A near-bit formation detection method, system, device, electronic device and medium

Abstract

The present invention relates to the technical field of logging while drilling, and discloses a near-bit formation detection method, system, device, electronic device and medium. The method includes: transmitting seismic waves to a formation to be measured in front of the drill bit through a downhole vibrator, and receiving the reflected wave signals of the formation to be measured through a geophone; calculating in real time the relative position of the formation in front of the drill bit from the drill bit according to the reflected wave signals; obtaining the first lithological seismic wave characteristic parameters of the reflected wave signals and the second lithological seismic wave characteristic parameters of known typical formations; matching the first lithological seismic wave characteristic parameters with the second lithological seismic wave characteristic parameters to determine the formation lithology and geological structure of the formation to be measured; outputting the formation information of the formation to be measured including the distance data, formation lithology and geological structure. The seismic waves in this application are not affected by various complex geological environments, and can accurately obtain the formation information of the formation to be measured, reducing the drilling risk and improving the drilling efficiency and safety.

Description

Background Art

[0002] The distance parameter of the forward exploration while drilling is a key geological parameter used in the geosteering drilling system for real-time formation prediction and providing drilling guidance information. The forward exploration while drilling technology is developed on the basis of the traditional logging while drilling technology and is a measurement method that can measure the formation information in front of the drill bit in real time during the drilling process. During the drilling process, the uncertainty of the geological body information in front of the drill bit is the main cause of drilling accidents. Therefore, it is particularly crucial to obtain the distance parameter of the geological body in front of the drill bit in real time. The forward exploration while drilling technology can not only predict the formation structure and formation characteristics to be measured in front of the drill bit during the drilling process; it can predict the high-pressure layer, depleted oil layer, and the top boundary of the complex salt layer in front of the drill bit, etc., to guide geological drilling suspension; it can optimize the cementing casing design (casing shoe position) and adjust the drilling fluid density in real time, reducing the number of times the downhole drill string assembly is taken out of the well; it can reduce drilling uncertainty, optimize the development plan, and make early judgments and disposals for drilling risks, etc.

[0003] In the process of oil drilling and geological exploration, accurately detecting the formation information in front of the drill bit is of great significance for optimizing the drilling path, preventing drilling risks, and improving drilling efficiency. Traditional logging while drilling technologies mostly rely on electromagnetic waves or acoustic waves for measurement. For example, the drill-bit seismic while drilling (R-VSP) technology is used to obtain the formation information in front of the drill bit. The drill-bit seismic while drilling (R-VSP) technology can detect the geological structure within a range of several hundred meters to several kilometers in front of the drill bit, which is helpful for geological prediction on a large scale. The seismic wave data can provide a continuous cross-section of the underground structure, which is helpful for identifying complex structures such as geological faults and salt domes. However, the resolution of the seismic while drilling signal is low, and it cannot guide the drilling engineering at the "meter-level accuracy", and the application effect is poor for deep wells, highly deviated wells, soft formations, etc.

[0004] Another example is to use the electromagnetic wave forward exploration while drilling technology to obtain the formation information in front of the drill bit. The electromagnetic wave forward exploration while drilling technology can provide high-resolution formation electrical property information, and its measurement accuracy is less than 1m, which is helpful for accurately identifying the formation and rock electrical property characteristics in front of the drill bit, has a wider adaptability to mud types, and can work in mud systems with different conductivities. However, the detection depth of electromagnetic waves is relatively shallow. At present, the maximum forward exploration distance of the mature tool is 30m, which is greatly affected by the formation conductivity and the electric field of downhole instruments. The forward exploration distance is unstable, and it is not applicable to formation environments with high resistivity and unclear electrical property characteristics.

[0005] For another example, the technology of acoustic wave forward detection while drilling is used to obtain the formation information in front of the drill bit. The data collected by the acoustic wave forward detection while drilling can be used to determine porosity, pore pressure, rock mechanical parameters, gas layer identification, fracture evaluation, and time-depth relationship correction of seismic data. However, at present, the technology of acoustic wave forward detection while drilling is still in the theoretical stage, and the key factor restricting its development is that there is no effective method to increase the acoustic wave energy radiated in front of the drill bit, which cannot meet the requirements of engineering forward detection.

[0006] In summary, the current traditional measurement-while-drilling technology cannot adapt to various complex geological environments during drilling and cannot obtain accurate formation information in front of the drill bit. Summary of the Invention

[0007] In order to overcome the problems that the current traditional measurement-while-drilling technology cannot adapt to various complex geological environments during drilling and cannot obtain accurate formation information in front of the drill bit, the present invention provides a near-bit formation detection method, system, device, electronic device, and medium.

