Mud pulse target signal extraction method and device based on multi-peak fitting
Through the multimodal fitting method, the mud pulse signal is processed and reconstructed, which solves the problems of signal baseline correction and noise interference in the prior art, and realizes efficient signal extraction and analysis, which is suitable for drilling measurement under deep and complex formation conditions.
Patent Information
- Application Number
- CN202311593985.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-11-27
- Publication Date
- 2025-05-27
AI Technical Summary
In the existing drilling-as-a-drilling measurement technology, mud pulse signals are disturbed by noise during transmission, signal baseline correction is difficult, and the flexibility and effectiveness of wavelet decomposition and EMD methods are reduced under deep complex formation conditions.
Using a multimodal fitting method, by obtaining the initial mud pulse signal and processing iteration, the unimodal function is selected as the fitting substrate, and multimodal fitting iteration is performed by combining the particle swarm optimization algorithm to obtain the optimal multimodal eigenvector and reconstructing the signal.
It effectively solves the problem of signal baseline drift, improves signal analyticity, retains the time domain characteristics of the original signal, improves the waveform, and expands the scope of application.
Smart Images
Figure CN120045835A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of measurement while drilling, and is a method and device for extracting target signals of mud pulses based on multi-peak fitting. Background Art
[0002] During drilling operations, the real-time transmission of drilling parameter data from the wellbore to the ground plays a crucial role in guiding drilling operations and making risk control decisions. The measurement while drilling (MWD) technology uses the circulating mud in the drill string as a transmission carrier, and the method of transmitting signals through mud pulses is an important means to realize the upload and analysis of downhole engineering data.
[0003] However, during the signal upload process of MWD, since the mud pressure sensor is at the ground drilling port and is very close to the circulating system such as the mud pump, the noise has complex and high-intensity characteristics, resulting in these noises directly drowning out the target signal. At the same time, the channel environment is harsh, and the long transmission distance also causes a large attenuation of the target signal, and the distortion and distortion phenomena are serious.
[0004] To address this problem, currently commonly used signal analysis and processing means such as wavelet transform and EMD empirical mode decomposition are used to extract target signals. However, these methods have problems such as difficulty in identifying the basic pump pressure and difficulty in signal baseline correction, and need to be further optimized by assisting other methods; in addition, with the current development of oil and gas exploration towards deeper and more complex formations, it is difficult to determine the wavelet decomposition basis and decomposition layers, and the iterative termination conditions of EMD decomposition, which greatly reduces their flexibility and effectiveness, and the applicable range is also greatly limited. Summary of the Invention
[0005] The present invention provides a method and device for extracting target signals of mud pulses based on multi-peak fitting, which overcomes the deficiencies of the above-mentioned prior art and can effectively solve the problems of difficult signal baseline correction and baseline drift in the waveforms processed by existing mud pulse signal analysis and processing methods.
[0006] One of the technical solutions of the present invention is achieved by the following measures: A method for extracting target signals of mud pulses based on multi-peak fitting, including:
[0007] Obtain the initial mud pulse signal at the surface riser and process it to obtain the mud pressure pulse signal in the second stage;
[0008] Select a fitting basis, use the fitting basis to perform multi-peak fitting on the mud pressure pulse signal in the second stage, and determine the maximum number of iterations, fitness function, and corresponding threshold, where the fitting basis is selected from unimodal functions;
[0009] Select an optimization algorithm, and perform multi-peak fitting iteration on the mud pulse signal in the second stage based on the maximum number of iterations, fitness function, and corresponding threshold to obtain the optimal multi-peak feature vector;
[0010] Use the optimal multi-peak feature vector and fitting basis to obtain the reconstructed mud pressure pulse signal.
