WO2021120454A1 - 一种随钻测量mwd系统噪声消除方法、装置及存储介质 - Google Patents

一种随钻测量mwd系统噪声消除方法、装置及存储介质 Download PDF

Info

Publication number
WO2021120454A1
WO2021120454A1 PCT/CN2020/083789 CN2020083789W WO2021120454A1 WO 2021120454 A1 WO2021120454 A1 WO 2021120454A1 CN 2020083789 W CN2020083789 W CN 2020083789W WO 2021120454 A1 WO2021120454 A1 WO 2021120454A1
Authority
WO
WIPO (PCT)
Prior art keywords
signal
mud
pumping
frequency
buffer
Prior art date
Application number
PCT/CN2020/083789
Other languages
English (en)
French (fr)
Inventor
王智明
张爽
陈伟
张松炜
顾庆水
杜小强
孙宝阳
程怀标
袁超
朱辉
杨波
Original Assignee
中国海洋石油集团有限公司
中海油田服务股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中国海洋石油集团有限公司, 中海油田服务股份有限公司 filed Critical 中国海洋石油集团有限公司
Publication of WO2021120454A1 publication Critical patent/WO2021120454A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
    • G01V1/00—Seismology; Seismic or acoustic prospecting or detecting
    • G01V1/40—Seismology; Seismic or acoustic prospecting or detecting specially adapted for well-logging
    • G01V1/44—Seismology; Seismic or acoustic prospecting or detecting specially adapted for well-logging using generators and receivers in the same well
    • G01V1/48—Processing data
    • G01V1/50—Analysing data
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
    • G01V2210/00—Details of seismic processing or analysis
    • G01V2210/60—Analysis
    • G01V2210/61—Analysis by combining or comparing a seismic data set with other data
    • G01V2210/616—Data from specific type of measurement
    • G01V2210/6169—Data from specific type of measurement using well-logging

