Self-adaptive offset data processing system and method based on magneto-electric signal polarization receiving

Abstract

The invention discloses a self-adaptive offset data processing system and method based on magneto-electric signal polarization receiving. The system comprises a signal acquisition module, a self-adaptive filtering module, a data processing module and an output module which are electrically connected together in sequence. The signal acquisition module comprises a multi-channel receiver, two electrodes and two magnetic sensors, wherein the two electrodes and the two magnetic sensors are electrically connected with the multi-channel receiver. According to the system and the method disclosed by the invention, a self-adaptive filtering technology is adopted, so that filter parameters can be adjusted in real time to adapt to change characteristics of signals and noise; and a high-precision filter is adopted, so that the polarization noise can be accurately extracted without influencing other useful signals.

Description

Technical Field

[0001] The present invention belongs to the field of receiving low-frequency electromagnetic wave logging while drilling and receiving low-frequency electromagnetic information in production wells, and particularly relates to an adaptive offset data processing system and method based on polarization reception of magnetoelectric signals in the field. It is used for wireless transmission of underground information in the process of exploration and development of mineral resources such as petroleum and coalbed methane, and for wireless transmission of casing information measurement in production wells, so as to monitor the status of production wells. Background Art

[0002] In the drilling project of geological exploration, timely and accurate monitoring of the wellbore trajectory and precise adjustment of the drilling posture are one of the key technologies. Electromagnetic wave transmission and communication while drilling can transmit the engineering parameters and geological parameters of the bottom of the well to the ground in real time, providing important data support for real-time monitoring and correction of the wellbore trajectory. Special drilling processes such as underbalanced drilling and gas drilling have special requirements for data transmission methods. The electromagnetic wave transmission method is more suitable for these special drilling processes because it is not affected by the drilling medium. The development of electromagnetic wave transmission and communication technology while drilling is an inevitable product of the progress of drilling engineering technology. Its technical characteristics and application background make it an important direction for future drilling data transmission. With the continuous maturity and improvement of technology, electromagnetic wave transmission and communication while drilling will play a more important role in drilling projects.

[0003] The production well environment is complex and changeable, including extreme conditions such as high temperature, high pressure, high humidity, and the diversity of formation media. Low-frequency electromagnetic information receiving technology needs to have good environmental adaptability to ensure stable operation under complex well conditions. Low-frequency electromagnetic information receiving technology is based on the principle of electromagnetic induction. It receives low-frequency electromagnetic wave signals from the well and converts them into recognizable electrical signals for processing and analysis. During the underground propagation process, these electromagnetic wave signals will interact with the formation media and carry rich geological information. The received low-frequency electromagnetic wave signals need to undergo a complex signal processing and analysis process, including data filtering, amplification, demodulation and other steps to extract useful geological information. At the same time, the signal needs to be corrected and compensated to eliminate the influence of environmental factors on the signal.

[0004] Although electromagnetic wave signal processing while drilling has been partially applied in engineering, it also has the following disadvantages:

[0005] (1) Signal attenuation and interference:

[0006] Signal attenuation: As electromagnetic waves propagate underground, they are affected by the formation medium, such as changes in resistivity and absorption of the formation, which causes the signal strength to gradually attenuate. This attenuation phenomenon is particularly obvious as the drilling depth increases, limiting the detection depth and accuracy of electromagnetic wave transmission while drilling.

[0007] External interference: Electromagnetic waves are susceptible to external electromagnetic interference during transmission, such as electromagnetic noise generated by radio waves and power lines. These interference signals may be mixed into the drilling electromagnetic wave signal, affecting the signal quality and analysis accuracy.

[0008] (2) Technical complexity and cost:

[0009] Equipment complexity: The electromagnetic wave signal processing system while drilling usually includes a transmitter, a receiver and a complex signal processing unit. The design, manufacture and maintenance of these devices are technically difficult and costly. In addition, in order to adapt to the harsh environment underground, the equipment must also have characteristics such as high temperature resistance, high pressure resistance and corrosion resistance, which further increases the technical complexity and cost.

[0010] Difficulty in signal processing: electromagnetic wave signals while drilling will be distorted, attenuated, and so on during transmission, so it is necessary to perform complex processing and analysis on the received signals, such as filtering, amplification, demodulation, and denoising. These processing processes require high-precision algorithms and powerful computing support, which increases the difficulty and cost of technical implementation.

