Hydraulic fracturing monitoring method based on Krauklis wave resonance

Abstract

The invention belongs to the technical field of oil-gas exploration, particularly relates to a hydraulic fracturing monitoring method based on Krauklis wave resonance, and aims at solving the problem that the monitoring effect is poor in the hydraulic fracturing process. Comprising the steps that a vibration signal of a target well in the hydraulic fracturing process is obtained; extracting a plurality of sub-signal segments of the vibration signal according to the time sequence, and performing time-frequency analysis on each sub-signal segment by adopting short-time Fourier transform to obtain a time-frequency spectrogram of the vibration signal; acquiring a resonance energy amplitude statistical chart of the vibration signal; under the condition that the resonance energy amplitude statistical chart meets preset conditions, it is determined that hydraulic fracturing succeeds, and the preset conditions include that after the resonance energy amplitude average value in the set duration is changed from the first amplitude to the second amplitude, the resonance energy amplitude average value is kept within the set range with the third amplitude as the center, and the second amplitude is higher than the first amplitude and the third amplitude. The hydraulic fracturing process can be visually and rapidly evaluated, and dynamic monitoring of the hydraulic fracturing process is achieved.

Description

Background Art

[0002] With the continuous development of unconventional oil and gas exploration and development, hydraulic fracturing technology, as one of the core technologies for unconventional oil and gas exploration and development, has been widely applied in recent years. Hydraulic fracturing technology injects high-pressure fluid (fracturing fluid composed of water, mud, etc.) into the wellbore, causing the reservoir rock to fracture and form fractures. Subsequently, proppants are injected to keep the fractures open, thereby increasing the permeability of the formation, enhancing the connectivity of the reservoir, and increasing the production of a single well, so as to achieve the purpose of increasing oil and gas production.

[0003] During the hydraulic fracturing process, a Krauklis wave will be generated in the fractures filled with fluid (such as fracturing fluid). The Krauklis wave propagates back and forth in the fractures and exhibits unique resonance characteristics, which can be used to evaluate hydraulic fracturing fractures. However, the current resonance research on the Krauklis wave still remains in the stage of numerical simulation and physical simulation, and the observation method of recording the Krauklis wave inside the fractures is impossible to achieve in practical applications. Summary of the Invention

[0004] To solve the above problems in the prior art, that is, the problem of poor monitoring effect during the hydraulic fracturing process, the present application proposes a hydraulic fracturing monitoring method based on Krauklis wave resonance, including: Step S10: Obtain the vibration signal during the hydraulic fracturing process of the target well, where the vibration signal is converted from the Krauklis wave generated during the hydraulic fracturing process; Step S20: Extract multiple sub-signal segments of the vibration signal according to time sequence, and perform time-frequency analysis on each sub-signal segment using the short-time Fourier transform to obtain the time-frequency spectrogram of the vibration signal; Step S30: Obtain the resonance energy amplitude statistical chart of the vibration signal by extracting the spectral energy amplitude of the time-frequency spectrogram, where the resonance energy amplitude statistical chart is used to characterize the intensity of the resonance energy at different time periods; Step S40: Determine that the hydraulic fracturing is successful when the resonance energy amplitude statistical chart meets the preset conditions, where the preset conditions include: after the average value of the resonance energy amplitude within the set duration changes from the first amplitude to the second amplitude, it remains within the set range centered on the third amplitude, and the second amplitude is higher than the first amplitude and the third amplitude.

[0005] As a possible implementation manner, extracting multiple sub-signal segments of the vibration signal according to time sequence and performing time-frequency analysis on each sub-signal segment using the short-time Fourier transform includes: Performing windowed short-time Fourier transform on the vibration signal, including: Wherein, Represents the transformation result of the vibration signal after windowed short-time Fourier transform, Represents the window function, Represents the vibration signal, t Represents the current time, Represents The conjugate of, Represents that the Fourier window function is shifted at time At, e Represents the natural constant, j Represents the imaginary unit, Represents the angular frequency; Converts the windowed short-time Fourier transform to its discrete form, and sets the discrete signal of the vibration signal To be , including: Wherein, Represents The discrete form of, N Represents the number of points of the Fourier transform, Represents the window function of the discrete signal, n Represents the number of time sampling points, Represents the Fourier window function, centered at the time point m , k Represents the discrete frequency index, Represents the frequency.

[0006] As a possible implementation, by extracting the spectral energy amplitude of the time-frequency spectrogram, a resonance energy amplitude statistical chart of the vibration signal is obtained, including: Extract the spectral energy amplitude at different times in the time-frequency spectrogram; Within the same time period, superimpose the spectral energy amplitudes within the preset frequency band range to generate the resonance energy amplitude; Based on the resonance energy amplitude and the corresponding time period, generate a resonance energy amplitude statistical chart.

[0007] As a possible implementation, when the resonance energy amplitude statistical chart meets the preset conditions, it is determined that the hydraulic fracturing is successful, including: Input the resonance energy amplitude statistical chart into the trained hydraulic fracturing monitoring neural network, and output the monitoring result of the hydraulic fracturing. Among them, the training data of the hydraulic fracturing monitoring neural network are the historical resonance energy amplitude statistical charts of multiple historical vibration signals during the historical fracturing process.