Definitions

  • the present disclosure relates to, but is not limited to, the field of logging, in particular to a method, device and storage medium for eliminating noise in a MWD measurement while drilling system.
  • the Measurement While Drilling (MWD) system is used to measure downhole geological parameters and engineering parameters in real time during the drilling process.
  • the mud pulse data while drilling demodulation system is used to transmit the data measured by the measurement while drilling system from downhole to the surface in real time.
  • the mud pulse while drilling data transmission system uses the movement of the pulser to generate the mud interception effect to cause mud pressure fluctuations, modulate the digital signal in the mud pressure wave and transmit it to the ground, so as to realize the real-time transmission of downhole measurement data to the ground.
  • the mud pulse data transmission system while drilling contains three main functional modules: mud pulse generator, mud signal transmission channel, and pressure wave acquisition and demodulation system.
  • the mud pulse generating mechanism is mainly a pulser, and the pulse signal generated can be divided into three categories: positive pulse, negative pulse, and continuous wave pulse.
  • the core principle of the mud pulse generating mechanism is the interception effect of the mud.
  • the mud signal transmission channel includes the entire drilling fluid circulation pipeline, upwards includes mud pumps, manifolds, risers, hoses, and various mud flow joints, and downwards includes drill pipes, multiple logging instruments, and runner converters. , Steering nubs, drill bits and bottom-hole reflective surfaces, etc., through the mud signal transmission path, mud pressure waves are transmitted from downhole to uphole, and each part of the mud pressure wave causes superimposed noise or attenuation characteristics.
  • the pressure wave acquisition and demodulation system includes a pressure sensor to convert the mud pressure wave into an electrical signal, and the demodulation system processes the electrical signal to demodulate it into meaningful data.
  • the mud signal transmission system contains a variety of equipment and tools with complex structures and mechanical motion characteristics. These equipment and tools will cause superimposed noise on the mud pulse signal and affect the quality of the mud pulse signal.
  • These superimposed noises are usually generated by specific mechanical structures, and mainly include three categories.
  • the first category is periodic noise, such as mud pump noise, motor rotating noise, top drive, turntable rotating noise, etc.; the second category is non-periodic noise.
  • Aperiodic noise is mainly manifested as the attenuation of the signal, such as the noise generated by the runner converter, the diameter of the runner, the water eye structure of the drill bit, etc.;
  • the third type is instantaneous noise, which is only generated under special circumstances, such as drill bits. The water eye is temporarily blocked and so on. These noises are mixed together to produce superimposed noises that affect the quality of mud pulse signals.
  • the noise elimination methods used include time domain elimination of pump noise, frequency domain elimination of high-frequency electrical noise, etc.
  • the above-mentioned techniques cannot achieve good periodic noise interference. Denoising treatment can easily distort the mud signal.
  • the embodiments of the present disclosure provide a noise elimination method for a Measurement While Drilling (MWD) system, which realizes the suppression of periodic noise in the frequency domain, and can effectively eliminate the interference of periodic noise.
  • MWD Measurement While Drilling
  • embodiments of the present disclosure provide a method for eliminating noise in a measurement while drilling MWD system, including:
  • a noise-removed mud frequency domain output signal is obtained.
  • the receiving the collected mud signal and determining the characteristic frequency of the periodic noise signal in the mud signal includes:
  • the average pumping frequency obtained according to the pumping signal in the pumping signal buffer is determined as the characteristic frequency of the periodic noise signal in the mud signal.
  • the receiving the collected mud signal and determining the characteristic frequency of the periodic noise signal in the mud signal includes:
  • the obtained preset frequency of any one or more periodic noise signals is determined as the period in the mud signal The characteristic frequency of the noise signal.
  • the receiving the collected mud signal and determining the characteristic frequency of the periodic noise signal in the mud signal includes:
  • the central frequency of the periodic noise signal in the mud signal is determined as the characteristic frequency of the periodic noise signal in the mud signal according to the frequency spectrum of the mud signal.
  • the method Before performing Fourier transform on the mud signal to obtain a mud signal in the frequency domain, the method further includes:
  • s b [s in (nM)... s in (n-2) s in (n-1) s in ] T
  • the pumping signal is:
  • p in [p in (n) p in (n+1)... p in (n+K-1)] T ;
  • p in represents an input signal vector pump stroke
  • p in (n) is the input pump stroke signal
  • K is the length of the input signal vector impulse pump
  • n denotes an input signal of the pump stroke starting sequence number.
  • the inputting the pumping signal into the pumping signal buffer when the inputted pumping signal can be obtained includes:
  • p b is the pump stroke signal buffer pump stroke signal
  • N is the length of the pump stroke signal buffer
  • p in (n) signal is input pump strokes
  • N K + M
  • M is a pump strokes signal buffer The length of the original signal vector in.
  • the determining the average pumping frequency obtained from the pumping signal in the pumping signal buffer as the characteristic frequency of the periodic noise signal in the mud signal includes:
  • the average pumping frequency of the pumping signal in the pumping signal buffer is calculated
  • the formula for calculating the average pumping frequency includes:
  • the performing Fourier transform on the mud signal to obtain the mud signal in the frequency domain includes:
  • the obtaining the denoised mud frequency domain output signal according to the characteristic frequency of the periodic noise signal in the mud signal, the mud signal in the frequency domain and a preset suppression factor includes:
  • the inhibitory factor includes:
  • ⁇ j is the suppression factor, the value of ⁇ j is a positive real number less than 1, j is the sample point number in the period, S is the number of samples in the frequency domain after Fourier transform; the suppression factor suppression frequency is The signal component of k is an integer and satisfies Is the characteristic frequency; f 0 is the center frequency of the effective mud signal; B s is the bandwidth of the effective mud signal in the frequency domain; ⁇ f is the frequency resolution of the Fourier transform of the mud signal, T s is the sampling period of the mud signal, and N is the number of sampling points in the period; Indicates rounding down.
  • the embodiments of the present disclosure also provide a noise cancellation device for a measurement while drilling MWD system, the device including: a memory and a processor;
  • the memory is set to save a program for noise elimination of the MWD system
  • the processor is configured to read and execute the program for MWD system noise cancellation, and perform the following operations:
  • a noise-removed mud frequency domain output signal is obtained.
  • the receiving the collected mud signal and determining the characteristic frequency of the periodic noise signal in the mud signal includes:
  • the average pumping frequency obtained according to the pumping signal in the pumping signal buffer is determined as the characteristic frequency of the periodic noise signal in the mud signal.
  • the receiving the collected mud signal and determining the characteristic frequency of the periodic noise signal in the mud signal includes:
  • the obtained preset frequency of any one or more periodic noise signals is determined as the period in the mud signal The characteristic frequency of the noise signal.
  • the receiving the collected mud signal and determining the characteristic frequency of the periodic noise signal in the mud signal includes:
  • the central frequency of the periodic noise signal in the mud signal is determined as the characteristic frequency of the periodic noise signal in the mud signal according to the frequency spectrum of the mud signal.
  • the mud signal is:
  • s in [s in (n) s in (n+1)... s in (n+K-1)] T ; among them, s in (n) represents the mud signal, s in represents the input mud signal vector, and K is The length of the input signal vector, n represents the starting number of the input signal;
  • the processor Before performing the Fourier transform on the mud signal to obtain the mud signal in the frequency domain, the processor further performs the following operations:
  • s b [s in (nM)... s in (n-2) s in (n-1) s in ] T ,
  • p in represents an input signal vector pump stroke
  • p in (n) is the input pump stroke signal
  • K is the length of the input signal vector impulse pump
  • n denotes an input signal of the pump stroke starting sequence number.
  • inputting the pumping signal into the pumping signal buffer includes:
  • p b is the pump stroke signal buffer pump stroke signal
  • N is the length of the pump stroke average signal buffer
  • p in (n) signal is input pump strokes
  • N K + M
  • M is a pump strokes signal buffering The length of the original signal vector in the device.
  • the determining the average pumping frequency obtained from the pumping signal in the pumping signal buffer as the characteristic frequency of the periodic noise signal in the mud signal includes:
  • the average pumping frequency of the pumping signal in the pumping signal buffer is calculated
  • the formula for calculating the average pumping frequency includes:
  • the performing Fourier transform on the mud signal to obtain the mud signal in the frequency domain includes:
  • the obtaining the denoised mud frequency domain output signal according to the characteristic frequency of the periodic noise signal in the mud signal, the mud signal in the frequency domain and a preset suppression factor includes:
  • the inhibitory factor includes:
  • ⁇ j is the suppression factor, the value of ⁇ j is a positive real number less than 1, j is the sample point number in the period, S is the number of samples in the frequency domain after Fourier transform; the suppression factor suppression frequency is The signal component of k is an integer and satisfies Is the characteristic frequency; f 0 is the center frequency of the effective mud signal; B s is the bandwidth of the effective mud signal in the frequency domain; ⁇ f is the frequency resolution of the Fourier transform of the mud signal, T s is the sampling period of the mud signal, and N is the number of sampling points in the period; Indicates rounding down; among them, the number of sampling points in the time domain period N and the number of sampling points in the frequency domain period S are the same.
  • a computer-readable storage medium stores a computer program and executable instructions. When the computer-executable instructions are executed by a processor, the MWD system noise elimination method is implemented.
  • Figure 1 is a schematic diagram of the overall structure of a drilling fluid MWD system in a technology
  • Figure 2 is a flow chart of signal processing at the ground receiving end in a technology
  • Fig. 3 is a flowchart of a method for eliminating noise in a measurement while drilling MWD system according to an embodiment of the present invention
  • Fig. 4 is a schematic diagram of a noise elimination device for a measurement while drilling MWD system according to an embodiment of the present invention
  • Figure 5 is a flow chart of periodic noise denoising in which periodic signals cannot be collected in a technique
  • Figure 6 is a periodic noise denoising process that can collect periodic signals in a technique
  • Fig. 7 is a flowchart of a method for eliminating noise in an MWD system according to an example of an embodiment of the present invention.
  • Fig. 8a is a schematic diagram of a power spectrum of a ground acquisition signal according to an example of an embodiment of the present invention.
  • Fig. 8b is a schematic diagram of a signal power spectrum after the pumping interference is eliminated according to an example of the embodiment of the present invention.
  • Measurement While Drilling is a technology that can measure and collect logging data near the drill bit during drilling, and transmit the collected logging data to the surface system in real time.
  • Logging data usually includes formation characteristic information and a variety of drilling engineering parameters.
  • the basic working principle of the drilling fluid pressure signal transmission method is to convert the information measured downhole into control information, and the control information is applied to the downhole drilling fluid pressure signal generation
  • the device changes the pressure of the drilling fluid in the transmission channel to generate drilling fluid pressure pulses.
  • the pressure pulses are transmitted to the surface through the drilling fluid in the transmission channel, and are processed by the surface processing system and converted into required downhole measurement information.
  • the overall structure of the drilling MWD system is shown in Figure 1.
  • the MWD system includes: mud pool, mud pump, ground receiving unit and downhole sending end; the mud pool is connected with the mud pump, and the mud pump is connected with the ground receiving.
  • the mud pump drives the circulation of the drilling fluid.
  • the downhole transmitter sends the data to the surface in the form of drilling fluid pressure pulses.
  • the pressure sensor converts the pressure changes of the drilling fluid into electrical signals and sends them to the surface receiving unit.
  • the surface receiving unit is used for decoding out of the well. The data sent by the sender.
  • the signal processing flow commonly used at the ground receiving end is shown in Figure 2.
  • the embodiments of the present disclosure are related to the technical solution of noise interference cancellation, which is carried out in the preprocessing stage, so that the subsequent data demodulation and decoding in the processing flow can be accurately solved. Data sent downhole.
  • the characteristics of drilling fluid signal transmission mainly include signal transmission speed, signal attenuation, signal reflection and so on. But for the drilling fluid pressure transmission system, the pressure pulse signal is transmitted from the bottom of the well to the wellhead in the drill string. Because the drilling fluid is a three-phase flow of gas, liquid, and solid, it contains solid phases such as clay, cuttings, and barite powder. There are also gaseous substances such as free gas. The strength of the drilling fluid pressure pulse signal generated by the pulse generator will be continuously attenuated. The attenuation degree is affected by the signal frequency and transmission distance, and is also related to the inner diameter of the drilling fluid channel and the drilling fluid channel. The type of liquid is related to internal parameters such as composition, viscosity, and volumetric gas content.
  • the drilling fluid channel is a channel with very complex transmission characteristics.
  • the output signal of the pressure sensor installed on the riser to detect the fluctuation of mud pressure includes not only the useful mud signal from downhole, but also Contains large-scale periodic pressure fluctuation noise caused by mud pump compression of mud, pressure fluctuation noise caused by various other mechanical actions, and random noise.
  • the interference appears as periodic pulses related to the pumping characteristics, and the noise appears as broadband white Noise, the amplitude of the noise is much larger than the amplitude of the useful signal, and the useful signal of mud at the wellhead is completely submerged in a variety of noises.
  • the statistical distribution of the drilling fluid noise signal in the time domain is normally distributed and contains a strong periodic component.
  • the periodic components include mud pump noise, motor rotation noise, top drive, turntable rotation noise, and so on.
  • FIG. 3 is a flowchart of a method for eliminating noise in a measurement while drilling MWD system according to an embodiment of the present disclosure, including steps 301 to 303.
  • Step 301 Receive the collected mud signal and determine the characteristic frequency of the periodic noise signal in the mud signal;
  • Step 302. Perform Fourier transform on the mud signal to obtain a mud signal in the frequency domain;
  • Step 303 Obtain a noise-removed mud frequency-domain output signal according to the characteristic frequency of the periodic noise signal in the mud signal, the mud signal in the frequency domain, and a preset suppression factor.
  • the periodic noise of the MWD MWD system is eliminated in the frequency domain.
  • the measurement while drilling MWD system can be used to measure multiple geological parameters downhole in real time during the drilling process, including: measurement while drilling borehole trajectory parameters, such as inclination angle, azimuth angle, Tool face angle and auxiliary parameters such as temperature.
  • the pulse signal generated by the mud pump has a strong amplitude.
  • the frequency component of the pulse signal generated by the mud pump is mixed with the mud signal sent downhole, it will interfere with the useful mud signal. .
  • Periodic noise in the mud signal can include mud pump noise, motor rotation noise, top drive, turntable rotation noise, etc.
  • the receiving the collected mud signal and determining the characteristic frequency of the periodic noise signal in the mud signal includes: receiving the collected mud signal; when the input pumping signal can be obtained, turning The pumping signal is input to a pumping signal buffer; the average pumping frequency obtained according to the pumping signal in the pumping signal buffer is determined as the characteristic frequency of the periodic noise signal in the mud signal.
  • the mud signal is input into the mud signal buffer to obtain the mud signal s b in the mud signal buffer:
  • s b [s in (nM)... s in (n-2) s in (n-1) s in ] T
  • N the length of the mud signal buffer
  • the pumping signal is:
  • p in [p in (n) p in (n+1)... p in (n+K-1)] T ;
  • p in represents an input signal vector pump stroke
  • p in (n) is the input pump stroke signal
  • K is the length of the input signal vector impulse pump
  • n denotes an input signal of the pump stroke starting sequence number.
  • inputting the pumping signal into the pumping signal buffer includes:
  • p b is the pump stroke signal buffer pump stroke signal
  • N is the length of the pump stroke signal buffer
  • p in (n) signal is input pump strokes
  • N K + M
  • M represents the original signal buffer Signal vector length.
  • determining the average pumping frequency obtained from the pumping signal in the pumping signal buffer as the characteristic frequency of the periodic noise signal in the mud signal includes:
  • the average pumping frequency of the pumping signal in the pumping signal buffer is calculated
  • the formula for calculating the average pumping frequency includes:
  • receiving the collected mud signal and determining the characteristic frequency of the periodic noise signal in the mud signal includes: receiving the collected mud signal; When there are multiple preset frequencies of periodic noise signals, the acquired preset frequencies of any one or more periodic noise signals are determined as the characteristic frequencies of the periodic noise signals in the mud signal.
  • the periodic noise signal may include top drive noise, and the characteristic frequency of the top drive noise signal can obtain the input frequency of the top drive through the top drive device, and the input frequency is the frequency of the top drive, that is, the top drive The characteristic frequency of the noise signal, where the periodic noise signal includes but is not limited to top drive noise, including any periodic noise that can be obtained with a preset frequency.
  • receiving the collected mud signal and determining the characteristic frequency of the periodic noise signal in the mud signal includes: receiving the collected mud signal; when any periodic noise signal cannot be obtained, according to the mud The frequency spectrum of the signal determines the central frequency of the periodic noise signal in the mud signal as the characteristic frequency of the periodic noise signal in the mud signal. In this embodiment, the case where the periodic noise frequency cannot be obtained is targeted.
  • step 302 the received mud signal is subjected to Fourier transform processing to obtain a mud signal in the frequency domain.
  • the method may further include:
  • s b [s in (nM)... s in (n-2) s in (n-1) s in ] T
  • N the length of the mud signal buffer
  • M the signal buffer The original signal vector length in the middle.
  • the performing Fourier transform on the mud signal to obtain the mud signal in the frequency domain includes: windowing the mud signal in the mud signal buffer to obtain the windowed mud Signal; Fourier transform is performed on the windowed mud signal to obtain a mud signal in the frequency domain.
  • the mud signal s b in the mud signal buffer is multiplied by a preset window function, and the windowed mud signal s w is calculated.
  • the windowing calculation formula is as follows:
  • s w is the mud signal after window treatment
  • w N is the preset window function
  • w N [w(0) w(1)... w(N-1)] T
  • N is the length of the window function
  • " ⁇ " in the formula means that two vectors are multiplied element by element.
  • the preset window function can be Hanning window or other window functions;
  • the window function vector of Hanning window is:
  • N is the length of the window function
  • n is the serial number of the sampling point in the window function.
  • the window function for windowing processing may be Hanning window or other window functions, which is not specifically limited, and different window functions may be selected according to the relevant information of the drilling data to be processed.
  • the denoised mud frequency domain output signal is obtained according to the average pumping frequency, the mud signal in the frequency domain and the preset suppression factor.
  • the value of the suppression factor can be set to a fixed value, or it can be selected according to the intensity of the pumping interference and the expected suppression effect; the smaller the value of the suppression factor, the more obvious the suppression of the pumping interference component of the corresponding frequency.
  • the obtaining the denoised mud frequency domain output signal according to the average pumping frequency, the mud signal in the frequency domain and a preset suppression factor includes:
