Underground crack propagation event identification method, device and equipment

Abstract

The embodiment of the invention relates to the technical field of hydraulic fracturing, in particular to an underground crack propagation event recognition method, device and equipment, and the method comprises the steps that fracturing signals in the fracturing operation process are obtained, the fracturing signals are used for representing the change relation between construction parameters and time, and the construction parameters at least comprise pressure and displacement; the fracturing signals are subjected to wavelet decomposition, pressure energy data and displacement energy data are obtained, the pressure energy data represent the change relation between pressure energy and time, and the displacement energy data represent the change relation between displacement energy and time; screening the pressure energy data by using the displacement energy data so as to screen out a pressure energy peak value corresponding to a non-rock event in the pressure energy data; and according to the screened pressure energy data, identifying an underground crack propagation event. According to the embodiment of the invention, the pressure energy peak value corresponding to the non-rock event in the pressure energy data is screened out by utilizing the displacement energy data, so that the reliability and the precision of identifying the underground crack propagation event are remarkably improved.

Description

Technical Field

[0001] The embodiments of this specification relate to the technical fields of oil and gas field development and hydraulic fracturing, and more particularly to a method, device, and apparatus for identifying downhole fracture propagation events. Background Art

[0002] Hydraulic fracturing is one of the most widely used and effective production-enhancing measures in shale oil and gas reservoir development. By injecting high-pressure fluid into the formation, a complex network of fractures is created in the rock surrounding the wellbore, thereby increasing reservoir permeability and boosting oil and gas production. To further optimize fracturing design and enhance reservoir stimulation effectiveness, real-time monitoring and assessment of downhole fracture expansion during the fracturing process is necessary.

[0003] Currently, the main methods for monitoring and identifying downhole fracture growth include microseismic monitoring and fiber optic monitoring. Microseismic monitoring captures microseismic signals released during rock fracture, providing a relatively intuitive picture of the distribution, extension direction, and complexity of fractures, providing a crucial basis for evaluating the effectiveness of hydraulic fracturing. Fiber optic monitoring, utilizing distributed fiber optic sensing technology, accurately measures changes in optical signals around the wellbore in real time, enabling accurate information on the dynamic growth of fractures, including parameters such as length, height, and width. Both technologies have demonstrated high monitoring accuracy and reliability in practical applications, providing crucial data support for oilfield development. However, despite their significant advantages in fracture growth monitoring, microseismic and fiber optic monitoring still face several challenges in their application. First, the equipment costs of both technologies are high. In particular, fiber optic monitoring requires a dedicated fiber optic sensing system with stringent installation and maintenance requirements, resulting in a high overall investment. Second, monitoring effectiveness is significantly affected by formation conditions. For example, in complex geological structures or low-permeability formations, the propagation of microseismic signals may be disrupted, resulting in reduced monitoring accuracy. Furthermore, fiber optic monitoring can also be affected by high temperatures, high pressures, or corrosive environments, compromising its stability and durability. In addition, the manually calibrated fracturing signal of the fracture extension stage, the accuracy and resolution of the instrument are crucial to the monitoring results. Relying on researchers to manually calibrate the fracture extension stage based on experience and interpret the analysis results based on experience may lead to data errors, thereby affecting the accurate assessment of fracture extension.

[0004] Therefore, how to solve the problems of existing methods such as reliance on manual experience calibration, strong requirements on formation conditions and instrument precision, and low recognition accuracy, and propose a downhole fracture propagation event identification method that does not rely on manual experience calibration, has low requirements on formation conditions and instrument precision, and has high accuracy, is a key issue that needs to be solved urgently. Summary of the Invention

[0005] The purpose of the embodiments of this specification is to provide a method, device and equipment for identifying downhole fracture propagation events to overcome the problems of existing downhole fracture propagation event identification methods that rely on manual experience calibration, have strong requirements on formation conditions and instrument accuracy, and have low identification accuracy.

[0006] On the one hand, an embodiment of the present specification provides a method for identifying downhole crack expansion events, including: obtaining a fracturing signal during a fracturing operation, wherein the fracturing signal is used to represent the changing relationship between construction parameters and time, and the construction parameters include at least pressure and displacement; performing wavelet decomposition on the fracturing signal to obtain pressure-energy data and displacement energy data, wherein the pressure-energy data represents the changing relationship between pressure energy and time, and the displacement energy data represents the changing relationship between displacement energy and time; using the displacement energy data to filter the pressure energy data to filter out pressure energy peaks corresponding to non-rock events in the pressure energy data; and identifying downhole crack expansion events based on the filtered pressure energy data.

[0007] On the other hand, an embodiment of the present specification provides a downhole crack expansion event identification device, including: an acquisition module, used to acquire a fracturing signal during a fracturing operation, wherein the fracturing signal is used to represent the changing relationship between construction parameters and time, and the construction parameters include at least pressure and displacement; a decomposition module, used to perform wavelet decomposition on the fracturing signal to obtain pressure energy data and displacement energy data, wherein the pressure energy data represents the changing relationship between pressure energy and time, and the displacement energy data represents the changing relationship between displacement energy and time; a screening module, used to use the displacement energy data to screen out the pressure energy data to screen out the pressure energy peaks corresponding to non-rock events in the pressure energy data; and an identification module, used to identify downhole crack expansion events based on the screened pressure energy data.

[0008] On the other hand, a computer device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the above-mentioned method for identifying downhole fracture propagation events.

[0009] It can be seen from the technical solutions provided in the above embodiments of this specification that the embodiments of this specification can obtain a fracturing signal during a fracturing operation, wherein the fracturing signal is used to represent the changing relationship between construction parameters and time, and the construction parameters include at least pressure and displacement; the fracturing signal is subjected to wavelet decomposition to obtain pressure energy data and displacement energy data, wherein the pressure energy data represents the changing relationship between pressure energy and time, and the displacement energy data represents the changing relationship between displacement energy and time; the pressure energy data is filtered using the displacement energy data to screen out the pressure energy peaks corresponding to non-rock events in the pressure energy data; and downhole crack expansion events are identified based on the filtered pressure energy data. Compared with existing methods, the embodiments of this specification can directly obtain the fracturing signal during the fracturing operation to identify downhole crack expansion events, without relying on refined data equipment and without refined data annotation, and the data acquisition cost is low; by performing wavelet decomposition on the fracturing signal, the time-frequency domain characteristics in the fracturing signal can be effectively extracted, and the signal-to-noise ratio of the fracturing signal can be improved, thereby reducing the interference of noise signals in the formation on the identification of downhole crack expansion events; by using displacement energy data to screen out the pressure energy peaks corresponding to non-rock events in the pressure energy data, the interference of non-rock events can be effectively eliminated, and the reliability and accuracy of downhole crack expansion event identification can be significantly improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the embodiments of this specification or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art.

[0011] Figure 1 This is a flow chart of a method for identifying downhole fracture propagation events provided by an embodiment of this specification;

[0012] Figure 2 This is a schematic diagram of the overall logic flow of a method for identifying downhole fracture propagation events provided by an embodiment of this specification;

[0013] Figure 3 Schematic diagram of a fracturing signal collected in Well A provided in an embodiment of this specification;

[0014] Figure 4 Schematic diagram of a sub-fracture signal corresponding to the fracture expansion stage captured from Well A provided in an embodiment of this specification;

[0015] Figure 5 It is the decomposed signal obtained by performing a 7-level wavelet decomposition on the sub-fracture signal corresponding to the fracture propagation stage of Well A using the optimal wavelet basis corresponding to Well A provided in the embodiments of this specification;

[0016] Figure 6It is a decomposed signal obtained by performing an 8-level wavelet decomposition on the sub-fracture signal corresponding to the fracture propagation stage of Well A using the optimal wavelet basis corresponding to Well A provided in the embodiments of this specification;

[0017] Figure 7 It is a decomposed signal obtained by performing a 9-level wavelet decomposition on the sub-fracture signal corresponding to the fracture propagation stage of Well A using the optimal wavelet basis corresponding to Well A provided in the embodiments of this specification;

