A method and system for demodulating a downhole mud pulse signal during a pumpdown pressure build-up period

Abstract

This invention proposes a method and system for demodulating downhole mud pulse signals during the pump start-up pressure build-up period. The method includes: S1, pump start-up detection and time window prediction; S2, adaptive condition discrimination; S3, transient demodulation processing; S4, synchronization matching and decoding; and S5, verification output. This invention can reliably identify the synchronization head and complete decoding even under unstable conditions such as pump pressure rise, waveform distortion during the pump start-up pressure build-up period, and when the drill bit is not completely lifted from the bottom of the hole. This improves the initial decoding success rate, reduces the error rate and the number of retests, minimizes pump stoppages and retests due to decoding failures, and improves the efficiency of directional drilling.

Description

Technical Field

[0001] This invention belongs to the field of underground drilling measurement technology in coal mines, and relates to a method and system for demodulating underground mud pulse signals during the pump start-up pressure build-up period. Background Technology

[0002] Directional drilling in coal mines is a key technology in areas such as mine gas extraction, water hazard prevention, and the investigation of hidden geological factors that could cause disasters. Measurement while drilling (MWD) devices measure borehole engineering parameters in real time during drilling, such as borehole inclination, azimuth, and the tool face angle of the screw motor, and transmit these parameters to the borehole opening for the driller to adjust the borehole trajectory. Early MWD systems in coal mines primarily used wired methods, employing specially designed central cable drill pipes to transmit signals. These methods had drawbacks such as easily damaged cable joints, high requirements for drill string consistency, inability to accommodate irregularly shaped drill pipes, and poor adaptability to complex geological formations.

[0003] In recent years, wireless drilling measurement technology using mud pulses has been gradually promoted and applied in underground coal mines. A typical system consists of an explosion-proof measuring sub, an explosion-proof rechargeable battery, an explosion-proof drive sub, and an explosion-proof pulse generator connected in series inside the borehole. The borehole opening section includes an explosion-proof pressure transmitter and an explosion-proof decoding computer installed at the mud pump outlet. The system operates in an intermittent mode: after drilling a single drill pipe, the pump is stopped, the flow switch determines the pump stop status, and the measuring sub acquires the static inclination and azimuth angles. When the pump is restarted, the flow switch outputs a pump start signal, the measuring sub synchronously acquires the dynamic tool face angle of the screw motor, and the drive sub controls the pulse generator to intermittently change the flow area of ​​the hydraulic channels within the drill string, generating positive pressure pulses. Using the flushing fluid within the drill string as the transmission channel, the encoded trajectory parameters are transmitted backflow to the borehole opening, where the explosion-proof computer performs filtering, de-trending, and demodulation display.

[0004] In actual construction, the flow switch determines that the device in the borehole will start emitting waves according to the agreed timing after the pump is started. At this time, the surface mud pump pressure is in the build-up period: the pump pressure gradually climbs from zero or low pressure to the rated working pressure. The alternating volumetric discharge of the reciprocating mud pump causes violent fluctuations in pump pressure, and the air chamber is often not fully pressurized. If the driller does not completely lift the drill bit from the bottom of the hole, the residual cutting torque and the vibration of the drill string against the bottom will further couple into the riser pressure. During the pump start-up pressure build-up period and under low lift-up conditions, the rising and falling edges of the pulses of the data frame synchronization head and the initial symbol are partially canceled out by the baseline rise, the edge slope is smoothed out by the pump pressure rise, and the reflection superposition and manifold water hammer further lengthen or distort the pulse shape, making the synchronization head easily submerged by background interference. Conventional steady-state demodulation algorithms at the orifice use energy detection or full-segment cross-correlation strategies designed for steady-pump operation, which are difficult to reliably identify the frame start position in the context of strong baseline drift, leading to synchronization loss, sharp increase in bit error rate, or even failure of whole-frame decoding. On-site, it is often necessary to stop the pump, re-establish the link, and wait for the next pump to start and re-measure, generating a lot of invalid auxiliary time, which seriously restricts the efficiency of directional drilling construction.

[0005] Existing orifice demodulation schemes mainly target the suppression of periodic pumping interference under stable pumping conditions. They do not specifically address baseline rise trends, synchronization head distortion, and low signal-to-noise ratio pulses during the pump start-up pressure build-up period and when the drill bit is not fully lifted from the bottom of the hole. Synchronization head searches are mostly blind searches within the entire pump pressure signal range, lacking protocol timing and hole depth propagation constraints, resulting in a high probability of false alarms. There is a lack of adaptive detrending, edge locking, and morphology repair methods for distorted pulse edges. A single steady-state process cannot simultaneously meet the decoding success rate of transient conditions and the real-time requirements of stable pumping conditions.

[0006] Therefore, there is an urgent need for a method and system for demodulating downhole mud pulse signals during the pump start-up pressure establishment period that is based on time-constrained search, adapts to unstable working condition distortions, and has working condition adaptive routing and closed-loop verification. Summary of the Invention

[0007] Based on the technical problems existing in the prior art, the present invention provides a method and system for demodulating downhole mud pulse signals during the pump start-up pressure establishment period.

[0008] According to a first aspect of the present invention, a downhole mud pulse signal demodulation system is provided during the pump start-up pressure build-up period.

[0009] The system includes: The system includes: an in-hole device for detecting the pump start-up or stop-up status of the mud pump through the flow switch, and an orifice device for acquiring manifold pressure signals through the explosion-proof pressure transmitter and transmitting them to the explosion-proof computer. The in-hole device is installed inside the drill string and includes, from the bottom of the hole to the opening, a directional drill bit, a screw motor, a non-magnetic outer tube, a pulse generator, and the drill string. The non-magnetic outer tube contains a measuring section, a rechargeable battery tube, and a drive section. The measuring section has a flow switch at its front end. The orifice device includes a drill rod, a water feeder, a high-pressure hose, a pump truck, an explosion-proof pressure transmitter, and an explosion-proof computer for demodulating signals during the pump start-up pressure build-up period and under unstable low lift-off conditions. The explosion-proof pressure transmitter is installed on the outlet side of the mud pump, and the explosion-proof computer is connected to the explosion-proof pressure transmitter via signal connection. The in-hole device collects the static tilt angle and azimuth angle when the pump is stopped, and collects the dynamic tool face angle of the screw motor when the pump is started, and encodes them according to a preset communication protocol; according to a preset wave delay, the pulse generator intermittently changes the hydraulic channel area inside the drill string to generate mud positive pressure pulses, and uses the flushing fluid inside the drill string as the signal transmission channel to transmit the encoded trajectory parameters backflow to the borehole opening.

[0010] According to a second aspect of the present invention, a method for demodulating downhole mud pulse signals during the pump start-up pressure build-up period is provided.