[0008] In a first aspect, to solve the above technical problems, the present invention provides a near-bit formation detection method, including:

[0009] Emitting seismic waves with a preset frequency and preset energy to the formation to be measured in front of the drill bit through a downhole vibrator, and receiving the reflected wave signals of the formation to be measured through a geophone;

[0010] Calculating the relative position of the formation in front of the drill bit from the drill bit in real time according to the reflected wave signals;

[0011] Obtaining the first lithology seismic wave characteristic parameters of the reflected wave signals and the second lithology seismic wave characteristic parameters of known typical formations;

[0012] Matching the first lithology seismic wave characteristic parameters with the second lithology seismic wave characteristic parameters to determine the formation lithology and geological structure of the formation to be measured;

[0013] Outputting the formation information of the formation to be measured including distance data, formation lithology, and geological structure.

[0014] In a second aspect, the present invention provides a near-bit formation detection system, including:

[0015] A reflected wave signal receiving module, configured to emit seismic waves with a preset frequency and preset energy to the formation to be measured in front of the drill bit through a downhole vibrator, and receive the reflected wave signals of the formation to be measured through a geophone;

[0016] A relative position determination module, configured to calculate the relative position of the formation in front of the drill bit from the drill bit in real time according to the reflected wave signals;

[0017] The lithology seismic wave characteristic parameter acquisition module is used to acquire the first lithology seismic wave characteristic parameters of the reflected wave signal and the second lithology seismic wave characteristic parameters of the known typical formations;

[0018] The formation identification module is used to match the first lithology seismic wave characteristic parameters with the second lithology seismic wave characteristic parameters to determine the formation lithology and geological structure of the formation to be measured;

[0019] The formation information output module is used to output the formation information of the formation to be measured including distance data, formation lithology and geological structure.

[0020] In a third aspect, the present invention provides a near-bit formation detection device, including: a downhole vibrator, a geophone and a terminal module; wherein, the geophone and the downhole vibrator are respectively connected to the terminal module, and the terminal module is used to execute a near-bit formation detection method as described above.

[0021] In a fourth aspect, the present invention provides a computing device, including a memory, a processor and a program stored on the memory and running on the processor, and the processor implements the steps of a near-bit formation detection method as described above when executing the program.

[0022] In a fifth aspect, the present invention provides a computer-readable storage medium, in which instructions are stored, and when the instructions are run on a terminal device, the terminal device is enabled to execute the steps of a near-bit formation detection method as described above.

[0023] The beneficial effects of the present invention are: the downhole vibrator is used to generate seismic waves, and the geophone is used to receive the reflected wave signals of the formation to be measured, so as to calculate the relative position of the formation in front of the drill bit from the reflected wave signals; and the first lithology seismic wave characteristic parameters are extracted from the reflected wave signals and matched with the second lithology seismic wave characteristic parameters of the known typical formations to obtain the formation information of the formation to be measured. The seismic waves of the present application are not affected by various complex geological environments, and the downhole vibrator can generate seismic waves with controllable frequencies and energy intensities, and the frequencies and energy intensities of the forward exploration seismic waves can be adjusted according to actual engineering needs and geological conditions, so as to increase the detection distance of the instrument and improve the signal-to-noise ratio of the received signals, which has the beneficial effect of improving the forward exploration ability of the logging-while-drilling instrument, can accurately obtain the formation information of the formation to be measured, reduce the drilling risk, and improve the drilling efficiency and safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following further describes the present invention with reference to the drawings and embodiments.

[0025] Figure 1 It is a schematic flow chart of a near-bit formation detection method according to an embodiment of the present invention;

[0026] Figure 2 It is a schematic flow chart for preprocessing the reflected wave signal;

[0027] Figure 3 It is a training flow chart of the lithology identification model;

[0028] Figure 4 It is a schematic structural diagram of a near-bit formation detection system according to an embodiment of the present invention;

[0029] Figure 5 It is a schematic structural diagram of a near-bit formation detection device according to an embodiment of the present invention. Specific implementation manners

[0030] The following embodiments are further explanations and supplements to the present invention and do not constitute any limitation to the present invention.

[0031] The following describes a near-bit formation detection method, system, device, electronic device and medium according to an embodiment of the present invention with reference to the accompanying drawings.

[0032] As Figure 1 shown, the present invention provides a near-bit formation detection method, including:

[0033] S1. Transmit seismic waves with a preset frequency and preset energy to the formation to be measured in front of the drill bit through a downhole vibrator, and receive the reflected wave signal of the formation to be measured through a geophone.

[0034] S2. Calculate the relative position of the formation in front of the drill bit from the drill bit in real time according to the reflected wave signal.

[0035] S3. Obtain the first lithology seismic wave characteristic parameters of the reflected wave signal and the second lithology seismic wave characteristic parameters of known typical formations.