[0008] As a possible implementation, the training process of the neural network includes: Obtain the historical vibration signals generated during the historical hydraulic fracturing process; By extracting the spectral energy amplitude from the historical time-frequency spectrogram of any historical vibration signal, a statistical graph of the historical resonance energy amplitude of any historical vibration signal is obtained; Using multiple statistical graphs of historical resonance energy amplitudes as the training set and the hydraulic fracturing evaluation results of the statistical graphs of historical resonance energy amplitudes as labels, a residual neural network is trained to obtain a trained hydraulic fracturing monitoring neural network, where the loss function of the residual neural network is the cross-entropy loss function.

[0009] As a possible implementation, before obtaining the historical vibration signal generated during the historical hydraulic fracturing process, it includes: Performing staged fracturing operations on the target well and recording the start time and end time of each staged fracturing operation.

[0010] As a possible implementation, after obtaining the historical vibration signal generated during the historical hydraulic fracturing process, it includes: Aligning the start time and end time of the historical vibration signal acquisition with the start time and end time of the staged fracturing operation respectively; Filtering and denoising the historical vibration signal.

[0011] As a possible implementation, the method further includes: In response to the end of the hydraulic fracturing stage, when the resonance energy amplitude exceeds the threshold, it is determined that the hydraulic fracturing is successful, where the threshold is determined by statistically analyzing the background noise of the time-frequency spectrogram.

[0012] As a possible implementation, when the vibration signal is a tube wave signal, after obtaining the vibration signal of the target well during the hydraulic fracturing process, it includes: Decoding the vibration signal from a waveform signal and storing it as a SEGY format file.

[0013] Advantages of this application: (1) Obtain the vibration signal during the hydraulic fracturing process of the target well, which can collect the relevant signals of Krauklis waves generated during hydraulic fracturing, avoiding the problem that the direct collection of Krauklis waves is affected by underground media; extract multiple sub-signal segments of the vibration signal according to time sequence, and perform time-frequency analysis on each sub-signal segment using the short-time Fourier transform to obtain the time-frequency spectrogram of the vibration signal. By extracting the sub-signal segments, the target signal segment to be analyzed can be analyzed separately, and the spectral characteristics at different times can be captured through the short-time Fourier transform; by extracting the spectral energy amplitude of the time-frequency spectrogram, obtain the resonance energy amplitude statistical chart of the vibration signal. Among them, the resonance energy amplitude statistical chart is used to characterize the intensity of resonance energy at different time periods, which can intuitively reflect the strength of resonance energy and facilitate signal analysis; when the resonance energy amplitude statistical chart meets the preset conditions, it is determined that the hydraulic fracturing is successful. Among them, the preset conditions include: after the average value of the resonance energy amplitude within the set duration changes from the first amplitude to the second amplitude, it remains within the set range centered on the third amplitude, and the second amplitude is higher than the first amplitude and the third amplitude. By statistically analyzing the change of resonance energy, the hydraulic fracturing process can be intuitively and quickly evaluated based on the resonance characteristics of Krauklis waves, realizing the dynamic monitoring of the hydraulic fracturing process, improving the monitoring effect of the hydraulic fracturing process, and providing guiding technology for the hydraulic fracturing operation. Description of the Drawings

[0014] By reading the detailed description of the non-restrictive embodiments with reference to the following drawings, other features, objectives and advantages of the present application will become more obvious: Figure 1 is a flow chart of a hydraulic fracturing monitoring method based on the resonance of Krauklis waves provided by an embodiment of the present application; Figure 2 is a schematic diagram of a "wellbore - fracture" model provided by an embodiment of the present application; Figure 3 is a visualization example diagram of a numerical simulation gather provided by an embodiment of the present application; Figure 4 is a single-trace waveform data example diagram provided by an embodiment of the present application; Figure 5 is a signal data example diagram after windowing provided by an embodiment of the present application; Figure 6 is a time-frequency spectrogram before hydraulic fracturing without fractures provided by an embodiment of the present application; Figure 7 is a time-frequency spectrogram after hydraulic fracturing with a single fracture provided by an embodiment of the present application; Figure 8It is the time-frequency spectrum diagram in the case of double fractures after hydraulic fracturing provided by an embodiment of the present application; Figure 9 It is the time-frequency spectrum diagram in the case of triple fractures after hydraulic fracturing provided by an embodiment of the present application; Figure 10 It is a well section design model diagram provided by an embodiment of the present application; Figure 11 It is an example diagram of a time-frequency spectrum provided by an embodiment of the present application; Figure 12 It is provided by an embodiment of the present application Figure 11 The corresponding resonance energy amplitude statistical chart; Figure 13 It is the system block diagram of a hydraulic fracturing monitoring system based on Krauklis wave resonance provided by an embodiment of the present application. Specific embodiments

[0015] The present application will be further described in detail below with reference to the accompanying drawings and embodiments. It can be understood that the specific embodiments described herein are only used to explain the related invention, rather than limiting the invention. Additionally, it should be noted that for the sake of description, only the parts related to the relevant invention are shown in the drawings.

[0016] It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments can be combined with each other. The present application will be described in detail below with reference to the drawings and embodiments.

[0017] To more clearly illustrate a hydraulic fracturing monitoring method based on Krauklis wave resonance of the present application, the following combines Figure 1 To elaborate on each step in the embodiments of the present application.

[0018] A hydraulic fracturing monitoring method based on Krauklis wave resonance in the first embodiment of the present application includes steps S10 - S40, and each step is described in detail as follows: Step S10, obtaining the vibration signal during the hydraulic fracturing process of the target well, where the vibration signal is converted from Krauklis waves generated during the hydraulic fracturing process.