[0011] The method is implemented through a downhole mud pulse signal demodulation system during the pump start-up pressure establishment period, and includes the following steps: S1. Pump start-up detection and time window prediction: Real-time monitoring of mud pump manifold pressure or flow switch status to detect mud pump start-up time. ; Delay for transmitting waves according to the user-defined protocol and the propagation time of the hole depth Calculate the theoretical arrival time of the synchronization head. And set the search time window. ; S2, Adaptive Working Condition Judgment: Within the time window described in step S1, mud pulse pressure signals are acquired. Calculate the slope of the pump pressure baseline rise. and volatility variance By Compared with the preset slope threshold Compare, Compared with the preset variance threshold Compare and determine whether the current operating condition label is a stable operating condition or an unstable operating condition; S3. Transient demodulation processing: Receive the pressure signal received in step S2 With operating condition labels; for signal distortion during the pump start-up pressure build-up period and under low lift-off unstable operating conditions, adaptive envelope extraction and baseline elimination, differential threshold edge coarse locking, subband decomposition and reconstruction enhancement, and morphological edge trimming are sequentially applied to output a repaired pulse sequence. ; S4. Synchronous matching and decoding: Receive the signal to be matched output from step S3, and slide the preset frame synchronization header template and the repaired pulse sequence point by point within the synchronization header arrival time window, using the sampling point index. To start a candidate frame, a feature vector is constructed by segmenting the code width into segments and calculating the average amplitude. Then, the correlation coefficient is calculated with the template. A one-dimensional normalized correlation coefficient or Hamming distance is calculated. The position corresponding to the maximum correlation coefficient or the minimum Hamming distance that is less than a set threshold is taken as the frame start position. The frame is then decoded according to a preset encoding protocol to obtain the drilling trajectory parameters such as the tilt angle, azimuth angle, and tool face angle. S5, Verification Output: Perform CRC or checksum verification on the decoding parameters output in step S4: if the verification passes, the drilling trajectory parameters such as inclination angle, azimuth angle, and tool face angle are parsed and displayed for the driller to adjust the drilling trajectory; if the verification fails or the start of a valid frame is not locked in step S4, the decoding is determined to have failed and a retransmission prompt is sent back to the device inside the hole.

[0012] Based on the above scheme, S1, pump start-up detection and time window prediction specifically includes the following steps: S11. Signal Acquisition and Preprocessing: Acquire steady-state pressure signals during the pump shutdown phase and calculate the steady-state baseline pressure during pump shutdown. Set the pump start-up threshold. The proportionality coefficient Pick Pressure Increment Bias All greater than At that time, the first to exceed The sampling point corresponding to the time is determined as the time when the mud pump starts. ; S12. Synchronization head arrival time window prediction: According to the protocol of the in-orifice device, after the flow switch detects the pump start-up, a preset wave emission delay will occur. Data frames have just begun to be sent; the propagation speed of the mud pulse signal in the flushing fluid inside the drill string. From the current hole depth Calculate propagation time Then the synchronization head arrives at the theoretical time. ,by Centered on the target and with a set tolerance ΔT as the radius, determine the arrival time window of the synchronization head. .

[0013] Based on the above scheme, S2, the adaptive judgment of working conditions, specifically includes: For time windows The digital pressure signal within the time window is extracted using a low-pass filter to obtain the macroscopic baseline of the pump pressure; the rise slope of the baseline within the time window is calculated through first-order linear fitting. And calculate the residual variance after filtering out the baseline. Set a discrimination threshold and The judgment logic is as follows; (1) Transient characteristics of pump start-up: The pump pressure is in the pressure rise period, which is manifested as the baseline rise slope. ( Pick Preferred ); (2) Low lifting condition characteristics: When the drill bit is not completely lifted from the bottom of the hole, the residual cutting torque causes the flushing fluid pipeline to generate through the impedance feedback of the screw motor. Low-frequency, large-amplitude pressure jitter, manifested as fluctuation variance ( Pick Preferred ).

[0014] Based on the above scheme, when the following conditions are met... and / or When the system determines that it is currently in an unstable operating condition such as the pump start-up pressure build-up period or low lift-off, it switches to transient demodulation branch processing. When not satisfied and / or If the condition is stable, the system switches to the conventional demodulation branch. The digital pressure signal is then filtered and denoised using adaptive notch / noise cancellation to remove baseline drift. The differential pressure is set according to the pulse amplitude to remove false pulses. Finally, the signal is decoded according to the preset encoding protocol to obtain the borehole trajectory parameters.

[0015] Based on the above scheme, the S3 transient demodulation processing specifically includes the following steps: S31. Adaptive envelope extraction: To address the physical characteristic of the upward abrupt change in the positive mud pulse pressure signal, an adaptive extraction of the pump pressure rise baseline envelope using a sliding window based on local minima is employed: the sliding window width is set to a preset symbol width. of Double, search for digital pressure signals within the sliding window. The sequence of local minima is obtained; cubic spline interpolation and smoothing filtering are performed on the local minima sequence to construct a lower envelope signal that fits the trend of pump start-up pressure rise. ; through subtraction Remove the low-frequency pump pressure rise trend and baseline drift caused by the pump start-up transient to obtain the pulse signal that brings the baseline to zero. ; S32, Differential Threshold Mutation Locking: Performing a first-order difference operation on the pulse signal s(n) yields the difference sequence: d(n) = s(n) - s(n-1); Where d(n) is the first-order difference sequence of the pulse signal s(n), that is, the difference between adjacent sampling points; Set the adaptive threshold θ = α·σd, where σd is the root mean square of the difference sequence. Take 2 to 3; for differential amplitude values ​​exceeding + The moment is used as a candidate point for the rising edge of the pulse, below The moment is used as a candidate point for the falling edge of the pulse; according to the preset symbol width Edge pairing is performed when the time interval between adjacent rising and falling edge candidate points is within [0.5]. 1.5 When the range is within the specified range, it is determined as a group of valid pulse edges. Each pair of valid edges is used as an anchor point, and its corresponding continuous time interval is determined according to the preset symbol width to obtain the pulse position coarse locking result containing the edge position and its corresponding time interval. S33. Subband Decomposition and Reconstruction: Based on the coarse pulse position locking result, pulse segments are extracted and multi-resolution subband decomposition is performed: discrete wavelet transform is used to select detailed subbands containing pulse edge frequency band features for reconstruction; or variational mode decomposition is used to select mode components whose center frequency is within the pulse edge frequency band and separated from the pump noise frequency band for reconstruction, thus obtaining enhanced pulse segments. S34. Morphological Repair: Select flat linear structural elements, perform morphological opening and closing operations on the enhanced pulse segments in sequence, remove pulse edge burrs and fill gaps, and then reshape the rising or falling edge according to the slope constraint of adjacent pulse edges to obtain the repaired pulse sequence.