[0018] Figure 8 This is a schematic diagram of the pressure energy peak in the pressure energy data corresponding to the 8-level wavelet decomposition of the sub-fracture signal corresponding to the fracture expansion stage of Well A provided in the embodiment of this specification;

[0019] Figure 9 This is a schematic diagram of the pressure energy peak in the pressure energy data after screening out the pressure energy peak corresponding to non-rock events after 8-level wavelet decomposition of the sub-fracture signal corresponding to the fracture propagation stage of Well A provided in the embodiment of this specification;

[0020] Figure 10 This is a schematic diagram of the pressure energy peak value in the pressure energy data corresponding to the 9-level wavelet decomposition of the sub-fracture signal corresponding to the fracture expansion stage of Well A provided in the embodiment of this specification;

[0021] Figure 11 Schematic diagram of the pressure energy peak value in the pressure energy data corresponding to the sub-fracture signal corresponding to the fracture expansion stage of Well A provided in the embodiment of this specification after the 9-level wavelet decomposition and the screening out of the pressure energy peak value corresponding to the random noise;

[0022] Figure 12 This is a schematic diagram of the pressure energy peak corresponding to the fracture extension event in the pressure energy data after 7-level wavelet decomposition of the sub-fracture signal corresponding to the fracture extension stage of Well A provided in the embodiment of this specification;

[0023] Figure 13 This is a schematic diagram of the pressure energy peak corresponding to the fracture extension event in the pressure energy data after 8-level wavelet decomposition of the sub-fracture signal corresponding to the fracture extension stage of Well A provided in the embodiment of this specification;

[0024] Figure 14 This is a schematic diagram of the structure of a downhole fracture propagation event identification device provided in an embodiment of this specification;

[0025] Figure 15 It is a schematic diagram of the structural composition of the computer device provided in the embodiment of this specification. DETAILED DESCRIPTION

[0026] The following will be combined with the drawings in the embodiments of this specification to clearly and completely describe the technical solutions in the embodiments of this specification. Obviously, the embodiments described are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments in this specification, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of this specification.

[0027] Figure 1 This is a flow chart of a method for identifying downhole fracture propagation events provided in an embodiment of this specification. Figure 2 This is a schematic diagram of the overall logic flow of a method for identifying downhole fracture propagation events provided in an embodiment of this specification. The specific implementation includes the following steps:

[0028] S101: Acquire a fracturing signal during a fracturing operation, where the fracturing signal is used to represent a relationship between operation parameters and time, and the operation parameters include at least pressure and displacement.

[0029] In some embodiments, a fracturing signal can be collected during the fracturing operation in a target well. The fracturing signal can be collected in real time by sensors installed on the target well, including pressure sensors and displacement sensors. The sampling frequency of the pressure and displacement sensors can be, for example, 1 Hz. The pressure sensor is used to monitor the pressure changes of the fracturing fluid in the wellbore and formation, while the displacement sensor is used to monitor the injection rate of the fracturing fluid. The fracturing signal is used to accurately characterize the temporal relationship between operation parameters. Operation parameters may include at least pressure and displacement, where pressure refers to the fluid pressure exerted by the fracturing fluid in the wellbore and formation, and displacement refers to the volume of fracturing fluid injected into the formation per unit time. Specifically, the fracturing signal can include pressure time series data collected by the pressure sensor and displacement time series data collected by the displacement sensor. These pressure and displacement time series data can reflect the dynamic characteristics and operation results of the fracturing operation. Furthermore, operation parameters may also include key parameters such as sand ratio, fluid viscosity, and temperature. Accordingly, the fracturing signal can include time series data of key parameters such as sand ratio, fluid viscosity, and temperature. Real-time monitoring and analysis of these fracturing signals allows for timely adjustments to construction strategies, optimizing fracturing results and ensuring operational safety and effectiveness. The changing trends and characteristics of fracturing signals can also be used to assess formation response and predict fracture propagation patterns, providing a scientific basis for subsequent fracturing design and construction.

[0030] S102: performing wavelet decomposition on the fracturing signal to obtain pressure energy data and displacement energy data, wherein the pressure energy data represents a changing relationship between pressure energy and time, and the displacement energy data represents a changing relationship between displacement energy and time.

[0031] By using wavelet decomposition on the fracturing signal, the time-frequency domain features in the fracturing signal can be effectively extracted, the signal-to-noise ratio of the fracturing signal can be improved, and the interference of noise signals in the formation on the identification of downhole fracture extension events can be reduced.

[0032] In some embodiments, the fracturing signal can be subjected to wavelet decomposition to obtain wavelet coefficients for each subband. The square sum of the wavelet coefficients for each subband is calculated to obtain the energy value for that subband. The energy values for each subband are arranged in chronological order to obtain energy data. Specifically, the fracturing signal can include pressure time series data. Wavelet decomposition is performed on the pressure time series data to obtain wavelet coefficients for each subband. The pressure energy values for each subband are arranged in chronological order to obtain pressure energy data. Pressure energy data represents the relationship between pressure energy and time. The fracturing signal can also include displacement time series data. Similarly, wavelet decomposition is performed on the displacement time series data to obtain wavelet coefficients for each subband. The displacement energy values for each subband are arranged in chronological order to obtain displacement energy data. Displacement energy data represents the relationship between displacement energy and time. By analyzing the pressure energy data and displacement energy data, key features in the fracturing signal, such as fracture propagation events and rock failure events, can be identified. This energy data can also be used to distinguish different types of fracturing events, such as rock-related events and non-rock-related events, thereby improving the accuracy and reliability of fracture propagation event identification. Rock events are events involving rocks and related geological bodies caused by geological processes (tectonic, magmatic, metamorphic, etc.). They primarily occur within the Earth's crust or on the surface and have long time spans. Non-rock events are events involving natural or man-made objects other than rocks, caused by meteorological, hydrological, biological, human, or astronomical activities. They primarily occur on the Earth's surface, in the atmosphere, in water bodies, or in space and have shorter time spans.

[0033] S103: Filtering the pressure energy data using the displacement energy data to remove pressure energy peaks corresponding to non-rock events in the pressure energy data.

[0034] In some embodiments, the maximum time delay of the pressure response caused by the displacement change can be obtained; for each pressure energy peak in the pressure energy data, a displacement energy peak that meets the adjacent conditions is matched in the displacement energy data; the adjacent conditions constrain that the time difference between the pressure energy peak and the matched displacement energy peak is less than or equal to the time difference between the pressure energy peak and any other displacement energy peak in the displacement energy data; for each pressure energy peak in the pressure energy data, the time difference between the pressure energy peak and the matched displacement energy peak is obtained, and if the time difference is less than the maximum time delay, it is determined that the pressure energy peak corresponds to a non-rock event; and the pressure energy peaks corresponding to non-rock events in the pressure energy data are screened out.

[0035] By obtaining the maximum time delay of the pressure response caused by displacement changes, the fluctuation of the pressure signal caused by non-rock events such as displacement changes can be effectively eliminated, avoiding the misidentification of the energy peaks corresponding to these non-rock events as fracture propagation events, and significantly improving the reliability and accuracy of downhole fracture propagation event identification.

[0036] The maximum time delay of the pressure response caused by the displacement change during the fracturing operation can be obtained. Specifically, by pre-analyzing the fracturing operation records of historical wells, the time relationship between the displacement change and the pressure response can be extracted. The maximum time delay refers to the maximum time interval between the displacement change and the pressure response in the historical data. The maximum time delay can reflect the response speed of the formation to the displacement change, which is usually affected by factors such as the formation permeability and the fracture expansion rate. By statistically analyzing the time difference distribution between the displacement change and the pressure response in the historical data, the value of the maximum time delay can be determined. For example, if the time difference between the displacement change and the pressure response in 95% of the historical data is less than 10 seconds, the maximum time delay is set to 10 seconds.

