block. threshold =100s, α=0.22, β=0.30, Verification results: ① ΔT = 260 - 185 = 75s ≤ 100s, which satisfies the condition.

[0061] ②ΔP drop =42.8-31.2=11.6MPa≥0.22×42.8=9.42MPa, which is satisfied.

[0062] ③ΔP rise =36.5-31.2=5.3MPa≥0.30×11.6=3.48MPa, which is satisfied.

[0063] S6. Field application effect: When t3=260s, the system determined in real time that the crack had successfully penetrated the bottom sandstone and entered the target coal seam. The pumping rate was maintained for 20 minutes before the pump was stopped. Later gas extraction data showed that the pure gas extraction volume of this borehole was 3.2 times higher than that of the adjacent borehole, which verified the reliability and practicality of the method of this invention in field engineering applications.

[0064] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for identifying coal-rock fracture interfaces based on water pressure derivative and bimodal distribution, characterized in that, Includes the following steps: S1. Real-time acquisition of construction data and conversion of bottom hole pressure: During fracturing operations or simulation experiments, real-time high-frequency acquisition of injection pressure data P at the wellhead is performed. surface The wellhead pressure is converted into the actual bottom hole injection pressure P(t) using the pump injection displacement data Q(t) and Q(t). S2. Data Preprocessing: The bottom-hole injection pressure P(t) is filtered and denoised, and the injection pressure is also normalized by displacement to obtain the effective pressure sequence P. smooth (t); S3. Calculation of transient pressure derivative: The effective pressure sequence after preprocessing is differentiated over time to obtain the transient derivative curve of pumping pressure D(t); S4. Bimodal Feature Extraction and Mechanical Mechanism Matching: Based on the zero-crossing point and positive-to-negative reversal characteristics of the transient derivative curve D(t), multiple feature points are extracted sequentially, namely the first peak value P. max1 Valley value P valley Second peak P max2 And complete the matching of each feature point with the mechanical behavior of crack propagation; S5. Quantitative discrimination of cross-interface behavior: Based on the extracted bimodal feature parameters, discrimination is performed through preset joint threshold criteria. When the real-time monitored pressure sequence simultaneously meets all joint threshold criteria, it is determined that the hydraulic fracture has successfully penetrated the coal-rock interface.

2. The method according to claim 1, characterized in that, In step S1, the conversion formula for the bottom water injection pressure P(t) is: In the formula, ρ is the density of the fracturing fluid, g is the acceleration due to gravity, H is the vertical depth of the fracturing well, and P f (Q) represents the manifold and wellbore friction related to the pump injection rate, which is obtained through field calibration or fitting of historical fracturing data.

3. The method according to claim 1, characterized in that, In step S2, the filtering and denoising process employs either a moving average filtering algorithm or a wavelet transform algorithm; the calculation formula for the moving average filtering algorithm is as follows: In the formula, P smooth (t) represents the effective pressure sequence, k is the sliding window radius, which is adaptively selected according to the sampling frequency, and Δt is the sampling time interval.

4. The method according to claim 1, characterized in that, In step S3, the physical meaning of the transient derivative curve D(t) of the pumping pressure is as follows: D(t)>0 indicates that the fracture system is in a state of pressure buildup; D(t)<0 indicates that the energy is released due to fluid loss or rapid expansion of the fracture volume; the point where D(t)=0 and the curve reverses is the pressure extreme point.

5. The method according to claim 1, characterized in that, In step S4, the first peak value P max1 Valley value P valley Second peak P max2 The extraction and mechanical mechanism matching are specifically as follows: S41. First Peak Extraction and Matching: When the transient derivative curve D(t) is detected to turn from positive to negative for the first time, i.e., D(t1) = 0 and D'(t1) < 0 in the neighborhood of t1, the pressure value corresponding to time t1 is recorded as the first peak value P. max1 This peak corresponds to the stage where, after the hydraulic fractures initiate and extend within the original rock, the fracture tip touches the coal-rock interface. The extension is hindered by the difference in geostress and the difference in rock mechanical properties, and the system energy accumulates to the limit for the first time at the interface stagnation / fracture initiation stage. S42. Valley Value Extraction and Matching: After the first peak appears, when the transient derivative curve D(t) is detected to change from negative to positive, i.e., D(t2) = 0 and D'(t2) > 0 in the neighborhood of t2, the pressure value corresponding to time t2 is recorded as the valley value P. valley This valley value corresponds to the interface opening / slipping stage in which high-pressure fluid forces the coal-rock interface to undergo shear slip or open the weak interface surface, resulting in a sudden increase in local volume and a rapid release of fluid energy. S43. Second Peak Extraction and Matching: After the valley appears, when the transient derivative curve D(t) is detected to turn from positive to negative again, i.e., D(t3) = 0 and D'(t3) < 0 in the neighborhood of t3, the pressure value corresponding to time t3 is recorded as the second peak value P. max2 This peak value corresponds to the secondary accumulation of energy by the fluid at the interface, reaching the rock fracture pressure on the other side of the interface, and the hydraulic fracture breaking through the coal-rock interface to achieve cross-layer expansion, which is the cross-interface breakthrough stage.

6. The method according to claim 5, characterized in that, In step S5, the joint threshold criterion includes the time window criterion, the energy release drop criterion, and the secondary pressure buildup breakthrough criterion, specifically: Time window criterion: Time difference between the two peaks In the formula, T threshold The preset threshold time; Criterion for energy release drop: First peak-to-valley pressure drop In the formula, α is the preset first pressure drop ratio coefficient; Criterion for breaking through secondary pressure: amplitude of secondary pressure buildup In the formula, β is the preset secondary pressure reduction ratio coefficient.

7. The method according to claim 1, characterized in that, The method further includes step S6: real-time optimization of construction parameters, adjusting the fracturing construction parameters in real time based on the judgment result of step S5 and the fracturing project objectives; If the project objective is to control the crack height and prevent cracks from penetrating the top and bottom slabs, then when the first peak value is detected or the value is in the trough range, measures such as reducing the pump injection rate, adding temporary plugging agent, or stopping the pump should be taken. If the engineering objective is to promote fracture penetration into the rock strata, then after monitoring and confirming that the joint threshold criterion is met and the fracture has crossed the interface, the pumping energy should be maintained or increased.

Images