[0016] Based on the above scheme, the specific calculation method for the frame start position in step S4 is a one-dimensional normalized correlation, as follows: Candidate starting sampling points Based on the symbol width Extract the average envelope features of the first 8 code points of the synchronization header. And calculate the one-dimensional normalized cross-correlation coefficient by sliding point by point with the preset 8-bit frame synchronization header template vector. :

[0017] In the formula: This is the index of the sampling point corresponding to the starting position of the candidate frame. This is the sequence number of the synchronization header element. The width of the symbol. The system sampling rate, The pulse sequence after morphological repair. Indicated by When the starting point is, the first Average pulse amplitude within a symbol time window; The sampling point number within the symbol time window, with a value range of 0 ≤ j ≤ Tb·fs-1;

[0018] In the formula: For the frame synchronization header template The theoretical binary logic value of 1 bit ( ), consistent with the preset coding protocol of the device inside the hole;

[0019] In the formula: For the template vector, the first The level value of each symbol; It is the arithmetic mean of the components of the template vector, i.e. ; As the current candidate starting point Next Average envelope characteristics of each code element; It is the arithmetic mean of these 8 code element features, i.e. ;

[0020] In the formula: For the synchronization head arrival time window The corresponding starting sampling point index in the sampling sequence, For the synchronization head arrival time window The index of the corresponding termination sampling point in the sampling sequence; This represents the total number of sampling points occupied by the 8-bit synchronization header. When the maximum correlation coefficient At that time, confirm The corresponding time is the precise start position of the data frame; if If the synchronization fails within the current time window, the system will output a retransmission prompt.

[0021] Based on the above scheme, the specific calculation method for the frame start position in step S4 is the Hamming distance, as follows:

[0022]

[0023] In the formula: H(s) is the Hamming distance; (·) is a 0 / 1 hard decision function; This is the decision threshold; This is the i-th component of the frame synchronization header template; This is the start position of the frame.

[0024] According to a third aspect of the present invention, a computer-readable storage medium is provided.

[0025] In some embodiments, the computer-readable storage medium includes a computer program for storing a step of the downhole mud pulse signal demodulation method during the pump start-up pressure build-up period, which, when executed by a processor, implements the steps of the method.

[0026] According to a fourth aspect of the present invention, a computer device is provided.

[0027] In some embodiments, the computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of the above-described method for demodulating downhole mud pulse signals during the pump start-up pressure establishment period.

[0028] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects: This invention uses pump start-up timing detection and synchronization head arrival time window prediction to strictly constrain the synchronization search within a narrow window. It only enters the transient demodulation branch when the baseline slope and fluctuation variance exceed the limits, while the steady-state process continues directly to the conventional process. This reduces the false alarm rate without increasing the computational burden under stable pump conditions. Within the transient branch, adaptive envelope detrending, differential threshold edge coarse locking, subband decomposition and reconstruction enhancement, and morphological edge trimming are executed sequentially. This process peels away the pump pressure rise baseline layer by layer, sharpens the smoothed and elongated pulse edges, and repairs the distorted symbol morphology, enabling reliable extraction of the frame synchronization head during the pump pressure setup period and under low lift-off conditions. Finally, normalized cross-correlation and sliding matching with a preset synchronization head template are used on the repaired signal. This approach is insensitive to pulse amplitude fluctuations and has strong anti-background interference capabilities, thereby improving the first frame decoding success rate under unstable conditions from near failure to a usable level, while simultaneously reducing bit errors and retransmission overhead.

[0029] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit the invention. Attached Figure Description

[0030] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0031] Figure 1 This is a flowchart illustrating a downhole mud pulse signal demodulation system during the pump start-up pressure build-up period, according to an exemplary embodiment. Figure 2 This is a flowchart illustrating a downhole mud pulse signal demodulation method during the pump start-up pressure establishment period, according to an exemplary embodiment. Figure 3 This is a diagram illustrating the relationship between signals and operating conditions according to an exemplary embodiment, wherein, Figure 3 (a) To stabilize pumping conditions, Figure 3 (b) is the condition during the pump start-up pressure build-up period / drill bit not completely lifted from the bottom of the hole; Figure 4 This is a waveform comparison of synchronization and decoding decisions between conventional demodulation and the method of the present invention under a low lift-off condition (15dB) according to an exemplary embodiment, wherein... Figure 4 (a) is the acquired raw orifice pressure waveform (in) (A real synchronization header exists at this location). Figure 4 (b) is the result of a routine mediation decision (in) (Detection value for erroneous synchronization head position). Figure 4 (c) is the judgment result of the method of the present invention; Figure 5 This is a schematic diagram of the structure of a computer device according to an exemplary embodiment. Detailed Implementation

[0032] The following description and accompanying drawings fully illustrate specific embodiments of this application to enable those skilled in the art to practice them. Some parts and features of some embodiments may be included in or replace parts and features of other embodiments.

[0033] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0034] Example 1

[0035] like Figure 1 As shown, the present invention provides a downhole mud pulse signal demodulation system during the pump start-up pressure establishment period.

[0036] Specifically, the system includes: an in-hole device for detecting the pump start / stop status of the mud pump through the flow switch and an orifice device for acquiring manifold pressure signals through the explosion-proof pressure transmitter and transmitting them to the explosion-proof computer. The in-hole device is installed inside the drill string and includes, from the bottom of the hole to the opening, a directional drill bit, a screw motor, a non-magnetic outer tube, a pulse generator, and the drill string. The non-magnetic outer tube contains a measuring section, a rechargeable battery tube, and a drive section. The measuring section has a flow switch at its front end. The orifice device includes a drill rod, a water feeder, a high-pressure hose, a pump truck, an explosion-proof pressure transmitter, and an explosion-proof computer for demodulating signals during the pump start-up pressure build-up period and under unstable low lift-off conditions. The explosion-proof pressure transmitter is installed on the outlet side of the mud pump, and the explosion-proof computer is connected to the explosion-proof pressure transmitter via signal connection. The in-hole device collects the static tilt angle and azimuth angle when the pump is stopped, and collects the dynamic tool face angle of the screw motor when the pump is started, and encodes them according to a preset communication protocol; according to a preset wave delay, the pulse generator intermittently changes the hydraulic channel area inside the drill string to generate mud positive pressure pulses, and uses the flushing fluid inside the drill string as the signal transmission channel to transmit the encoded trajectory parameters backflow to the borehole opening.