[0036] S4. Match the first lithology seismic wave characteristic parameters with the second lithology seismic wave characteristic parameters to determine the formation lithology and geological structure of the formation to be measured.

[0037] S5. Output the formation information of the formation to be measured including distance data, formation lithology and geological structure.

[0038] In this embodiment, a downhole vibrator is used to generate seismic waves, and seismic geophones are used to receive the reflected wave signals of the formation to be measured, so as to calculate the relative position of the formation in front of the drill bit based on the reflected wave signals; and the first lithological seismic wave characteristic parameters are extracted from the reflected wave signals and matched with the second lithological seismic wave characteristic parameters of known typical formations to obtain the formation information of the formation to be measured. The seismic waves of this application are not affected by various complex geological environments, and the downhole vibrator can generate seismic waves with controllable frequencies and energy intensities, and the frequencies and energy intensities of the forward exploration seismic waves are adjusted according to actual engineering needs and geological conditions, thereby increasing the detection distance of the instrument and improving the signal-to-noise ratio of the received signals, having the beneficial effect of improving the forward exploration ability of the logging-while-drilling instrument, being able to accurately obtain the formation information of the formation to be measured, reducing the drilling risk, and improving the drilling efficiency and safety.

[0039] In this embodiment, a downhole vibrator is deployed near the drill bit to emit seismic waves with a specific intensity and a specific frequency. By using its propagation characteristics in the underground structure, effective reflected waves of the formation in front are obtained through seismic geophones, and the geological information of the formation in front is analyzed in real time. This includes but is not limited to key parameters such as the distance, thickness, lithology, fractures, porosity, fluid properties, etc. of the formation. Through high-precision data processing and analysis techniques, accurate information about the formation to be measured in front can be provided for drilling, helping to optimize the drilling plan, reduce the drilling risk, and improve the drilling efficiency and safety.

[0040] In this embodiment, the preset frequency and preset energy are set according to the actual situation.

[0041] In this embodiment, the seismic geophone can adopt a three-component array geophone. The three-component array geophone is a special geophone used in multi-wave exploration. Three mutually perpendicular sensors are installed inside, which respectively record the three components of the particle vibration velocity vector, that is, the seismic geophone can simultaneously receive vibrations in three directions (usually the X, Y, and Z axes), which are respectively:

[0042] Amplitude in the X-axis direction: The amplitude that the seismic geophone can receive in the X-axis direction, which represents a component in the horizontal direction.

[0043] Amplitude in the Y-axis direction: Similar to the X-axis, the Y-axis direction is also another component in the horizontal direction and can capture the amplitude in this direction.

[0044] Amplitude in the Z-axis direction: The Z-axis direction is usually the vertical direction, and the seismic geophone can also receive and measure the amplitude in this direction.

[0045] Based on the above, the three-component seismic data are the amplitudes in three directions (the X, Y, and Z axes), and it is a prior art for the seismic geophone to determine the three-component seismic data based on the reflected wave signals, so it will not be elaborated here.

[0046] In this embodiment, after obtaining the formation information corresponding to the formation to be measured, only the predicted formation information is obtained. The true formation information needs to iteratively update the predicted formation information according to the true formation information measured by the measurement while drilling.

[0047] Optionally, extracting the first lithology seismic wave characteristic parameters of the reflected wave signal includes:

[0048] Performing signal amplification, baseline correction, time synchronization, noise identification, filtering operation, denoising processing, and signal extraction on the reflected wave signal in sequence to determine the target reflected wave signal;

[0049] Extracting the first lithology seismic wave characteristic parameters of the target reflected wave signal.

[0050] As Figure 2 shown, it is necessary to perform signal preprocessing (signal amplification, baseline correction, time synchronization), noise identification, filtering operation (designing a filter, filtering operation), effect evaluation and adjustment (denoising processing), and signal extraction on the reflected wave signal in sequence. The specific process is as follows:

[0051] (1) Performing signal amplification, baseline correction, and time synchronization on the reflected signal can enhance the useful signal and suppress noise to improve the quality of the seismic signal.

[0052] In this embodiment, performing noise identification on the reflected signal can identify and distinguish background noise and effective signals to improve the signal-to-noise ratio of the signal.

[0053] Among them, methods such as spectral analysis can be used to determine the frequency and intensity ranges where the noise in the reflected signal is mainly concentrated.

[0054] (2) Performing a filtering operation on the reflected signal can further purify the effective signal.

[0055] Among them, it is necessary to first design a filter. According to the results of noise identification, an appropriate filter is designed to remove noise. This includes low-pass, high-pass, band-pass, or band-stop filters, or even more complex adaptive filters or multiple filtering schemes to specifically suppress noise in a specific frequency range.