[0019] Optionally, the target well is an oil and gas well undergoing hydraulic fracturing. By performing hydraulic fracturing on the target well, the generated vibration signal is collected by signal acquisition equipment.

[0020] In the embodiments of the present application, the hydraulic fracturing operation is performed using the staged fracturing technique, and a horizontal well is selected as the target well for hydraulic fracturing. In other embodiments, other oil and gas wells can also be selected for hydraulic fracturing, such as extended reach wells, multilateral wells, etc.

[0021] In the embodiments of the present application, the ball-drop sliding sleeve staged fracturing technology is adopted to achieve staged fracturing, and the mechanical operation of downhole tools is utilized to achieve precise staged fracturing of different formations, thereby significantly improving the production efficiency of oil and gas wells.

[0022] It should be noted that the main stages of staged fracturing are perforation, fracturing fluid injection, fracture generation by fracturing, and proppant injection. The ball-drop sliding sleeve staged fracturing is achieved by dropping balls of different diameters. First-stage fracturing: It is carried out inside the pressure sliding sleeve at the end of the fracturing pipe. When the pressure rises, the sliding sleeve is opened for perforation, the fracturing fluid is injected into the fracturing formation, the proppant is injected, the sliding sleeve is closed, and the proppant is flushed to complete the operation. Staged fracturing step by step: Balls of different diameters are dropped to separate each stage of the fracturing section. When the pressure rises, the ball sliding sleeve is opened for perforation, the fracturing fluid is injected to form fractures, the proppant is injected, and then flushed. By repeating the operation, staged fracturing step by step is achieved. The fracturing sections are separated by simply dropping balls, realizing high-efficiency production of the fracturing construction.

[0023] Optionally, a signal acquisition device is installed at the wellhead of the target well to collect the vibration signals of the target well during the hydraulic fracturing process. For example, a single-component geophone or a three-component geophone.

[0024] In the embodiments of the present application, by installing a signal acquisition device at the wellhead and receiving vibration signals only at the wellhead, the signal reception cost can be reduced.

[0025] It should be noted that only the vertical component data needs to be analyzed in this application. If a three-component geophone is selected as the data acquisition device, only the vertical component data needs to be analyzed during the data analysis.

[0026] Optionally, the collected vibration signal is a waveform signal, mainly dominated by pipe waves.

[0027] It should be noted that during the hydraulic fracturing process, due to the action of the fracturing pressure, Krauklis waves will be generated in the cracks. When the Krauklis waves are converted into pipe waves at the connection between the cracks and the well, that is, when the propagation medium of the Krauklis waves changes from the cracks to the wellbore wall, vibration signals can be generated. The vibration signals carry the resonance information of the Krauklis waves and propagate along the wellbore wall of the target well and are transmitted to the geophone at the wellhead.

[0028] Receiving signals through the wellbore wall can obtain vibration signals more directly because if the waves are directly transmitted to the ground through the underground medium, the underground medium will have a greater impact on the waves, resulting in the inability to identify the Krauklis waves.

[0029] It should be noted that during the staged fracturing process, the geophone at the wellhead needs to record the waveform data of each fracturing section throughout the process.

[0030] In the embodiments of the present application, before collecting vibration signals, it is necessary to deploy signal collection instruments. To better receive wellhead data, geophones are selected to be deployed at the wellhead. During the deployment process, the receiving end (bottom end) of the instrument is closely contacted with the pipe wall to reduce signal loss. The instrument itself needs to be fixed with tape or rope, and the instrument is not allowed to shake during the measurement. At the same time, considering environmental factors around the wellhead, such as environmental noise, temperature, humidity, and potential electromagnetic interference, corresponding measures need to be taken to protect the geophones to ensure the accuracy of the data. For environmental noise, personnel movement and vehicle driving near the wellhead must be avoided during the fracturing monitoring process. For temperature, the working temperature range of commonly used geophones on the market (taking the SmartSolo seismograph as an example) is -40°C to 70°C, and the specific temperature range depends on the instrument model. During the monitoring process, the measurement must be ensured within the specified temperature range. For humidity, since the instrument wires and interfaces are exposed, they cannot come into contact with liquids such as water to avoid damaging the instrument. Therefore, it is necessary to avoid measuring in rainy days. For potential electromagnetic interference, the detection construction location needs to avoid high-voltage transmission lines, mains networks, telluric currents, solar panels, etc. to avoid electromagnetic crosstalk.

[0031] Furthermore, to ensure the integrity and reliability of the data during the experiment, spare geophones need to be prepared. The specific quantity is determined according to the actual monitoring time and the instrument power. The quantity is ensured to have more than 3 spare geophones available for replacement each time of monitoring. So that during the experiment, if a certain geophone fails or runs out of power (power < 5%), it can be quickly replaced to ensure the continuity of the experiment. All geophones need to be clock-synchronized so that they can be accurately corresponded to the same time point when recording data. For example, clock synchronization can be performed through GPS.

[0032] Furthermore, it is also necessary to perform performance tests on the geophones to ensure the consistency of the instruments when recording the same seismic event: place each geophone at equal intervals along a 5m-radius ring, strike the ground with a heavy hammer in the center of the ring, export the records of all geophones, check the waveform amplitude and sensitivity response of each geophone, ensure that the amplitude changes are consistent after normalization, and at the same time, the time difference between the first recorded seismic events between adjacent instruments is less than 3ms. If it exceeds this range, the accuracy is unqualified.