[0037] In one optional embodiment, the sources of the above-mentioned devices are explained as follows: The explosion-proof computer can be the explosion-proof computer disclosed in the authorized utility model patent of our company, "A Mining Explosion-proof and Intrinsically Safe Computer" (authorization announcement number CN222927006U), and the demodulation method described in this invention is executed by the pulse decoding board of the computer; the pulse generator can be the device disclosed in the authorized utility model patent of our company, "A Rotary Valve Mud Pulse While Drilling Measurement Device" (authorization announcement number CN223136125U); the measuring sub, rechargeable battery tube, and drive sub can be the ZSZM-16T mining intrinsically safe mud pulse while drilling measurement probe of our company; the directional drill bit, screw motor, non-magnetic outer tube, drill string, drill rod, water supply device, high-pressure hose, pump truck, and explosion-proof pressure transmitter can all be corresponding models of our company's products, or conventional commercially available products in the field, which are not innovative points of this invention and will not be elaborated here.

[0038] Example 2

[0039] Based on Example 1, this embodiment of the invention provides a method for demodulating downhole mud pulse signals during the pump start-up pressure establishment period.

[0040] Specifically, such as Figure 2 As shown, the method includes the following steps: S1. Pump start-up detection and time window prediction: Real-time monitoring of mud pump manifold pressure or flow switch status to detect mud pump start-up time. ; Delay for transmitting waves according to the user-defined protocol and the propagation time of the hole depth Calculate the theoretical arrival time of the synchronization head. And set the search time window. ; Specifically, it includes the following steps: S11. Signal Acquisition and Preprocessing: Acquire steady-state pressure signals during the pump shutdown phase and calculate the steady-state baseline pressure during pump shutdown. (Continued after pump shutdown) (average pressure); set pump start-up threshold. The proportionality coefficient Pick (Preferred) Pressure Increment Bias Pick (Preferred) ), to avoid pump stop baseline near Time-multiplication threshold fails; when the filtered digital pressure signal continuous sampling points ( Corresponding duration Preferred point / All are greater than At that time, the first to exceed The sampling point corresponding to the time is determined as the time when the mud pump starts. ; S12. Synchronization head arrival time window prediction: According to the protocol of the in-orifice device, after the flow switch detects the pump start-up, a preset wave emission delay will occur. (3-10 seconds) is acceptable before data frames are sent; the propagation speed of the mud pulse signal in the flushing fluid within the drill string. Approximately 1366–1445 m / s (1400 m / s can be taken), based on the current borehole depth. Calculate propagation time Then the synchronization head arrives at the theoretical time. ,by Centered on the target and with a set tolerance ΔT (which can be 1 to 3 seconds) as the radius, determine the arrival time window of the synchronization head. (Note: ΔT is the half-width of the synchronization head arrival time window. Its meaning and value are consistent throughout the text and can be taken from 1 to 3 seconds. In this embodiment, it is taken as 3 seconds.) S2, Adaptive Working Condition Judgment: Within the time window described in step S1, mud pulse pressure signals are acquired. Calculate the slope of the pump pressure baseline rise. and volatility variance By Compared with the preset slope threshold Compare, Compared with the preset variance threshold Compare and determine whether the current operating condition is a stable or unstable condition; Specifically, regarding time windows The internal digital pressure signal uses a low-pass filter (cutoff frequency). Preferred Extract the macroscopic baseline of pump pressure; calculate the rise slope of the baseline within the time window using first-order linear fitting. And calculate the residual variance after filtering out the baseline. Set a discrimination threshold and The judgment logic is as follows; (1) Transient characteristics of pump start-up: The pump pressure is in the pressure rise period, which is manifested as the baseline rise slope. ( Pick Preferred ); (2) Low lifting condition characteristics: When the drill bit is not completely lifted from the bottom of the hole, the residual cutting torque causes the flushing fluid pipeline to generate through the impedance feedback of the screw motor. Low-frequency, large-amplitude pressure jitter, manifested as fluctuation variance ( Pick Preferred ).

[0041] The results of the judgment are as follows: When satisfied and / or When the system determines that it is currently in an unstable operating condition such as the pump start-up pressure build-up period or low lift-off, it switches to transient demodulation branch processing. If neither condition is met, the condition is determined to be steady-state and the system switches to the conventional demodulation branch.

[0042] Specifically, when the condition is determined to be stable, the digital pressure signal is filtered and denoised using adaptive notch / noise cancellation to remove baseline drift, the differential pressure is set according to the pulse amplitude to remove false pulses, and then decoded according to a preset encoding protocol to obtain the borehole trajectory parameters.

[0043] S3. Transient demodulation processing: Receive the pressure signal received in step S2 With operating condition labels; for signal distortion during the pump start-up pressure build-up period and under low lift-off unstable operating conditions, adaptive envelope extraction and baseline elimination, differential threshold edge coarse locking, subband decomposition and reconstruction enhancement, and morphological edge trimming are sequentially applied to output a repaired pulse sequence. ; Specifically, the following steps are included: S31. Adaptive Envelope Extraction: Addressing the physical characteristic of upward abrupt changes in the positive mud pulse pressure signal, an adaptive extraction of the pump pressure rise baseline envelope is employed using a sliding window based on local minima. The sliding window width is set to a preset symbol width. of times (preferred) Search for digital pressure signals within a sliding window. The sequence of local minima is obtained; cubic spline interpolation and smoothing filtering are performed on the local minima sequence to construct a lower envelope signal that fits the trend of pump start-up pressure rise. ; through subtraction Remove the low-frequency pump pressure rise trend and baseline drift caused by the pump start-up transient to obtain the pulse signal that brings the baseline to zero. .