[0056] In addition, applying the designed filter to the data to perform the actual filtering operation. Using digital signal processing software and algorithms, through techniques such as convolution and Fourier transform, the data is filtered in the frequency domain or time domain.

[0057] (3) Performing denoising processing on the reflected signal can perform quality evaluation on the filtered signal to achieve the optimal filtering effect.

[0058] Among them, after filtering, it is necessary to evaluate the filtering effect, confirm whether the noise is effectively removed, and at the same time ensure that the effective signal is not over-attenuated or distorted. Compare the signal spectra, waveforms and other parameters before and after filtering. According to the evaluation results, adjust the filter parameters to achieve the best filtering effect.

[0059] (4) Signal extraction of the reflected signal can extract the effective reflected wave signal for further lithologic formation characteristic analysis and interpretation.

[0060] Among them, after the filtering process, the background noise and interference signals are effectively removed, and at this time, the effective formation reflected wave signal can be more clearly identified. Further signal extraction is carried out to facilitate the analysis and interpretation of formation information.

[0061] In addition, the entire filtering process needs to comprehensively consider the characteristics of the signal, the nature of the noise, and the analysis objectives. Through iterative adjustment and optimization, ensure the accurate extraction of the effective signal, providing a basis for subsequent geological interpretation.

[0062] Optionally, extract the first lithologic seismic wave characteristic parameters of the target reflected wave signal, including:

[0063] Extract the FFT amplitude spectrum, energy spectrum and complex cepstrum from the target reflected wave signal to determine the spectral characteristic data;

[0064] Extract the frequency-wavenumber spectrum and time-frequency analysis parameters from the target reflected wave signal to determine the time-frequency characteristic data;

[0065] The first lithologic seismic wave characteristic parameters include spectral characteristic data and time-frequency characteristic data.

[0066] In this embodiment, the target reflected wave signal is converted into common seismic wave data such as spectral characteristic data and time-frequency characteristic data, so as to better match the seismic wave data corresponding to known typical formations.

[0067] Among them, the spectral characteristic data includes frequency-domain characteristic data such as FFT amplitude spectrum, energy spectrum, and complex cepstrum extracted from the reflected wave signal, as follows:

[0068] (1) Through FFT transformation, the reflected wave signal is converted from the time domain to the frequency domain, and the amplitude spectrum and energy spectrum are extracted, thereby revealing the frequency content and energy distribution of the reflected wave signal, which is very important for identifying formation interfaces and rock properties. The calculation formula is:

[0069]

[0070] Among them, x(n) is the reflected wave signal corresponding to the nth sampling point in the time domain, X(k) is the kth frequency component in the frequency domain, and N is the total number of sampling points.

[0071] The spectral distribution can be obtained by analyzing all frequency components X(k), that is, the frequency content (amplitude spectrum) and energy distribution (energy spectrum) of the reflected wave signal.

[0072] (2) Cepstrum analysis can separate the reverberation components in seismic signals, helping to extract the periodic characteristics of the reflection sequence, which is useful for identifying waveforms in multi-layer media. The cepstrum is obtained by performing an inverse Fourier transform on the logarithmic amplitude spectrum of the reflected wave signal, and the formula is as follows:

[0073]

[0074] c(n) represents the cepstrum, represents the inverse Fourier transform.

[0075] In addition, the time-frequency characteristic data mainly includes time-frequency domain characteristic data extracted from the reflected wave signal, such as the frequency-wavenumber spectrum (f-k spectrum), time-frequency analysis parameters (such as the short-time Fourier transform STFT), etc., which are as follows:

[0076] (1) The frequency-wavenumber spectrum (f-k spectrum) can be used to analyze the propagation characteristics of seismic waves in detail, such as wave velocity analysis and determination of wave propagation direction, by analyzing the distribution of the reflected wave signal in the frequency-wavenumber domain. It is based on the two-dimensional Fourier transform. For the two-dimensional wave function u(t,x), its relationship with the corresponding frequency-wavenumber spectrum function U(f,k) is:

[0077]

[0078] (2) Time-frequency analysis parameters, such as time-frequency analysis techniques like the short-time Fourier transform (STFT), can be used to describe the time-frequency characteristics of the reflected wave signal and provide information on the changes of seismic waves in time and frequency:

[0079] STFT(x)(τ,ω) = ∫x(t)ω(t - τ)e -iωt dt

[0080] where ω(t) is the window function, τ and ω are variables in the time and frequency domains respectively, and STFT(x)(τ,ω) represents the short-time Fourier transform function.