  • the inhibitory factor includes:
  • ⁇ j is the suppression factor
  • the value of ⁇ j is a positive real number less than 1
  • j is the number of samples in the period
  • S is the number of samples in the frequency domain after Fourier transform
  • the suppression frequency of the suppression factor is The signal component of k is an integer and satisfies Is the characteristic frequency
  • f 0 is the center frequency of the effective signal of the mud signal in the frequency domain
  • B s is the bandwidth of the effective signal of the mud signal in the frequency domain
  • ⁇ f is the frequency resolution of the Fourier transform of the mud signal
  • T s is the sampling period of the mud signal
  • N is the number of sampling points in the period
  • S w may be a mud signal in the frequency domain or a mud signal in the frequency domain after windowing.
  • the suppression frequency is the signal component of k is an integer and satisfies f 0 is the center frequency of the effective mud signal in the frequency domain.
  • the effective mud signal of this frequency is the signal generated by the downhole pulser and transmitted to the ground acquisition system for demodulation.
  • f 0 is determined by the modulation method and frequency of the effective signal, namely After determining the modulation method and the frequency used by the downhole pulser, the effective mud signal center frequency f 0 in the frequency domain is obtained;
  • B s is the effective mud signal bandwidth in the frequency domain.
  • the frequency resolution of DFT Fourier transform is:
  • T s is the sampling period of the mud signal
  • N is the number of sampling points in the period.
  • a suppression vector is defined according to the sampling period of the mud signal and the number of sampling points in the period:
  • a [a(0) a(1) ... a(S-1)] T , where S is the length of the suppression vector, and the value of S is the same as the value of N.
  • S dn S w ⁇ a
  • S w is the mud signal in the frequency domain or the mud signal in the frequency domain after windowing
  • a is the suppression factor vector
  • S dn is the mud signal in the frequency domain after noise reduction.
  • the denoised frequency domain mud signal S dn is transformed into a time domain denoised mud signal by using an inverse discrete Fourier transform (Inverse Discrete Fourier Transform, IDFT). Mud signal, get the time-domain denoised mud signal s dn :
  • K is the number of sample points intercepted
  • N is the number of sample points in the period.
  • the present disclosure provides a method for eliminating noise in a measurement while drilling MWD system.
  • the method includes: receiving a collected mud signal and determining the characteristic frequency of a periodic noise signal in the mud signal; performing Fourier transform on the mud signal to obtain a frequency domain Mud signal; according to the characteristic frequency of the periodic noise signal in the mud signal, the mud signal in the frequency domain and a preset suppression factor to obtain a noise-removed mud frequency domain output signal.
  • an embodiment of the present disclosure also provides a noise cancellation device for a measurement while drilling MWD system, including: a memory and a processor;
  • the memory is set to save a program for noise elimination of the MWD system
  • the processor is configured to read and execute the program for MWD system noise cancellation, and perform the following operations:
  • a noise-removed mud frequency domain output signal is obtained.
  • receiving the collected mud signal and determining the characteristic frequency of the periodic noise signal in the mud signal includes:
  • the average pumping frequency obtained according to the pumping signal in the pumping signal buffer is determined as the characteristic frequency of the periodic noise signal in the mud signal.
  • receiving the collected mud signal and determining the characteristic frequency of the periodic noise signal in the mud signal includes:
  • the obtained preset frequency of any one or more periodic noise signals is determined as the period in the mud signal The characteristic frequency of the noise signal.
  • receiving the collected mud signal and determining the characteristic frequency of the periodic noise signal in the mud signal includes:
  • the central frequency of the periodic noise signal in the mud signal is determined as the characteristic frequency of the periodic noise signal in the mud signal according to the frequency spectrum of the mud signal.
  • the mud signal is:
  • s in [s in (n) s in (n+1)... s in (n+K-1)] T ; among them, s in (n) represents the mud signal, s in represents the input mud signal vector, and K is The length of the input signal vector, n represents the starting sequence number of the input signal;
  • the processor Before the processor performs Fourier transform on the mud signal to obtain the mud signal in the frequency domain, the processor further performs the following operations:
  • s b [s in (nM)... s in (n-2) s in (n-1) s in ] T
  • N the length of the mud signal buffer
  • p in represents an input signal vector pump stroke
  • p in (n) is the input pump stroke signal
  • K is the length of the input signal vector impulse pump
  • n denotes an input signal of the pump stroke starting sequence number.
  • inputting the pumping signal into the pumping signal buffer includes:
  • p b is the pumping signal in the pumping signal buffer
  • N is the length of the pumping signal buffer
  • N K+M
  • M is the length of the original signal vector in the pumping signal buffer.
  • the determining the average pumping frequency obtained from the pumping signal in the pumping signal buffer as the characteristic frequency of the periodic noise signal in the mud signal includes:
  • the average pumping frequency of the pumping signal in the pumping signal buffer is calculated
  • the formula for calculating the average pumping frequency includes:
  • the performing Fourier transform on the mud signal to obtain the mud signal in the frequency domain includes:
  • Window processing is performed on the mud signal in the mud signal buffer to obtain a windowed mud signal; Fourier transform is performed on the windowed mud signal to obtain a mud signal in the frequency domain.
  • the obtaining the denoised mud frequency domain output signal according to the characteristic frequency of the periodic noise signal in the mud signal, the mud signal in the frequency domain and a preset suppression factor includes:
  • the inhibitory factor includes:
  • ⁇ j is the suppression factor, the value of ⁇ j is a positive real number less than 1, j is the sample point number in the period, S is the number of samples in the period; the suppression factor suppression frequency is The signal component of k is an integer and satisfies Is the characteristic frequency; f 0 is the center frequency of the effective signal of the mud signal in the frequency domain; B s is the bandwidth of the effective signal of the mud signal in the frequency domain; ⁇ f is the frequency resolution of the Fourier transform of the mud signal, T s is the sampling period of the mud signal, and N is the number of sampling points in the period; Indicates rounding down.
  • S w may be a mud signal in the frequency domain or a mud signal in the frequency domain after windowing.
  • the suppression frequency is the signal component of k is an integer and satisfies f 0 is the center frequency of the effective mud signal in the frequency domain.
  • the effective mud signal of this frequency is the signal generated by the downhole pulser and transmitted to the ground acquisition system for demodulation.
  • f 0 is determined by the modulation method and frequency of the effective signal, namely After determining the modulation method and the frequency used by the downhole pulser, the effective mud signal center frequency f 0 in the frequency domain is obtained;
  • B s is the effective mud signal bandwidth in the frequency domain.
  • the frequency resolution of DFT Fourier transform is:
  • T s is the sampling period of the mud signal
  • N is the number of sampling points in the period.
  • a suppression vector is defined according to the sampling period of the mud signal and the number of sampling points in the period:
  • a [a(0) a(1) ... a(S-1)] T ;
  • S is the length of the suppression vector, and the value of S is consistent with the number of sampling points N in the period.
  • the mud signal in the frequency domain is denoised.
  • the calculation formula for denoising is:
  • S dn S w ⁇ a
  • S w is the mud signal in the frequency domain or the mud signal in the frequency domain after windowing
  • a is the suppression factor vector
  • S dn is the mud signal in the frequency domain after noise reduction.
  • the denoised frequency domain mud signal S dn is transformed into a time domain denoised mud signal by using an inverse discrete Fourier transform (Inverse Discrete Fourier Transform, IDFT). Mud signal, get the time-domain denoised mud signal s dn :
  • K is the number of sampling points intercepted, and N is the sampling point in the period.
  • the pulse signal generated by the mud pump has a strong amplitude. When its frequency component is mixed with the mud signal sent downhole, it will cause a strong interference to the useful mud signal, and it is difficult to remove .
  • the periodic noise denoising process for non-collectable periodic signals is shown in Figure 5; the periodic noise denoising process for periodic signals that can be collected is shown in Figure 6.
  • the method for eliminating periodic pump noise is shown in FIG. 7.
  • the method implementation process includes steps 701 to 709:
  • Step 701. Receive the collected mud signal, and obtain the input pumping signal.
  • the amplitude of the mud pump pumping interference is strong, but the pumping interference signal has periodic characteristics.
  • the collected mud signal can be expressed as s in (n), and the input pumping signal can be expressed as p in (n);
  • s in (n) indicates the mud signal
  • s in represents the input mud signal vector
  • p in (n) represents the pump stroke signal
  • p in represents the input pump stroke signal vector
  • K represents the length of the input mud signal and mud signal vector.
  • Step 702. Input the mud signal and the pumping signal into the mud signal buffer and the pumping signal buffer, respectively.
  • s b [s in (nM)... s in (n-2) s in (n-1) s in ] T
  • s b is the mud signal in the mud signal buffer
  • the length of the mud signal buffer is N
  • N K+M
  • M is the length of the original signal vector in the mud signal buffer.
  • p b [p in (nM)... p in (n-2) p in (n-1) p in ] T ;
  • p b is the pump stroke signal buffer pump stroke signal
  • N is the length of the pump stroke average signal buffer
  • p in (n) signal is input pump strokes
  • N K + M
  • M is a pump strokes signal buffering The length of the original signal vector in the device.
  • Step 703 Obtain an average pumping frequency according to the pumping signal in the pumping signal buffer.
  • the average pumping frequency of the pumping signal in the pumping signal buffer is calculated according to the time corresponding to the rising edge of the pumping signal in the pumping signal buffer and the average pumping frequency calculation formula;
  • the formula for calculating the average pumping frequency includes:
  • f p represents the average pumping frequency
  • t k is the time corresponding to the rising edge.
  • Step 704. Perform windowing processing on the mud signal in the mud signal buffer.
  • the mud signal in the mud signal buffer is windowed.
  • the formula for windowing is as follows:
  • s w is the mud signal after windowing
  • w N [w(0) w(1)... w(N-1)] T is the Hanning window function of length N;
  • the window function vector of the Hanning window is:
  • Step 705. Perform Fourier transform on the mud signal in the windowed mud signal buffer to obtain a mud signal in the frequency domain.
  • the mud signal s w in the windowed mud signal buffer is subjected to DFT transformation to obtain the frequency domain mud signal S w ,
  • Step 706 Set the suppression factor and determine the suppression vector.
  • Inhibitors include:
  • ⁇ j is the suppression factor
  • the value of ⁇ j is a positive real number less than 1
  • j is the sample point number in the period
  • S is the number of samples in the frequency domain after Fourier transform
  • the suppression factor suppression frequency is The signal component of k is an integer and satisfies Is the average pumping frequency
  • f 0 is the center frequency of the effective signal of the mud signal in the frequency domain
  • B s is the bandwidth of the effective signal of the mud signal in the frequency domain
  • ⁇ f is the frequency resolution of the Fourier transform of the mud signal:
  • T s is the sampling period of the mud signal, and N is the number of sampling points in the period; Indicates rounding down.
  • the suppression factor define the suppression vector:
  • a [a(0) a(1)... a(S-1)] T , where S is the length of the suppression vector, and the length value of the suppression vector is the same as the value of the number of sampling points N in the period.
  • Step 707 Multiply the frequency domain mud signal by the suppression factor to obtain a noise-eliminated mud frequency domain output signal.
  • the mud signal in the frequency domain is denoised.
  • the calculation formula for denoising is:
  • S dn S w ⁇ a
  • s w the mud signal in the frequency domain or the mud signal in the frequency domain after windowing
  • a the suppression factor
  • S dn the mud signal in the frequency domain after noise reduction.
  • Step 708 (not shown).
  • the denoised frequency domain mud signal S dn is transformed into the time domain using Inverse Discrete Fourier Transform (IDFT) to obtain the time domain denoised mud signal s dn :
  • IDFT Inverse Discrete Fourier Transform
  • Step 709 (not shown). Obtain the denoised output mud signal according to the time domain denoised mud signal.
  • s dn [s dn (0) s dn (1) ... s dn (N-1)] T , intercept the time domain denoised mud signal s dn center K sample points, namely Obtain the output mud signal vector after the pumping interference is eliminated:
  • K is the number of samples taken
  • N is the number of sampling points in one cycle
  • s o is a pump stroke output interference signal vector after cancellation mud.
  • the MWD system it is proposed for the MWD system to eliminate pump noise in the frequency domain, and suppress the periodic component of pumping interference in the frequency domain, so as to obtain a method for eliminating pumping interference in the frequency domain, which can effectively eliminate periodic noise , To obtain a better denoising effect.
  • the noise elimination method of the measurement while drilling MWD system of this exemplary embodiment is aimed at the noise elimination method that can obtain the frequency of the periodic noise signal.
  • the method implementation process includes steps 801 to 809:
  • Step 801. Receive the collected mud signal.
  • Step 802. Input the mud signal into the mud signal buffer.
  • s b [s in (nM)... s in (n-2) s in (n-1) s in ] T
  • s b is the mud signal in the mud signal buffer
  • the length of the mud signal buffer is N
  • N K+M
  • M is the original signal vector length in the mud signal buffer.
  • Step 803. Determine the characteristic frequency of the periodic noise signal in the mud signal.
  • determining the characteristic frequency of the periodic noise signal in the mud signal may include obtaining any one or more preset frequencies of the periodic noise signal, and determining the obtained preset frequency as the mud signal.
  • the characteristic frequency of a periodic noise signal For example: when the periodic noise is top-drive noise, the frequency of the top-drive noise can be directly obtained.
  • the frequency of obtaining the top drive noise is a common technical method used by those skilled in the art, and this is not limited.
  • Step 804. Perform windowing processing on the mud signal in the mud signal buffer.
  • the mud signal in the mud signal buffer is windowed.
  • the formula for windowing is as follows:
  • s w is the mud signal after windowing
  • w N [w(0) w(1)... w(N-1)] T
  • w N is the Hanning window function of length N
  • the window function vector of the Hanning window is:
  • Step 805. Perform Fourier transform on the mud signal in the windowed mud signal buffer to obtain a mud signal in the frequency domain.
  • the mud signal s w in the windowed mud signal buffer is subjected to DFT transformation to obtain the frequency domain mud signal S w ,
  • Step 806. Set the suppression factor and determine the suppression vector.
  • Inhibitors including:
  • ⁇ j is the suppression factor
  • the value of ⁇ j is a positive real number less than 1
  • j is the number of samples in the period
  • S is the number of samples in the frequency domain after Fourier transform
  • the suppression frequency of the suppression factor is The signal component of k is an integer and satisfies Is the preset frequency of the periodic noise signal
  • f 0 is the center frequency of the effective signal of the mud signal in the frequency domain
  • B s is the bandwidth of the effective signal of the mud signal in the frequency domain
  • ⁇ f is the frequency resolution of the mud signal for DFT transformation
  • T s is the sampling period of the mud signal
  • N is the number of sampling points in the period
  • a [a(0) a(1) ... a(S-1)] T , where S is the length of the suppression vector, and the value is consistent with the number of sampling points in the period.
  • Step 807 Multiply the mud signal in the frequency domain by the suppression factor to obtain a noise-eliminated mud frequency domain output signal.
  • the mud signal in the frequency domain is denoised.
  • the calculation formula for denoising is:
  • S dn S w ⁇ a
  • s w the mud signal in the frequency domain or the mud signal in the frequency domain after windowing
  • a the suppression factor
  • S dn the mud signal in the frequency domain after noise reduction.
  • Step 808 Transform the denoised mud signal S dn in the frequency domain into the time domain using Inverse Discrete Fourier Transform (IDFT) to obtain the denoised mud signal s dn in the time domain:
  • IDFT Inverse Discrete Fourier Transform
  • Step 809 Obtain the denoised output mud signal according to the denoised mud signal in the time domain.
  • s dn [s dn (0) s dn (1) ... s dn (N-1)] T , intercept the denoised mud in the time domain
  • K samples in the center of the signal s dn are the signal vectors of the output mud after the pumping interference is eliminated:
  • K is the number of samples taken
  • N is the number of sampling points in one cycle
  • s o is a pump stroke output interference signal vector after cancellation mud.
  • the noise elimination method of the MWD system of the measurement while drilling of this exemplary embodiment includes steps 901 to 909:
  • Step 901. Receive the collected mud signal.
  • Step 902. Input the mud signal into the mud signal buffer.
  • s b [s in (nM)... s in (n-2) s in (n-1) s in ] T
  • N The length of the mud signal buffer
  • N K+M
  • M the central part of the signal buffer
  • Step 903. Determine the characteristic frequency of the periodic noise signal in the mud signal.
  • the central frequency of the periodic noise signal in the mud signal is determined as the characteristic frequency of the periodic noise signal in the mud signal.
  • the central frequency of the periodic noise signal in the mud signal can be used to determine the characteristic frequency of the periodic noise signal in the mud signal.
  • Step 904. Perform windowing processing on the mud signal in the mud signal buffer.
  • the mud signal in the mud signal buffer is windowed.
  • the formula for windowing is as follows:
  • w N [w(0) w(1)... w(N-1)] T
  • w N is the Hanning window function of length N
  • the window function vector of the Hanning window is:
  • Step 905 Perform Fourier transform on the mud signal in the windowed mud signal buffer to obtain a mud signal in the frequency domain.
  • the mud signal s w in the windowed mud signal buffer is subjected to DFT transformation to obtain the frequency domain mud signal S w ,
  • Step 906. Set the suppression factor and determine the suppression vector.
  • Inhibitors including:
  • ⁇ j is the suppression factor
  • the value of ⁇ j is a positive real number less than 1
  • j is the sample number in the period
  • S is the number of samples in the frequency domain after Fourier transform
  • the frequency of the suppression factor is The signal component of k is an integer and satisfies Is the characteristic frequency
  • f 0 is the center frequency of the effective signal of the mud signal in the frequency domain
  • B s is the bandwidth of the effective signal of the mud signal in the frequency domain
  • ⁇ f is the frequency resolution of the mud signal for DFT transformation: T s is the sampling period of the mud signal, and N is the number of sampling points in the period; Indicates rounding down. .
  • a [a(0) a(1) ... a(N-1)] T , where N is the length of the suppression vector.
  • Step 907 Multiply the frequency domain mud signal by the suppression factor to obtain a denoised mud frequency domain output signal.
  • the mud signal in the frequency domain is denoised.
  • the calculation formula for denoising is:
  • S dn S w ⁇ a
  • s w the mud signal in the frequency domain or the mud signal in the frequency domain after windowing
  • a the suppression factor
  • S dn the mud signal in the frequency domain after noise reduction.
  • Step 908 The denoised frequency domain mud signal S dn is transformed into the time domain using Inverse Discrete Fourier Transform (IDFT) to obtain the denoised mud signal s dn in the time domain:
  • IDFT Inverse Discrete Fourier Transform
  • Step 909 Obtain the denoised output mud signal according to the denoised mud signal in the time domain.
  • s dn [s dn (0) s dn (1) ... s dn (N-1)] T , intercept the denoised mud in the time domain
  • K samples in the center of the signal s dn are the signal vectors of the output mud after the pumping interference is eliminated:
  • K is the number of samples taken
  • N is the number of sampling points in one cycle
  • s o is a pump stroke output interference signal vector after cancellation mud.
  • Such software may be distributed on a computer-readable medium, and the computer-readable medium may include a computer storage medium (or a non-transitory medium) and a communication medium (or a transitory medium).
  • the term computer storage medium includes volatile and non-volatile data implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Sexual, removable and non-removable media.
  • Computer storage media include but are not limited to RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disk (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or Any other medium used to store desired information and that can be accessed by a computer.
  • communication media usually contain computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as carrier waves or other transmission mechanisms, and may include any information delivery media. .