[0033] Furthermore, it is also necessary to store the data. To ensure the integrity of the data, the embodiments of the present application use large-capacity data storage devices and back up regularly to prevent data loss.

[0034] Step S20, extract multiple sub-signal segments of the vibration signal according to the time sequence, and perform time-frequency analysis on each sub-signal segment using the short-time Fourier transform to obtain the time-frequency spectrogram of the vibration signal.

[0035] Optionally, by multiplying the window function with the vibration signal, individual sub-signal segments can be extracted for efficient analysis of effective data. Further, by translating this window function on the time axis, multiple sub-signal segments can be extracted in sequence.

[0036] Optionally, perform short-time Fourier transform and discretization processing on each sub-signal segment to achieve time-frequency analysis of any sub-signal segment.

[0037] In the embodiments of the present application, a window function is added on the basis of the short-time Fourier transform, and time-frequency analysis of the vibration signal is performed through windowed short-time Fourier transform, which can be achieved by the following formula: Wherein, represents the transformation result of the vibration signal after windowed short-time Fourier transform, represents the window function, represents the vibration signal, t represents the current time, represents the conjugate of, represents that the Fourier window function is translated at time , e represents the natural constant, j represents the imaginary unit, represents the angular frequency.

[0038] In the embodiments of the present application, the window function is a Hanning window function. In other embodiments, it can also be other window functions such as a Gaussian window function, which can further narrow the window extracted by the short-time Fourier transform to obtain a more accurate and interference-free sub-signal segment. By performing short-time Fourier transform on the vibration signal within the local area covered by the Hanning window function and continuously translating the Hanning window on the time axis, the spectral characteristics of the signal at different times can be captured, thereby realizing time-frequency analysis of the vibration signal.

[0039] Since the vibration signal is composed of signals at continuous time sampling points and is essentially a discrete signal, it is necessary to convert the windowed short-time Fourier transform into a discrete form of the windowed short-time Fourier transform. Set the discrete signal of the vibration signal as , then the discrete form of the windowed short-time Fourier transform is: Wherein, represents the discrete form of, N represents the number of points of the Fourier transform, represents the window function of the discrete signal, n represents the number of time sampling points, denotes the Fourier window function, centered at the time point m , k denotes the discrete frequency index, denotes the frequency.

[0040] In the embodiment of the present application, taking the waveform data of numerical simulation as an example, the above time-frequency analysis is carried out to show the process and results of data processing. Among them, the numerical simulation method adopts the elastic wave staggered grid finite difference method. According to the actual fracturing mechanism, a "wellbore-fracture" model is set up, as Figure 2 shown. The model size is 20m×20m. Assuming that the underground medium is uniform, the longitudinal wave velocity is set to 2300m / s, the transverse wave velocity is set to 1550m / s, and the density is set to 2500kg / m3. The pipeline and fractures are filled with fluid. Assuming that the filled fluid is water, the longitudinal wave velocity of water is 1490m / s, the transverse wave velocity is 0m / s, and the density is 1000kg / m3. As the hydraulic fracturing process progresses, more fractures may be generated. Therefore, the situation where the number of fractures gradually increases with the fracturing process is further simulated in the model.

[0041] During the simulation process of the embodiment of the present application, a water hammer source is used to excite multiple times at the wellhead. The source function adopts the Ricker wavelet function commonly used in elastic wave numerical simulation. The source function is: Among them, s ( t ) denotes the source function, f 0 denotes the main frequency of the source, t 0 denotes the initial excitation time, π denotes the pi, and e denotes the natural constant.

[0042] Obtain the shot gather records received from the surface to the depth in the whole well, and obtain the numerical simulation gather as Figure 3 shown. The arrow in the figure indicates the response generated when the Krauklis wave generated in the fracture propagates to the connection between the fracture and the wellbore.

[0043] In the embodiment of the present application, the main frequency of the numerical simulation is set to 1000Hz.

[0044] It should be noted that the shot gather record refers to the seismic wave data record received by multiple geophones after being excited by one source.

[0045] Furthermore, select Figure 3 the first trace waveform data in Figure 4 as the simulated single-trace waveform data collected at the wellhead, as Figure 3From the record, it can be seen that the response of the Krauklis wave appears only after 0.04 s. Before 0.04 s, they are all the spectra of the source or its reflection waveform at the crack. Therefore, only the data from 0.04 s to 0.2 s are analyzed. The time range of the window function is from 0.04 s to 0.2 s. The windowed signal is as Figure 5 shown.

[0046] As the fracturing process progresses, the number of cracks will increase synchronously. Therefore, in the embodiments of the present application, the time-frequency response characteristics under different numbers of cracks are analyzed, and the time-frequency spectrogram as shown in Figures 6 to 9 is obtained. Among them, Figure 6 is the time-frequency spectrogram without cracks before hydraulic fracturing, Figure 7 is the time-frequency spectrogram with a single crack after hydraulic fracturing, Figure 8 is the time-frequency spectrogram with double cracks after hydraulic fracturing, Figure 9 is the time-frequency spectrogram with triple cracks after hydraulic fracturing.