[0081] In this embodiment, the commonly used characteristic sequence length and dimension depend on the purpose of data processing and the required resolution. For example, for frequency domain characteristics, the characteristic sequence length of the FFT amplitude spectrum and energy spectrum may be between 128 and 512 data sampling points. For more detailed characteristics, such as the phase spectrum or time-frequency characteristics, different dimension selections may be available.

[0082] Optionally, the relative position of the formation in front of the drill bit with respect to the drill bit is calculated in real time according to the reflected wave signal, including:

[0083] Obtain the propagation velocity of seismic waves in the formation to be measured, and the two-way travel time of seismic waves from the downhole vibrator until the seismic detector receives the reflected wave signal;

[0084] Determine the distance between the drill bit and the formation to be measured according to the propagation velocity and the two-way travel time;

[0085] Extract the three-component seismic data of the reflected wave signal, and determine the formation dip angle and azimuth angle of the formation to be measured according to the three-component seismic data.

[0086] In this embodiment, the three-component seismic data includes the first amplitude generated in the X-axis direction, the second amplitude generated in the Y-axis direction, and the third amplitude generated in the Z-axis direction after the seismic waves reach the formation to be measured;

[0087] In this embodiment, according to the first amplitude, the second amplitude, and the third amplitude, determine the formation dip angle of the formation to be measured, and the formula is as follows:

[0088]

[0089] tan(θ) represents the formation dip angle, A x 、A y and A z are the first amplitude, the second amplitude, and the third amplitude respectively.

[0090] According to the first amplitude and the second amplitude, determine the azimuth angle of the formation to be measured, and the formula is as follows:

[0091]

[0092] α represents the azimuth angle.

[0093] According to the propagation velocity and the two-way travel time, determine the distance between the drill bit and the formation to be measured, and the formula is as follows:

[0094]

[0095] Among them, d represents the distance, v represents the propagation velocity, and t represents the two-way travel time.

[0096] In this embodiment, when seismic waves radiate to the formation to be measured in front of the drill bit, if they encounter geological structures such as fractures, faults, and interfaces, the seismic waves are reflected or scattered back from the geological structures and received by the seismic detector. Based on this principle, by receiving and analyzing these reflected waves, the lithological characteristics and interface information of the formation can be accurately detected.

[0097] Optionally, match the first lithological seismic wave characteristic parameter with the second lithological seismic wave characteristic parameter to determine the formation lithology and geological structure of the formation to be measured, including:

[0098] Input the first lithology seismic wave characteristic parameters of the formation to be measured into the lithology identification model;

[0099] Use the lithology identification model to match the first lithology seismic wave characteristic parameters of the formation to be measured with the second lithology seismic wave characteristic parameters of known typical formations, and determine the target known typical formation;

[0100] Take the formation lithology and geological structure corresponding to the target known typical formation as the formation lithology and geological structure of the formation to be measured.

[0101] In this embodiment, the use of machine learning and artificial intelligence technologies can effectively learn and extract complex features in seismic data, and then predict the lithology of unknown formations, which has important value in improving the accuracy of seismic data interpretation, reducing risks, and guiding oil and gas exploration and development activities.

[0102] Optionally, the training process of the lithology identification model includes:

[0103] Obtain the reflected wave signals, formation lithologies, and geological structures corresponding to different known typical formations;

[0104] Use the reflected wave signals, formation lithologies, and geological structures corresponding to each known typical formation to construct a training set and a test set, and perform preprocessing on both the training set and the test set by data cleaning, filtering, resampling, removing multiple waves, and removing interference waves;

[0105] Extract spectral feature data and time-frequency feature data from the reflected wave signals in the training set, and take the spectral feature data, time-frequency feature data, formation lithology, and geological structure corresponding to each known typical formation in the training set as the target training set;

[0106] Extract spectral feature data and time-frequency feature data from the reflected wave signals in the test set, and take the spectral feature data, time-frequency feature data, formation lithology, and geological structure corresponding to each known typical formation in the test set as the target test set;

[0107] Input the target training set into the deep neural network model for training, and test the trained deep neural network model through the target test set to determine the lithology identification model.

[0108] As Figure 3 shown, select the typical formation lithology as the known typical formation, and the typical seismic wave data (reflected wave signals, formation lithology, and geological structure) encountered in the same work area and adjacent wells should be stored as early as possible. The typical formation lithology can include, for example, sandstone, mudstone, shale, limestone, dolomite, granite, carbonate rock, igneous rock, coal seam, etc. In the case of unconditional early data acquisition, sufficient typical rock seismic wave data in adjacent areas should be collected and stored.

[0109] After collecting the seismic wave data, it is necessary to construct a training set and a test set. The training set and the test set can be allocated according to a ratio of 4:1 or 3:1 to ensure that there is enough data to train the deep neural network model, and at the same time, the validation set can evaluate the generalization ability of the model.