[0044] As an alternative implementation: the local maxima and minima sequences can be extracted separately and the mean envelope can be calculated, or the Hilbert transform can be used to extract the analytic signal modulus as the baseline envelope for detrending processing; S32, Differential Threshold Abrupt Locking: Perform a first-order difference operation on the pulse signal s(n) to obtain the difference sequence d(n) = s(n) - s(n-1) (or take the first and second level detail coefficients of the discrete wavelet transform as the differential feature sequence), where d(n) is the first-order difference sequence of the pulse signal s(n), i.e., the difference between adjacent sampling points; s(n) is the baseline zeroing pulse signal output by S31, with dimensions in MPa / sampling point. Set the adaptive threshold θ = α·σd (σd is the root mean square of the difference sequence, Take 2 to 3); the differential amplitude exceeding + The moment is used as a candidate point for the rising edge of the pulse, below The moment is used as a candidate point for the falling edge of the pulse; according to the preset symbol width Edge pairing is performed when the time interval between adjacent rising and falling edge candidate points is within [0.5]. 1.5 When the range is within the specified range, it is determined as a group of valid pulse edges. Each pair of valid edges is used as an anchor point, and its corresponding continuous time interval is determined according to the preset symbol width to obtain the pulse position coarse locking result containing the edge position and its corresponding time interval. S33. Subband Decomposition and Reconstruction: Based on the coarse pulse position locking result, pulse segments are extracted and multi-resolution subband decomposition is performed: Discrete wavelet transform (such as db4 wavelet, with 3 to 5 decomposition levels) is used to select detailed subbands containing pulse edge frequency band features (approximately 0.5 to 3 Hz) for reconstruction; or variational mode decomposition (with K being 5 to 8 modes) is used to select modal components whose center frequency is within the pulse edge frequency band and separated from the pump noise frequency band for reconstruction, thus obtaining enhanced pulse segments; S34. Morphological repair: Select flat linear structural elements (length is 1 / 4 to 1 / 2 of the preset symbol width), and perform morphological opening (erosion and then dilation) and closing (dilation and then erosion) operations on the enhanced pulse segments in sequence to remove pulse edge burrs and fill gaps. Then, shape the rising / falling edges according to the slope constraints of adjacent pulse edges to obtain the repaired pulse sequence. S4. Synchronous matching and decoding: Receive the signal to be matched output from step S3, and slide the preset frame synchronization header template and the repaired pulse sequence point by point within the synchronization header arrival time window, using the sampling point index. To start a candidate frame, a feature vector is constructed by segmenting the code width into segments and calculating the average amplitude. Then, the correlation coefficient is calculated with the template, and a one-dimensional normalized correlation coefficient (or Hamming distance) is calculated. The position corresponding to the maximum value of the correlation coefficient (or the minimum Hamming distance and less than the set threshold) is taken as the frame start position. The frame is then decoded according to the preset encoding protocol to obtain the drilling trajectory parameters such as the tilt angle, azimuth angle, and tool face angle. The specific calculation method for the frame start position in step S4 is as follows: Method 1 (One-dimensional normalized correlation): Using candidate starting sampling points Based on the symbol width Extract the average envelope features of the first 8 code points of the synchronization header. And calculate the one-dimensional normalized cross-correlation coefficient by sliding point by point with the preset 8-bit frame synchronization header template vector. :

[0045] In the formula: This is the index of the sampling point corresponding to the starting position of the candidate frame; Synchronization header code element sequence number (value) (A total of 8 code elements) The symbol width (in this embodiment, it is) 1s is the symbol width The value of is related to the data frame structure [synchronization header 8 bits + data bits 13 bits, symbol width]. =1s] and the parameters in Example 3 are consistent); The system sampling rate (in this embodiment, it is taken as...) ); The pulse sequence after morphological repair; Indicated by When the starting point is, the first Average pulse amplitude within a symbol time window; This represents the sampling point number within the symbol time window, with a value range of 0 ≤ j ≤ Tb·fs-1.

[0046]

[0047] In the formula: For the frame synchronization header template The theoretical binary logic value of 1 bit ( ), consistent with the preset coding protocol of the device inside the hole.

[0048]

[0049] In the formula: For the template vector, the first The level value of each symbol; It is the arithmetic mean of the components of the template vector, i.e. (Regarding templates) , ); As the current candidate starting point Next Average envelope characteristics of each code element; It is the arithmetic mean of these 8 code element features, i.e. .

[0050]

[0051] In the formula: , These are the synchronization head arrival time windows. The indices of the start and end sampling points in the sampling sequence; This represents the total number of sampling points occupied by the 8-bit synchronization header.

[0052] When the maximum correlation coefficient (This embodiment sets relevant thresholds) When confirming The corresponding time is the precise start position of the data frame; if If the synchronization fails within the current time window, the system will output a retransmission prompt.

[0053] Method 2 (Hamming distance):

[0054]

[0055] In the formula: H(s) is the Hamming distance; (·) is a 0 / 1 hard decision function; The decision threshold is the average of the maximum and minimum values ​​of the synchronization header symbol features. This is the i-th component of the frame synchronization header template; This is the start position of the frame.

[0056] S5, Verification Output: Perform CRC or checksum verification on the decoding parameters output in step S4: if the verification passes, the drilling trajectory parameters such as inclination angle, azimuth angle, and tool face angle are parsed and displayed for the driller to adjust the drilling trajectory; if the verification fails or the start of a valid frame is not locked in step S4, the decoding is determined to have failed and a retransmission prompt is sent back to the device inside the hole.

[0057] Example 3

[0058] Based on Example 2, this study focuses on the mud pulse pressure signal under unstable operating conditions (pump start-up transient / low lift). The parameters include a borehole depth L = 1500m and a symbol width... =1s, sampling rate 100Hz, pulse amplitude 0.5MPa, pulse width 0.6s; transmission delay =5s, propagation speed v = 1400m / s (Tp ≈ 1.07s), tolerance ΔT = 3s; power-on judgment threshold =1.1· ( =3MPa, or 3.3MPa, determined to last for 0.3s); =0.05MPa / s =0.01MPa 2 Using parameters as typical environmental conditions, this embodiment of the invention provides a specific implementation plan for a downhole mud pulse signal demodulation method during the pump start-up pressure establishment period.

[0059] like Figure 3 The diagram shows the pressure fluctuation of the mud pulse signal involved in this invention under different operating conditions, and the data frame structure is explained below: Data frame structure: This embodiment uses a fixed format data frame, including an 8-bit synchronization header and 13-bit data bits, with a symbol width of... =1s, positive pulse amplitude is about 0.5MPa, pulse width is about 0.6s.

[0060] (a) Stable pumping conditions: such as Figure 3 As shown in (a), the pump pressure baseline is stable (approximately 3.0 MPa), the synchronization header features are clear, and the periodicity and amplitude of subsequent symbols are consistent. Normal decoding can be achieved with conventional demodulation branches.

[0061] (b) Pump pressure build-up period and low lift-off conditions (drill bit not completely lifted from the bottom of the hole): such as Figure 3 As shown in (b), there is a double distortion under this unstable operating condition: First, the pump pressure rises from zero and the reciprocating pump pressure fluctuates violently (the air bag is not fully pressurized), resulting in high-frequency jitter superimposed on the pump pressure rise baseline; Second, the synchronization head and the edge of the initial pulse are severely distorted or even submerged by noise, and the rising edge distortion is significant; If the low lift-off state is superimposed, the residual cutting torque causes low-frequency fluctuations in the pump pressure, further aggravating the pulse edge distortion, resulting in the loss of the synchronization head and the failure of conventional decoding.

[0062] The above comparison of operating conditions reveals the necessity of setting up an adaptive discrimination and transient demodulation branch for the operating conditions in this invention: stable operating conditions pass through lightweight decoding to save computing power, while unstable operating conditions (pumping transient / low lift-off) must go through transient processing such as envelope detrending and edge repair to restore the synchronization header and symbol form.

[0063] Specifically, transient condition discrimination uses an adaptive envelope as the pump pressure baseline, and calculates its rise slope and fluctuation variance; the adaptive envelope uses a 1s sliding window envelope and is smoothed for 0.2s; the differential threshold coefficient α = 2.5; subband reconstruction is achieved using a 0.3-6Hz bandpass filter (covering the 0.5-3Hz pulse edge frequency band); the morphological structure element length is 0.4s (40 sampling points); and the synchronization correlation coefficient threshold is 0.25.