[0011] The following is a further optimization or / and improvement of the above technical solution of the invention:
[0012] The above selection of the fitting basis, performing multi-peak fitting on the mud pressure pulse signal in the second stage using the fitting basis, and determining the maximum number of iterations, fitness function, and corresponding threshold include:
[0013] Perform differential analysis on the mud pressure pulse signal in the second stage to obtain the number, position, and value of the maximum points, and use the number of maximum points as the maximum number of iterations of the optimization algorithm;
[0014] Select a fitting basis, perform multi-peak fitting on the mud pressure pulse signal in the second stage using the fitting basis to obtain the multi-peak fitted mud pressure pulse signal, where the fitting basis is selected from unimodal functions;
[0015] Based on the mud pressure pulse signal in the second stage and the multi-peak fitted mud pressure pulse signal, obtain the fitting residual, use the fitting residual as the fitness function of the optimization algorithm, and set the corresponding fitness value as the threshold;
[0016]
[0017] Among them, h(x) is the fitting residual; F(x) is the multi-peak fitted mud pressure pulse signal; G(x) is the mud pressure pulse signal in the second stage.
[0018] If the above fitting basis is a Gaussian unimodal function, then perform multi-peak fitting on the mud pressure pulse signal in the second stage using the Gaussian unimodal function to obtain the multi-peak fitted mud pressure pulse signal, where the multi-peak fitted mud pressure pulse signal is as follows:
[0019]
[0020] Among them, F() is the fitted mud pressure pulse signal; n is the number of fitted peaks; α, β, γ are multi-peak feature vectors.
[0021] If the selected optimization algorithm is the particle swarm optimization algorithm, then perform multi-peak fitting on the mud pulse signal in the second stage based on the particle swarm optimization algorithm, maximum number of iterations, fitness function, and corresponding threshold to obtain the optimal multi-peak feature vector, including:
[0022] Initialize the particle swarm parameters;
[0023] Calculate the fitness value of the particles, and update the individual extreme value Pbest, the global extreme value gbest, the particle positions, and the particle velocities;
[0024] Determine whether the current fitness value is less than the threshold;
[0025] If so, end the optimization process; if not, determine whether the current iteration number is less than the maximum number of iterations;
[0026] If so, end the optimization process; if not, increment the iteration number by one and continue to calculate the fitness value of the particles.
[0027] The above-mentioned obtaining the initial mud pulse signal at the surface riser and processing it to obtain the mud pressure pulse signal in the second stage includes:
[0028] Obtain the initial mud pulse signal obtained by the mud pressure pulse sensor arranged at the surface riser;
[0029] Perform Savitzky-Golay convolution smoothing filtering on the initial mud pulse signal to obtain the mud pressure pulse signal in the second stage.
[0030] The second technical solution of the present invention is achieved by the following measures: A mud pulse target signal extraction device based on multi-peak fitting, including:
[0031] A signal acquisition unit, which acquires the initial mud pulse signal at the surface riser and processes it to obtain the mud pressure pulse signal in the second stage;
[0032] A first analysis unit, which selects a fitting basis, performs multi-peak fitting on the mud pressure pulse signal in the second stage using the fitting basis, determines the maximum number of iterations, the fitness function, and the corresponding threshold, where the fitting basis is selected from unimodal functions;
[0033] A second analysis unit, which selects an optimization algorithm and performs multi-peak fitting iteration on the mud pulse signal in the second stage based on the maximum number of iterations, the fitness function, and the corresponding threshold to obtain the optimal multi-peak feature vector;
[0034] A third analysis unit, which uses the optimal multi-peak feature vector and the fitting basis to obtain the reconstructed mud pressure pulse signal.
[0035] The following is a further optimization and / or improvement of the above-mentioned invention technical solution:
[0036] The above-mentioned first analysis unit includes:
[0037] The first analysis module performs differential analysis on the mud pressure pulse signal in the second stage to obtain the number, positions, and values of the maximum points, and uses the number of maximum points as the maximum number of iterations of the optimization algorithm;
[0038] The second analysis module selects a fitting basis and uses the fitting basis to perform multi-peak fitting on the mud pressure pulse signal in the second stage to obtain the multi-peak fitted mud pressure pulse signal;
[0039] The third analysis module obtains the fitting residual based on the mud pressure pulse signal in the second stage and the multi-peak fitted mud pressure pulse signal, uses the fitting residual as the fitness function of the optimization algorithm, and sets the corresponding fitness value as the threshold;
[0040]
[0041] where h(x) is the fitting residual; F(x) is the multi-peak fitted mud pressure pulse signal; G(x) is the mud pressure pulse signal in the second stage.