[0037] After obtaining the maximum time delay, the pressure energy data and displacement energy data can be time-domain aligned. The purpose of time-domain alignment is to eliminate time differences caused by asynchronous data acquisition or signal transmission delays. The specific method includes: interpolating the pressure energy data and displacement energy data to ensure consistent time resolution. By calculating the cross-correlation function between the two curves to determine their time offset, one of the curves is time-shifted to align the two curves on the time axis. After time-domain alignment, based on the pressure energy peaks in the pressure energy data, displacement energy peaks that meet the adjacent condition can be matched in the displacement energy data. The adjacent condition means that the time difference between the current pressure energy peak and the current displacement energy peak is less than or equal to the time difference between the current pressure energy peak and any displacement energy peak in the displacement energy data. The specific steps are as follows: Iterate through all energy peaks in the pressure energy data. For each pressure energy peak, search for the displacement energy peak in the displacement energy data with the smallest time difference. Specifically, calculate the time difference between the current pressure energy peak and the currently matched displacement energy peak. Determine whether the time difference meets the adjacent conditions, that is, whether it is less than or equal to the time difference between the current pressure energy peak and other displacement energy peaks in the displacement energy data. For pressure energy peaks that meet the adjacent conditions, further determine whether their time difference is less than the maximum delay. If the time difference is less than the maximum delay, the current pressure energy peak is determined to be the pressure energy peak corresponding to the non-rock event. This is because the pressure response caused by displacement changes usually has a shorter time delay, while the pressure response caused by rock events (such as crack expansion) usually has a longer time delay. Through this judgment, the pressure fluctuations caused by displacement changes and the pressure fluctuations caused by rock events can be effectively distinguished. The pressure energy peaks corresponding to non-rock events in the pressure energy data can be screened out. The specific method includes: removing the pressure energy peaks determined to be non-rock events from the pressure energy data. Reorder and mark the remaining pressure energy peaks, and retain the pressure energy peaks corresponding to rock events.

[0038] S104: Identify downhole fracture propagation events based on the filtered pressure-energy data.

[0039] In some embodiments, the time of occurrence of the downhole fracture propagation event of the target well is identified based on the time points of multiple pressure energy peaks corresponding to the rock events in the screened pressure energy data; the intensity of the downhole fracture propagation event of the target well is identified based on the multiple pressure energy peaks corresponding to the rock events in the screened pressure energy data.

[0040] By analyzing the pressure energy peaks representing rock events, not only can the occurrence time of downhole fracture propagation events be accurately identified, but also the intensity of downhole fracture propagation events can be accurately quantified and evaluated, laying a good data foundation for a more in-depth analysis of the downhole fracture propagation effect of the target well.

[0041] The time of occurrence of downhole fracture propagation events in the target well can be determined based on the time points corresponding to multiple pressure energy peaks corresponding to rock-related events in the filtered pressure energy data. Specifically, the pressure energy peaks corresponding to rock-related events reflect the pressure fluctuations caused by rock fracturing during fracture propagation. Each pressure energy peak corresponds to a time point, indicating the time of occurrence of the rock fracturing event. By extracting these time points, the time sequence of downhole fracture propagation events in the target well can be determined. For example, if the pressure energy peaks correspond to time points t1, t2, and t3, then the fracture propagation events occurred at times t1, t2, and t3.

[0042] In some embodiments, after determining the sub-fracture signals corresponding to the fracture propagation events, the multiple pressure energy peaks in the pressure energy data of the sub-fracture signals can be normalized. Normalization can eliminate dimensional differences between the different peaks, making them comparable. A specific method includes calculating the maximum value E_max and the minimum value E_min of all pressure energy peaks. For each pressure energy peak E_i, normalization is performed using the formula E^_i = *(E_i - E_min) / (E_max - E_min), where E^_i is the normalized pressure energy peak value and ranges from [0, 1]. Based on the multiple normalized pressure energy peaks corresponding to the rock-related events, the intensity of the downhole fracture propagation event in the target well can be quantified. The intensity of the fracture propagation event reflects the scale of the rock failure and its impact on the formation. For example, a normalized pressure wavelet energy greater than 0.5 represents a high-intensity fracture propagation event, while a normalized pressure wavelet energy ≤ 0.5 represents a low-intensity fracture propagation event. This quantification method allows for accurate quantitative assessment of the intensity characteristics of the downhole fracture propagation events in the target well.

[0043] In some embodiments, the fracturing operation process may include at least a crack initiation stage, a crack expansion stage, and a pump stop stage; the method may further include: sliding the fracturing signal using a window of preset size; calculating the fluctuation degree values of the fracturing signal in multiple windows during the sliding process; selecting a sub-fracturing signal corresponding to the crack expansion stage from the fracturing signal based on the fluctuation degree values of the fracturing signal in multiple windows and a preset fluctuation degree threshold; performing wavelet decomposition on the fracturing signal to obtain pressure energy data and displacement energy data, including: performing wavelet decomposition on the sub-fracturing signal corresponding to the crack expansion stage to obtain pressure energy data and displacement energy data.

[0044] By presetting the window size and fluctuation threshold, the sub-fracturing signal corresponding to the crack expansion stage can be accurately intercepted, avoiding interference from signals in other stages, and laying a good data foundation for subsequent feature extraction and identification of downhole crack expansion events.

[0045] The fracturing process is a complex dynamic system, encompassing at least the fracture initiation phase, fracture expansion phase, and pump-off phase. During the fracture initiation phase, fracturing fluid is initially injected into the formation, gradually increasing the formation pressure until it reaches the fracture pressure, at which point fractures begin to form. During the fracture expansion phase, as the fracturing fluid continues to be injected, the fractures continue to extend and expand within the formation, with operational parameters (such as pressure and displacement) exhibiting specific trends. During the pump-off phase, fracturing fluid injection ceases, formation pressure gradually decreases, and the fractures stabilize. In order to more accurately analyze and identify these stages, a method for identifying downhole crack extension events provided in an embodiment of the present specification may also include the following steps: first, sliding processing is performed on the collected fracturing signal using a window of preset size. The size of the window can be optimized according to the actual construction conditions and data sampling frequency to ensure that both the local characteristics of the signal can be captured and the overall trend can be reflected; second, during the sliding process, the fluctuation degree value of the fracturing signal in each window is calculated. The fluctuation degree value can be quantified by statistical methods, where the statistical method can be variance, standard deviation or range, etc.; finally, based on the fluctuation degree values of the fracturing signals in multiple windows and the preset fluctuation degree threshold, the sub-fracturing signal corresponding to the crack extension stage is selected from the fracturing signal.

[0046] In some embodiments, wavelet decomposition can be performed on the sub-fracture signals corresponding to the fracture propagation stage to obtain pressure-energy data and displacement-energy data. Fluctuation thresholds can be determined through statistical analysis of historical data or through machine learning model training to ensure accurate differentiation of fracture signal characteristics at different stages. This approach can effectively identify fracture propagation stages, providing reliable data support for real-time monitoring and optimization of fracturing operations. It also facilitates the evaluation of fracturing effectiveness and the prediction of fracture morphology, thereby improving the scientific and economic efficiency of fracturing operations.

[0047] In some embodiments, the fluctuation degree threshold may include at least a pressure fluctuation degree threshold and a displacement fluctuation degree threshold; a first fracturing signal may be selected from the fracturing signal based on the fluctuation degree values of the pressure data in multiple windows and a preset pressure fluctuation degree threshold; a second fracturing signal may be selected from the fracturing signal based on the fluctuation degree values of the displacement data in multiple windows and a preset displacement fluctuation degree threshold; and a sub-fracturing signal corresponding to the crack expansion stage may be determined based on the intersection of the first fracturing signal and the second fracturing signal.

[0048] By determining the sub-signal corresponding to the fracture propagation stage based on the intersection of the first and second fracturing signals, the fracture propagation stage can be effectively identified, avoiding misjudgment caused by single parameter analysis, while improving the accuracy and reliability of fracturing operation analysis.