[0011] (3) Comprehensiveness and accuracy of information collection:

[0012] Incomplete information collection: Although the electromagnetic wave technology while drilling can obtain electrical parameter information such as formation resistivity, its detection ability for other physical and chemical properties of the formation (such as lithology, porosity, permeability, etc.) is relatively limited. Therefore, under complex geological conditions, a single electromagnetic wave technology while drilling may not meet the needs of comprehensive and accurate geological information collection.

[0013] Limited data analysis accuracy: Due to the complexity and uncertainty of electromagnetic wave signals propagating underground, as well as the limitations of signal processing technology, the analyzed geological information data may have certain errors and uncertainties, which may affect the accuracy of geological interpretation and oil and gas reservoir evaluation.

[0014] (4) Real-time performance and stability:

[0015] Insufficient real-time performance: During the drilling operation, the real-time performance of geological information is required to be high. However, due to the complexity of electromagnetic wave signal transmission and processing and the limitation of equipment performance, the real-time performance of the drilling electromagnetic wave signal processing system may not meet the actual needs.

[0016] Stability issues: Various complex geological and engineering conditions may be encountered during drilling operations, such as sudden changes in formations, changes in drilling fluid properties, etc. These changes may affect the stability and reliability of the drilling electromagnetic wave signal processing system, resulting in system failures or data anomalies.

[0017] In summary, the existing electromagnetic wave signal processing while drilling has certain shortcomings in signal attenuation and interference, technical complexity and cost, comprehensiveness and accuracy of information collection, real-time performance and stability. Summary of the invention

[0018] The technical problem to be solved by the present invention is to analyze the distribution characteristics of the while-drilling electromagnetic wave transmission signal on the surface, and propose an adaptive cancellation data processing system and method based on polarization reception of magnetoelectric signals according to the distribution characteristics of the electric field and the magnetic field. The system can effectively improve the terminal reception signal-to-noise ratio, further serve the while-drilling electromagnetic wave transmission communication system and the low-frequency information transmission and receiving system of the production well, and can also be applied to communication or signal detection systems in similar scenarios, such as through-the-ground two-way communication systems and underground mineral resource exploration.

[0019] The present invention adopts the following technical solution:

[0020] An adaptive cancellation data processing system based on polarization reception of magnetoelectric signals, the improvement of which lies in: comprising a signal acquisition module, an adaptive filtering module, a data processing module and an output module electrically connected in sequence; the signal acquisition module comprises a multi-channel receiver and two electrodes and two magnetic sensors electrically connected to the multi-channel receiver.

[0021] Furthermore, the magnetic sensor is a magnetic flux negative feedback magnetic sensor, one magnetic sensor is placed at the well site or near the wellhead, and the other is placed 100 meters away.

[0022] Furthermore, the electrodes are copper rods, one electrode is placed at the wellhead, and the other is placed 50 meters away.

[0023] An adaptive cancellation data processing method based on polarization reception of magnetoelectric signals uses the above-mentioned system, and its improvement lies in: the signals received by the magnetic sensor and the electrode are synchronously input into a multi-channel receiver, and after signal conditioning and amplification, hardware filtering, and synchronous data acquisition, the signals are uploaded to a host computer, and demodulated and decoded after being processed by an adaptive algorithm.

[0024] Furthermore, the adaptive filter adopts an adaptive filter based on feedforward control; a high-order low-pass filter is used to extract polarization noise from the amplified signal in real time and accurately; the cutoff frequency of the filter is adjusted by an external capacitor; the polarization noise reference end is received from the magnetic sensor; the extracted polarization noise is used as a feedforward amount and subtracted from the original signal to achieve adaptive cancellation of the polarization noise; data processing includes signal recovery and filtering and denoising.