[0042] The third technical solution of the present invention is achieved by the following measures: A mud pulse target signal extraction system based on multi-peak fitting, comprising:
[0043] A mud pulse target signal extraction device based on multi-peak fitting, wherein the mud pulse target signal extraction device based on multi-peak fitting is the mud pulse target signal extraction device according to claims 6 to 7;
[0044] An interaction device, which is used for an operator to communicate with the mud pulse target signal extraction device based on multi-peak fitting, so that the operator can provide the data required for mud pulse target signal extraction for the mud pulse target signal extraction device based on multi-peak fitting through the interaction unit.
[0045] The fourth technical solution of the present invention is achieved by the following measures: A storage medium, characterized in that a computer program readable by a computer is stored on the storage medium, and the computer program is set to execute the mud pulse target signal extraction method based on multi-peak fitting when running.
[0046] The fifth technical solution of the present invention is achieved by the following measures: An electronic device, characterized in that it includes a processor and a memory, and a computer program is stored in the memory, and the computer program is loaded and executed by the processor to implement the mud pulse target signal extraction method based on multi-peak fitting.
[0047] The present invention uses a unimodal function as the fitting basis, determines the maximum number of iterations, fitness function and corresponding thresholds, and combines an optimization algorithm to perform multi-peak fitting iteration on the mud pulse signal in the second stage to obtain the optimal multi-peak feature vector. Then, the reconstructed mud pressure pulse signal is obtained using the optimal multi-peak feature vector and the fitting basis. Thus, the position, shape and intensity points of each peak can be accurately captured, the baseline drift problem that is difficult to solve by EMD iteration and wavelet decomposition can be solved, and the time-domain characteristics of the original signal are retained to the greatest extent, and the waveform is greatly improved, making the reconstructed mud pressure pulse signal more analyzable. In addition, the multi-peak parameters obtained by multi-peak fitting include pulse width and height information, which can be directly analyzed and processed by the decoding system to complete the analysis of downhole drilling parameters. BRIEF DESCRIPTION OF THE DRAWINGS
[0048] Attached Figure 1 is a schematic flow chart of the method of the present invention.
[0049] Attached Figure 2 is a schematic flow chart of the method for determining relevant data based on the fitting basis in the present invention.
[0050] Attached Figure 3 is a schematic flow chart of the operation method of the particle swarm optimization algorithm in the present invention.
[0051] Attached Figure 4 is the Savitzky-Golay convolution smoothing effect diagram of the initial mud pressure pulse signal in Embodiment 3 of the present invention.
[0052] Attached Figure 5 is the effect diagram of each unimodal signal finally obtained by optimization in Embodiment 3 of the present invention.
[0053] Attached Figure 6 is the effect diagram of the multi-peak reconstructed signal in Embodiment 3 of the present invention.
[0054] Attached Figure 7 is the EMD iteration processing effect diagram in Embodiment 3 of the present invention.
[0055] Attached Figure 8 is the db8 wavelet decomposition processing effect diagram in Embodiment 3 of the present invention.
[0056] Attached Figure 9 is a schematic diagram of the device structure of the present invention.
[0057] Attached Figure 10 is a schematic diagram of the system structure of the present invention. DETAILED DESCRIPTION OF THE INVENTION
[0058] The present invention is not limited by the following embodiments, and the specific implementation manners can be determined according to the technical solutions of the present invention and the actual situation.