[0049] The fracturing signal may include pressure time series data, and the magnitude of the pressure signal change during the fracturing process can be measured by setting a pressure fluctuation threshold. Based on the magnitude of the pressure signal change during the fracturing process, the different stages of the fracturing operation can be distinguished. The fracturing signal may also include displacement time series data, and the magnitude of the displacement signal change during the fracturing process can be measured by setting a displacement fluctuation threshold. Based on the magnitude of the displacement signal change during the fracturing process, the different stages of the fracturing operation can also be distinguished. A single pressure fluctuation threshold / displacement fluctuation threshold can be used to determine the sub-fracturing signal corresponding to the fracture propagation stage, or a combination of, but not limited to, a pressure fluctuation threshold and a displacement fluctuation threshold can be used to determine the sub-fracturing signal corresponding to the fracture propagation stage.

[0050] In a specific implementation, a sub-fracture signal corresponding to the fracture propagation stage can be selected from the fracture signal based on the fluctuation values of the fracture signal within multiple windows and a preset fluctuation threshold. For example, a first fracture signal can be selected from the fracture signal based on the fluctuation values of the pressure time series data within multiple windows and a preset pressure fluctuation threshold. Specifically, by calculating the fluctuation value of the pressure time series data within each window and comparing it with the preset pressure fluctuation threshold, windows whose fluctuation values match the fracture propagation stage characteristics are selected, thereby obtaining the first fracture signal. The first fracture signal can reflect the typical pressure variation trend during the fracture propagation stage, manifested as relatively stable fluctuations or a specific rising / falling pattern. Similarly, a second fracture signal can be selected from the fracture signal based on the fluctuation values of the displacement time series data within multiple windows and a preset displacement fluctuation threshold. By calculating the fluctuation value of the displacement time series data within each window and comparing it with the displacement fluctuation threshold, windows that match the fracture propagation stage characteristics are selected, thereby obtaining the second fracture signal. The second fracture signal can reflect the typical characteristics of the displacement during the fracture propagation stage, manifested as a relatively stable injection rate or a specific variation pattern. The intersection of the first and second fracturing signals allows the identification of sub-signals corresponding to the fracture propagation phase. Specifically, by comparing the time ranges of the first and second fracturing signals, the overlapping time period is extracted as the sub-signal for the fracture propagation phase. These sub-signals comprehensively reflect the coordinated changes in pressure and displacement during the fracture propagation phase, providing accurate data support for real-time monitoring and optimization of fracturing operations.

[0051] In some embodiments, a window of a preset size may be used, using the formula I=L-l+1 calculates the fluctuation degree of the fracturing signal in multiple windows during the sliding process; where L is the length of the fracturing signal; l is the preset window length; I is the total number of windows in the fracturing signal; μ i is the fluctuation value of the fracturing signal in the i-th window; σ i is the standard deviation of the fracturing signal in the i-th window; σ min is the minimum value of the standard deviation of the fracturing signal within I windows; σ max is the maximum value of the standard deviation of the fracturing signal within I window.

[0052] By calculating the corresponding fluctuation degree value based on the standard deviation of the fracturing signal in multiple windows during the sliding process, the changing trends and characteristics of the fracturing signal at different stages can be accurately characterized, and the sub-fracturing signal corresponding to the crack expansion stage can be accurately intercepted, avoiding interference from signals in other stages, and laying an accurate data foundation for subsequent feature extraction and identification of downhole crack expansion events.

[0053] After obtaining the fracturing signal, the standard deviation of the data in multiple windows of the fracturing signal is calculated based on the windows of preset size. The specific calculation method includes: using the windows of preset size to slide the collected fracturing signal, and during the sliding process, calculating the standard deviation of the fracturing signal in each window. According to the maximum standard deviation and minimum standard deviation of the fracturing signal in multiple windows of the fracturing signal, the fluctuation degree value of each window in the sliding process of the fracturing signal is calculated. The fluctuation degree value can be used to quantify the fluctuation characteristics of the fracturing signal in each window, reflecting the changes in pressure and displacement during the fracturing operation. For example, the following formula can be used to calculate the fluctuation degree value of the fracturing signal in multiple windows during the sliding process:

[0054]

[0055] Where L is the length of the fracturing signal; l is the preset window length; I is the total number of windows in the fracturing signal; μ i is the fluctuation value of the fracturing signal in the i-th window; σ i is the standard deviation of the fracturing signal in the i-th window; σ min is the minimum value of the standard deviation of the fracturing signal within I windows; σ max is the maximum value of the standard deviation of the fracturing signal within I window.

[0056] In some embodiments, a set of window size values and a set of fluctuation threshold value values are combined to construct a multidimensional grid space; each grid in the multidimensional grid space corresponds to a window size value and a fluctuation threshold value, and each window size value and a fluctuation threshold value together correspond to a candidate sub-fracture signal; the degree of matching between the candidate sub-fracture signal corresponding to each grid and the preset crack extension feature is calculated; a multidimensional grid search algorithm is used to maximize the matching degree to obtain a grid with the maximum matching degree as the optimal grid; and the window size and the fluctuation threshold value are determined based on the window size value and the fluctuation threshold value corresponding to the optimal grid.

[0057] By using a multidimensional grid search algorithm to obtain a set of values of window size and fluctuation threshold corresponding to the optimal grid in the multidimensional grid space, adaptive optimization of multiple parameters of window size and fluctuation threshold is achieved without relying on manual experience calibration, and the selected results are more objective and accurate.

[0058] The set of window size values and the set of fluctuation threshold values are combined to obtain all possible combinations of window size and fluctuation threshold values. A multidimensional grid space is then constructed based on all possible combinations. Specifically, the window size can be used to define the time period for analyzing fracturing signals, and the pressure fluctuation threshold can be used to screen the pressure / displacement signal characteristics during the fracture propagation phase. Each grid point in the multidimensional grid space represents a specific set of window size and fluctuation threshold values. The coordinates of each grid point represent the values of multiple parameters, forming a multidimensional parameter space covering all possible parameter combinations. For example, the window size range is [L_min, L_max], the fluctuation threshold includes the pressure fluctuation threshold and the displacement fluctuation threshold, with the pressure fluctuation threshold range being [P_min, P_max] and the displacement fluctuation threshold range being [Q_min, Q_max]. The value range of each parameter is discretized to form evenly distributed grid points. Specifically, the window length is set to 10 discrete values, the pressure fluctuation threshold is set to 10 discrete values, and the displacement fluctuation threshold is set to 10 discrete values. Then, there are 10×10×10=1000 grid points in the three-dimensional grid space.

[0059] After constructing the multidimensional grid space, the degree of match between the candidate sub-fracture signal corresponding to each grid point and the preset crack extension feature is calculated. Specifically, for each grid point, the collected fracturing signal is slid using the window size, and then the candidate sub-fracture signal of the crack extension stage is extracted from the fracturing signal based on the fluctuation degree threshold. The extracted sub-fracture signal segment can be compared with the preset crack extension feature to calculate the degree of match between the two. The degree of match can be quantified by comprehensive correlation analysis, energy distribution similarity, or the probability value output by the machine learning model to measure the consistency between the extracted sub-fracture signal segment and the crack extension feature under the current multi-tuple parameters. For example, for each grid point, after extracting the fracturing signal segment, its degree of match with the preset crack extension feature is calculated. The preset crack extension feature can be a fracturing signal of a known crack extension event. The matching degree can be calculated by the formula T = ω1·SIM+ω2·POW+ω3·POS, where T represents the matching degree between the extracted fracturing signal segment and the preset crack extension feature, SIM represents the correlation coefficient between the extracted fracturing signal segment and the fracturing signal of the known crack extension event, POW represents the energy distribution similarity between the extracted fracturing signal segment and the fracturing signal of the known crack extension event, POS represents the matching probability between the extracted fracturing signal segment and the fracturing signal of the known crack extension event output using the preset SVM model, ω1, ω2 and ω3 are weight coefficients used to balance the influence of different matching degree indicators, which will not be repeated here.