[0110] Then, it is necessary to preprocess the seismic wave data in the training set and the test set, including data cleaning, filtering, resampling, removing multiple waves, and removing interference waves, where:

[0111] Data cleaning: This step involves checking the consistency of the seismic wave data, including identifying and processing outliers, invalid values, and missing values. This can be done by comparing the reflected wave signals received by different three-component array geophones or using statistical methods. Data cleaning ensures the reliability and consistency of the data and lays the foundation for accurate analysis.

[0112] Filtering: Filtering is a key step in the preprocessing of seismic wave data, aiming to remove the noise unrelated to the reflection signals of the geological structure. The types of filtering can include: ① High-pass filtering: Removing the noise components with frequencies lower than a specific threshold, such as slow seismic waves and microseismic noise. ② Low-pass filtering: Removing high-frequency noise, such as circuit noise generated by electronic devices or high-frequency vibrations. ③ Band-pass filtering: Combining high-pass and low-pass filtering to retain only the data within a specific frequency band. The band-pass filtering frequency range needs to be optimized according to the characteristics of the target signal and the test results of the field environment. Generally, for seismic wave data, it may be between 10 Hz and several hundred Hz.

[0113] Resampling: The seismic wave data may need to be resampled to meet the requirements of subsequent processing steps. If the original sampling rate is too high, the sampling rate can be reduced to reduce the data volume and improve the processing efficiency. Conversely, if the original data sampling rate is insufficient to capture the reflected wave signals reflected by the target formation, the sampling rate may need to be increased.

[0114] Removing multiple waves and removing interference waves: The preprocessing of seismic wave data requires removing multiple waves and interference waves. These are usually caused by multiple reflections of waves in the underground structure and will interfere with the identification of the main reflection signals. After removing the above noise, signal enhancement techniques are applied to improve the signal-to-noise ratio of the effective seismic reflection signals.

[0115] Secondly, after the seismic wave data is preprocessed as above, the spectral feature data and time-frequency feature data of the reflected wave signals in the training set and the test set are extracted. The specific algorithms have been described above, so they will not be elaborated here.

[0116] Finally, the target training set and the target test set of the typical formation lithology are obtained and input into the deep neural network to train the lithology identification model, and the target known typical formation matching the formation to be measured can be obtained through the lithology identification model.

[0117] In this embodiment, appropriate initialization parameters need to be selected during model training, including the learning rate, loss function, optimization algorithm, etc. According to the characteristics of seismic waves in the constructed target training set and the requirements of the intelligent model, the length of the input feature vector is adjusted. The length may affect the training efficiency and recognition accuracy of the model. Through a large number of model trainings and comparative tests, the algorithm parameters are optimized, and the accuracy of the model is verified.

[0118] In this embodiment, different types of machine learning algorithms can be selected, such as traditional methods like BP neural network, support vector machine, random forest, etc., or modern intelligent algorithms like convolutional neural network, deep learning, transfer learning, etc. Multiple seismic wave features can be used as the input data for the algorithm, and multiple intelligent lithology recognition algorithms can be combined to improve the recognition accuracy through model fusion. Statistical analysis methods are used to evaluate and refine the final lithology recognition results, improving the reliability of lithology judgment of the formation to be measured in front of the drill bit.

[0119] As Figure 4 shown, the present invention provides a near-bit formation detection system, including:

[0120] A reflected wave signal receiving module, configured to emit seismic waves with a preset frequency and preset energy to the formation to be measured in front of the drill bit through a downhole vibrator, and receive the reflected wave signal of the formation to be measured through a geophone;

[0121] A relative position determination module, configured to calculate the relative position of the formation in front of the drill bit from the drill bit in real time according to the reflected wave signal;

[0122] A lithology seismic wave characteristic parameter acquisition module, configured to acquire the first lithology seismic wave characteristic parameter of the reflected wave signal and the second lithology seismic wave characteristic parameter of a known typical formation;

[0123] A formation identification module, configured to match the first lithology seismic wave characteristic parameter with the second lithology seismic wave characteristic parameter to determine the formation lithology and geological structure of the formation to be measured;

[0124] A formation information output module, configured to output the formation information of the formation to be measured including distance data, formation lithology and geological structure.

[0125] Optionally, the acquisition of the lithology seismic wave characteristic parameter is specifically used for:

[0126] Performing signal amplification, baseline correction, time synchronization, noise identification, filtering operation, denoising processing and signal extraction on the reflected wave signal in sequence to determine the target reflected wave signal;

[0127] Extracting the first lithology seismic wave characteristic parameter of the target reflected wave signal.