Landscapes

  • Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Acoustics & Sound (AREA)
  • Environmental & Geological Engineering (AREA)
  • Geology (AREA)
  • Remote Sensing (AREA)
  • General Life Sciences & Earth Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Geophysics (AREA)
  • Noise Elimination (AREA)

Abstract

一种随钻测量MWD系统噪声消除方法，包括：接收采集的泥浆信号并确定泥浆信号中周期噪声信号的特征频率（301）；对泥浆信号进行傅里叶变换得到频域的泥浆信号（302）；根据泥浆信号中周期噪声信号的特征频率、频域的泥浆信号和预先设置的抑制因子得到消躁后的泥浆频域输出信号（303）。

Description

一种随钻测量MWD系统噪声消除方法、装置及存储介质 技术领域
本公开涉及但不限于测井领域，尤指一种随钻测量MWD系统噪声消除方法、装置及存储介质。
背景技术
随钻测量(MWD，Measurement While Drilling)系统用于在钻井过程中实时测量井下地质参数及工程参数。泥浆脉冲随钻数据解调系统用于将随钻测量系统所测量的数据由井下实时传输至地面。泥浆脉冲随钻数据传输系统利用脉冲器运动产生泥浆截流效应造成泥浆压力波动，将数字信号调制在泥浆压力波中传输至地面，从而实现将井下测量数据实时传输至地面。泥浆脉冲随钻数据传输系统包含三个主要功能模块：泥浆脉冲发生机构、泥浆信号传输信道，压力波采集解调系统。泥浆脉冲发生机构主要为脉冲器，产生的脉冲信号可分为正脉冲、负脉冲、连续波脉冲三大类，泥浆脉冲发生机构的核心原理为对泥浆的截流效应。泥浆信号传输信道包括整个钻井液循环管路，向上包含泥浆泵、管汇、立管、软管多种泥浆流过的接头等，向下包含钻杆、多个测井仪器、流道转换器、导向短节、钻头及井底反射面等，通过该泥浆信号传输通路，泥浆压力波由井下传输至井上，并且每个部分对泥浆压力波造成叠加噪声或衰减特性。压力波采集解调系统包括压力传感器将泥浆压力波转换成电信号，解调系统对电信号进行处理，解调成为有意义的数据。
泥浆信号传输系统中包含多种具有复杂结构及机械运动特性的设备和工具，这些设备和工具将对泥浆脉冲信号造成叠加噪声，影响泥浆脉冲信号质量。这些叠加噪声通常是由特定机械结构产生，主要包含三类，第一类为周期性噪声，如泥浆泵噪声，马达旋转噪声，顶驱、转盘旋转噪声等；第二类为非周期噪声，该非周期噪声主要表现为对信号的衰减，如流道转换器，流道内变径，钻头水眼结构等产生的噪声；第三类为瞬时噪声，该瞬时噪声只在特殊情况下产生，如钻头水眼短暂堵塞等。这些噪声混合在一起产生叠加 噪声共同影响泥浆脉冲信号质量。
一些技术中，针对于泥浆脉冲随钻数据解调系统，采用的消噪方法包括时域消除泵噪方法、频率域消除高频电噪声方法等，上述技术对于周期性噪声干扰不能实现很好的去噪处理，容易使泥浆信号发生畸变。
发明概述
以下是对本文详细描述的主题的概述。本概述并非是为了限制权利要求的保护范围。
本公开实施例提供了一种随钻测量MWD(MWD，Measurement While Drilling)系统噪声消除方法，实现了在频域对周期性噪声进行抑制，可有效消除周期性噪声的干扰。
一方面，本公开实施例提供了一种随钻测量MWD系统噪声消除方法，包括：
接收采集的泥浆信号并确定所述泥浆信号中周期噪声信号的特征频率；
对所述泥浆信号进行傅里叶变换得到频域的泥浆信号；
根据所述泥浆信号中周期噪声信号的特征频率、所述频域的泥浆信号和预先设置的抑制因子得到消噪后的泥浆频域输出信号。
一种示例性的实施例中，所述接收采集的泥浆信号并确定所述泥浆信号中周期噪声信号的特征频率包括：
接收采集的泥浆信号；
当能够获取到所输入的泵冲信号时，将所述泵冲信号输入泵冲信号缓冲器；
将根据所述泵冲信号缓冲器中的泵冲信号所得到的平均泵冲频率，确定为所述泥浆信号中周期噪声信号的特征频率。
一种示例性的实施例中，所述接收采集的泥浆信号并确定所述泥浆信号中周期噪声信号的特征频率包括：
接收采集的泥浆信号；
当无法获取泵冲信号但能获取到任一种或多种周期噪声信号的预置频率时，将所获取的任一种或多种周期噪声信号的预置频率确定为所述泥浆信号中周期噪声信号的特征频率。
一种示例性的实施例中，所述接收采集的泥浆信号并确定所述泥浆信号中周期噪声信号的特征频率，包括：
接收采集的泥浆信号；
当无法获取任一种周期噪声信号时，根据所述泥浆信号的频谱，将所述泥浆信号中的周期噪声信号的中心频率确定为所述泥浆信号中周期噪声信号的特征频率。
一种示例性的实施例中，
所述泥浆信号为：s in＝[s in(n) s in(n+1) … s in(n+K-1)] T；其中，s in(n)表示泥浆信号，s in表示输入泥浆信号向量，K是输入信号向量的长度，n表示输入信号的起始序号；
所述对所述泥浆信号进行傅里叶变换得到频域的泥浆信号之前，方法还包括：
将所述泥浆信号输入泥浆信号缓冲器中，得到泥浆信号缓冲器中的泥浆信号s b：
s b＝[s in(n-M) … s in(n-2) s in(n-1) s in] T
其中，泥浆信号缓冲器的长度为N，N＝K+M，M为泥浆信号缓冲器中原有信号向量的长度。
一种示例性的实施例中，所述泵冲信号为：
p in＝[p in(n) p in(n+1) … p in(n+K-1)] T；
其中，p in表示输入泵冲信号向量，p in(n)为输入泵冲信号，K是输入泵冲信号向量的长度，n表示输入泵冲信号的起始序号。
一种示例性的实施例中，所述当能够获取到所输入的泵冲信号时，将所 述泵冲信号输入泵冲信号缓冲器，包括：
当能够获取到所输入的泵冲信号时，将所述泵冲信号输入泵冲信号缓冲器中，得到泵冲信号缓冲器中的泵冲信号p b：p b＝[p in(n-M) … p in(n-2) p in(n-1) p in] T；
其中，p b为泵冲信号缓冲器中的泵冲信号，N为泵冲信号缓冲器的长度，p in(n)为输入泵冲信号，N＝K+M，M为泵冲信号缓冲器中原有信号向量的长度。
一种示例性的实施例中，所述将根据所述泵冲信号缓冲器中的泵冲信号所得到的平均泵冲频率，确定为所述泥浆信号中周期噪声信号的特征频率，包括：
根据泵冲信号缓冲器中的泵冲信号上升沿所对应的时刻和平均泵冲频率计算公式，计算得到所述泵冲信号缓冲器中的泵冲信号的平均泵冲频率；
将所得到的泵冲信号的平均泵冲频率确定为所述泥浆信号中周期噪声信号的特征频率；
其中，所述平均泵冲频率计算公式，包括：
Figure PCTCN2020083789-appb-000001
Figure PCTCN2020083789-appb-000002
表示泵冲信号的平均泵冲频率，泵冲信号缓冲器中的泵冲信号是包括L+1个上升沿的信号，上升沿所对应的时刻为t k；t k为上升沿所对应的时刻，k＝0,1,2,...,L，L+1表示上升沿的个数。
一种示例性的实施例中，所述对所述泥浆信号进行傅里叶变换得到频域的泥浆信号，包括：
对泥浆信号缓冲器中的泥浆信号进行加窗处理，获得加窗处理后的泥浆信号；
对所述加窗处理后的泥浆信号进行傅里叶变换得到频域的泥浆信号。
一种示例性的实施例中，所述根据所述泥浆信号中周期噪声信号的特征频率、所述频域的泥浆信号和预先设置的抑制因子得到消噪后的泥浆频域输出信号，包括：
根据所述泥浆信号中周期噪声信号的特征频率确定抑制因子的频率；
将频域的泥浆信号与所述抑制因子相乘获得消噪后的泥浆频域输出信号；
其中，所述抑制因子包括：
Figure PCTCN2020083789-appb-000003
α j是抑制因子，α j的值为小于1的正实数，j是周期内的样点序号，S是傅里叶变换后频域内的样点数；抑制因子抑制频率为
Figure PCTCN2020083789-appb-000004
的信号分量，k为整数且满足
Figure PCTCN2020083789-appb-000005
为特征频率；f 0为泥浆信号有效信号的中心频率；B s为频域的泥浆信号有效信号的带宽；Δf为泥浆信号傅里叶变换的频率分辨率，
Figure PCTCN2020083789-appb-000006
T s为泥浆信号的采样周期，N为周期内的采样点数；
Figure PCTCN2020083789-appb-000007
表示向下取整。
另一方面，本公开实施例还提供了一种随钻测量MWD系统噪声消除装置，所述装置包括：存储器和处理器；
所述存储器，设置为保存用于MWD系统噪声消除的程序；
所述处理器，设置为读取执行所述用于MWD系统噪声消除的程序，执行如下操作：
接收采集的泥浆信号并确定所述泥浆信号中周期噪声信号的特征频率；
对所述泥浆信号进行傅里叶变换得到频域的泥浆信号；
根据所述泥浆信号中周期噪声信号的特征频率、所述频域的泥浆信号和预先设置的抑制因子得到消噪后的泥浆频域输出信号。
一种示例性的实施例中，所述接收采集的泥浆信号并确定所述泥浆信号中周期噪声信号的特征频率包括：
接收采集的泥浆信号；
当能够获取到所输入的泵冲信号时，将所述泵冲信号输入泵冲信号缓冲器；
将根据所述泵冲信号缓冲器中的泵冲信号所得到的平均泵冲频率，确定为所述泥浆信号中周期噪声信号的特征频率。
一种示例性的实施例中，所述接收采集的泥浆信号并确定所述泥浆信号中周期噪声信号的特征频率包括：
接收采集的泥浆信号；
当无法获取泵冲信号但能获取到任一种或多种周期噪声信号的预置频率时，将所获取的任一种或多种周期噪声信号的预置频率确定为所述泥浆信号中周期噪声信号的特征频率。
一种示例性的实施例中，所述接收采集的泥浆信号并确定所述泥浆信号中周期噪声信号的特征频率包括：
接收采集的泥浆信号；
当无法获取任一种周期噪声信号时，根据所述泥浆信号的频谱，将所述泥浆信号中的周期噪声信号的中心频率确定为所述泥浆信号中周期噪声信号的特征频率。
一种示例性的实施例中，所述泥浆信号为：
s in＝[s in(n) s in(n+1) … s in(n+K-1)] T；其中，s in(n)表示泥浆信号，s in表示输入泥浆信号向量，K是输入信号向量的长度，n表示输入信号的起始序号；
所述对所述泥浆信号进行傅里叶变换得到频域的泥浆信号之前，所述处 理器还执行以下操作：
将所述泥浆信号输入泥浆信号缓冲器中，得到泥浆信号缓冲器中的泥浆信号s b：
s b＝[s in(n-M) … s in(n-2) s in(n-1) s in] T，
其中，泥浆信号缓冲器的长度为N，N＝K+M，M为泥浆信号缓冲器中原有信号向量的长度。
一种示例性的实施例中，
所述泵冲信号为：p in＝[p in(n) p in(n+1) … p in(n+K-1)] T；
其中，p in表示输入泵冲信号向量，p in(n)为输入泵冲信号，K是输入泵冲信号向量的长度，n表示输入泵冲信号的起始序号。
一种示例性的实施例中，所述当能够获取到所输入的泵冲信号时，将所述泵冲信号输入泵冲信号缓冲器，包括：
当能够获取到所输入的泵冲信号时，将所述泵冲信号输入泵冲信号缓冲器中，得到泵冲信号缓冲器中的泵冲信号p b：p b＝[p in(n-M) … p in(n-2) p in(n-1) p in] T；
其中，p b为泵冲信号缓冲器中的泵冲信号，N为泵冲信号缓冲器的长度均，p in(n)为输入泵冲信号，N＝K+M，M为泵冲信号缓冲器中原有信号向量的长度。
一种示例性的实施例中，所述将根据所述泵冲信号缓冲器中的泵冲信号所得到的平均泵冲频率，确定为所述泥浆信号中周期噪声信号的特征频率，包括：
根据泵冲信号缓冲器中的泵冲信号上升沿所对应的时刻和平均泵冲频率计算公式，计算得到所述泵冲信号缓冲器中的泵冲信号的平均泵冲频率；
将所得到的泵冲信号的平均泵冲频率确定为所述泥浆信号中周期噪声信号的特征频率；
其中，所述平均泵冲频率计算公式，包括：
Figure PCTCN2020083789-appb-000008
Figure PCTCN2020083789-appb-000009
表示平均泵冲频率；泵冲信号缓冲器中的泵冲信号是包括L+1个上升沿的信号，上升沿所对应的时刻为t k；t k为上升沿所对应的时刻，k＝0,1,2,...,L，L+1表示上升沿的个数。
一种示例性的实施例中，所述对所述泥浆信号进行傅里叶变换得到频域的泥浆信号，包括：
对泥浆信号缓冲器中的泥浆信号进行加窗处理，获得加窗处理后的泥浆信号；
对所述加窗处理后的泥浆信号进行傅里叶变换得到频域的泥浆信号。
一种示例性的实施例中，所述根据所述泥浆信号中周期噪声信号的特征频率、所述频域的泥浆信号和预先设置的抑制因子得到消噪后的泥浆频域输出信号，包括：
根据所述泥浆信号中周期噪声信号的特征频率确定抑制因子的频率；
将频域的泥浆信号与所述抑制因子相乘获得消噪后的泥浆频域输出信号；
其中，所述抑制因子包括：
Figure PCTCN2020083789-appb-000010
α j是抑制因子，α j值为小于1的正实数，j是周期内的样点序号，S是傅里叶变换后频域内的样点数；抑制因子抑制频率为
Figure PCTCN2020083789-appb-000011
的信号分量，k为整数且满足
Figure PCTCN2020083789-appb-000012
为特征频率；f 0为泥浆信号有效信号的中心频 率；B s为频域的泥浆信号有效信号的带宽；Δf为泥浆信号傅里叶变换的频率分辨率，
Figure PCTCN2020083789-appb-000013
T s为泥浆信号的采样周期，N为周期内的采样点数；
Figure PCTCN2020083789-appb-000014
表示向下取整；其中，时间域周期内的采样点数N和频域周期内的样点数S两个数值相同。
一种计算机可读存储介质，存储有计算机程序，可执行指令，所述计算机可执行指令被处理器执行时实现所述的MWD系统噪声消除方法。
在阅读并理解了附图和详细描述后，可以明白其他方面。
附图概述
附图用来提供对本申请技术方案的进一步理解，并且构成说明书的一部分，与本申请的实施例一起用于解释本申请的技术方案，并不构成对本申请技术方案的限制。
图1为一种技术中钻井液MWD系统的总体结构示意图；
图2为一种技术中地面接收端信号处理流程图；
图3是本发明实施例的随钻测量MWD系统噪声消除方法流程图；
图4是本发明实施例的随钻测量MWD系统噪声消除装置示意图；
图5是一种技术中不可采集周期信号的周期性噪声消噪流程图；
图6是一种技术中可采集周期信号的周期性噪声消噪流程；
图7是本发明实施例一示例的MWD系统噪声消除方法流程图；
图8a是本发明实施例一示例的地面采集信号的功率谱示意图；
图8b是本发明实施例一示例的泵冲干扰消除后的信号功率谱示意图。
详述
下文中将结合附图对本申请的实施例进行详细说明。在不冲突的情况下， 本公开实施例及实施例中的特征可以相互任意组合。
在附图的流程图示出的步骤可以在诸如一组计算机可执行指令的计算机系统中执行。并且，虽然在流程图中示出了逻辑顺序，但是在某些情况下，可以以不同于此处的顺序执行所示出或描述的步骤。
随钻测量(MWD，Measurement While Drilling)是一种能在钻头钻井过程中测量、采集钻头附近测井数据，并将采集测井数据实时传输到地面系统的技术。测井数据通常包括地层特性信息和多种钻井工程参数。作为一种用于钻井随钻测量中的信息传输技术，钻井液压力信号传输方式的基本工作原理是将井下测得的信息转换成控制信息，并将控制信息作用于井下的钻井液压力信号发生器，使传输信道中的钻井液压力发生变化，从而产生钻井液压力脉冲，压力脉冲通过传输信道中的钻井液传递到地面，经地面处理系统处理而转换成所需的井下测量信息。钻井MWD系统总体结构如图1所示，MWD系统包括：泥浆池、泥浆泵、地面接收单元和井下发送端；泥浆池与泥浆泵相连通，泥浆泵与地面接收相连通。
泥浆泵驱动钻井液循环，井下发送端将数据以钻井液压力脉冲形式发送到地面，地面通过压力传感器将钻井液的压力变化转换为电信号送入地面接收单元，地面接收单元用于解码出井下发送端发送的数据。
地面接收端常用的信号处理流程如图2所示，本公开实施例是关于噪声干扰消除的技术方案，在预处理阶段进行，可以使得处理流程中后面的数据解调和解码可以准确地解出井下发送的数据。
钻井液信号传输特性主要包含信号的传输速度、信号衰减、信号的反射等。但是对于钻井液压力传输系统，压力脉冲信号在钻柱中从井底向井口传输过程当中，由于钻井液属于气、液、固三相流，其间含有粘土、岩屑、重晶石粉等固相物质，还存在着游离状态的气体等气相物质，通过脉冲发生器产生的钻井液压力脉冲信号强度会不断地衰减，衰减程度受信号频率及传输距离的影响，也与钻井液信道的内径、钻井液类型和组分、粘度、体积含气率等内部参数有关。总之，钻井液信道是一个传输特性非常复杂的信道。结合钻井液信道环境十分恶劣，在此信息传输方式中，由于现场测量条件的影 响，安装在立管上的检测泥浆压力波动的压力传感器的输出信号，不仅包含井下传来的泥浆有用信号，而且含有由于泥浆泵压缩泥浆而引起的大幅度周期性压力波动噪声和其他多种机械作用所引起的压力波动噪声以及随机噪声，干扰表现为与泵冲特性相关的周期性脉冲，噪声表现为宽带白噪声，该噪声幅度远大于有用信号幅度，井口处泥浆有用信号完全淹没在多种噪声中。