[0060] In order to find the optimal grid, that is, the optimal parameter combination, a multidimensional grid search algorithm can be used to maximize the matching degree. The multidimensional grid search algorithm can be an exhaustive search, a random search, or an intelligent search method based on an optimization algorithm (such as a genetic algorithm, a particle swarm optimization algorithm). Through continuous iteration and optimization, a multi-tuple that maximizes the matching degree is found. Specifically, the multidimensional grid search algorithm traverses all grid points in the multidimensional grid space, calculates the matching degree corresponding to each grid point, and records the grid point with the highest matching degree. A set of values for the window size and the fluctuation degree threshold corresponding to the optimal grid point is the optimal parameter combination of the window size and the fluctuation degree threshold, and then the sub-fracture signal corresponding to the fracture extension stage can be accurately selected from the fracture signal. In this way, it can be ensured that the extracted sub-fracture signal segment has the highest matching degree with the preset fracture extension feature, thereby improving the accuracy of fracture extension event identification.

[0061] In some embodiments, the fracturing signal can be decomposed using a plurality of preset wavelet bases to obtain a plurality of first signals corresponding to the plurality of wavelet bases; one wavelet base corresponds to one first signal; the wavelet base corresponding to the first signal with the largest energy peak is determined as the optimal wavelet base for the target well; the fracturing signal is decomposed according to a plurality of preset wavelet decomposition layers to obtain a plurality of second signals corresponding to the plurality of wavelet decomposition layers; one wavelet decomposition layer corresponds to one second signal; based on the energy variance of the plurality of second signals, the degree of fit between the plurality of second signals and the preset fracture extension characteristics is calculated; the wavelet decomposition layer corresponding to the second signal with the largest degree of fit is determined as the optimal wavelet decomposition layer for the target well.

[0062] By selecting the wavelet basis corresponding to the first signal with the largest energy peak as the optimal wavelet basis for the target well, the ability to identify downhole fracture propagation events can be maintained even under relatively weak pressure fluctuations. By selecting the wavelet decomposition level corresponding to the second signal with the highest degree of fit to the pre-set fracture propagation characteristics as the optimal wavelet decomposition level for the target well, the effectiveness of feature extraction for the fracturing signal can be further enhanced.

[0063] The fracturing signal can be decomposed using multiple preset wavelet bases to obtain multiple first signals corresponding to the multiple wavelet bases. The preset wavelet bases include a variety of commonly used wavelet functions, such as Daubechies wavelet (dbN), Symlets wavelet (symN), Coiflets wavelet (coifN), and Haar wavelet. Each wavelet base has different time-frequency localization characteristics and smoothness, capable of capturing the characteristics of the fracturing signal from different perspectives. For each wavelet base, the fracturing signal is subjected to wavelet decomposition to obtain the corresponding first signal. The first signal is a multi-scale subband signal generated by the wavelet decomposition, reflecting the energy distribution characteristics of the fracturing signal within different frequency ranges. After obtaining the multiple first signals, the energy peak of each first signal is calculated. The energy peak refers to the maximum value of the sum of the squares of the wavelet coefficients of each subband in the first signal and is used to measure the energy concentration of the signal. By comparing the energy peaks of the first signals, the wavelet base corresponding to the first signal with the largest energy peak is determined as the optimal wavelet base for the target well. The optimal wavelet base can maximize the concentration of the energy of the fracturing signal, thereby more effectively extracting key features from the fracturing signal.

[0064] The fracturing signal can be decomposed according to multiple preset wavelet decomposition levels to obtain multiple second signals corresponding to the multiple wavelet decomposition levels. The wavelet decomposition level refers to the number of wavelet decomposition stages, ranging from 3 to 10. Each wavelet decomposition level corresponds to a second signal, which is a multi-scale subband signal generated by that decomposition level. Different wavelet decomposition levels can capture different frequency ranges of the signal. Lower levels primarily reflect high-frequency details, while higher levels primarily reflect low-frequency trends. After obtaining multiple second signals, the energy variance of each second signal is calculated. Energy variance refers to the variance of the energy of the wavelet coefficients in each subband of the second signal and is used to measure the uniformity of the signal's energy distribution. By calculating the energy variance of each second signal, its consistency with the preset fracture propagation characteristic can be evaluated. The preset fracture propagation characteristic can specifically be the time of a known fracture propagation event. The consistency of the change in the energy variance of each second signal with the time of the known fracture propagation event can be calculated. The wavelet decomposition level corresponding to the second signal with the highest consistency can be determined as the optimal wavelet decomposition level for the target well. The optimal number of wavelet decomposition layers can match the preset crack extension characteristics to the greatest extent, thereby providing the optimal multi-scale decomposition results for subsequent signal analysis and event identification.

[0065] In some embodiments, the optimal wavelet basis for the target well can be used to decompose the fracturing signal according to the optimal wavelet decomposition layer number for the target well, thereby obtaining pressure-energy data and displacement-energy data. The pressure-energy data represents the relationship between pressure-energy and time under the optimal wavelet basis and optimal wavelet decomposition layer number. The displacement-energy data represents the relationship between displacement-energy and time under the optimal wavelet basis and optimal wavelet decomposition layer number.

[0066] In some embodiments, a pressure energy peak determination function can be constructed based on the peak parameters of the pressure energy peak in the pressure energy data; the pressure energy peak determination function is used to determine whether the pressure energy peak corresponds to a rock event; the degree of matching between the time data corresponding to the rock event determined by the pressure energy peak determination function and the monitored microseismic event time data is calculated; a multi-point mutation genetic algorithm is used to maximize the matching degree to optimize the peak parameters in the pressure energy peak determination function; the optimized pressure energy peak determination function is used to determine whether the pressure energy peak in the screened pressure energy data corresponds to a rock event; and multiple pressure energy peaks corresponding to rock events are screened out.

[0067] By constructing a pressure energy peak determination function and using a multi-point variation anomaly algorithm to optimize the pressure energy peak determination function, it is helpful to accurately and quickly determine multiple pressure energy peaks corresponding to rock events in pressure energy data, thereby further effectively screening out signal noise and non-rock events in pressure energy data.

[0068] Based on the peak parameters of the pressure energy peak, such as peak height, peak width, peak significance and peak platform size, a pressure energy peak determination function can be constructed. The pressure energy peak determination function can be a multi-parameter comprehensive evaluation function used to quantify the characteristics of each pressure energy peak and determine whether it corresponds to a rock-type event. Specifically, the peak height reflects the intensity of the pressure fluctuation, the peak width reflects the duration of the pressure fluctuation, the peak significance reflects the degree of contrast between the pressure fluctuation and the background noise, and the peak platform size reflects the stability of the pressure fluctuation. These parameters together form the basis of the determination function and can comprehensively describe the characteristics of the pressure energy peak. For example, the form of the pressure energy peak determination function can be expressed as F = α1·H+α2·W+α3·S+α4·P, where F is the effective value of the pressure energy peak, H is the peak height, W is the peak width, S is the peak significance, P is the peak platform size, and α1, α2, α3 and α4 are weight coefficients used to balance the influence of each parameter. The effectiveness threshold F can be set th , according to the validity threshold F th Determine whether the pressure energy peak corresponds to a rock event. Specifically, for the pressure energy peak A, if F(A)≥F th , it is determined that the pressure energy peak A corresponds to a rock event; otherwise, it is determined that the pressure energy peak A corresponds to a non-rock event.

[0069] In order to optimize the pressure energy peak judgment function, the weight coefficients α1, α2, α3, α4 and the validity threshold F can be optimized using the preset multi-point mutation genetic algorithm. thOptimization can be performed. Specifically, the time point data of microseismic events monitored during the fracture propagation phase of the target well can be obtained. This microseismic event time point data can be obtained through microseismic monitoring technology, which records the time of rock failure events during the fracture propagation process. Microseismic monitoring technology can use surface or downhole sensor arrays to capture elastic wave signals generated by rock failure and determine the time point of the microseismic event through signal processing and analysis. After obtaining the microseismic event time point data, the degree of match between the time data corresponding to rock events identified by the pressure energy peak judgment function and the monitored microseismic event time data can be calculated. The pressure energy peak judgment function is derived by quantitatively analyzing the peak characteristics in the pressure energy data and is used to determine which pressure energy peaks correspond to rock events. Specifically, the pressure energy peak judgment function scores each pressure energy peak based on peak parameters such as peak height, peak width, peak significance, and peak platform size, and determines whether it is a rock event based on the scoring result. The degree of match can be calculated by comparing the consistency of the rock event time data identified by the pressure energy peak judgment function with the microseismic event time data. Specifically, for each identified rock event time point, the microseismic event time point with the smallest time difference is found in the microseismic event time data. The time differences between all matching time point pairs are calculated, and the number of matching pairs with time differences less than a preset threshold is counted. The ratio of the number of matching pairs to the total number of identified pairs is used as the matching degree.