[0064] In stable operating conditions, the conventional demodulation branch is used; in pump start-up transient / low lift-off operating conditions, the transient demodulation branch is activated. (1) Stabilize pumping conditions: Follow the steps below in sequence: ①Signal acquisition and preprocessing: The explosion-proof pressure transmitter converts the pressure signal into a 4~20mA current signal, which is then converted into a 1~5V voltage signal by the isolation circuit board. The signal is then converted into a digital signal by the 16-bit analog-to-digital converter at a sampling rate of 100Hz. After low-pass filtering and DC removal, the digital pressure signal p(n) is obtained. ② Pump start-up detection: Take the average pressure value for one consecutive minute after pump shutdown as the baseline. (This embodiment) Set the power-on threshold. (This embodiment takes) ,Right now ),when Amplitude of 30 consecutive sampling points (corresponding to) )Exceed The pump start-up time is defined as the moment corresponding to the first point exceeding the limit. ; ③ Synchronization head arrival time window prediction: transmit / receive waveform delay according to protocol =5s, take the propagation speed v = 1400m / s, and calculate the propagation time based on the hole depth L = 1500m. ≈1.07s, theoretical arrival time of the synchronization head = +6.07s, with a tolerance ΔT = 3s to determine the synchronization head arrival time window. =[ +3.07s, +9.07s]; ④ Transient condition identification: The pump pressure baseline is obtained by low-pass filtering the digital pressure signal within the time window (cutoff frequency 0.2Hz), and its rise slope is calculated. With fluctuation variance σ 2 Under stable operating conditions ≈0.02MPa / s (0.05MPa / s) and σ 2 ≈0.004MPa 2 < (0.01MPa) 2 This condition is considered a stable operating condition. ⑤ Conventional demodulation branch: The digital pressure signal is filtered and denoised using adaptive notch / noise cancellation, baseline drift is removed, false pulses are removed by setting the detection pressure difference according to the pulse amplitude, and then decoded according to the preset encoding protocol to obtain drilling trajectory parameters such as inclination angle, azimuth angle, and tool face angle; ⑥ Output display: Display the drilling trajectory parameters on the LCD screen and draw the drilling trajectory profile.

[0065] (2) Pump start-up transient / low lift-off condition: Perform the following steps in sequence: ①~③: Consistent with stable operating conditions, digital pressure signals are obtained. Pump start-up time and the synchronization head arrival time window [ +3.07s, +9.07s]; ④ Transient operating condition identification: During the pump start-up pressure build-up period, the pump pressure rises, and the baseline rise slope k≈0.12MPa / s > (0.05MPa / s), fluctuation variance σ 2 ≈0.025MPa 2 > (0.01MPa) 2 ), and determine that the current demodulation window is in the pump start-up pressure establishment period (including low lift-off), and switch to transient demodulation branch ⑤~⑧; ⑤ Adaptive envelope extraction (S31): A sliding window method is used (the window width is taken as the symbol width). This embodiment takes times, Extract the local minimum sequence of the digital pressure signal, perform cubic spline interpolation, and smooth to construct the lower envelope. ,by Eliminate the trend of pump start-up pressure rise; ⑥ Differential threshold mutation locking (S32): Obtain the difference sequence d(n) = s(n) - s(n-1) by first-order difference of s(n), calculate the root mean square σd, and set an adaptive threshold θ = α·σd (α is 2~3, 2.5 in this embodiment), and lock the differential amplitude if it exceeds + The moment is used as a candidate point for the rising edge, below - The moment is used as a candidate point for the falling edge, according to the symbol width. Perform edge pairing (time interval is [0.5)). 1.5 The range is determined as the effective pulse edge, resulting in a coarse pulse position locking result; ⑦ Subband decomposition and reconstruction (S33): Based on the coarse locking result, the pulse segment is truncated, and the detailed subband containing the pulse edge frequency band (0.5~3Hz) features is selected for reconstruction using discrete wavelet transform (db4 wavelet, decomposition level 3~5). Alternatively, variational mode decomposition (number of modes K is 5~8) is used to select the modal components whose center frequency is within the pulse edge frequency band and separated from the pump noise frequency band for reconstruction, thus obtaining the enhanced pulse segment. ⑧ Morphological repair (S34): Select flat linear structural elements (length is 1 / 4 to 1 / 2 of the preset symbol width, in this embodiment, 0.4s corresponds to 40 sampling points), and perform morphological opening operation (erosion followed by dilation) and closing operation (dilation followed by erosion) on the enhanced pulse segment in sequence to remove pulse edge burrs and fill gaps. Then, shape the pulse according to the slope constraint of the adjacent pulse edge to obtain the repaired pulse sequence. ⑨ Synchronization Matching and Decoding (S4): Using the frame synchronization header template within the synchronization header arrival time window [ +3.07s, Within +9.07s, the repaired pulse sequence is slid point by point to calculate the one-dimensional normalized correlation coefficient (see step S4 for the specific formula). The maximum correlation coefficient that is not lower than the threshold is used. The position corresponding to (0.25) is used as the frame start position, and the drilling trajectory parameters are obtained by decoding according to the preset encoding protocol; ⑩ Output display: Displays drilling trajectory parameters; if decoding fails, outputs a retransmission prompt, and the device inside the hole will automatically retransmit the data frame in the next pump start cycle according to the preset protocol.

[0066] Monte Carlo simulations were performed using the above-mentioned optimized parameters, with four operating conditions set (stable pumping, pump start-up transient, low lift-off, and pump start-up transient + low lift-off superposition). The signal-to-noise ratio was set to 5 / 10 / 15dB. Each operating condition × signal-to-noise ratio combination was used for 100 simulations (with a fixed random seed of 2026). The synchronization detection rate, frame decoding success rate, bit error rate, and average number of retests were statistically analyzed.

[0067] Simulation results are shown in the table below (data format: "conventional demodulation / method of this invention"):

[0068] Note: The four columns "Synchronization Detection Rate / Frame Decoding Success Rate / Bit Error Rate / Average Retest Count" in the table are comparison values ​​between "Conventional Demodulation / This Invention's Method".

[0069] As shown in the table, at SNRs of 10dB and 15dB, the frame decoding success rate of the method of this invention is no lower than that of the conventional demodulation method under all operating conditions. The improvement is significant under the following conditions: pumping transient condition (from 8% to 85% at 15dB), low lift-off condition (from 0% to 98% at 15dB), and pumping transient + low lift-off combined condition (from 0% to 32% at 15dB). The bit error rate and average retest count decrease simultaneously and significantly. Under stable pumping conditions, the performance of the two methods is comparable (frame decoding success rates of 86% and 99% respectively at 15dB). At 5dB, which is an extremely low SNR condition, both methods are close to failure and fall within the normal range.