根据实际测量分析，钻井液噪声信号在时域的统计分布呈正态分布，并且包含有很强的周期分量。该周期分量包括泥浆泵噪声，马达旋转噪声，顶驱、转盘旋转噪声等。
图3是本公开实施例的随钻测量MWD系统噪声消除方法流程图，包括步骤301至303。
步骤301.接收采集的泥浆信号并确定所述泥浆信号中周期噪声信号的特征频率；
步骤302.对所述泥浆信号进行傅里叶变换得到频域的泥浆信号；
步骤303.根据泥浆信号中周期噪声信号的特征频率、频域的泥浆信号和预先设置的抑制因子得到消噪后的泥浆频域输出信号。
通过本公开实施例的技术方案，实现了在频域中消除随钻测量MWD系统的周期噪音。在本实施例中，随钻测量MWD系统随钻测井系统可以用于在钻井过程中实时测量井下多个地质参数，包括：随钻测量井眼轨迹参数，例如：井斜角、方位角、工具面角及辅助参数如温度等。
钻井液随钻数据传输系统的多种干扰和噪声中，泥浆泵产生的脉冲信号幅度较强，当泥浆泵产生的脉冲信号频率成分与井下发送的泥浆信号混叠时会对泥浆有用信号形成干扰。
泥浆信号中周期性噪声可以包括泥浆泵噪声，马达旋转噪声，顶驱，转盘旋转噪声等。
一种示例性的实施例中，所述接收采集的泥浆信号并确定所述泥浆信号中周期噪声信号的特征频率包括：接收采集的泥浆信号；当能够获取到所输入的泵冲信号时，将所述泵冲信号输入泵冲信号缓冲器；将根据所述泵冲信 号缓冲器中的泵冲信号所得到的平均泵冲频率，确定为所述泥浆信号中周期噪声信号的特征频率。
一种示例性的实施例中，泥浆信号为：s in＝[s in(n) s in(n+1) … s in(n+K-1)] T；其中，s in(n)表示泥浆信号，s in表示输入泥浆信号向量，K是输入信号向量的长度，n表示输入信号的起始序号；
一种示例性的实施例中，将所述泥浆信号输入泥浆信号缓冲器中，得到泥浆信号缓冲器中的泥浆信号s b：
s b＝[s in(n-M) … s in(n-2) s in(n-1) s in] T其中，泥浆信号缓冲器的长度为N，N＝K+M，M为泥浆信号缓冲器中原有信号向量的长度。
一种示例性的实施例中，所述泵冲信号为：
p in＝[p in(n) p in(n+1) … p in(n+K-1)] T；
其中，p in表示输入泵冲信号向量，p in(n)为输入泵冲信号，K是输入泵冲信号向量的长度，n表示输入泵冲信号的起始序号。
一种示例性的实施例中，所述当能够获取到所输入的泵冲信号时，将所述泵冲信号输入泵冲信号缓冲器包括：
当能够获取到所输入的泵冲信号时，将所述泵冲信号输入泵冲信号缓冲器中，得到泵冲信号缓冲器中的泵冲信号p b：p b＝[p in(n-M) … p in(n-2) p in(n-1) p in] T；
其中，p b为泵冲信号缓冲器中的泵冲信号，N为泵冲信号缓冲器的长度，p in(n)为输入泵冲信号，N＝K+M，M表示信号缓冲器中原有信号向量长度。
一种示例性的实施例中，所述将根据所述泵冲信号缓冲器中的泵冲信号所得到的平均泵冲频率，确定为所述泥浆信号中周期噪声信号的特征频率包括：
根据泵冲信号缓冲器中的泵冲信号上升沿所对应的时刻和平均泵冲频率计算公式，计算得到所述泵冲信号缓冲器中的泵冲信号的平均泵冲频率；
将所得到的泵冲信号的平均泵冲频率确定为所述泥浆信号中周期噪声信号的特征频率；
其中，所述平均泵冲频率计算公式，包括：
Figure PCTCN2020083789-appb-000015
Figure PCTCN2020083789-appb-000016
表示泵冲信号的平均泵冲频率；泵冲信号缓冲器中的泵冲信号是L+1个上升沿，上升沿所对应的时刻为t k；t k为上升沿所对应的时刻，k＝0,1,2,...,L，L+1表示上升沿的个数。
一种示例性的实施例中，接收采集的泥浆信号并确定所述泥浆信号中周期噪声信号的特征频率，包括：接收采集的泥浆信号；当无法获取泵冲信号但能获取到任一种或多种周期噪声信号的预置频率时，将所获取的任一种或多种周期噪声信号的预置频率确定为所述泥浆信号中周期噪声信号的特征频率。在本实施例中，该周期噪声信号可包括顶驱噪声，该顶驱噪声信号的特征频率可以通过顶驱装置获得顶驱的输入频率，该输入频率为该顶驱的频率，也就是顶驱噪声信号的特征频率，其中，该周期噪声信号包括但不限于顶驱噪声，包括任何一种可以获取到预置频率的周期噪声。
一种示例性的实施例中，接收采集的泥浆信号并确定所述泥浆信号中周期噪声信号的特征频率包括：接收采集的泥浆信号；当无法获取任一种周期噪声信号时，根据所述泥浆信号的频谱，将所述泥浆信号中的周期噪声信号的中心频率确定为所述泥浆信号中周期噪声信号的特征频率。在本实施例中，针对无法获得周期噪声频率的情况。
在步骤302中，将接收的泥浆信号进行傅里叶变换处理得到频域的泥浆信号。
一种示例性的实施例中，泥浆信号为：s in＝[s in(n) s in(n+1) … s in(n+K-1)] T；其中，s in(n)表示泥浆信号，s in表示输入泥浆信号向量，K是输入信号向量的长度，n表示输入信号的起始序号；
对所述泥浆信号进行傅里叶变换得到频域的泥浆信号之前，方法还可包括：
将所述泥浆信号输入泥浆信号缓冲器中，得到泥浆信号缓冲器中的泥浆信号s b：
s b＝[s in(n-M) … s in(n-2) s in(n-1) s in] T其中，泥浆信号缓冲器的长度为N，N＝K+M，M表示信号缓冲器中原有信号向量长度。
一种示例性的实施例中，所述对所述泥浆信号进行傅里叶变换得到频域的泥浆信号包括：对泥浆信号缓冲器中的泥浆信号进行加窗处理，获得加窗处理后的泥浆信号；对所述加窗处理后的泥浆信号进行傅里叶变换得到频域的泥浆信号。在本实施例中，对泥浆信号缓冲器中的泥浆信号s b与预先设置的窗函数进行相乘运算，计算得到加窗处理后的泥浆信号s w，加窗处理的计算公式如下：
s w＝s b·w N，
其中，s w为加窗处理后的泥浆信号，w N为预先设置的窗函数，w N＝[w(0) w(1) … w(N-1)] T，N为窗函数的长度，公式中的“·”表示两个向量逐元素对应相乘。
该预先设置的窗函数可以为汉宁窗，也可以选择其它的窗函数；汉宁窗的窗函数向量为：
Figure PCTCN2020083789-appb-000017
其中，N为窗函数的长度，n为窗函数中采样点的序号。
在本实施例中，加窗处理的窗函数，可以为汉宁窗，也可以选择其它的窗函数，对此并不进行具体限定，可以根据待处理的钻井资料相关信息选择不同的窗函数。
在本实施例中，根据平均泵冲频率、频域的泥浆信号和预先设置的抑制因子得到消噪后的泥浆频域输出信号。关于抑制因子的取值可以设为固定值，也可以根据泵冲干扰的强度和预期达到的抑制效果来选择；抑制因子的取值越小，对相应频率的泵冲干扰分量抑制越明显。
一种示例性的实施例中，所述根据所述平均泵冲频率、频域的泥浆信号 和预先设置的抑制因子得到消噪后的泥浆频域输出信号，包括：
将频域的泥浆信号与抑制因子相乘获得消噪后的泥浆频域输出信号；
其中，所述抑制因子包括：
Figure PCTCN2020083789-appb-000018
其中，α j是抑制因子，α j的值为小于1的正实数，j是周期内的样点数，S是傅里叶变换后频域内的样点数；抑制因子抑制频率为
Figure PCTCN2020083789-appb-000019
的信号分量，k为整数且满足
Figure PCTCN2020083789-appb-000020
为特征频率；f 0为频域的泥浆信号有效信号的中心频率；B s为频域的泥浆信号有效信号的带宽；Δf为泥浆信号进行傅里叶变换的频率分辨率，
Figure PCTCN2020083789-appb-000021
T s为泥浆信号的采样周期，N为周期内的采样点数；
Figure PCTCN2020083789-appb-000022
表示向下取整。在本实施例中，确定抑制因子的实现过程可以如下：对泥浆信号s w进行傅里叶变换处理得到频域泥浆信号S w，其中，傅里叶变换公式如下：S w＝DFT{s w}，该傅里叶变换公式中DFT{·}表示离散傅里叶变换(DiscreteFourierTransform，DFT)运算。S w可为频域的泥浆信号也可以是加窗处理后的频域的泥浆信号。在频域中，对频域的泥浆信号S w中所包含的有效信号带宽内的泵冲干扰分量进行抑制，即抑制频率为
Figure PCTCN2020083789-appb-000023
的信号分量，k为整数且满足
Figure PCTCN2020083789-appb-000024
f 0为频域的泥浆有效信号中心频率，该频率的泥浆有效信号是由井下脉冲器产生、传输至地面采集系统供解 调处理的信号，f 0由有效信号的调制方式及频率决定，即在确定了调制方式及井下脉冲器所用频率后，即获得频域的泥浆有效信号中心频率f 0；B s为频域的泥浆有效信号带宽。DFT傅里叶变换的频率分辨率为：
Figure PCTCN2020083789-appb-000025
其中，T s是泥浆信号的采样周期，N为周期内的采样点数。
在本实施例中，根据泥浆信号的采样周期和周期内的采样点数，定义一个抑制向量：
a＝[a(0) a(1) ... a(S-1)] T，其中，S为抑制向量的长度，S的数值与N的数值相同。在定义抑制因子向量后，对频域的泥浆信号进行消噪处理，消噪处理的计算公式为：
S dn＝S w·a其中，S w为频域的泥浆信号或加窗处理后频域的泥浆信号，a为抑制因子向量；S dn为消噪后的频域的泥浆信号。
一种示例性的实施例中，将消噪后频域的泥浆信号S dn采用逆离散傅里叶变换(InverseDiscreteFourierTransform，IDFT)将消噪后的频域的泥浆信号变换为时域的去噪后泥浆信号，得到时域的去噪后的泥浆信号s dn：
s dn＝IDFT{S dn}
当s dn＝[s dn(0) s dn(1) ... s dn(N-1)] T，截取时域的去噪后的泥浆信号s dn中心K个样点即得到泵冲干扰消除后的输出泥浆的信号向量：
Figure PCTCN2020083789-appb-000026
K为所截取的样点的个数，N为周期内的采样点数。
本公开提供一种随钻测量MWD系统噪声消除方法，方法包括：接收采集的泥浆信号并确定所述泥浆信号中周期噪声信号的特征频率；对所述泥浆信号进行傅里叶变换得到频域的泥浆信号；根据所述泥浆信号中周期噪声信 号的特征频率、所述频域的泥浆信号和预先设置的抑制因子得到消噪后的泥浆频域输出信号。如图4所示，本公开实施例还提供了一种随钻测量MWD系统噪声消除装置，包括：存储器和处理器；
所述存储器，设置为保存用于MWD系统噪声消除的程序；
所述处理器，设置为读取执行所述用于MWD系统噪声消除的程序，执行如下操作：
接收采集的泥浆信号并确定所述泥浆信号中周期噪声信号的特征频率；
对所述泥浆信号进行傅里叶变换得到频域的泥浆信号；
根据所述泥浆信号中周期噪声信号的特征频率、所述频域的泥浆信号和预先设置的抑制因子得到消噪后的泥浆频域输出信号。
一种示例性的实施例中，接收采集的泥浆信号并确定所述泥浆信号中周期噪声信号的特征频率包括：
接收采集的泥浆信号；
当能够获取到所输入的泵冲信号时，将所述泵冲信号输入泵冲信号缓冲器；
将根据所述泵冲信号缓冲器中的泵冲信号所得到的平均泵冲频率，确定为所述泥浆信号中周期噪声信号的特征频率。
一种示例性的实施例中，接收采集的泥浆信号并确定所述泥浆信号中周期噪声信号的特征频率包括：
接收采集的泥浆信号；
当无法获取泵冲信号但能获取到任一种或多种周期噪声信号的预置频率时，将所获取的任一种或多种周期噪声信号的预置频率确定为所述泥浆信号中周期噪声信号的特征频率。
一种示例性的实施例中，接收采集的泥浆信号并确定所述泥浆信号中周期噪声信号的特征频率包括：
接收采集的泥浆信号；
当无法获取任一种周期噪声信号时，根据所述泥浆信号的频谱，将所述泥浆信号中的周期噪声信号的中心频率确定为所述泥浆信号中周期噪声信号的特征频率。
一种示例性的实施例中，所述泥浆信号为：
s in＝[s in(n) s in(n+1) … s in(n+K-1)] T；其中，s in(n)表示泥浆信号，s in表示输入泥浆信号向量，K是输入信号向量的长度，n表示输入信号的起始序号；
处理器对所述泥浆信号进行傅里叶变换得到频域的泥浆信号之前，所述处理器还执行以下操作：
将所述泥浆信号输入泥浆信号缓冲器中，得到泥浆信号缓冲器中的泥浆信号s b：
s b＝[s in(n-M) … s in(n-2) s in(n-1) s in] T其中，泥浆信号缓冲器的长度为N，N＝K+M，M为泥浆信号缓冲器中原有信号向量的长度。
一种示例性的实施例中，
所述泵冲信号为：p in＝[p in(n) p in(n+1) … p in(n+K-1)] T；
其中，p in表示输入泵冲信号向量，p in(n)为输入泵冲信号，K是输入泵冲信号向量的长度，n表示输入泵冲信号的起始序号。
一种示例性的实施例中，当能够获取到所输入的泵冲信号时，将泵冲信号输入泵冲信号缓冲器，包括：
当能够获取到所输入的泵冲信号时，将所述泵冲信号输入泵冲信号缓冲器中，得到泵冲信号缓冲器中的泵冲信号p b：p b＝[p in(n-M) … p in(n-2) p in(n-1) p in] T；
其中，p b为泵冲信号缓冲器中的泵冲信号，N为泵冲信号缓冲器的长度，N＝K+M，M为泵冲信号缓冲器中原有信号向量的长度。
一种示例性的实施例中，所述将根据所述泵冲信号缓冲器中的泵冲信号所得到的平均泵冲频率，确定为所述泥浆信号中周期噪声信号的特征频率， 包括：
根据泵冲信号缓冲器中的泵冲信号上升沿所对应的时刻和平均泵冲频率计算公式，计算得到所述泵冲信号缓冲器中的泵冲信号的平均泵冲频率；
将所得到的泵冲信号的平均泵冲频率确定为所述泥浆信号中周期噪声信号的特征频率；
其中，所述平均泵冲频率计算公式，包括：
Figure PCTCN2020083789-appb-000027
Figure PCTCN2020083789-appb-000028
表示泵冲信号的平均泵冲频率；泵冲信号缓冲器中的泵冲信号是L+1个上升沿，上升沿所对应的时刻为t k；t k为上升沿所对应的时刻，k＝0,1,2,...,L，L+1表示上升沿的个数。
一种示例性的实施例中，所述对所述泥浆信号进行傅里叶变换得到频域的泥浆信号包括：
对泥浆信号缓冲器中的泥浆信号进行加窗处理，获得加窗处理后的泥浆信号；对所述加窗处理后的泥浆信号进行傅里叶变换得到频域的泥浆信号。
一种示例性的实施例中，所述根据所述泥浆信号中周期噪声信号的特征频率、所述频域的泥浆信号和预先设置的抑制因子得到消噪后的泥浆频域输出信号，包括：
根据所述泥浆信号中周期噪声信号的特征频率确定抑制因子的频率；
将频域的泥浆信号与所述抑制因子相乘获得消噪后的泥浆频域输出信号；
其中，所述抑制因子包括：
Figure PCTCN2020083789-appb-000029
α j是抑制因子，α j的值为小于1的正实数，j是周期内的样点序号，S是周期内的样点数；抑制因子抑制频率为
Figure PCTCN2020083789-appb-000030
的信号分量，k为整数且满足
Figure PCTCN2020083789-appb-000031
为特征频率；f 0为频域的泥浆信号有效信号的中心频率；B s为频域的泥浆信号有效信号的带宽；Δf为泥浆信号进行傅里叶变换的频率分辨率，
Figure PCTCN2020083789-appb-000032
T s为泥浆信号的采样周期，N为周期内的采样点数；
Figure PCTCN2020083789-appb-000033
表示向下取整。在本实施例中，确定抑制因子的实现过程可以如下：对泥浆信号s w进行傅里叶变换处理得到频域泥浆信号S w，其中，傅里叶变换公式如下：S w＝DFT{s w}，该傅里叶变换公式中DFT{·}表示离散傅里叶变换(Discrete Fourier Transform，DFT)运算。S w可为频域的泥浆信号也可以是加窗处理后的频域的泥浆信号。在频域中，对频域的泥浆信号S w中所包含的有效信号带宽内的泵冲干扰分量进行抑制，即抑制频率为
Figure PCTCN2020083789-appb-000034
的信号分量，k为整数且满足
Figure PCTCN2020083789-appb-000035
f 0为频域的泥浆有效信号中心频率，该频率的泥浆有效信号是由井下脉冲器产生、传输至地面采集系统供解调处理的信号，f 0由有效信号的调制方式及频率决定，即在确定了调制方式及井下脉冲器所用频率后，即获得频域的泥浆有效信号中心频率f 0；B s为频域的泥浆有效信号带宽。DFT傅里叶变换的频率分辨率为：
Figure PCTCN2020083789-appb-000036
其中，T s是泥浆信号的采样周期，N为周期内的采样点数。
在本实施例中，根据泥浆信号的采样周期和周期内的采样点数，定义一个抑制向量：
a＝[a(0) a(1) ... a(S-1)] T；其中，S为抑制向量的长度，S的数值与周期内的采样点数N一致。