[0025] Furthermore, the adaptive algorithm uses an adaptive filter:

[0026] From e(n)=s(n)+u1(n)-y(n), we can get:

[0027] e 2 (n) = s 2 (n)+[u1(n)-y(n)] 2 +2s(n)[u1(n)-y(n)]

[0028] Take the ensemble average on both sides of the above equation, and consider that s(n), u1(n) and y(n) are unrelated, so we have:

[0029] E{e 2 (n)}=E{s 2 (n)}+E{[u1(n)-y(n)] 2}

[0030] In the above formula, E{s 2 (n)} is a deterministic quantity, let E{e 2 (n)} is minimized, which is equivalent to making E{[u1(n)-y(n)] 2} is minimum. When H(z) converges to the optimal solution, y(n) approaches u1(n), and the error sequence e(n) approaches the signal s(n).

[0031] Furthermore, the adaptive algorithm adopts the least mean square algorithm, which includes the following steps:

[0032] (1) Initialization: Assign a set of small random non-zero values ​​to the filter weight coefficients and set the global step size parameters and the filter order;

[0033] (2) Filter output: Calculate the filter output based on the current input signal and the filter weight coefficients;

[0034] (3) Error calculation: Calculate the error between the desired signal and the filter output;

[0035] (4) Weight coefficient update: According to the error signal and the input signal, the weight coefficient of the filter is adjusted according to the update rule of the LMS algorithm;

[0036] (5) Iteration: Repeat (2) to (4) until a certain convergence condition is met or the preset number of iterations is reached.

[0037] Furthermore, the adaptive algorithm adopts a recursive least squares algorithm, which includes the following steps:

[0038] (1) Initialization: Assign initial values ​​to the weight coefficients and inverse correlation matrix of the filter. The weight coefficients are initialized to a zero vector or a small random vector, and the inverse correlation matrix is ​​initialized to a large positive definite matrix.

[0039] (2) Calculating the intermediate quantity: Calculating the intermediate quantity used to update the weight coefficient based on the input signal at the current moment and the value of the inverse correlation matrix at the previous moment;

[0040] (3) Calculate the gain vector: Calculate the gain vector using the intermediate quantity and the input signal;

[0041] (4) Calculate the prior estimation error: Calculate the prior estimation error based on the expected signal at the current moment and the output of the filter at the previous moment;

[0042] (5) Update filter coefficients: Use the gain vector and the prior estimation error to update the filter weight coefficients;

[0043] (6) Updating the inverse correlation matrix: updating the inverse correlation matrix according to the gain vector, the input signal and the previous moment value of the inverse correlation matrix;

[0044] (7) Iteration: Repeat (2) to (6) until a certain convergence condition is met or a preset number of iterations is reached.

[0045] The beneficial effects of the present invention are:

[0046] The system and method disclosed in the present invention adopt adaptive filtering technology, which can adjust the filter parameters in real time to adapt to the changing characteristics of signals and noise; adopt high-precision filters to ensure that polarization noise can be accurately extracted without affecting other useful signals; adopt noise cancellation and amplification circuits to achieve effective cancellation of polarization noise and signal amplification processing; adopt digital signal processing technology to further improve the signal-to-noise ratio and measurement accuracy of the signal.

[0047] The system and method disclosed in the present invention can improve the signal-to-noise ratio. The adaptive cancellation algorithm can estimate and cancel the background noise and interference signals in real time, thereby significantly improving the signal-to-noise ratio of the target signal. This is crucial for extracting useful signals in a complex electromagnetic environment. Through adaptive processing, the original target signal can be highly restored, signal distortion can be reduced, and signal accuracy can be improved.

[0048] The system and method disclosed in the present invention can improve system performance. The adaptive cancellation algorithm can cope with changes in signal paths and uncertainties in interference sources, making the system more robust and stable. The technology is not limited to specific environments or conditions and can be widely used in various occasions requiring high-precision signal processing.

[0049] The system and method disclosed in the present invention can reduce hardware costs, realize noise cancellation through software algorithms, reduce the dependence on high-precision hardware, and reduce the overall hardware cost of the system. In the background application of the present invention, the adaptive polarization noise cancellation scheme can simplify the design of signal conditioning circuits and reduce circuit complexity and power consumption.

[0050] The system and method disclosed in the present invention can improve the efficiency of data processing. The adaptive algorithm can track signal changes in real time, quickly adjust parameters to adapt to new environments, and ensure the real-time and effectiveness of data processing. By optimizing the algorithm design, the amount of calculation can be reduced while ensuring the processing effect, thereby improving the efficiency of data processing.