[0059] The present invention will be further described below in conjunction with embodiments and the accompanying drawings:
[0060] Embodiment 1: As shown in the Figure 1 accompanying drawings, an embodiment of the present invention discloses a method for extracting a mud pulse target signal based on multi-peak fitting, including:
[0061] Step S110: Obtain the initial mud pulse signal at the surface riser and process it to obtain the mud pressure pulse signal in the second stage;
[0062] Step S120: Select a fitting basis, perform multi-peak fitting on the mud pressure pulse signal in the second stage using the fitting basis, and determine the maximum number of iterations, fitness function, and corresponding threshold, where the fitting basis is selected from unimodal functions;
[0063] Select a fitting basis, perform multi-peak fitting on the mud pressure pulse signal in the second stage using the fitting basis, and determine the maximum number of iterations, fitness function, and corresponding threshold, where the fitting basis is selected from unimodal functions;
[0064] Step S130: Select an optimization algorithm and perform multi-peak fitting iteration on the mud pulse signal in the second stage based on the maximum number of iterations, fitness function, and corresponding threshold to obtain the optimal multi-peak feature vector;
[0065] Step S140: Use the optimal multi-peak feature vector and the fitting basis to obtain the reconstructed mud pressure pulse signal.
[0066] The present invention discloses a method for extracting a mud pulse target signal based on multi-peak fitting. Using a unimodal function as the fitting basis, the maximum number of iterations, fitness function, and corresponding threshold are determined, and an optimization algorithm is combined to perform multi-peak fitting iteration on the mud pulse signal in the second stage to obtain the optimal multi-peak feature vector. Then, the optimal multi-peak feature vector and the fitting basis are used to obtain the reconstructed mud pressure pulse signal. Thus, the position, shape, and intensity points of each peak can be accurately captured, the baseline drift problem that is difficult to solve by EMD iteration and wavelet decomposition can be solved, and the time-domain characteristics of the original signal are retained to the greatest extent, and the waveform is greatly improved, making the reconstructed mud pressure pulse signal more analyzable. In addition, the multi-peak parameters obtained by multi-peak fitting include pulse width and height information, which can be directly analyzed and processed by the decoding system to complete the analysis of downhole drilling parameters.
[0067] Embodiment 2: An embodiment of the present invention discloses a method for extracting a mud pulse target signal based on multi-peak fitting, including:
[0068] Step S210: Obtain the initial mud pulse signal at the surface riser and process it to obtain the mud pressure pulse signal in the second stage.
[0069] The above steps specifically include:
[0070] (1) Obtain the initial mud pulse signal obtained by the mud pressure pulse sensor set at the surface riser;
[0071] (2) Perform filtering and smoothing processing on the initial mud pulse signal to obtain the mud pressure pulse signal in the second stage. In this embodiment, Savitzky-Golay convolution smoothing filtering can be used. Savitzky-Golay convolution smoothing filtering can ensure that the shape and width of the signal remain unchanged while filtering out noise.
[0072] Step S220: Select a fitting basis, and use the fitting basis to perform multi-peak fitting on the mud pressure pulse signal in the second stage to determine the maximum number of iterations, fitness function, and corresponding threshold values, where the fitting basis is selected from unimodal functions.
[0073] As shown in the appendix Figure 2 The above steps specifically include:
[0074] Step S221: Perform differential analysis on the mud pressure pulse signal in the second stage to obtain the number, position, and value of the maximum points, and use the number of maximum points as the maximum number of iterations of the optimization algorithm.
[0075] Step S222: Select a fitting basis, and use the fitting basis to perform multi-peak fitting on the mud pressure pulse signal in the second stage to obtain the multi-peak fitted mud pressure pulse signal, where the fitting basis is selected from unimodal functions;
[0076] Here, according to different drilled formations and different drilling conditions, different fitting bases such as Gaussian unimodal function, Lorentz unimodal function, and exponential unimodal function can be selected to perform multi-peak fitting on the mud pressure pulse signal in the second stage; the characteristic parameters of the fitting basis include peak height, full width at half maximum (frequency), and peak position.
[0077] In this embodiment, if the Gaussian unimodal function is selected as the fitting basis, the multi-peak fitted mud pressure pulse signal is as follows:
[0078]
[0079] Among them, F() is the fitted mud pressure pulse signal; n is the number of fitted peaks; α, β, and γ are multi-peak characteristic vectors.
[0080] The above Gaussian unimodal function is as follows:
[0081]
[0082] Among them, α, β, and γ are the characteristic parameters of the unimodal function. α contains the unimodal intensity information, β contains the peak position information, and γ contains the peak width information.