[0047] In some embodiments, before data analysis, it is first necessary to preprocess the signal, which is crucial for the accuracy and reliability of subsequent analysis. This step includes decoding data, matching time, and eliminating noise. (1) Decoding data: Decode the recorded waveform data into a SEGY format file commonly used in exploration seismology and save the output. (2) Matching time: This step aligns the time points of the seismic records with the start and end times of the fracturing operation to facilitate the analysis of the target frequency band range. (3) Eliminating noise: Noise is the unwanted interference in the signal, which will mask or distort the useful signal. To improve the signal quality, it is necessary to eliminate the obvious interference signals during the recording process. The filtering method can be used to remove high-frequency noise and limit the analysis frequency range to the empirical low-frequency region.

[0048] Step S30: Obtain the resonance energy amplitude statistical chart of the vibration signal by extracting the spectral energy amplitude of the time-frequency spectrogram, where the resonance energy amplitude statistical chart is used to characterize the intensity of the resonance energy at different time periods.

[0049] Optionally, the time-frequency spectrogram shows the energy distribution of the vibration signal at different times and frequencies. By extracting the energy amplitude, the intensity of the resonance energy is obtained, and the resonance energy amplitude statistical chart is generated.

[0050] As a possible implementation manner, extract the spectral energy amplitude at different times in the time-frequency spectrogram; within the same time period, superimpose the spectral energy amplitudes within the preset frequency band range to generate the resonance energy amplitude; generate the resonance energy amplitude statistical chart based on the resonance energy amplitude and the corresponding time period.

[0051] Among them, the extraction of the spectral energy amplitude can be: calculating the amplitude spectrum of the windowed short-time Fourier result, and calculating the energy spectrum based on the amplitude spectrum, that is, calculating the square of the amplitude as the energy. The specific calculation formula is the prior art and will not be elaborated here.

[0052] Furthermore, within the same time period, the spectral energy amplitudes within the preset frequency band range are superimposed to generate the resonance energy amplitude.

[0053] As a possible implementation manner, the resonance energy amplitude can be determined by the following formula: Wherein, represents the resonance energy amplitude with the time parameter being , f 1 represents the lowest frequency of the preset frequency band range, f 2 represents the highest frequency of the preset frequency band range, represents the spectral energy amplitude with the time parameter being .

[0054] As an example, the preset frequency band range ( f 1 , f 2 ) can be (0, 3000 Hz). In other embodiments, the preset frequency band range can be determined according to the actual hydraulic fracturing situation.

[0055] Furthermore, taking time and the resonance energy amplitude as coordinate axes, a resonance energy amplitude statistical graph is constructed.

[0056] Step S40, when the resonance energy amplitude statistical graph meets the preset conditions, it is determined that the hydraulic fracturing is successful. Among them, the preset conditions include: after the average value of the resonance energy amplitude within the set duration changes from the first amplitude to the second amplitude, it remains within the set range centered on the third amplitude, and the second amplitude is higher than the first amplitude and the third amplitude.

[0057] It should be noted that during the process of staged fracturing, in the perforation stage, natural fractures dominate in the formation, and the effective fracture energy is relatively low; in the fracturing stage, as the pressure increases, the scale of the induced fractures in the formation gradually expands, and the resonance energy increases; as the injection mud rate stabilizes, the fracturing pressure decreases, proppants are gradually injected, and the relative scale of the effective fractures tends to be stable, and the resonance energy tends to be stable. Therefore, the curve change trend of the resonance energy first rises, then falls, and finally tends to be stable.

[0058] Optionally, evaluate the changing trend of the resonance energy amplitude statistical graph. If it conforms to the changing trend of the staged fracturing process, determine that the hydraulic fracturing is successful; otherwise, the hydraulic fracturing fails and the hydraulic fracturing operation needs to be carried out again.

[0059] As a possible implementation manner, the change of the resonance energy amplitude statistical graph is reflected by monitoring the amplitude average value of the resonance energy amplitude statistical graph within a preset time period, that is, the average value of the resonance energy amplitude within the set time period is statistically calculated. After the energy amplitude average value changes from the first amplitude to the second amplitude and then remains within the set range centered on the third amplitude, it indicates that the resonance energy change process conforms to the changing trend of the staged fracturing, where the second amplitude is higher than the first amplitude and the third amplitude.

[0060] It can be understood that during the normal staged fracturing process, the changing trend of the resonance energy is to increase first, then decrease, and finally tend to be stable. By statistically calculating the energy amplitude average value to reflect the change of the resonance energy, if the energy amplitude average value increases from the first amplitude to the second amplitude, it indicates that the resonance energy changes to increase. Until the energy amplitude average value decreases from the second amplitude to the third amplitude and remains within a certain fluctuation range of the third amplitude, it indicates that the resonance energy tends to be stable, and this change conforms to the changing trend of the resonance energy.

[0061] Among them, the set range is (P 3 -P 0 , P 3 +P 0 ), P 3 is the third amplitude, and P 0 is the allowable fluctuation value. The energy amplitude average value fluctuates within the set range, indicating that the resonance energy has tended to be stable.

[0062] As a possible implementation manner, input the resonance energy amplitude statistical graph into the trained hydraulic fracturing monitoring neural network, and output the monitoring result of the hydraulic fracturing. Among them, the training data of the hydraulic fracturing monitoring neural network is the historical resonance energy amplitude statistical graphs of multiple historical vibration signals during the historical fracturing process.