[0051] In summary, the system and method disclosed in the present invention have multiple beneficial effects such as improving signal-to-noise ratio, enhancing system performance, reducing hardware costs, improving data processing efficiency and expanding application fields. These effects make the system and method have broad application prospects and important practical value in multiple fields. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 This is a schematic diagram of receiving ground signals of electromagnetic waves while drilling;

[0053] Figure 2 It is a block diagram of the implementation of noise adaptive cancellation;

[0054] Figure 3 It is a schematic diagram of collecting electromagnetic wave receiving signals and reference noise while drilling. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0056] Embodiment 1, this embodiment discloses an adaptive cancellation data processing system based on polarization reception of magnetoelectric signals, which mainly uses adaptive filtering technology to extract and cancel polarization noise in magnetoelectric signals in real time. It includes a signal acquisition module, an adaptive filtering module, a data processing module and an output module which are electrically connected in sequence; the signal acquisition module includes a multi-channel receiver and two electrodes and two magnetic sensors electrically connected to the multi-channel receiver.

[0057] The magnetic sensor is a magnetic flux negative feedback magnetic sensor, one magnetic sensor is placed at the well site or near the wellhead, and the other is placed 100 meters away. The electrodes are copper rods, one electrode is placed at the wellhead, and the other is placed 50 meters away.

[0058] Taking the electromagnetic wave communication system while drilling as an example, the on-site implementation scenario of the present invention is as follows: Figure 1As shown: (1) Electric field signal receiving method: a metal electrode with good conductivity is buried in the ground, one end of which is placed at the far end (usually more than 50 meters in the project), and the other end is set at the wellhead. The potential difference signal received by the two electrodes is fed back to the receiver through the cable. (2) Magnetic field signal receiving method: the magnetic sensor is fixed on the ground to receive the vertical upward signal (i.e., Hz direction), the 1# magnetic sensor is placed at the well site or near the wellhead (mainly used to receive interference signals such as surface generators at the project site), and the 2# sensor is placed at a longer distance (usually 100 meters away, mainly used to receive external interference such as lightning). The signals received by the magnetic sensor and the electrode are synchronously input into the multi-channel receiver, and after signal conditioning and amplification, hardware filtering, and synchronous data acquisition, the signal is uploaded to the host computer software, and demodulated and decoded after being processed by the adaptive algorithm of the software or receiver hardware.

[0059] In the adaptive algorithm application, the signal received by the magnetic sensor is used as the reference noise end to perform adaptive noise reduction with the electric field signal received by the electrode (the electric field signal contains the communication signal and the noise signal, and the magnetic field signal only contains the noise signal in the Hz direction). The adaptive noise reduction algorithm uses the least squares algorithm and the least mean square algorithm, and also includes the improved algorithms of these two algorithms.

[0060] This embodiment also discloses an adaptive cancellation data processing method based on polarization reception of magnetoelectric signals, which mainly involves the identification, extraction and cancellation of polarization noise received in magnetoelectric signals. Using the above system, the signals received by the magnetic sensor and the electrode are synchronously input into a multi-channel receiver, and after signal conditioning and amplification, hardware filtering, and synchronous data acquisition, the signals are uploaded to the host computer, and demodulated and decoded after being processed by an adaptive algorithm. This processing method can effectively improve the signal-to-noise ratio of the signal, which is very important for communication and measurement information extraction.

[0061] (1) Signal acquisition:

[0062] Sensor selection: Use highly sensitive magnetoelectric sensors (such as flux negative feedback magnetic sensors, etc.) to collect signals containing polarization noise; electrode sensors use metal materials such as copper rods with good conductivity to feed back signals to the receiver through cables;

[0063] Signal amplification: Amplify and condition the weak signal output by the sensor for subsequent processing.

[0064] (2) Adaptive filtering:

[0065] Adaptive filter design: An adaptive filter based on feedforward control is used, which can automatically adjust the parameters of the current moment according to the filter parameters of the previous moment to adapt to the statistical characteristics of the signal or noise changing over time.

[0066] Polarization noise extraction: A high-order low-pass filter with a very narrow transition band (such as an eighth-order Butterworth low-pass filter) is used to accurately extract polarization noise from the amplified signal in real time. The cutoff frequency of the filter can be adjusted by an external capacitor to ensure that the noise in the required frequency band can be accurately extracted. The polarization noise reference end mainly comes from the magnetic sensor reception.