[0070] The simulation uses the Monte Carlo method: 4 operating conditions × 3 signal-to-noise ratios (5 / 10 / 15dB) × 100 trials per combination, with a fixed random seed of 2026. The synchronization detection rate (a frame synchronization header position error of less than 0.5 times the symbol width is considered a detection), frame decoding success rate (all 13 data bits are correct), bit error rate, and average number of retests are statistically analyzed.

[0071] Simulation statistics show that, at signal-to-noise ratios of 10dB and 15dB, the method of this invention significantly outperforms conventional methods in frame decoding success rates under pump-on transient, low lift-off, and combined operating conditions (improvements of 30% to 98%), and exhibits excellent performance under steady-state conditions (reaching 99% at 15dB). This invention has practical industrial applicability at signal-to-noise ratios of (…). Significant technological advantages and advancements under unstable operating conditions.

[0072] Figure 4 The waveforms show a comparison of synchronization and decoding decisions between conventional demodulation and the method of this invention under low lift-off conditions (15dB). Wherein: Figure 4 (a) The raw orifice pressure waveform is collected (a real synchronization head exists at t≈6.07 s); the vertical axis "Decision Value" represents the binary level output sequence (level 1.0 represents logic '1' / pulse high level, level 0.2 represents logic '0' / pulse low level), the dashed line represents the actual transmitted 8-bit synchronization head + 13-bit data sequence, and the solid line represents the algorithm demodulation decision output sequence. Conventional demodulation is affected by low-frequency fluctuations due to low lift-off, resulting in missynchronization locking at t≈11.90 s, leading to full-frame decoding errors; the method of this invention accurately locks the synchronization head at t≈5.75 s under time window constraints (error less than 0.5). The waveform was completely restored, and all 13-bit data was decoded correctly.

[0073] In one embodiment, such as Figure 5 The schematic diagram of the computer device shown illustrates a computer device, which can be a server, and its internal structure can be as follows. Figure 5 As shown, the computer device includes a processor, memory, and a network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database stores static and dynamic information data. The network interface communicates with external terminals via a network connection. When the computer program is executed by the processor, it implements the steps in the above method embodiments.

[0074] Those skilled in the art will understand that Figure 5 The structure shown is merely a block diagram of a portion of the structure related to the present invention and does not constitute a limitation on the computer device to which the present invention is applied. A specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0075] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the methods described above. Any references to memory, storage, databases, or other media used in the embodiments provided by this invention can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical storage, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0076] Finally, it should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0077] The above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them; although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications can still be made to the specific implementation of the present invention or equivalent substitutions can be made to some technical features without departing from the spirit of the technical solutions of the present invention, and all such modifications and substitutions should be covered within the scope of the technical solutions claimed in the present invention.

Claims

1. A downhole mud pulse signal demodulation system for the pump start-up pressure establishment period, characterized in that, The system includes: an in-hole device for detecting the pump start-up or stop-up status of the mud pump through the flow switch, and an orifice device for acquiring manifold pressure signals through an explosion-proof pressure transmitter and transmitting them to an explosion-proof computer. The in-hole device is installed inside the drill string and includes, from the bottom of the hole to the opening, a directional drill bit, a screw motor, a non-magnetic outer tube, a pulse generator, and the drill string. The non-magnetic outer tube contains a measuring section, a rechargeable battery tube, and a drive section. The measuring section has a flow switch at its front end. The orifice device includes a drill rod, a water feeder, a high-pressure hose, a pump truck, an explosion-proof pressure transmitter, and an explosion-proof computer for demodulating signals during the pump start-up pressure build-up period and under unstable low lift-off conditions. The explosion-proof pressure transmitter is installed on the outlet side of the mud pump, and the explosion-proof computer is connected to the explosion-proof pressure transmitter via signal connection. The in-hole device collects the static tilt angle and azimuth angle when the pump is stopped, and collects the dynamic tool face angle of the screw motor when the pump is started, and encodes them according to a preset communication protocol; according to a preset wave delay, the pulse generator intermittently changes the hydraulic channel area inside the drill string to generate mud positive pressure pulses, and uses the flushing fluid inside the drill string as the signal transmission channel to transmit the encoded trajectory parameters backflow to the borehole opening.

2. A method for demodulating downhole mud pulse signals during the pump start-up pressure establishment period, characterized in that, The method is implemented using the downhole mud pulse signal demodulation system for the pump start-up pressure establishment period as described in claim 1, and includes the following steps: S1. Pump start-up detection and time window prediction: Real-time monitoring of mud pump manifold pressure or flow switch status to detect mud pump start-up time. ; Delay for transmitting waves according to the user-defined protocol and the propagation time of the hole depth Calculate the theoretical arrival time of the synchronization head. And set the search time window. ; S2, Adaptive Working Condition Judgment: Within the time window described in step S1, mud pulse pressure signals are acquired. Calculate the slope of the pump pressure baseline rise. and volatility variance By Compared with the preset slope threshold Compare, Compared with the preset variance threshold Compare and determine whether the current operating condition label is a stable operating condition or an unstable operating condition; S3. Transient demodulation processing: Receive the pressure signal received in step S2 With operating condition labels; for signal distortion during the pump start-up pressure build-up period and under low lift-off unstable operating conditions, adaptive envelope extraction and baseline elimination, differential threshold edge coarse locking, subband decomposition and reconstruction enhancement, and morphological edge trimming are sequentially applied to output a repaired pulse sequence. ; S4. Synchronous matching and decoding: Receive the signal to be matched output from step S3, and slide the preset frame synchronization header template and the repaired pulse sequence point by point within the synchronization header arrival time window, using the sampling point index. To start a candidate frame, a feature vector is constructed by segmenting the code width into segments and calculating the average amplitude. Then, the correlation coefficient is calculated with the template. A one-dimensional normalized correlation coefficient or Hamming distance is calculated. The position corresponding to the maximum correlation coefficient or the minimum Hamming distance that is less than a set threshold is taken as the frame start position. The frame is then decoded according to a preset encoding protocol to obtain the drilling trajectory parameters such as the tilt angle, azimuth angle, and tool face angle. S5, Verification Output: Perform CRC or checksum verification on the decoding parameters output in step S4: if the verification passes, the drilling trajectory parameters such as inclination angle, azimuth angle, and tool face angle are parsed and displayed for the driller to adjust the drilling trajectory; if the verification fails or the start of a valid frame is not locked in step S4, the decoding is determined to have failed and a retransmission prompt is sent back to the device inside the hole.