[0128] Optionally, the acquisition of the lithology seismic wave characteristic parameter is specifically used for:

[0129] Extract the FFT amplitude spectrum, energy spectrum, and complex cepstrum from the target reflected wave signal to determine the spectral characteristic data;

[0130] Extract the frequency-wavenumber spectrum and time-frequency analysis parameters from the target reflected wave signal to determine the time-frequency characteristic data;

[0131] The first lithologic seismic wave characteristic parameters include spectral characteristic data and time-frequency characteristic data.

[0132] Optionally, the relative position determination module is specifically used for:

[0133] Obtain the propagation velocity of the seismic wave in the formation to be measured, and the two-way travel time of the seismic wave from the downhole vibrator until the reflected wave signal is received by the geophone;

[0134] Determine the distance between the drill bit and the formation to be measured according to the propagation velocity and the two-way travel time;

[0135] Extract the three-component seismic data of the reflected wave signal, and determine the formation dip angle and azimuth angle of the formation to be measured according to the three-component seismic data.

[0136] Optionally, the formation identification module is specifically used for:

[0137] Input the first lithologic seismic wave characteristic parameters of the formation to be measured into the lithology identification model;

[0138] Use the lithology identification model to match the first lithologic seismic wave characteristic parameters of the formation to be measured with the second lithologic seismic wave characteristic parameters of the known typical formations to determine the target known typical formation;

[0139] Take the formation lithology and geological structure corresponding to the target known typical formation as the formation lithology and geological structure of the formation to be measured.

[0140] Optionally, the system further includes a model training module, which is specifically used for:

[0141] Obtain the reflected wave signals, formation lithologies, and geological structures corresponding to different known typical formations;

[0142] Use the reflected wave signals, formation lithologies, and geological structures corresponding to each known typical formation to construct a training set and a test set, and perform preprocessing on both the training set and the test set, including data cleaning, filtering, resampling, removing multiple waves, and removing interference waves;

[0143] Extract the spectral characteristic data and time-frequency characteristic data from the reflected wave signals in the training set, and take the spectral characteristic data, time-frequency characteristic data, formation lithologies, and geological structures corresponding to each known typical formation in the training set as the target training set;

[0144] Extract spectral feature data and time-frequency feature data from the reflected wave signals in the test set, and use the spectral feature data, time-frequency feature data, formation lithology, and geological structure corresponding to each known typical formation in the test set as the target test set.

[0145] Input the target training set into the deep neural network model for training, and test the trained deep neural network model with the target test set to determine the lithology identification model.

[0146] As Figure 5 shown, the present invention provides a near-bit formation detection device, including: a downhole vibrator O, a geophone R, and a terminal module (not shown); wherein, the geophone R and the downhole vibrator O are respectively connected to the terminal module, and the terminal module is used to execute any one of the near-bit formation detection methods in the above embodiments.

[0147] In addition, Figure 5 in which A and B are both formations, and C is the formation interface between the two formations.

[0148] The embodiment of the present invention also provides a computing device, including a memory, a processor, and a program stored on the memory and running on the manager. When the manager executes the program, it implements some or all of the steps of the above near-bit formation detection method.

[0149] Among them, the computing device can be a computer. Correspondingly, its program is computer software, and the above parameters and steps in the computing device of the present invention can refer to the parameters and steps in the embodiments of the near-bit formation detection method in the above text, which will not be elaborated here.

[0150] Those skilled in the art know that the present invention can be implemented as a system, a method, or a computer program product. Therefore, the present disclosure can be specifically implemented in the following forms: it can be completely hardware, can also be completely software (including firmware, resident software, microcode, etc.), or can also be a combination of hardware and software, which is generally referred to as "circuit", "module" or "system" in this article. In addition, in some embodiments, the present invention can also be implemented in the form of a computer program product in one or more computer-readable media, which contains computer-readable program code. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or component, or any combination of the above.

[0151] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples", etc. means that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, without contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0152] Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present invention.

Claims

1. A near-bit formation detection method, characterized in that, Including: Emitting seismic waves with a preset frequency and preset energy from an underground vibrator to the formation to be measured in front of the drill bit, and receiving the reflected wave signals of the formation to be measured through a geophone; Calculating in real time the relative position of the formation in front of the drill bit from the drill bit according to the reflected wave signals; Obtaining the first lithologic seismic wave characteristic parameters of the reflected wave signals and the second lithologic seismic wave characteristic parameters of known typical formations; Matching the first lithologic seismic wave characteristic parameters with the second lithologic seismic wave characteristic parameters to determine the formation lithology and geological structure of the formation to be measured; Outputting the formation information of the formation to be measured including the distance data, formation lithology and geological structure; The matching the first lithologic seismic wave characteristic parameters with the second lithologic seismic wave characteristic parameters to determine the formation lithology and geological structure of the formation to be measured includes: Inputting the first lithologic seismic wave characteristic parameters of the formation to be measured into a lithology identification model; Using the lithology identification model to match the first lithologic seismic wave characteristic parameters of the formation to be measured with the second lithologic seismic wave characteristic parameters of known typical formations to determine the target known typical formation; Taking the formation lithology and geological structure corresponding to the target known typical formation as the formation lithology and geological structure of the formation to be measured.