在定义抑制因子向量后，对频域的泥浆信号进行消噪处理，消噪处理的计算公式为：
S dn＝S w·a其中，S w为频域的泥浆信号或加窗处理后频域的泥浆信号，a为抑制因子向量；S dn为消噪后的频域的泥浆信号。
一种示例性的实施例中，将消噪后频域的泥浆信号S dn采用逆离散傅里叶变换(InverseDiscreteFourierTransform，IDFT)将消噪后的频域的泥浆信号变换为时域的去噪后泥浆信号，得到时域的去噪后的泥浆信号s dn：
s dn＝IDFT{S dn}
当s dn＝[s dn(0) s dn(1) ... s dn(N-1)] T，截取时域的去噪后的泥浆信号s dn中心K个样点即得到泵冲干扰消除后的输出泥浆的信号向量：
Figure PCTCN2020083789-appb-000037
K为所截取样点的个数，N为周期内的采样点。一种示例性实施例
随钻测量传输系统的多种干扰和噪声中，泥浆泵产生的脉冲信号幅度较强，当其频率成分与井下发送的泥浆信号混叠时会对泥浆有用信号形成较强的干扰，且难以去除。一些技术中，对于不可采集周期信号的周期性噪声消噪流程如图5所示；可采集周期信号的周期性噪声消噪流程如图6所示。在本示例性实施例针对于周期性泵噪消除方法如图7所示，方法实现过程包括步骤701至709：
步骤701.接收采集的泥浆信号，获取输入的泵冲信号。
在本步骤中，泥浆泵泵冲干扰幅度较强，但泵冲干扰信号具有周期特性。可以将采集的泥浆信号表示为s in(n)，输入的泵冲信号表示为p in(n)；
采集的泥浆信号为：s in＝[s in(n) s in(n+1) … s in(n+K-1)] T；
输入的泵冲信号为：p in＝[p in(n) p in(n+1) … p in(n+K-1)] T；
其中，s in(n)表示泥浆信号，s in表示输入泥浆信号向量，p in(n)表示泵冲信号，p in表示输入泵冲信号向量，K表示输入泥浆信号和泥浆信号向量的长度。
步骤702.将泥浆信号和泵冲信号分别输入泥浆信号缓冲器和泵冲信号缓冲器。
在本步骤中，将所述泥浆信号输入泥浆信号缓冲器中，得到泥浆信号缓冲器中的泥浆信号s b：
s b＝[s in(n-M) … s in(n-2) s in(n-1) s in] T其中，s b为泥浆信号缓冲器中的泥浆信号，泥浆信号缓冲器的长度为N，N＝K+M，M为泥浆信号缓冲器中原有信号向量的长度。将所述泵冲信号输入泵冲信号缓冲器中，得到泵冲信号缓冲器中的泵冲信号p b：
p b＝[p in(n-M) … p in(n-2) p in(n-1) p in] T；
其中，p b为泵冲信号缓冲器中的泵冲信号，N为泵冲信号缓冲器的长度均，p in(n)为输入泵冲信号，N＝K+M，M为泵冲信号缓冲器中原有信号向量的长度。
步骤703.根据所述泵冲信号缓冲器中的泵冲信号得到平均泵冲频率。
在本步骤中，根据泵冲信号缓冲器中的泵冲信号上升沿所对应的时刻和平均泵冲频率计算公式，计算得到所述泵冲信号缓冲器中泵冲信号的平均泵冲频率；
其中，所述平均泵冲频率计算公式包括：
Figure PCTCN2020083789-appb-000038
泵冲信号缓冲器中泵冲信号是包括L+1个上升沿的信号，上升沿所对应的时刻为t k,k＝0,1,2,...,L；
其中，f p表示平均泵冲频率；t k为上升沿所对应的时刻。
步骤704.对所述泥浆信号缓冲器中的泥浆信号进行加窗处理。
在本步骤中，对于泥浆信号缓冲器中的泥浆信号进行加窗处理，加窗处理的公式如下：
s w＝s b·w N，
其中，s w为加窗处理后的泥浆信号，w N＝[w(0) w(1) … w(N-1)] T为长度为N的汉宁窗窗函数；
该汉宁窗的窗函数向量为：
Figure PCTCN2020083789-appb-000039
其中，s w＝s b·w N公式中的“·”表示两个向量逐元素对应相乘。步骤705.对加窗处理后的泥浆信号缓冲器中的泥浆信号进行傅里叶变换得到频域的泥浆信号。
在本步骤中，对加窗处理后的泥浆信号缓冲器中的泥浆信号s w做DFT变换得到频域泥浆信号S w，
S w＝DFT{s w}，这里DFT{·}表示傅里叶变换DFT运算。
步骤706.设置抑制因子，确定抑制向量。
抑制因子包括：
Figure PCTCN2020083789-appb-000040
其中，α j是抑制因子，α j值为小于1的正实数，j是周期内的样点序号，S是傅里叶变换后频域内的样点数；抑制因子抑制频率为
Figure PCTCN2020083789-appb-000041
的信号分量，k为整数且满足
Figure PCTCN2020083789-appb-000042
为平均泵冲频率；f 0为频域的泥浆信号有效信号的中心频率；B s为频域的泥浆信号有效信号的带宽；Δf为泥浆信号进行傅里叶变换的频率分辨率：
Figure PCTCN2020083789-appb-000043
T s为泥浆信号的采样周期，N为周期内的采样点数；
Figure PCTCN2020083789-appb-000044
表示向下取整。根据所设置的抑制因子，定义抑制向量：
a＝[a(0) a(1) ... a(S-1)] T，其中，S为抑制向量的长度，该抑制向量的长度数值与周期内的采样点数N的数值相同。
步骤707.将频域的泥浆信号与抑制因子相乘获得消噪后的泥浆频域输出信号。
在本步骤中，在定义抑制因子后，对频域的泥浆信号进行消噪处理，消噪处理的计算公式为：
S dn＝S w·a其中，s w为频域的泥浆信号或加窗处理后的频域的泥浆信号，a为抑制因子；S dn为消噪后的频域的泥浆信号。
步骤708(未示出).将消噪后的频域的泥浆信号S dn采用逆离散傅里叶变换(Inverse Discrete Fourier Transform，IDFT)变换为时域，得到时域的去噪后的泥浆信号s dn：
s dn＝IDFT{S dn}
步骤709(未示出).根据时域的去噪后的泥浆信号得到去噪后的输出泥浆的信号。
在本步骤中，s dn＝[s dn(0) s dn(1) ... s dn(N-1)] T，截取时域的去噪后的泥浆信号s dn中心K个样点即得到泵冲干扰消除后的输出泥浆的信号向量：
Figure PCTCN2020083789-appb-000045
K为所截取的样点的个数，N为周期内的采样点数，s o为泵冲干扰消除后的输出泥浆的信号向量。采用本实施例的消噪处理流程，在频域对泥浆信号的消噪效果对比展示如图8a和图8b所示。示例中泥浆信号的有效信号是4bps的2FSK调制信号，两个载波频率分别为4Hz和8Hz，信号频带范围为2Hz～10Hz。从图8a地面采集信号的功率谱中可以看出，地面采集的泥浆信号包含大量泵冲干扰的分量。从图8b泵冲干扰消除后的信号功率谱中可以看到，消噪后的泥浆信号在2Hz～10Hz范围内泵冲干扰分量已经被抑制，只剩下4Hz和8Hz的载波分量。
本实施例中，对MWD系统提出了在频域中进行泵噪消除，在频域对泵冲干扰的周期性分量进行抑制，从而获得频率域消除泵冲干扰的方法，可以有效消除周期性噪声，获得较好去噪效果。
一种示例性实施例
本示例性实施例随钻测量MWD系统噪声消除方法，针对可以获取到周期噪声信号的频率的噪声消除方法，方法实现过程包括步骤801至809：
步骤801.接收采集的泥浆信号。
在本步骤中，可以将采集的泥浆信号为s in(n)，采集的泥浆信号为：s in＝[s in(n) s in(n+1) … s in(n+K-1)] T；其中，s in(n)表示泥浆信号，s in表示输入泥浆信号向量，K是输入信号向量的长度，n表示输入信号的起始序号。
步骤802.将所述泥浆信号输入泥浆信号缓冲器。
在本步骤中，将所述泥浆信号输入泥浆信号缓冲器中，得到泥浆信号缓冲器中的泥浆信号s b：
s b＝[s in(n-M) … s in(n-2) s in(n-1) s in] T其中，s b为泥浆信号缓冲器中的泥浆信号，泥浆信号缓冲器的长度为N，N＝K+M，M为泥浆信号缓冲器中原有信号向量长度。
步骤803.确定所述泥浆信号中周期噪声信号的特征频率。
在本步骤中，确定所述泥浆信号中周期噪声信号的特征频率可包括获取的任一种或多种周期噪声信号的预置频率，并将所获取的预置频率确定为所述泥浆信号中周期噪声信号的特征频率。例如：当周期噪声为顶驱噪声时，该顶驱噪声的频率是可以直接获取到的。
关于获得顶驱噪声的频率是本领域技术人员常用技术手段，对此并不进行限定。
步骤804.对所述泥浆信号缓冲器中的泥浆信号进行加窗处理。
在本步骤中，对于泥浆信号缓冲器中的泥浆信号进行加窗处理，加窗处理的公式如下：
s w＝s b·w N，
其中，s w为加窗处理后的泥浆信号，w N＝[w(0) w(1) … w(N-1)] T，w N是长度为N的汉宁窗窗函数，
该汉宁窗的窗函数向量为：
Figure PCTCN2020083789-appb-000046
其中，s w＝s b·w N公式中的“·”表示两个向量逐元素对应相乘。步骤805.对加窗处理后的泥浆信号缓冲器中的泥浆信号进行傅里叶变换得到频域的泥浆信号。
在本步骤中，对加窗处理后的泥浆信号缓冲器中的泥浆信号s w做DFT变 换得到频域泥浆信号S w，
S w＝DFT{s w}，这里DFT{·}表示DFT运算。
步骤806.设置抑制因子，确定抑制向量。
抑制因子，包括：
Figure PCTCN2020083789-appb-000047
其中，α j是抑制因子，α j值为小于1的正实数，j是周期内的样点数，S是傅里叶变换后频域内的样点数；抑制因子抑制频率为
Figure PCTCN2020083789-appb-000048
的信号分量，k为整数且满足
Figure PCTCN2020083789-appb-000049
为周期噪声信号的预置频率；f 0为频域的泥浆信号有效信号的中心频率；B s为频域的泥浆信号有效信号的带宽；Δf为泥浆信号进行DFT变换的频率分辨率，
Figure PCTCN2020083789-appb-000050
T s为泥浆信号的采样周期，N为周期内的采样点数；
Figure PCTCN2020083789-appb-000051
表示向下取整。
根据所设置的抑制因子，定义抑制向量：
a＝[a(0) a(1) ... a(S-1)] T，其中，S为抑制向量的长度，数值与周期内的采样点数一致。
步骤807.将所频域的泥浆信号与抑制因子相乘获得消噪后的泥浆频域输出信号。
在本步骤中，在定义抑制因子后，对频域的泥浆信号进行消噪处理，消噪处理的计算公式为：
S dn＝S w·a其中，s w为频域的泥浆信号或加 窗处理后的频域的泥浆信号，a为抑制因子；S dn为消噪后的频域的泥浆信号。
步骤808.将消噪后的频域的泥浆信号S dn采用逆离散傅里叶变换(Inverse Discrete Fourier Transform，IDFT)变换为时域，得到时域的去噪后的泥浆信号s dn：
s dn＝IDFT{S dn}
步骤809.根据时域的去噪后的泥浆信号得到去噪后的输出泥浆的信号。
在本步骤中，s dn＝[s dn(0) s dn(1) ... s dn(N-1)] T，截取时域的去噪后的泥浆
信号s dn中心K个样点即得到泵冲干扰消除后的输出泥浆的信号向量：
Figure PCTCN2020083789-appb-000052
K为所截取的样点的个数，N为周期内的采样点数，s o为泵冲干扰消除后的输出泥浆的信号向量。本实施例中，对MWD系统提出了在频域中进行噪声消除，在频域对周期性分量进行抑制，从而达到消除周期性噪声干扰的目，能够取得较好去噪效果。
一种示例性实施例
本示例性实施例随钻测量MWD系统噪声消除方法，方法实现过程包括步骤901至909：
步骤901.接收采集的泥浆信号。
在本步骤中，可以将采集的泥浆信号为s in(n)，采集的泥浆信号为：s in＝[s in(n) s in(n+1) … s in(n+K-1)] T；其中，s in(n)表示泥浆信号，s in表示输入泥浆信号向量，K是输入信号向量的长度，n表示输入信号的起始序号。
步骤902.将所述泥浆信号输入泥浆信号缓冲器。
在本步骤中，将所述泥浆信号输入泥浆信号缓冲器中，得到泥浆信号缓冲器中的泥浆信号s b：
s b＝[s in(n-M) … s in(n-2) s in(n-1) s in] T泥浆信号缓冲器的长度均为N，N＝K+M，M表示信号缓冲器中原有信号向量长度。
步骤903.确定泥浆信号中周期噪声信号的特征频率。
在本步骤中，将泥浆信号中周期噪声信号的中心频率确定为所述泥浆信号中周期噪声信号的特征频率。
针对于无法确定噪声的频率和无法采集到噪声信号的情况，可以利用泥浆信号中周期噪声信号的中心频率确定为所述泥浆信号中周期噪声信号的特征频率。
步骤904.对泥浆信号缓冲器中的泥浆信号进行加窗处理。
在本步骤中，对于泥浆信号缓冲器中的泥浆信号进行加窗处理，加窗处理的公式如下：
s w＝s b·w N，
其中，w N＝[w(0) w(1) … w(N-1)] T，w N是长度为N的汉宁窗窗函数，
该汉宁窗的窗函数向量为：
Figure PCTCN2020083789-appb-000053
其中，s w＝s b·w N公式中的“·”表示两个向量逐元素对应相乘。步骤905.对加窗处理后的泥浆信号缓冲器中的泥浆信号进行傅里叶变换得到频域的泥浆信号。
在本步骤中，对加窗处理后的泥浆信号缓冲器中的泥浆信号s w做DFT变换得到频域泥浆信号S w，
S w＝DFT{s w}，这里DFT{·}表示DFT运算。
步骤906.设置抑制因子，确定抑制向量。
抑制因子，包括：
Figure PCTCN2020083789-appb-000054
其中，α j是抑制因子，α j值为小于1的正实数，j是周期内的样点序号，S是傅里叶变换后频域内的样点数；抑制因子的频率为
Figure PCTCN2020083789-appb-000055
的信号分量，k为整数且满足
Figure PCTCN2020083789-appb-000056
为特征频率；f 0为频域的泥浆信号有效信号的中心频率；B s为频域的泥浆信号有效信号的带宽；Δf为泥浆信号进行DFT变换的频率分辨率：
Figure PCTCN2020083789-appb-000057
T s为泥浆信号的采样周期，N为周期内的采样点数；
Figure PCTCN2020083789-appb-000058
表示向下取整。。
根据所设置的抑制因子，定义抑制向量：
a＝[a(0) a(1) ... a(N-1)] T，其中，N为抑制向量的长度。
步骤907.将频域的泥浆信号与抑制因子相乘获得消噪后的泥浆频域输出信号。
在本步骤中，在定义抑制因子后，对频域的泥浆信号进行消噪处理，消噪处理的计算公式为：
S dn＝S w·a其中，s w为频域的泥浆信号或加窗处理后的频域的泥浆信号，a为抑制因子；S dn为消噪后的频域的泥浆信号。
步骤908.将消噪后的频域的泥浆信号S dn采用逆离散傅里叶变换(Inverse Discrete Fourier Transform，IDFT)变换为时域，得到时域的去噪后的泥浆信号s dn：
s dn＝IDFT{S dn}
步骤909.根据时域的去噪后的泥浆信号得到去噪后的输出泥浆的信号。
在本步骤中，s dn＝[s dn(0) s dn(1) ... s dn(N-1)] T，截取时域的去噪后的泥浆
信号s dn中心K个样点即得到泵冲干扰消除后的输出泥浆的信号向量：
Figure PCTCN2020083789-appb-000059
K为所截取的样点的个数，N为周期内的采样点数，s o为泵冲干扰消除后的输出泥浆的信号向量。本实施例中，对MWD系统提出了在频域中进行噪声消除，在频域对周期性分量进行抑制，从而达到消除噪声干扰的目，能够取得较好去噪效果。
本领域普通技术人员可以理解，上文中所公开方法中的全部或某些步骤、系统、装置中的功能模块/单元可以被实施为软件、固件、硬件及其适当的组合。在硬件实施方式中，在以上描述中提及的功能模块/单元之间的划分不一定对应于物理组件的划分；例如，一个物理组件可以具有多个功能，或者一个功能或步骤可以由若干物理组件合作执行。某些组件或所有组件可以被实施为由处理器，如数字信号处理器或微处理器执行的软件，或者被实施为硬件，或者被实施为集成电路，如专用集成电路。这样的软件可以分布在计算机可读介质上，计算机可读介质可以包括计算机存储介质(或非暂时性介质)和通信介质(或暂时性介质)。如本领域普通技术人员公知的，术语计算机存储介质包括在用于存储信息(诸如计算机可读指令、数据结构、程序模块或其他数据)的任何方法或技术中实施的易失性和非易失性、可移除和不可移除介质。计算机存储介质包括但不限于RAM、ROM、EEPROM、闪存或其他存储器技术、CD-ROM、数字多功能盘(DVD)或其他光盘存储、磁盒、磁带、磁盘存储或其他磁存储装置、或者可以用于存储期望的信息并且可以被计算机访问的任何其他的介质。此外，本领域普通技术人员公知的是，通信介质通常包含计算机可读指令、数据结构、程序模块或者诸如载波或其他传输机制之类的调制数据信号中的其他数据，并且可包括任何信息递送介质。