3. The method for demodulating downhole mud pulse signals during the pump start-up pressure establishment period according to claim 2, characterized in that, The S1 step, pump start-up detection and time window prediction, specifically includes the following steps: S11. Signal Acquisition and Preprocessing: Acquire steady-state pressure signals during the pump shutdown phase and calculate the steady-state baseline pressure during pump shutdown. Set the pump start-up threshold. The proportionality coefficient Pick Pressure Increment Bias All greater than At that time, the first to exceed The sampling point corresponding to the time is determined as the time when the mud pump starts. ; S12. Synchronization head arrival time window prediction: After the flow switch detects pump start-up, a preset waveform delay is elapsed. Data frames have just begun to be sent; the propagation speed of the mud pulse signal in the flushing fluid inside the drill string. From the current hole depth Calculate propagation time Then the synchronization head arrives at the theoretical time. ,by Centered on the target and with a set tolerance ΔT as the radius, determine the arrival time window of the synchronization head. .

4. The method for demodulating downhole mud pulse signals during the pump start-up pressure establishment period according to claim 2, characterized in that, The S2, adaptive working condition discrimination, specifically includes: For time windows The digital pressure signal within the time window is extracted using a low-pass filter to obtain the macroscopic baseline of the pump pressure; the rise slope of the baseline within the time window is calculated through first-order linear fitting. And calculate the residual variance after filtering out the baseline. Set a discrimination threshold and The judgment logic is as follows; (1) Transient characteristics of pump start-up: The pump pressure is in the pressure rise period, which is manifested as the baseline rise slope. ; (2) Low lifting condition characteristics: When the drill bit is not completely lifted from the bottom of the hole, the residual cutting torque causes the flushing fluid pipeline to generate through the impedance feedback of the screw motor. Low-frequency, large-amplitude pressure jitter, manifested as fluctuation variance .

5. The method for demodulating downhole mud pulse signals during the pump start-up pressure establishment period according to claim 4, characterized in that, When satisfied and / or When the system determines that it is currently in an unstable operating condition such as the pump start-up pressure build-up period or low lift-off, it switches to transient demodulation branch processing. When not satisfied and / or If the condition is stable, the system switches to the conventional demodulation branch. The digital pressure signal is then filtered and denoised using adaptive notch / noise cancellation to remove baseline drift. The differential pressure is set according to the pulse amplitude to remove false pulses. Finally, the signal is decoded according to the preset encoding protocol to obtain the borehole trajectory parameters.

6. The method for demodulating downhole mud pulse signals during the pump start-up pressure establishment period according to claim 2, characterized in that, The S3 transient demodulation process specifically includes the following steps: S31. Adaptive envelope extraction: To address the physical characteristic of the upward abrupt change in the positive mud pulse pressure signal, an adaptive extraction of the pump pressure rise baseline envelope using a sliding window based on local minima is employed: the sliding window width is set to a preset symbol width. of Double, search for digital pressure signals within the sliding window. The sequence of local minima is obtained; cubic spline interpolation and smoothing filtering are performed on the local minima sequence to construct a lower envelope signal that fits the trend of pump start-up pressure rise. ; through subtraction Remove the low-frequency pump pressure rise trend and baseline drift caused by the pump start-up transient to obtain the pulse signal that brings the baseline to zero. ; S32, Differential Threshold Mutation Locking: Performing a first-order difference operation on the pulse signal s(n) yields the difference sequence: d(n) = s(n) - s(n-1); Where d(n) is the first-order difference sequence of the pulse signal s(n), that is, the difference between adjacent sampling points; Set the adaptive threshold θ = α·σd, where σd is the root mean square of the difference sequence. Take 2 to 3; for differential amplitude values ​​exceeding + The moment is used as a candidate point for the rising edge of the pulse, below The moment is used as a candidate point for the falling edge of the pulse; according to the preset symbol width Edge pairing is performed when the time interval between adjacent rising and falling edge candidate points is within [0.5]. 1.5 When the range is within the specified range, it is determined as a group of valid pulse edges. Each pair of valid edges is used as an anchor point, and its corresponding continuous time interval is determined according to the preset symbol width to obtain the pulse position coarse locking result containing the edge position and its corresponding time interval. S33. Subband Decomposition and Reconstruction: Based on the coarse pulse position locking result, pulse segments are extracted and multi-resolution subband decomposition is performed: discrete wavelet transform is used to select detailed subbands containing pulse edge frequency band features for reconstruction; or variational mode decomposition is used to select mode components whose center frequency is within the pulse edge frequency band and separated from the pump noise frequency band for reconstruction, thus obtaining enhanced pulse segments. S34. Morphological Repair: Select flat linear structural elements, perform morphological opening and closing operations on the enhanced pulse segments in sequence, remove pulse edge burrs and fill gaps, and then reshape the rising or falling edge according to the slope constraint of adjacent pulse edges to obtain the repaired pulse sequence.

7. The method for demodulating downhole mud pulse signals during the pump start-up pressure establishment period according to claim 2, characterized in that, The specific calculation method for the frame start position in step S4 is a one-dimensional normalized correlation, as follows: Candidate starting sampling points Based on the symbol width Extract the average envelope features of the first 8 code points of the synchronization header. And calculate the one-dimensional normalized cross-correlation coefficient by sliding point by point with the preset 8-bit frame synchronization header template vector. : In the formula: This is the index of the sampling point corresponding to the starting position of the candidate frame. This is the sequence number of the synchronization header element. The width of the symbol. The system sampling rate, The pulse sequence after morphological repair. Indicated by When the starting point is, the first Average pulse amplitude within a symbol time window; The sampling point number within the symbol time window, with a value range of 0 ≤ j ≤ Tb·fs-1; In the formula: For the frame synchronization header template The theoretical binary logic value of 1 bit ( ), consistent with the preset coding protocol of the device inside the hole; In the formula: For the template vector, the first The level value of each symbol; It is the arithmetic mean of the components of the template vector, i.e. ; As the current candidate starting point Next Average envelope characteristics of each code element; It is the arithmetic mean of these 8 code element features, i.e. ; In the formula: For the synchronization head arrival time window The corresponding starting sampling point index in the sampling sequence, For the synchronization head arrival time window The index of the corresponding termination sampling point in the sampling sequence; This represents the total number of sampling points occupied by the 8-bit synchronization header. When the maximum correlation coefficient At that time, confirm The corresponding time is the precise start position of the data frame; if If the synchronization fails within the current time window, the system will output a retransmission prompt.

8. The method for demodulating downhole mud pulse signals during the pump start-up pressure establishment period according to claim 2, characterized in that, The specific method for calculating the frame start position in step S4 is the Hamming distance, as follows: In the formula: H(s) is the Hamming distance; (·) is a 0 / 1 hard decision function; This is the decision threshold; This is the i-th component of the frame synchronization header template; This is the start position of the frame.

9. A computer-readable storage medium, characterized in that, Used to store a computer program, wherein the computer program, when executed by a processor, implements the downhole mud pulse signal demodulation method as described in any one of claims 2-8 during the pump start-up pressure establishment period.

10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 2-8.

Images