[0083] Step S223: Based on the mud pressure pulse signal in the second stage and the multi-peak fitting mud pressure pulse signal, obtain the fitting residual, use the fitting residual as the fitness function of the optimization algorithm, and set the corresponding fitness value as the threshold value.
[0084]
[0085] Among them, h(x) is the fitting residual; F(x) is the multi-peak fitting mud pressure pulse signal; G(x) is the mud pressure pulse signal in the second stage.
[0086] Step S230: Select an optimization algorithm, and perform multi-peak fitting iteration on the mud pulse signal in the second stage based on the maximum number of iterations, the fitness function, and the corresponding threshold value to obtain the optimal multi-peak feature vector.
[0087] The above optimization algorithm may include a genetic algorithm, a particle swarm optimization algorithm, etc., which can be selected according to needs.
[0088] As shown in the appendix Figure 3 If the particle swarm optimization algorithm is selected, the above steps specifically include:
[0089] Step S231: Initialize the particle swarm parameters;
[0090] Here, initializing the particle swarm parameters includes initializing the number of particles, the number of iterations, the particle dimension, the velocity range, the position range, and the learning factor;
[0091] Step S232: Calculate the fitness value of the particles, update the individual extreme value Pbest, the population extreme value gbest, the particle position, and the particle velocity;
[0092] The update of the particle position is as follows:
[0093] Denote the velocity vector of the j-th particle at the t-th iteration as Use the following formula to update the position vector at the (t + 1)-th iteration
[0094]
[0095] Among them, w is the inertia weight, μ and ρ are random numbers between 0 and 1, used to limit the maximum self-velocity of the particles, pbest represents the local extreme value, and gbest represents the global extreme value.
[0096] Denote the position vector of the j-th particle at the t-th iteration as Use the following formula to update the position vector at the (t + 1)-th iteration
[0097]
[0098] Step S233: Determine whether the current fitness value is less than the threshold;
[0099] Step S234: If the response is yes, end the optimization process; if the response is no, determine whether the current iteration number is less than the maximum iteration number;
[0100] Step S235: If the response is yes, end the optimization process; if the response is no, increment the iteration number by one and continue to calculate the fitness value of the particle.
[0101] Step S240: Obtain the reconstructed mud pressure pulse signal by using the optimal multi-peak feature vector and the fitting basis.
[0102] Embodiment 3: Select the fitting basis as the Gaussian unimodal function, select the optimization algorithm as the particle swarm optimization algorithm, and extract the mud pulse target signal by the method of Embodiment 2, specifically as follows:
[0103] (1) Perform Savitzky-Golay convolution smoothing filtering on the initial mud pulse signal. The Savitzky-Golay convolution smoothing effect of the initial mud pressure pulse signal is as shown in the appendix Figure 4 as follows;
[0104] (2) Perform differential analysis on the mud pressure pulse signal in the second stage to obtain the number, positions, and values of the maximum points, and use the number of maximum points as the maximum iteration number of the optimization algorithm;
[0105] (3) Select the Gaussian unimodal function to perform multi-peak fitting on the mud pressure pulse signal in the second stage to obtain the multi-peak fitted mud pressure pulse signal;
[0106] (4) Based on the mud pressure pulse signal in the second stage and the multi-peak fitted mud pressure pulse signal, obtain the fitting residual, use the fitting residual as the fitness function of the optimization algorithm, and set the corresponding fitness value as the threshold;
[0107]
[0108] (5) Select the particle swarm optimization algorithm, and perform multi-peak fitting iteration on the mud pulse signal in the second stage based on the maximum iteration number, fitness function, and the corresponding threshold to obtain the optimal multi-peak feature vector; here, the effect diagrams of each unimodal signal finally obtained by optimization are as shown in the appendix Figure 5 as follows;
[0109] (6) Obtain the reconstructed mud pressure pulse signal by using the optimal multi-peak feature vector and the fitting basis, that is, substitute the optimal multi-peak feature vector into the multi-peak fitted mud pressure pulse signal to reconstruct the mud pressure pulse signal. The effect diagram of the multi-peak reconstruction signal is as shown in the appendixFigure 6 As shown, the multi-peak fitting mud pressure pulse signal is as follows:
[0110]
[0111] Furthermore, the effect diagrams of the mud pressure pulse signal after EMD iterative processing and db8 wavelet decomposition processing are respectively as shown in Appendix Figure 7 and 8 As shown, comparing it with Appendix Figure 6 it can be found that the method disclosed in the present invention solves the baseline drift problem that is difficult to solve by EMD iteration and wavelet decomposition, and retains the time-domain characteristics of the original signal to the greatest extent. The waveform is greatly improved and the analyzability is stronger.