Claims (15)

  1. 一种随钻测量MWD系统噪声消除方法，包括：
    接收采集的泥浆信号并确定所述泥浆信号中周期噪声信号的特征频率；
    对所述泥浆信号进行傅里叶变换得到频域的泥浆信号；
    根据所述泥浆信号中周期噪声信号的特征频率、所述频域的泥浆信号和预先设置的抑制因子得到消噪后的泥浆频域输出信号。
  2. 根据权利要求1所述的MWD系统噪声消除方法，其中，所述接收采集的泥浆信号并确定所述泥浆信号中周期噪声信号的特征频率包括：
    接收采集的泥浆信号；
    当能够获取到所输入的泵冲信号时，将所述泵冲信号输入泵冲信号缓冲器；
    将根据所述泵冲信号缓冲器中的泵冲信号所得到的平均泵冲频率，确定为所述泥浆信号中周期噪声信号的特征频率。
  3. 根据权利要求1所述的MWD系统噪声消除方法，
    所述泥浆信号为：s in＝[s in(n) s in(n+1) … s in(n+K-1)] T；其中，s in(n)表示泥浆信号，s in表示输入泥浆信号向量，K是输入信号向量的长度，n表示输入信号的起始序号；
    所述对所述泥浆信号进行傅里叶变换得到频域的泥浆信号之前，还包括：
    将所述泥浆信号输入泥浆信号缓冲器中，得到泥浆信号缓冲器中的泥浆信号s b：
    s b＝[s in(n-M) … s in(n-2) s in(n-1) s in] T其中，泥浆信号缓冲器的长度为N，N＝K+M，M为泥浆信号缓冲器中原有信号向量的长度。
  4. 根据权利要求2所述的MWD系统噪声消除方法，
    所述泵冲信号为：p in＝[p in(n) p in(n+1) … p in(n+K-1)] T；
    其中，p in表示输入泵冲信号向量，p in(n)为输入泵冲信号，K是输入泵冲信号向量的长度，n表示输入泵冲信号的起始序号。
  5. 根据权利要求4所述的MWD系统噪声消除方法，其中，所述当能够获取到所输入的泵冲信号时，将所述泵冲信号输入泵冲信号缓冲器，包括：
    当能够获取到所输入的泵冲信号时，将所述泵冲信号输入泵冲信号缓冲器中，得到泵冲信号缓冲器中的泵冲信号p b：
    p b＝[p in(n-M) … p in(n-2) p in(n-1) p in] T；
    其中，p b为泵冲信号缓冲器中的泵冲信号，N为泵冲信号缓冲器的长度，N＝K+M，M为泵冲信号缓冲器中原有信号向量的长度。
  6. 根据权利要求5所述的MWD系统噪声消除方法，其中，所述将根据所述泵冲信号缓冲器中的泵冲信号所得到的平均泵冲频率，确定为所述泥浆信号中周期噪声信号的特征频率，包括：
    根据泵冲信号缓冲器中的泵冲信号上升沿所对应的时刻和平均泵冲频率计算公式，计算得到所述泵冲信号缓冲器中的泵冲信号的平均泵冲频率；
    将所得到的泵冲信号的平均泵冲频率确定为所述泥浆信号中周期噪声信号的特征频率；
    其中，所述平均泵冲频率计算公式，包括：
    Figure PCTCN2020083789-appb-100001
    Figure PCTCN2020083789-appb-100002
    表示泵冲信号的平均泵冲频率，泵冲信号缓冲器中的泵冲信号是包括L+1个上升沿的信号，上升沿所对应的时刻为t k；t k为上升沿所对应的时刻，k＝0,1,2,...,L，L+1表示上升沿的个数。
  7. 根据权利要求1所述的MWD系统噪声消除方法，其中，所述根据所述泥浆信号中周期噪声信号的特征频率、所述频域的泥浆信号和预先设置的抑制因子得到消噪后的泥浆频域输出信号包括：
    根据所述泥浆信号中周期噪声信号的特征频率确定抑制因子的频率；
    将频域的泥浆信号与所述抑制因子相乘获得消噪后的泥浆频域输出信号；
    其中，所述抑制因子包括：
    Figure PCTCN2020083789-appb-100003
    α j是抑制因子，α j的值为小于1的正实数，j是周期内的样点序号，S是傅里叶变换后频域内的样点数；抑制因子抑制频率为
    Figure PCTCN2020083789-appb-100004
    的信号分量，k为整数且满足
    Figure PCTCN2020083789-appb-100005
    为特征频率；f 0为泥浆信号有效信号的中心频率；B s为频域的泥浆信号有效信号的带宽；Δf为泥浆信号傅里叶变换的频率分辨率，
    Figure PCTCN2020083789-appb-100006
    T s为泥浆信号的采样周期，N为周期内的采样点数；
    Figure PCTCN2020083789-appb-100007
    表示向下取整。
  8. 一种MWD系统噪声消除装置，存储器和处理器；
    所述存储器，设置为保存用于MWD系统噪声消除的程序；
    所述处理器，设置为读取执行所述用于MWD系统噪声消除的程序，执行如下操作：
    接收采集的泥浆信号并确定所述泥浆信号中周期噪声信号的特征频率；
    对所述泥浆信号进行傅里叶变换得到频域的泥浆信号；
    根据所述泥浆信号中周期噪声信号的特征频率、所述频域的泥浆信号和预先设置的抑制因子得到消噪后的泥浆频域输出信号。
  9. 根据权利要求8所述的MWD系统噪声消除装置，其中，所述接收采 集的泥浆信号并确定所述泥浆信号中周期噪声信号的特征频率包括：
    接收采集的泥浆信号；
    当能够获取到所输入的泵冲信号时，将所述泵冲信号输入泵冲信号缓冲器；
    将根据所述泵冲信号缓冲器中的泵冲信号所得到的平均泵冲频率，确定为所述泥浆信号中周期噪声信号的特征频率。
  10. 根据权利要求8所述的MWD系统噪声消除装置，
    所述泥浆信号为：s in＝[s in(n) s in(n+1) … s in(n+K-1)] T；其中，s in(n)表示泥浆信号，s in表示输入泥浆信号向量，K是输入信号向量的长度，n表示输入信号的起始序号；
    所述对所述泥浆信号进行傅里叶变换得到频域的泥浆信号之前，所述处理器还执行以下操作：
    将所述泥浆信号输入泥浆信号缓冲器中，得到泥浆信号缓冲器中的泥浆信号s b：
    s b＝[s in(n-M) … s in(n-2) s in(n-1) s in] T其中，泥浆信号缓冲器的长度为N，N＝K+M，M为泥浆信号缓冲器中原有信号向量的长度。
  11. 根据权利要求9所述的MWD系统噪声消除装置，
    所述泵冲信号为：p in＝[p in(n) p in(n+1) … p in(n+K-1)] T；
    其中，p in表示输入泵冲信号向量，p in(n)为输入泵冲信号，K是输入泵冲信号向量的长度，n表示输入泵冲信号的起始序号。
  12. 根据权利要求11所述的MWD系统噪声消除装置，所述当能够获取到所输入的泵冲信号时，将所述泵冲信号输入泵冲信号缓冲器，包括：
    当能够获取到所输入的泵冲信号时，将所述泵冲信号输入泵冲信号缓冲器中，得到泵冲信号缓冲器中的泵冲信号p b：
    p b＝[p in(n-M) … p in(n-2) p in(n-1) p in] T；
    其中，p b为泵冲信号缓冲器中的泵冲信号，N为泵冲信号缓冲器的长度均，p in(n)为输入泵冲信号，N＝K+M，M为泵冲信号缓冲器中原有信号向量的长度。
  13. 根据权利要求12所述的MWD系统噪声消除装置，其中，所述将根据所述泵冲信号缓冲器中的泵冲信号所得到的平均泵冲频率，确定为所述泥浆信号中周期噪声信号的特征频率，包括：
    根据泵冲信号缓冲器中的泵冲信号上升沿所对应的时刻和平均泵冲频率计算公式，计算得到所述泵冲信号缓冲器中的泵冲信号的平均泵冲频率；
    将所得到的泵冲信号的平均泵冲频率确定为所述泥浆信号中周期噪声信号的特征频率；
    其中，所述平均泵冲频率计算公式，包括：
    Figure PCTCN2020083789-appb-100008
    Figure PCTCN2020083789-appb-100009
    表示泵冲信号的平均泵冲频率；泵冲信号缓冲器中的泵冲信号是L+1个上升沿的信号，上升沿所对应的时刻为t k；t k为上升沿所对应的时刻，k＝0,1,2,...,L，L+1表示上升沿的个数。
  14. 根据权利要求9所述的MWD系统噪声消除装置，其中，所述根据所述泥浆信号中周期噪声信号的特征频率、所述频域的泥浆信号和预先设置的抑制因子得到消噪后的泥浆频域输出信号，包括：
    根据所述泥浆信号中周期噪声信号的特征频率确定抑制因子的频率；
    将频域的泥浆信号与所述抑制因子相乘获得消噪后的泥浆频域输出信号；
    其中，所述抑制因子包括：
    Figure PCTCN2020083789-appb-100010
    α j是抑制因子，α j的值为小于1的正实数，j是周期内的样点序号，S是傅里叶变换后频域内的样点数；抑制因子抑制频率为
    Figure PCTCN2020083789-appb-100011
    的信号分量，k为整数且满足
    Figure PCTCN2020083789-appb-100012
    为特征频率；f 0为泥浆信号有效信号的中心频率；B s为频域的泥浆信号有效信号的带宽；Δf为泥浆信号傅里叶变换的频率分辨率
    Figure PCTCN2020083789-appb-100013
    T s为泥浆信号的采样周期，N为周期内的采样点数；
    Figure PCTCN2020083789-appb-100014
    表示向下取整。
  15. 一种计算机可读存储介质，存储有计算机可执行指令，所述计算机可执行指令被处理器执行时实现如权利要求1-7任一项所述的MWD系统噪声消除方法的步骤。
PCT/CN2020/083789 2019-12-18 2020-04-08 一种随钻测量mwd系统噪声消除方法、装置及存储介质 WO2021120454A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201911309266.3A CN111308560B (zh) 2019-12-18 2019-12-18 一种随钻测量mwd系统噪声消除方法和装置
CN201911309266.3 2019-12-18