[0063] In the embodiments of the present application, the monitoring result of the hydraulic fracturing is success or failure. According to the model prediction result, provide real-time decision support for the on-site fracturing operation. For example: if the prediction is "success", it can be considered that the fracturing of this fracturing stage is completed, and the fracturing operation of the subsequent fracturing stage can be carried out; if the prediction is "failure", adjust the fracturing parameters in time (such as increasing the injection volume of the fracturing fluid, adjusting the fracturing pressure, etc.) and re-perform the fracturing to improve the fracturing effect.

[0064] Among them, the training process of the neural network includes: Obtain historical vibration signals generated during historical hydraulic fracturing processes; extract the spectral energy amplitudes from the historical time-frequency spectrograms of any historical vibration signal to obtain the historical resonance energy amplitude statistical graphs of any historical vibration signal; use multiple historical resonance energy amplitude statistical graphs as the training set and the hydraulic fracturing evaluation results of the historical resonance energy amplitude statistical graphs as labels to train a residual neural network to obtain a trained hydraulic fracturing monitoring neural network, where the loss function of the residual neural network is the cross-entropy loss function.

[0065] Optionally, the historical vibration signals are collected by geophones during historical hydraulic fracturing processes, and also include the resonance information of Krauklis waves. The number of historical vibration signals is multiple.

[0066] In the embodiments of the present application, simulated vibration signals can also be obtained by simulating a "wellbore-fracture" model as historical vibration signals to expand the data volume.

[0067] It should be noted that the specific steps for obtaining the historical resonance energy amplitude statistical graph based on the historical vibration signal refer to the processes of steps S10 to S30 and will not be elaborated here.

[0068] In the embodiments of the present application, the residual neural network selects the ResNet-18 network and is implemented using the PyTorch architecture, replacing the last fully connected layer to adapt to the classification task.

[0069] Furthermore, data annotation is performed on any historical resonance energy amplitude statistical graph, and the labels are divided into two categories: "success" and "failure". For example, if the oil and gas production increases significantly after fracturing and the energy statistical curve meets the expectations (such as the average energy amplitude increases from the first amplitude to the second amplitude, then decreases from the second amplitude to the third amplitude and remains within a certain fluctuation range of the third amplitude), it is labeled as "success"; conversely, if the production after fracturing fails to meet the expectations and the energy curve is abnormal (such as the energy does not increase significantly or fluctuates greatly and is unstable), it is labeled as "failure".

[0070] Optionally, use multiple historical resonance energy amplitude statistical graphs as the training set and the data annotation results of the historical resonance energy amplitude statistical graphs as labels. The label data is divided into two categories: "success" and "failure". Use the cross-entropy loss function and the Adam optimizer to train the model until the loss value of the loss function is less than 0.01 to obtain a trained hydraulic fracturing monitoring neural network.

[0071] Optionally, before obtaining the historical vibration signals generated during the historical hydraulic fracturing process, perform staged fracturing operations on the target well, and record the start time and end time of each staged fracturing operation. After obtaining the historical vibration signals generated during the historical hydraulic fracturing process, align the start time and end time of the collected historical vibration signals with the start time and end time of the staged fracturing operations respectively.

[0072] Optionally, filter and denoise the historical vibration signals.

[0073] As another possible implementation, in response to the end of the hydraulic fracturing stage, when the resonance energy amplitude exceeds a threshold, it is determined that the hydraulic fracturing is successful, where the threshold is determined by statistically analyzing the background noise of the time-frequency spectrogram.

[0074] In the embodiments of the present application, the threshold is 5 times the average value of the background noise amplitude of the time-frequency spectrogram. When the resonance energy amplitude exceeds the threshold, it indicates that a fracture propagation signal with sufficient intensity is generated during the fracturing process, indicating good fracturing effect. Otherwise, it means that the fracturing is incomplete and subsequent re-fracturing needs to be considered.

[0075] It should be noted that the method of the present application has also been applied in actual fracturing scenarios and achieved good results. The experimental site is a fracturing well in a desert and gobi area in Xinjiang. The depth of the fracturing well is 1300m - 1400m, the horizontal length is 841.4m, and the fracturing process is recorded sequentially by staged fracturing. There are a total of 26 fracturing stages, and the average length of each fracturing stage is 38.4m. The well section design model is as Figure 10 shown.

[0076] Select the data of the 15th fracturing stage as an example for analysis. After preprocessing the vibration signals, input them into the windowed short-time Fourier transform program for time-frequency analysis, output the time-frequency spectrogram, perform resonance energy statistics, and obtain the time-frequency spectrogram example diagram as shown in Figure 11 and the resonance energy amplitude statistical chart as shown in Figure 12 . Input the resonance energy amplitude statistical chart as shown in Figure 12 into the trained neural network for hydraulic fracturing monitoring, and the output prediction result is "successful".

[0077] It should be noted that from Figure 11 it can be seen that the energy in the A - B section is relatively low, which is very likely the perforation stage. The energy in the B - C section gradually increases, which is very likely the fracturing stage. The energy in the C - D section tends to be stable, which is very likely the stage of proppant injection.

[0078] Furthermore, a visualization interface can be developed to display the resonance energy amplitude statistical chart, the model prediction result, and other key parameters during the fracturing process (such as fracturing pressure, proppant injection volume, etc.) in real time, facilitating on-site engineers to intuitively understand the hydraulic fracturing process.

[0079] Although the steps are described in the above - mentioned order in the above - mentioned embodiments, those skilled in the art can understand that, in order to achieve the effects of this embodiment, different steps do not have to be executed in such an order. They can be executed simultaneously (in parallel) or in a reversed order, and these simple changes are within the protection scope of this application.