[0112] Based on the above embodiments, in the multi-peak iteration process of the present invention, according to different drilled formations and different drilling conditions, different fitting bases such as Gaussian single-peak function, Lorentz single-peak function, and exponential single-peak function can be selected to perform multi-peak decomposition on the filtered signal (the mud pulse signal in the second stage). The initial value used (i.e., the mud pulse signal in the second stage) is related to the extreme point characteristics of the filtered signal, which makes the initial value easy to determine, converges quickly, has low calculation delay, and the real-time performance of the analysis system is stronger. And the reconstructed signal retains the time-domain characteristics of the original signal to the greatest extent, the waveform is greatly improved, and the analyzability is stronger. In addition, the multi-peak parameters obtained by decomposition contain pulse width and height information, which can be directly analyzed and processed by the decoding system to complete the analysis of downhole drilling parameters.
[0113] Embodiment 4: As shown in Appendix Figure 9 The embodiment of the present invention discloses a device for extracting a mud pulse target signal based on multi-peak fitting, including:
[0114] A signal acquisition unit, which acquires the initial mud pulse signal at the surface riser and processes it to obtain the mud pressure pulse signal in the second stage;
[0115] A first analysis unit, which selects a fitting base and performs multi-peak fitting on the mud pressure pulse signal in the second stage using the fitting base to determine the maximum number of iterations, the fitness function, and the corresponding threshold, where the fitting base is selected from single-peak functions;
[0116] A second analysis unit, which selects an optimization algorithm and performs multi-peak fitting iteration on the mud pulse signal in the second stage based on the maximum number of iterations, the fitness function, and the corresponding threshold to obtain an optimal multi-peak feature vector;
[0117] A third analysis unit, which uses the optimal multi-peak feature vector and the fitting base to obtain the reconstructed mud pressure pulse signal.
[0118] Among them, the first analysis unit includes:
[0119] The first analysis module performs differential analysis on the mud pressure pulse signal in the second stage to obtain the number, positions, and values of the maximum points, and uses the number of maximum points as the maximum number of iterations of the optimization algorithm;
[0120] The second analysis module selects a fitting basis and uses the fitting basis to perform multi-peak fitting on the mud pressure pulse signal in the second stage to obtain the multi-peak fitted mud pressure pulse signal;
[0121] The third analysis module obtains the fitting residual based on the mud pressure pulse signal in the second stage and the multi-peak fitted mud pressure pulse signal, uses the fitting residual as the fitness function of the optimization algorithm, and sets the corresponding fitness value as the threshold;
[0122]
[0123] where h(x) is the fitting residual; F(x) is the multi-peak fitted mud pressure pulse signal; G(x) is the mud pressure pulse signal in the second stage.
[0124] Example 5: As shown in the appendix Figure 10 The embodiment of the present invention discloses a mud pulse target signal extraction system based on multi-peak fitting, including:
[0125] A mud pulse target signal extraction device based on multi-peak fitting, where the mud pulse target signal extraction device based on multi-peak fitting is the mud pulse target signal extraction device based on multi-peak fitting as described in the above embodiment;
[0126] An interaction device, which is used for an operator to communicate with the mud pulse target signal extraction device based on multi-peak fitting, so that the operator can provide the data required for mud pulse target signal extraction for the mud pulse target signal extraction device based on multi-peak fitting through the interaction unit.
[0127] The above interaction device includes terminal devices such as a computer and a touch screen, and is used for an operator to input the data required for mud pulse target signal extraction by the mud pulse target signal extraction device based on multi-peak fitting.