Publications (1)

Publication Number Publication Date
WO2021120454A1 true WO2021120454A1 (zh) 2021-06-24

Family

ID=71156253

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/083789 WO2021120454A1 (zh) 2019-12-18 2020-04-08 一种随钻测量mwd系统噪声消除方法、装置及存储介质

Country Status (2)

Country Link
CN (1) CN111308560B (zh)
WO (1) WO2021120454A1 (zh)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11940586B2 (en) 2020-12-16 2024-03-26 Halliburton Energy Services, Inc. Noise elimination or reduction in drilling operation measurements using machine learning
CN114088195B (zh) * 2021-11-17 2024-04-02 西安石油大学 一种钻井井场噪声的分析方法、采集装置、电子设备及介质

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020180613A1 (en) * 2000-05-08 2002-12-05 Pengyu Shi Digital signal receiver for measurement while drilling system having noise cancellation
CN102857194A (zh) * 2012-09-16 2013-01-02 中国石油大学(华东) 钻井液连续压力波信号的频带噪声消除方法
CN104133982A (zh) * 2014-06-27 2014-11-05 中天启明石油技术有限公司 一种泥浆脉冲信号泵冲噪声的消除方法
CN104265278A (zh) * 2014-07-30 2015-01-07 中天启明石油技术有限公司 一种利用回音抵消技术消除随钻测井中的泵冲噪声的方法
CN105041303A (zh) * 2015-07-27 2015-11-11 电子科技大学 钻井液随钻数据传输系统的泵冲干扰信号消除方法
CN106089188A (zh) * 2016-06-02 2016-11-09 中国石油大学(华东) 一种泥浆脉冲信号泵噪声实时去除方法
CN106301289A (zh) * 2016-08-03 2017-01-04 广东工业大学 利用自适应滤波算法消除泥浆脉冲信号中的泵冲噪声的方法
CN107083957A (zh) * 2017-05-25 2017-08-22 电子科技大学 钻井液随钻信号的泵冲干扰消除方法及系统

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5073877A (en) * 1986-05-19 1991-12-17 Schlumberger Canada Limited Signal pressure pulse generator
US4825421A (en) * 1986-05-19 1989-04-25 Jeter John D Signal pressure pulse generator
EP1217010B1 (en) * 2000-12-20 2004-08-18 Agfa-Gevaert Method of emulsion polymerization
CN105041304B (zh) * 2015-07-27 2017-09-26 电子科技大学 基于二维dct的泵冲干扰信号消除方法
CN106125148B (zh) * 2016-06-13 2017-07-14 中南大学 一种针对有源周期电磁信号的降噪方法及装置
CN109061736A (zh) * 2018-08-09 2018-12-21 中国石油天然气股份有限公司 一种测井资料周期性噪音消除方法及系统

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020180613A1 (en) * 2000-05-08 2002-12-05 Pengyu Shi Digital signal receiver for measurement while drilling system having noise cancellation
CN102857194A (zh) * 2012-09-16 2013-01-02 中国石油大学(华东) 钻井液连续压力波信号的频带噪声消除方法
CN104133982A (zh) * 2014-06-27 2014-11-05 中天启明石油技术有限公司 一种泥浆脉冲信号泵冲噪声的消除方法
CN104265278A (zh) * 2014-07-30 2015-01-07 中天启明石油技术有限公司 一种利用回音抵消技术消除随钻测井中的泵冲噪声的方法
CN105041303A (zh) * 2015-07-27 2015-11-11 电子科技大学 钻井液随钻数据传输系统的泵冲干扰信号消除方法
CN106089188A (zh) * 2016-06-02 2016-11-09 中国石油大学(华东) 一种泥浆脉冲信号泵噪声实时去除方法
CN106301289A (zh) * 2016-08-03 2017-01-04 广东工业大学 利用自适应滤波算法消除泥浆脉冲信号中的泵冲噪声的方法
CN107083957A (zh) * 2017-05-25 2017-08-22 电子科技大学 钻井液随钻信号的泵冲干扰消除方法及系统

Also Published As

Publication number Publication date
CN111308560A (zh) 2020-06-19
CN111308560B (zh) 2021-06-04

Similar Documents

Publication Publication Date Title
CN106089188B (zh) 一种泥浆脉冲信号泵噪声实时去除方法
RU2419996C2 (ru) Система и способ связи по зашумленным каналам связи
WO2021120454A1 (zh) 一种随钻测量mwd系统噪声消除方法、装置及存储介质
US7089118B2 (en) Shear wave velocity determination using circumferentially aligned transmitter and receiver elements
CA2483592C (en) Method of detecting signals in acoustic drill string telemetry
US5151882A (en) Method for deconvolution of non-ideal frequency response of pipe structures to acoustic signals
Qu et al. Adaptive dual-sensor noise cancellation method for continuous wave mud pulse telemetry
CN110187388B (zh) 一种基于变分模态分解的稳定地震品质因子q估计方法
US12006815B2 (en) Method and system for monitoring a wellbore object using a reflected pressure signal
US20210032978A1 (en) Method of determining depths of wellbore reflectors
CN113685172B (zh) 一种随钻声波固井质量评价方法及处理装置
CN111503527B (zh) 基于自适应多元变分模态分解的流体管道泄漏定位方法
US8942330B2 (en) Interference reduction method for downhole telemetry systems
Wu et al. Mathematical model and analysis of characteristics of downhole continuous pressure wave signal
Yan et al. Parametric study on pump noise processing method of continuous wave mud pulse signal based on dual-sensor
CA2882259A1 (en) Measurement transformation apparatus, methods, and systems
Ahmad et al. Performance analysis of downhole acoustic communication in multiphase flow
CN113568044B (zh) 阵列声波测井首波到时确定方法和装置
RU2787653C9 (ru) Способ и устройство для устранения шума для системы измерения во время бурения (MWD), и носитель данных
RU2787653C1 (ru) Способ и устройство для устранения шума для системы измерения во время бурения (MWD), и носитель данных
Brandon et al. Adaptive compensation of the mud pump noise in a measurement-while-drilling system
CN110924940B (zh) 一种mwd系统泵噪自适应预测消除方法及装置
US20180267190A1 (en) Methods and systems employing windowed frequency spectra analysis to derive a slowness log
Zhou et al. Measurement of sound velocity in oil wells based on fast adaptive median filtering
CN109459788B (zh) 地层品质因子计算方法及系统

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20900809

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20900809

Country of ref document: EP

Kind code of ref document: A1