[0080] The embodiment of this application acquires the vibration signal during the hydraulic fracturing process of the target well, can collect the relevant signals of Krauklis waves generated during the hydraulic fracturing process, and avoids the problem that the direct acquisition of Krauklis waves is affected by underground media; extracts multiple sub - signal segments of the vibration signal according to time sequence, and performs time - frequency analysis on each sub - signal segment by using short - time Fourier transform to obtain the time - frequency spectrogram of the vibration signal. By extracting the sub - signal segments, the target signal segment to be analyzed is analyzed separately, and the spectral characteristics at different moments are captured through short - time Fourier transform; by extracting the spectral energy amplitude of the time - frequency spectrogram, a resonance energy amplitude statistical chart of the vibration signal is obtained. Among them, the resonance energy amplitude statistical chart is used to characterize the intensity of resonance energy at different time periods, can intuitively reflect the strength of resonance energy, and is convenient for signal analysis; when the resonance energy amplitude statistical chart meets the preset conditions, it is determined that the hydraulic fracturing is successful. Among them, the preset conditions include: after the average value of the resonance energy amplitude within a set time period changes from the first amplitude to the second amplitude, it remains within the set range centered on the third amplitude, and the second amplitude is higher than the first amplitude and the third amplitude. By statistically analyzing the change of resonance energy, it is possible to intuitively and quickly evaluate the hydraulic fracturing process based on the resonance characteristics of Krauklis waves, realize the dynamic monitoring of the hydraulic fracturing process, improve the monitoring effect of the hydraulic fracturing process, and provide a guiding technology for the hydraulic fracturing operation.

[0081] Please refer to Figure 13 a hydraulic fracturing monitoring system based on Krauklis wave resonance according to the second embodiment of this application, including: a vibration signal acquisition module 100, a time - frequency analysis module 200, a resonance energy amplitude statistical chart acquisition module 300, and a hydraulic fracturing monitoring module 400.

[0082] The vibration signal acquisition module 100 is used to acquire the vibration signal during the hydraulic fracturing process of the target well, where the vibration signal is converted from Krauklis waves, and Krauklis waves are generated during the hydraulic fracturing process; The time - frequency analysis module 200 is used to extract multiple sub - signal segments of the vibration signal according to time sequence, and perform time - frequency analysis on each sub - signal segment by using short - time Fourier transform to obtain the time - frequency spectrogram of the vibration signal; The resonance energy amplitude statistical graph acquisition module 300 is configured to obtain a resonance energy amplitude statistical graph of the vibration signal by extracting the spectral energy amplitude of the time-frequency spectrogram, where the resonance energy amplitude statistical graph is used to characterize the intensity of the resonance energy at different time periods; The hydraulic fracturing monitoring module 400 is configured to determine that the hydraulic fracturing is successful when the resonance energy amplitude statistical graph meets a preset condition, where the preset condition includes: after the average value of the resonance energy amplitude within a set time period changes from a first amplitude to a second amplitude and then remains within a set range centered on a third amplitude, and the second amplitude is higher than the first amplitude and the third amplitude.

[0083] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working process and related explanations of the above-described system can refer to the corresponding process in the foregoing method embodiments, and will not be elaborated herein.

[0084] It should be noted that for the above-described hydraulic fracturing monitoring system based on Krauklis wave resonance provided in the above embodiments, only the above-described division of each functional module is used as an example for illustration. In practical applications, the above functions can be allocated to different functional modules as needed, that is, the modules or steps in the embodiments of the present application can be further decomposed or combined. For example, the modules in the above embodiments can be combined into one module, or further split into multiple sub-modules to complete all or part of the functions described above. For the names of the modules and steps involved in the embodiments of the present application, they are only used to distinguish each module or step, and are not regarded as an improper limitation of the present application.

[0085] An electronic device according to a third embodiment of the present application includes: At least one processor; and A memory communicatively connected to at least one of the processors; where The memory stores instructions executable by the processor, and the instructions are used to be executed by the processor to implement the above-described hydraulic fracturing monitoring method based on Krauklis wave resonance.

[0086] A computer-readable storage medium according to a fourth embodiment of the present application, where the computer-readable storage medium stores computer instructions, and the computer instructions are used to be executed by the computer to implement the above-described hydraulic fracturing monitoring method based on Krauklis wave resonance.

[0087] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working process and related explanations of the above-described electronic device and computer-readable storage medium can refer to the corresponding process in the foregoing method embodiments, and will not be elaborated herein.

[0088] Those skilled in the art should be able to realize that the modules and method steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of both. The programs corresponding to the software modules and method steps can be placed in a random access memory (RAM), internal memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the technical field. To clearly illustrate the interchangeability of electronic hardware and software, the composition and steps of each example have been generally described according to functions in the above description. Whether these functions are executed in the form of electronic hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0089] Computer program code for performing the operations of this application can be written in one or more programming languages or combinations thereof. The above programming languages include object-oriented programming languages such as Java, Smalltalk, C++, and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, executed as an independent software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computer (for example, by using an Internet service provider to connect through the Internet).

[0090] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in the flowchart or block diagram can represent a module, a program segment, or a part of code that contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than that marked in the accompanying drawings. For example, two consecutive blocks shown can actually be executed substantially in parallel, and they can sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, as well as the combination of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0091] The terms "first", "second", etc. are used to distinguish similar objects, rather than to describe or indicate a specific order or sequence.