[0070] To maximize the matching degree, a multi-point mutation genetic algorithm can be used to optimize the peak height, peak width, peak significance, and peak platform size parameters in the pressure-energy peak judgment function. This multi-point mutation genetic algorithm is an optimization algorithm based on the principles of biological evolution. It searches for the optimal solution in parameter space by simulating operations such as natural selection, crossover, and mutation. The specific optimization process includes the following steps: Population initialization: A set of initial parameter combinations is randomly generated, each parameter combination consisting of . Fitness evaluation: For each parameter combination, the fitness value of the corresponding pressure-energy peak judgment function is calculated. The fitness value is determined by comparing the consistency of the judgment results with the microseismic event time data. The specific calculation method is: Fitness = number of matching pairs / total number of judgments, where the number of matching pairs refers to the number of matches between the pressure-energy peak time points of rock-related events and the time points of microseismic events. Selection: Based on the fitness value, the best-performing parameter combinations are selected as parents. Crossover: A crossover operation is performed on the selected parent parameter combinations to generate new child parameter combinations. Mutation: The child parameter combinations are randomly mutated to increase the diversity of the population. Iterative optimization: Fitness evaluation, selection, crossover, and mutation operations are repeated until the preset number of iterations is reached or fitness converges. Through optimization using a multi-point mutation genetic algorithm, the parameter combination that maximizes fitness is found, thereby optimizing the performance of the pressure energy peak determination function. This optimized determination function can more accurately identify rock events and improve the accuracy and reliability of fracture propagation event identification.

[0071] By optimizing the time data of monitored microseismic events and using a multi-point mutation genetic algorithm to optimize peak parameters in the pressure energy peak determination function, such as peak height, peak width, peak significance, and peak platform size, the adaptive adjustment of the pressure energy peak determination function was achieved, enabling more accurate identification of pressure energy peaks representing rock failure events. Furthermore, the adaptive adjustment of the pressure energy peak determination function using the multi-point mutation genetic algorithm eliminates the need for manual intervention, effectively improving its adaptability under different geological conditions.

[0072] According to the optimized pressure energy peak determination function, multiple pressure energy peaks corresponding to rock events in the pressure energy data after screening can be determined. The specific method includes: for each pressure energy peak in the pressure energy data after screening, using the optimized pressure energy peak determination function to calculate its corresponding effective value. According to the optimized threshold F th , determine whether the pressure energy peak corresponds to a rock event. Extract the pressure energy peaks determined to be rock events to form a set of pressure energy peaks corresponding to rock events.

[0073] In some embodiments, the time and intensity of downhole fracture propagation events in a target well can be determined based on a set of pressure energy peaks corresponding to rock events generated after optimization using a multi-point mutation genetic algorithm. Specifically, the pressure energy peaks corresponding to rock events reflect the pressure fluctuations caused by rock fracturing during fracture propagation. Each pressure energy peak corresponds to a time point and a normalized value, representing the time and intensity of the rock fracturing event, respectively. By extracting these time points and normalized values, the time series and intensity series of downhole fracture propagation events in the target well can be determined.

[0074] A specific embodiment of this specification is provided below:

[0075] 1. Collect the fracturing signal during the hydraulic fracturing operation of Well A, refer to Figure 3 During the hydraulic fracturing operation in Well A, pressure sensors and flow meters were used to collect real-time data on pressure, flow rate, and sand concentration in the wellbore at a sampling frequency of 1 Hz.

[0076] 2. According to the pressure and displacement changes in the fracturing signal of Well A, the construction process is divided into the crack initiation stage, the crack expansion stage and the pump stop stage. The sub-fracturing signal corresponding to the crack expansion stage is obtained, and the reference Figure 4 .

[0077] 3. Perform multi-level wavelet decomposition on the sub-fracture signal corresponding to the fracture expansion stage of Well A, referring to Figure 5 、 Figure 6 and Figure 7 .

[0078] 4. According to the energy relationship principle of pressure and energy, the pressure energy peak corresponding to non-rock events in the pressure energy data after 8-level wavelet decomposition of the sub-fracturing signal corresponding to the fracture expansion stage of Well A was screened out, and the reference Figure 8 and Figure 9 .

[0079] 5. Use the multi-point mutation genetic algorithm to determine the pressure energy peak corresponding to the rock event in the pressure energy data after 8-level wavelet decomposition of the sub-fracturing signal corresponding to the fracture expansion stage of Well A, referring to Figure 10 and Figure 11 .

[0080] 6. According to the pressure capacity peak value after multi-level wavelet decomposition of the sub-fracture signal corresponding to the fracture expansion stage of Well A, the fracture expansion event in Well A is identified. Figure 12 and Figure 13 The time when the pressure energy peak occurs is the time when the crack propagation event occurs. The normalized pressure energy > 0.5 represents a crack propagation event with a larger intensity, and the normalized pressure energy ≤ 0.5 represents a crack propagation event with a smaller intensity.

[0081] The method for identifying downhole crack extension events provided in the embodiments of this specification can obtain a fracturing signal during a fracturing operation, wherein the fracturing signal is used to represent the changing relationship between construction parameters and time, wherein the construction parameters include at least pressure and displacement; the fracturing signal is subjected to wavelet decomposition to obtain pressure-energy data and displacement-energy data, wherein the pressure-energy data represents the changing relationship between pressure energy and time, and the displacement-energy data represents the changing relationship between displacement energy and time; the pressure-energy data is screened using the displacement-energy data to screen out pressure-energy peaks corresponding to non-rock events in the pressure-energy data; and downhole crack extension events are identified based on the screened pressure-energy data. Compared with existing methods, the embodiments of this specification can directly obtain the fracturing signal during the fracturing operation to identify downhole crack expansion events, without relying on refined data equipment and without refined data annotation, and the data acquisition cost is low; by performing wavelet decomposition on the fracturing signal, the time-frequency domain characteristics in the fracturing signal can be effectively extracted, and the signal-to-noise ratio of the fracturing signal can be improved, thereby reducing the interference of noise signals in the formation on the identification of downhole crack expansion events; by using displacement energy data to screen out the pressure energy peaks corresponding to non-rock events in the pressure energy data, the interference of non-rock events can be effectively eliminated, and the reliability and accuracy of downhole crack expansion event identification can be significantly improved.

[0082] Based on the above-mentioned method for identifying downhole crack propagation events, this specification also proposes an embodiment of a downhole crack propagation event identification device. Figure 14 As shown, the downhole crack propagation event identification device 1400 may specifically include the following modules:

[0083] An acquisition module 1401 may be used to acquire a fracturing signal during a fracturing operation, wherein the fracturing signal is used to represent a relationship between operation parameters and time, wherein the operation parameters include at least pressure and displacement;

[0084] The decomposition module 1402 may be configured to perform wavelet decomposition on the fracturing signal to obtain pressure energy data and displacement energy data, wherein the pressure energy data represents a relationship between pressure energy and time, and the displacement energy data represents a relationship between displacement energy and time;

[0085] The screening module 1403 can be used to screen the pressure energy data using the displacement energy data to filter out the pressure energy peaks corresponding to non-rock events in the pressure energy data;

[0086] The identification module 1404 may be used to identify downhole fracture propagation events based on the filtered pressure-energy data.