[0128] Example 6: The embodiment of the present invention discloses a storage medium, on which a computer program readable by a computer is stored, and the computer program is set to perform mud pulse target signal extraction based on multi-peak fitting when running.
[0129] The above storage medium may include, but is not limited to: various media such as a USB flash drive, a read-only memory, a mobile hard disk, a magnetic disk, or an optical disc that can store a computer program.
[0130] Example 7: An embodiment of the present invention discloses an electronic device, including a processor and a memory. A computer program is stored in the memory and is loaded and executed by the processor to implement the extraction of mud pulse target signals based on multi-peak fitting.
[0131] The above-mentioned processor may be a central processing unit (CPU), a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute various exemplary logical blocks, modules, and circuits described in connection with the disclosure of the present application. It can also be a combination that implements computing functions, such as a combination including one or more microprocessors, a combination of DSP and microprocessors, and so on. The memory may include, but is not limited to: various media that can store computer programs, such as USB flash drives, read-only memories, mobile hard disks, magnetic disks, or optical discs.
[0132] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing computer-usable program codes. The solutions in the embodiments of the present application can be implemented using various computer languages. For example, object-oriented programming languages such as Java and interpreted scripting languages such as JavaScript.
[0133] The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.
[0134] These computer program instructions can 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 generate a manufactured article including an instruction device, and the instruction device implements the functions in Figure 1 one process or multiple processes and / or blocks Figure 1The functions specified in one or more boxes.
[0135] The above technical features constitute the best embodiment of the present invention, which has strong adaptability and the best implementation effect. Non-essential technical features can be added or subtracted according to actual needs to meet the requirements of different situations.
Claims
1. A method for extracting the target signal of mud pulse based on multi-peak fitting, characterized in that, it includes: Obtain the initial mud pulse signal at the surface riser and process it to obtain the mud pressure pulse signal in the second stage; Select a fitting basis, use the fitting basis to perform multi-peak fitting on the mud pressure pulse signal in the second stage, determine the maximum number of iterations, fitness function and corresponding threshold, where the fitting basis is selected from unimodal functions; Select an optimization algorithm, and perform multi-peak fitting iteration on the mud pulse signal in the second stage based on the maximum number of iterations, fitness function and corresponding threshold to obtain the optimal multi-peak feature vector; Use the optimal multi-peak feature vector and the fitting basis to obtain the reconstructed mud pressure pulse signal.
2. The method for extracting the target signal of mud pulse based on multi-peak fitting according to claim 1, characterized in that, The step of selecting a fitting basis, using the fitting basis to perform multi-peak fitting on the mud pressure pulse signal in the second stage, and determining the maximum number of iterations, fitness function and corresponding threshold includes: Perform differential analysis on the mud pressure pulse signal in the second stage to obtain the number, position and value of the maximum points, and use the number of maximum points as the maximum number of iterations of the optimization algorithm; Select a fitting basis, use the fitting basis to perform multi-peak fitting on the mud pressure pulse signal in the second stage to obtain the multi-peak fitting mud pressure pulse signal, where the fitting basis is selected from unimodal functions; Based on the mud pressure pulse signal in the second stage and the multi-peak fitting mud pressure pulse signal, obtain the fitting residual, use the fitting residual as the fitness function of the optimization algorithm, and set the corresponding fitness value as the threshold; where h(x) is the fitting residual; F(x) is the multi-peak fitting mud pressure pulse signal; G(x) is the mud pressure pulse signal in the second stage.
3. The method for extracting the target signal of mud pulse based on multi-peak fitting according to claim 2, characterized in that, The fitting basis is a Gaussian unimodal function, then use the Gaussian unimodal function to perform multi-peak fitting on the mud pressure pulse signal in the second stage to obtain the multi-peak fitting mud pressure pulse signal, where the multi-peak fitting mud pressure pulse signal is as follows: where F() is the fitted mud pressure pulse signal; n is the number of fitting peaks; α, β, γ are multi-peak feature vectors.