[0092] The term "comprising" or any other similar term is intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus / device that comprises a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to those process, method, article, or apparatus / device.

[0093] So far, the technical solution of the present application has been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present application is obviously not limited to these specific embodiments. Without departing from the principle of the present application, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the protection scope of the present application.

Claims

1. A hydraulic fracturing monitoring method based on Krauklis wave resonance, characterized in that: The following steps are involved: Step S10, obtaining a vibration signal of the target well during the hydraulic fracturing process, wherein the vibration signal is obtained by converting a Krauklis wave, and the Krauklis wave is generated during the hydraulic fracturing process; Step S20, extracting multiple sub-signal segments of the vibration signal in time sequence, and performing time-frequency analysis on each sub-signal segment using short-time Fourier transform to obtain a time-frequency spectrum of the vibration signal; Step S30, obtaining a resonance energy amplitude statistical graph of the vibration signal by extracting the spectrum energy amplitude of the time spectrum graph, wherein the resonance energy amplitude statistical graph is used to characterize the intensity of the resonance energy in different time periods; Step S40, when the resonance energy amplitude statistical graph meets the preset conditions, it is determined that the hydraulic fracturing is successful, wherein the preset conditions include: after the average value of the resonance energy amplitude within a set time period changes from the first amplitude to the second amplitude, it remains within a set range centered on the third amplitude, and the second amplitude is higher than the first amplitude and the third amplitude.

2. A hydraulic fracturing monitoring method based on Krauklis wave resonance according to claim 1, characterized in that: The step of extracting a plurality of sub-signal segments of the vibration signal in time sequence and performing time-frequency analysis on each sub-signal segment by using short-time Fourier transform includes: Performing a windowed short-time Fourier transform on the vibration signal includes: ； in, represents the transformation result of windowed short-time Fourier transform of the vibration signal, represents the window function, represents the vibration signal, t Indicates the current moment, express The conjugate of Represents the Fourier window function in time is translated, e represents a natural constant, j represents the imaginary unit, represents the angular frequency; Convert the windowed short-time Fourier transform into a discrete form of the windowed short-time Fourier transform, and set the vibration signal The discrete signal is ,include: ； in, express The discrete form of N represents the number of points of Fourier transform, Represents the window function of a discrete signal, n Indicates the number of time sampling points, represents the Fourier window function, centered at the time point m , k represents a discrete frequency index, Indicates frequency.

3. The hydraulic fracturing monitoring method based on Krauklis wave resonance according to claim 1, characterized in that: The step of obtaining a resonance energy amplitude statistical graph of the vibration signal by extracting the spectrum energy amplitude of the time-frequency spectrum graph comprises: Extracting spectrum energy amplitudes at different times in the time-spectrum diagram; In the same period, the spectrum energy amplitudes within the preset frequency band are superimposed to generate a resonance energy amplitude; The resonance energy amplitude statistical graph is generated based on the resonance energy amplitude and the corresponding time period.

4. The hydraulic fracturing monitoring method based on Krauklis wave resonance according to claim 1, characterized in that: When the resonance energy amplitude statistical graph meets the preset conditions, determining that the hydraulic fracturing is successful includes: The resonance energy amplitude statistical graph is input into a trained hydraulic fracturing monitoring neural network, and a monitoring result of hydraulic fracturing is output, wherein the training data of the hydraulic fracturing monitoring neural network is a historical resonance energy amplitude statistical graph of multiple historical vibration signals in historical fracturing processes.

5. A hydraulic fracturing monitoring method based on Krauklis wave resonance according to claim 4, characterized in that: The training process of the neural network includes: Obtain historical vibration signals generated during historical hydraulic fracturing; By extracting the spectrum energy amplitude of a historical time-frequency spectrum diagram of any historical vibration signal, a historical resonance energy amplitude statistical diagram of any historical vibration signal is obtained; Taking multiple historical resonance energy amplitude statistical graphs as training sets and hydraulic fracturing assessment results of the historical resonance energy amplitude statistical graphs as labels, a residual neural network is trained to obtain a trained hydraulic fracturing monitoring neural network, wherein the loss function of the residual neural network is a cross entropy loss function.

6. A hydraulic fracturing monitoring method based on Krauklis wave resonance according to claim 5, characterized in that: Before obtaining the historical vibration signal generated in the historical hydraulic fracturing process, the method includes: The target well is subjected to staged fracturing operations, and the start time and end time of each staged fracturing operation are recorded.

7. A hydraulic fracturing monitoring method based on Krauklis wave resonance according to claim 6, characterized in that: After obtaining the historical vibration signal generated during the historical hydraulic fracturing process, the method includes: Aligning the start time and the end time of the historical vibration signal collection with the start time and the end time of the staged fracturing operation respectively; The historical vibration signal is filtered and denoised.

8. The hydraulic fracturing monitoring method based on Krauklis wave resonance according to claim 1, characterized in that: The method further comprises: In response to the hydraulic fracturing stage ending, when the resonance energy amplitude exceeds a threshold, it is determined that the hydraulic fracturing is successful, wherein the threshold is determined by counting the background noise of the time-frequency spectrum.

9. The method for hydraulic fracturing monitoring based on Krauklis wave resonance according to claim 1, characterized in that: The vibration signal is a tube wave signal. After obtaining the vibration signal of the target well during the hydraulic fracturing process, the method includes: The vibration signal is decoded from a waveform signal into a SEGY format file for storage.

Images