[0087] In some embodiments, the acquisition module 1401 can be specifically used to slide the fracturing signal using a window of a preset size; calculate the fluctuation degree values of the fracturing signal in multiple windows during the sliding process; and select the sub-fracturing signal corresponding to the crack expansion stage from the fracturing signal based on the fluctuation degree values of the fracturing signal in multiple windows and a preset fluctuation degree threshold.

[0088] In some embodiments, the acquisition module 1401 may be further configured to calculate the fluctuation degree values of the fracturing signals in multiple windows during the sliding process using the following formula using a window of a preset size:

[0089]

[0090] Where L is the length of the fracturing signal; l is the preset window length; I is the total number of windows in the fracturing signal; μ i is the fluctuation value of the fracturing signal in the i-th window; σ i is the standard deviation of the fracturing signal in the i-th window; σ min is the minimum value of the standard deviation of the fracturing signal within I windows; σ max is the maximum value of the standard deviation of the fracturing signal within I window.

[0091] In some embodiments, the acquisition module 1401 can also be used to combine a set of window size values and a set of fluctuation threshold values to construct a multidimensional grid space; each grid in the multidimensional grid space corresponds to a window size value and a fluctuation threshold value, and a window size value and a fluctuation threshold value can jointly correspond to a candidate sub-fracture signal; the degree of matching between the candidate sub-fracture signal corresponding to each grid and the preset crack extension feature is calculated; the degree of matching is maximized using a multidimensional grid search algorithm to obtain the grid corresponding to the maximum degree of matching as the optimal grid; the window size and the fluctuation threshold value are determined based on the window size value and the fluctuation threshold value corresponding to the optimal grid.

[0092] In some embodiments, the above-mentioned decomposition module 1402 can be specifically used to use a preset plurality of wavelet bases to decompose the fracturing signal respectively, and obtain a plurality of first signals corresponding to the plurality of wavelet bases; one said wavelet base corresponds to one said first signal; the wavelet base corresponding to the first signal with the largest energy peak is determined as the optimal wavelet base of the target well; the fracturing signal is decomposed according to a preset plurality of wavelet decomposition layers, and obtain a plurality of second signals corresponding to the plurality of wavelet decomposition layers; one said wavelet decomposition layer corresponds to one said second signal; according to the energy variance of the plurality of second signals, the degree of fit between the plurality of second signals and the preset fracture extension characteristics is calculated; the wavelet decomposition layer corresponding to the second signal with the largest degree of fit is determined as the optimal wavelet decomposition layer for the target well.

[0093] In some embodiments, the above-mentioned decomposition module 1402 can also be specifically used to use the optimal wavelet basis of the target well to decompose the fracturing signal according to the optimal wavelet decomposition layer number of the target well to obtain pressure energy data and displacement energy data. The pressure energy data represents the relationship between the change of pressure energy and time under the optimal wavelet basis and the optimal wavelet decomposition layer number, and the displacement energy data represents the relationship between displacement energy and time under the optimal wavelet basis and the optimal wavelet decomposition layer number.

[0094] In some embodiments, the above-mentioned screening module 1403 can be specifically used to obtain the maximum time delay of the pressure response caused by the displacement change; for each pressure energy peak in the pressure energy data, the displacement energy peak that meets the adjacent conditions is matched in the displacement energy data; the adjacent conditions are used to constrain the time difference between the pressure energy peak and the matched displacement energy peak to be less than or equal to the time difference between the pressure energy peak and any other displacement energy peak in the displacement energy data; for each pressure energy peak in the pressure energy data, if the time difference between the pressure energy peak and the matched displacement energy peak is less than the maximum time delay, it is determined that the pressure energy peak corresponds to a non-rock event; and the pressure energy peaks corresponding to non-rock events in the pressure energy data are screened out.

[0095] In some embodiments, the above-mentioned screening module 1403 can also be specifically used to construct a pressure energy peak determination function based on the peak parameters of the pressure energy peak in the pressure energy data; the pressure energy peak determination function is used to determine whether the pressure energy peak corresponds to a rock event; calculate the matching degree between the time data corresponding to the rock event determined by the pressure energy peak determination function and the monitored microseismic event time data; use a multi-point mutation genetic algorithm to maximize the matching degree to optimize the peak parameters in the pressure energy peak determination function; use the optimized pressure energy peak determination function to determine whether the pressure energy peak in the screened pressure energy data corresponds to a rock event; and screen out multiple pressure energy peaks corresponding to rock events.

[0096] In some embodiments, the above-mentioned identification module 1404 can be specifically used to identify the time when the downhole fracture expansion event of the target well occurs based on the time points of multiple pressure energy peaks corresponding to the rock events in the filtered pressure energy data; and identify the intensity of the downhole fracture expansion event of the target well based on the multiple pressure energy peaks corresponding to the rock events in the filtered pressure energy data.

[0097] As can be seen from the above, the downhole crack expansion event identification device provided in the embodiment of this specification can obtain a fracturing signal during the fracturing operation, wherein the fracturing signal is used to represent the changing relationship between construction parameters and time, and the construction parameters include at least pressure and displacement; the fracturing signal is subjected to wavelet decomposition to obtain pressure energy data and displacement energy data, wherein the pressure energy data represents the changing relationship between pressure energy and time, and the displacement energy data represents the changing relationship between displacement energy and time; the pressure energy data is filtered using the displacement energy data to filter out the pressure energy peaks corresponding to non-rock events in the pressure energy data; and the downhole crack expansion event is identified based on the filtered pressure energy data. Compared with existing methods, the embodiments of this specification can directly obtain the fracturing signal during the fracturing operation to identify downhole crack expansion events, without relying on refined data equipment and without refined data annotation, and the data acquisition cost is low; by performing wavelet decomposition on the fracturing signal, the time-frequency domain characteristics in the fracturing signal can be effectively extracted, and the signal-to-noise ratio of the fracturing signal can be improved, thereby reducing the interference of noise signals in the formation on the identification of downhole crack expansion events; by using displacement energy data to screen out the pressure energy peaks corresponding to non-rock events in the pressure energy data, the interference of non-rock events can be effectively eliminated, and the reliability and accuracy of downhole crack expansion event identification can be significantly improved.

[0098] It should be noted that the units, devices or modules described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. For the convenience of description, the above devices are described in terms of functions and are divided into various modules and described separately. Of course, when implementing this specification, the functions of each module can be implemented in the same or multiple software and / or hardware, or the module that implements the same function can be implemented by a combination of multiple sub-modules or sub-units. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0099] An embodiment of the present specification also provides a computer device for a method of identifying downhole fracture propagation events, comprising a processor and a memory for storing instructions executable by the processor. When the processor is specifically implemented, the following steps can be performed according to the instructions: obtaining a fracturing signal during a fracturing operation, wherein the fracturing signal is used to represent the changing relationship between construction parameters and time, wherein the construction parameters include at least pressure and displacement; performing wavelet decomposition on the fracturing signal to obtain pressure-energy data and displacement-energy data, wherein the pressure-energy data represents the changing relationship between pressure energy and time, and the displacement-energy data represents the changing relationship between displacement energy and time; using the displacement-energy data to filter the pressure-energy data to filter out pressure-energy peaks corresponding to non-rock events in the pressure-energy data; and identifying downhole fracture propagation events based on the filtered pressure-energy data.

[0100] In order to complete the above instructions more accurately, refer to Figure 15 As shown, the embodiment of this specification also provides another specific computer device 1500, wherein the computer device 1500 includes a network communication port 1501, a processor 1502 and a memory 1503, and the above structures are connected through internal cables so that each structure can perform specific data interaction.

[0101] The processor 1502 can be specifically used to obtain a fracturing signal during a fracturing operation, where the fracturing signal is used to represent the changing relationship between construction parameters and time, where the construction parameters include at least pressure and displacement; perform wavelet decomposition on the fracturing signal to obtain pressure-energy data and displacement-energy data, where the pressure-energy data represents the changing relationship between pressure-energy and time, and the displacement-energy data represents the changing relationship between displacement-energy and time; use the displacement-energy data to filter the pressure-energy data to filter out pressure-energy peaks corresponding to non-rock events in the pressure-energy data; and identify downhole fracture propagation events based on the filtered pressure-energy data.