4. The method for extracting the target signal of mud pulse based on multi-peak fitting according to any one of claims 1 to 3, characterized in that, The selected optimization algorithm is the particle swarm optimization algorithm, then perform multi-peak fitting on the mud pulse signal in the second stage based on the particle swarm optimization algorithm, the maximum number of iterations, the fitness function and the corresponding threshold to obtain the optimal multi-peak feature vector, including: Initialize the particle swarm parameters; Calculate the fitness value of the particles, update the individual extreme value Pbest, the global extreme value gbest, the particle position and the particle velocity; Judge whether the current fitness value is less than the threshold; If so, end the optimization process; if not, judge whether the current number of iterations is less than the maximum number; If so, end the optimization process; if not, add one to the number of iterations and continue to calculate the fitness value of the particles.
5. The method for extracting mud pulse target signals based on multi-peak fitting according to any one of claims 1 to 4, characterized in that, the obtaining of the initial mud pulse signal at the surface riser and the processing thereof to obtain the mud pressure pulse signal in the second stage includes: obtaining the initial mud pulse signal obtained by the mud pressure pulse sensor arranged at the surface riser; performing Savitzky-Golay convolutional smoothing filtering on the initial mud pulse signal to obtain the mud pressure pulse signal in the second stage.
6. An apparatus for extracting mud pulse target signals based on multi-peak fitting applying the method according to any one of claims 1 to 5, characterized in that, it includes: a signal acquisition unit for acquiring the initial mud pulse signal at the surface riser and processing it to obtain the mud pressure pulse signal in the second stage; a first analysis unit for selecting a fitting basis, performing multi-peak fitting on the mud pressure pulse signal in the second stage using the fitting basis, determining the maximum number of iterations, fitness function and corresponding threshold, wherein the fitting basis is selected from unimodal functions; a second analysis unit for selecting an optimization algorithm and performing multi-peak fitting iteration on the mud pulse signal in the second stage based on the maximum number of iterations, fitness function and corresponding threshold to obtain an optimal multi-peak feature vector; a third analysis unit for obtaining the reconstructed mud pressure pulse signal using the optimal multi-peak feature vector and the fitting basis.
7. The apparatus for extracting mud pulse target signals based on multi-peak fitting according to claim 6, characterized in that, the first analysis unit includes: a first analysis module for performing differential analysis on the mud pressure pulse signal in the second stage to obtain the number, positions and values of the maximum points, and taking the number of maximum points as the maximum number of iterations of the optimization algorithm; a second analysis module for selecting a fitting basis and performing multi-peak fitting on the mud pressure pulse signal in the second stage using the fitting basis to obtain the mud pressure pulse signal of multi-peak fitting; a third analysis module for obtaining a fitting residual based on the mud pressure pulse signal in the second stage and the mud pressure pulse signal of multi-peak fitting, taking the fitting residual as the fitness function of the optimization algorithm, and setting the corresponding fitness value as the threshold; wherein, h(x) is the fitting residual; F(x) is the mud pressure pulse signal of multi-peak fitting; G(x) is the mud pressure pulse signal in the second stage.
8. A system for extracting mud pulse target signals based on multi-peak fitting, characterized in that, it includes: an apparatus for extracting mud pulse target signals based on multi-peak fitting, wherein the apparatus for extracting mud pulse target signals based on multi-peak fitting is the apparatus for extracting mud pulse target signals based on multi-peak fitting according to claims 6 to 7; an interaction device for communicating between the operator and the apparatus for extracting mud pulse target signals based on multi-peak fitting, so that the operator can provide the data required for mud pulse target signal extraction for the apparatus for extracting mud pulse target signals based on multi-peak fitting through the interaction unit.
9. A storage medium, characterized in that, A computer program readable by a computer is stored on the storage medium, and the computer program is configured to execute the method for extracting a mud pulse target signal based on multi-peak fitting according to any one of claims 1 to 5 when running.
10. An electronic device, characterized in that, it includes a processor and a memory, and a computer program is stored in the memory. The computer program is loaded and executed by the processor to implement the method for extracting a mud pulse target signal based on multi-peak fitting according to any one of claims 1 to 5.