2. The method according to claim 1, wherein The obtaining the first lithologic seismic wave characteristic parameters of the reflected wave signals includes: Successively performing signal amplification, baseline correction, time synchronization, noise identification, filtering operation, denoising processing and signal extraction on the reflected wave signals to determine the target reflected wave signals; Extracting the first lithologic seismic wave characteristic parameters of the target reflected wave signals.

3. The method according to claim 2, wherein The extracting the first lithologic seismic wave characteristic parameters of the target reflected wave signals includes: Extracting the FFT amplitude spectrum, energy spectrum and complex cepstrum from the target reflected wave signals to determine the spectral characteristic data; Extracting the frequency-wavenumber spectrum and time-frequency analysis parameters from the target reflected wave signals to determine the time-frequency characteristic data; The first lithologic seismic wave characteristic parameters include the spectral characteristic data and the time-frequency characteristic data.

4. The method according to claim 1, characterized in that, The calculating in real time the relative position of the formation in front of the drill bit from the drill bit according to the reflected wave signals includes: Obtaining the propagation speed of the seismic waves in the formation to be measured and the two-way time from when the seismic waves are emitted from the underground vibrator until the reflected wave signals are received by the geophone; Determining the distance between the drill bit and the formation to be measured according to the propagation speed and the two-way time; Extracting the three-component seismic data of the reflected wave signals and determining the formation dip angle and azimuth angle of the formation to be measured according to the three-component seismic data.

5. The method according to claim 1, wherein The training process of the lithology identification model includes: Obtaining the reflected wave signals, formation lithologies and geological structures corresponding to different known typical formations; Using the reflected wave signals, formation lithologies and geological structures corresponding to each known typical formation to construct a training set and a test set, and preprocessing the training set and the test set; Extracting spectral characteristic data and time-frequency characteristic data from the reflected wave signals in the training set, and taking the spectral characteristic data, time-frequency characteristic data, formation lithologies and geological structures corresponding to each known typical formation in the training set as the target training set; Extract spectral feature data and time-frequency feature data from the reflected wave signals in the test set, and use the spectral feature data, time-frequency feature data, formation lithology, and geological structure corresponding to each known typical formation in the test set as the target test set; Input the target training set into a deep neural network model for training, and test the trained deep neural network model with the target test set to determine a lithology identification model.

6. A near-bit formation detection system, characterized in that, Including: A reflected wave signal receiving module, configured to emit seismic waves with a preset frequency and preset energy to the formation to be measured in front of the drill bit through a downhole vibrator, and receive the reflected wave signals of the formation to be measured through a geophone; A relative position determination module, configured to calculate the relative position of the formation in front of the drill bit from the drill bit in real time according to the reflected wave signals; A lithology seismic wave feature parameter acquisition module, configured to acquire first lithology seismic wave feature parameters of the reflected wave signals and second lithology seismic wave feature parameters of known typical formations; A formation identification module, configured to match the first lithology seismic wave feature parameters with the second lithology seismic wave feature parameters to determine the formation lithology and geological structure of the formation to be measured; A formation information output module, configured to output the formation information of the formation to be measured including the distance data, formation lithology, and geological structure; The formation identification module is specifically configured to: Input the first lithology seismic wave feature parameters of the formation to be measured into the lithology identification model; Use the lithology identification model to match the first lithology seismic wave feature parameters of the formation to be measured with the second lithology seismic wave feature parameters of known typical formations to determine the target known typical formation; Use the formation lithology and geological structure corresponding to the target known typical formation as the formation lithology and geological structure of the formation to be measured.

7. A near-bit formation detection device, characterized in that, Including: A downhole vibrator, a geophone, and a terminal module; wherein, the geophone and the downhole vibrator are respectively connected to the terminal module, and the terminal module is configured to execute a near-bit formation detection method according to any one of claims 1-5.

8. A computing device, comprising a memory, a processor, and a program stored on the memory and running on the processor, characterized in that, When the processor executes the program, it implements the steps of a near-bit formation detection method according to any one of claims 1-5.

9. A computer-readable storage medium, characterized in that, Instructions are stored in a computer-readable storage medium, and when the instructions are run on a terminal device, the terminal device is caused to execute the steps of a near-bit formation detection method according to any one of claims 1-5.

Images