[0102] The memory 1503 may be specifically used to store corresponding instruction programs.

[0103] In this embodiment, the network communication port 1501 can be a virtual port that is bound to different communication protocols, thereby being capable of sending or receiving different data. For example, the network communication port can be a port responsible for web data communication, a port responsible for FTP data communication, or a port responsible for email data communication. Furthermore, the network communication port can also be a physical communication interface or communication chip. For example, it can be a wireless mobile network communication chip, such as GSM or CDMA; it can also be a Wi-Fi chip; or it can be a Bluetooth chip.

[0104] In this embodiment, the processor 1502 may be implemented in any suitable manner. For example, the processor may take the form of a microprocessor or a processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro)processor, a logic gate, a switch, an application-specific integrated circuit (ASIC), a programmable logic controller, an embedded microcontroller, etc. This specification is not intended to limit this.

[0105] In this embodiment, the memory 1503 includes volatile memory and non-volatile memory. The memory 1503 can include multiple levels. In digital systems, anything that can store binary data can be considered a memory. In integrated circuits, a circuit with a storage function that does not have a physical form is also called a memory, such as RAM and FIFO. In systems, a physical storage device is also called a memory, such as a memory stick or TF card.

[0106] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0107] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0108] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0109] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0110] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for identifying downhole fracture propagation events, characterized in that: The method comprises: Acquire a fracturing signal during a fracturing operation, wherein the fracturing signal is used to represent a relationship between operation parameters and time, wherein the operation parameters include at least pressure and displacement; Performing wavelet decomposition on the fracturing signal to obtain pressure energy data and displacement energy data, wherein the pressure energy data represents a changing relationship between pressure energy and time, and the displacement energy data represents a changing relationship between displacement energy and time; The pressure energy data are screened using the displacement energy data to remove the pressure energy peaks corresponding to non-rock events in the pressure energy data; Identify downhole fracture propagation events based on filtered pressure-energy data.

2. The method according to claim 1, characterized in that The method further comprises: Sliding the fracturing signal using a window of a preset size; calculating the fluctuation degree values of the fracturing signal in multiple windows during the sliding process; selecting a sub-fracturing signal corresponding to the crack propagation stage from the fracturing signal based on the fluctuation degree values of the fracturing signal in the multiple windows and a preset fluctuation degree threshold; The step of performing wavelet decomposition on the fracturing signal to obtain pressure energy data and displacement energy data includes: The sub-fracturing signals corresponding to the crack expansion stage are decomposed by wavelet to obtain pressure energy data and displacement energy data.

3. The method according to claim 2, characterized in that The calculation of the fluctuation degree values of the fracturing signals in multiple windows during the sliding process includes: Using a window of preset size, the following formula is used to calculate the fluctuation degree of the fracturing signal in multiple windows during the sliding process: Where L is the length of the fracturing signal; l is the preset window length; I is the total number of windows in the fracturing signal; μ i is the fluctuation value of the fracturing signal in the i-th window; σ i is the standard deviation of the fracturing signal in the i-th window; σ min is the minimum value of the standard deviation of the fracturing signal within I windows; σ max is the maximum value of the standard deviation of the fracturing signal within I window.

4. The method according to claim 2, characterized in that The method for determining the size of the window and the fluctuation degree threshold includes: Combining a set of window size values and a set of fluctuation threshold values to construct a multidimensional grid space; each grid in the multidimensional grid space corresponds to a value of the window size and a value of the fluctuation threshold, and each of the window size value and the fluctuation threshold value can jointly correspond to a candidate sub-fracturing signal; Calculate the matching degree between the candidate sub-fracturing signal corresponding to each grid and the preset fracture propagation characteristics; Maximizing the matching degree using a multidimensional grid search algorithm to obtain a grid corresponding to the maximum matching degree as the optimal grid; The window size and the fluctuation degree threshold are determined according to the window size value and the fluctuation degree threshold value corresponding to the optimal grid.

5. The method according to claim 1, characterized in that: The method further comprises: Decomposing the fracturing signal using a plurality of preset wavelet bases to obtain a plurality of first signals corresponding to the plurality of wavelet bases; one wavelet base corresponds to one first signal; Determine the wavelet basis corresponding to the first signal with the maximum energy peak as the optimal wavelet basis of the target well; Decomposing the fracturing signal according to a plurality of preset wavelet decomposition levels to obtain a plurality of second signals corresponding to the plurality of wavelet decomposition levels; one wavelet decomposition level corresponds to one second signal; Calculating the degree of agreement between the plurality of second signals and a preset crack extension feature based on energy variances of the plurality of second signals; Determine the wavelet decomposition layer number corresponding to the second signal with the maximum degree of agreement as the optimal wavelet decomposition layer number of the target well; The step of performing wavelet decomposition on the fracturing signal to obtain pressure energy data and displacement energy data includes: The fracturing signal is decomposed according to the optimal wavelet decomposition layer number of the target well using the optimal wavelet basis of the target well to obtain pressure energy data and displacement energy data. The pressure energy data represents the relationship between the pressure energy and time under the optimal wavelet basis and the optimal wavelet decomposition layer number, and the displacement energy data represents the relationship between the displacement energy and time under the optimal wavelet basis and the optimal wavelet decomposition layer number.

6. The method according to claim 1, characterized in that The method of screening the pressure energy data using the displacement energy data to remove the pressure energy peaks corresponding to non-rock events in the pressure energy data includes: Get the maximum time delay of pressure response caused by displacement change; For each pressure energy peak in the pressure energy data, matching a displacement energy peak in the displacement energy data that satisfies a neighboring condition; the neighboring condition is used to constrain the time difference between the pressure energy peak and the matched displacement energy peak to be less than or equal to the time difference between the pressure energy peak and any other displacement energy peak in the displacement energy data; For each pressure energy peak in the pressure energy data, if the time difference between the pressure energy peak and the matched displacement energy peak is less than the maximum time delay, determining that the pressure energy peak corresponds to a non-rock event; Screen out the pressure energy peaks corresponding to non-rock events in the pressure energy data.

7. The method according to claim 1, characterized in that: The method further comprises: Constructing a pressure energy peak determination function based on peak parameters of the pressure energy peak in the pressure energy data; the pressure energy peak determination function is used to determine whether the pressure energy peak corresponds to a rock event; Calculating the matching degree between the time data corresponding to the rock event determined by the pressure energy peak determination function and the time data of the monitored microseismic event; Maximizing the matching degree using a multi-point mutation genetic algorithm to optimize the peak parameters in the pressure energy peak determination function; Using the optimized pressure energy peak determination function, determining whether the pressure energy peak in the filtered pressure energy data corresponds to a rock event; Multiple pressure energy peaks corresponding to rock-type events are screened out.

8. The method according to claim 1, characterized in that: The identifying of downhole fracture propagation events based on the filtered pressure energy data includes: Identify the time of occurrence of downhole fracture propagation events in the target well based on the time points of multiple pressure energy peaks corresponding to rock events in the filtered pressure energy data; The intensity of the downhole fracture propagation event of the target well is identified based on multiple pressure energy peaks corresponding to rock events in the screened pressure energy data.

9. A downhole crack propagation event identification device, characterized in that: The device comprises: An acquisition module is used to acquire a fracturing signal during a fracturing operation, wherein the fracturing signal is used to represent a relationship between operation parameters and time, and the operation parameters include at least pressure and displacement; a decomposition module, configured to perform wavelet decomposition on the fracturing signal to obtain pressure energy data and displacement energy data, wherein the pressure energy data represents a changing relationship between pressure energy and time, and the displacement energy data represents a changing relationship between displacement energy and time; A screening module is used to screen the pressure energy data using the displacement energy data to remove the pressure energy peaks corresponding to non-rock events in the pressure energy data; The identification module is used to identify downhole fracture propagation events based on the filtered pressure energy data.

10. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the method according to any one of claims 1 to 8 is implemented.

Images