[0067] Noise cancellation: The extracted polarization noise is used as the feedforward quantity and subtracted from the original signal to achieve adaptive cancellation of polarization noise. This step is usually performed in the noise cancellation and amplification circuit to ensure the signal quality after cancellation.

[0068] (3) Data processing:

[0069] Signal recovery: Process the canceled signal to recover the electromagnetic wave signal with the reference noise filtered out.

[0070] Filtering and denoising: Digital signal processing technology (such as comb bandpass filtering, etc.) is further used to filter out other noise sources such as power frequency interference and differential interference.

[0071] (4) Output and display:

[0072] Result output: Convert the processed signal into readable measurement results, such as downhole engineering parameters.

[0073] Display and record: The measurement results are displayed in real time on the display screen and can be remotely transmitted and recorded through data interfaces (such as RS485, Ethernet, etc.).

[0074] Adaptive Noise Cancellation Technology:

[0075] Since adaptive filters have strong learning and tracking capabilities and adapt to the objective needs of unknown and time-varying statistical characteristics of input signals, they have been widely used in practical applications. These application areas include communications, radar, sonar, automatic control, geological exploration, navigation, and biomedical engineering. Adaptive noise cancellation is an application extension algorithm of adaptive filters. It uses a compensation cancellation method to collect noise signals in advance, cancel them with contaminated signals, thereby suppressing interference and extracting effective signals. Its implementation block diagram is shown below. Figure 2 As shown:

[0076] From e(n)=s(n)+u1(n)-y(n), we can get:

[0077] e 2 (n) = s 2 (n)+[u1(n)-y(n)] 2 +2s(n)[u1(n)-y(n)] (1)

[0078] Take the ensemble average on both sides of the above equation, and consider that s(n), u1(n) and y(n) are unrelated, so we have:

[0079] E{e 2 (n)}=E{s 2 (n)}+E{[u1(n)-y(n)] 2} (2)

[0080] In the above formula, E{s 2 (n)} is a deterministic quantity. Let E{e 2 (n)} is minimized, which is equivalent to making E{[u1(n)-y(n)] 2} is the minimum. When H(z) converges to its optimal solution, y(n) approaches u1(n), and the error sequence e(n) approaches the signal s(n), thus achieving the purpose of noise adaptive cancellation.

[0081] The input signal x(n) contains the required signal s(n) and noise u1(n), which is collected by a sensor close to the signal source s(n). Among them, u1(n) is the interference noise mixed into the sensor by the noise source u(n). Another sensor is used to make it close to the noise source u(n), and the noise signal collected by the sensor is recorded as u2(n). Obviously, since u1(n) and s(n) come from two signal sources, they are theoretically unrelated, but u1(n) and u2(n) come from the same source, so they should be very related. In order to remove u1(n) from x(n), since the statistical characteristics of the signal x(n) are unknown, and the performance of the sensor is also unknown, and it is likely to be time-varying, it is necessary to use an adaptive filter with adjustable coefficients to achieve the above purpose.

[0082] The input of the adaptive filter H(z) is u2(n), and the output is y(n). Let the desired signal d(n) be x(n), and it is hoped that by adjusting H(z), y(n) can be the maximum approximation of u1(n). In this way, subtracting d(n) from y(n) can effectively remove u1(n), and the error sequence e(n) after subtraction will be the required useful signal s(n). The above process is the adaptive cancellation of noise, which has a wide range of applications in many engineering fields. For example, when recording human physiological signals (electrocardiogram, electroencephalogram, electromyography, heart sounds, etc.), it is inevitable to be interfered by the power frequency. At this time, the adaptive cancellation method can be used to eliminate the power frequency interference. In addition, there are applications of adaptive noise cancellation in the cancellation of noise in the cab, echo cancellation in long-distance telephone lines, and interference cancellation in headphones and hearing aids. The following simulation example illustrates the specific application.

[0083] The layout of the adaptive noise cancellation algorithm used in the electromagnetic wave signal processing while drilling is as follows Figure 3 As shown, one end of this arrangement is used to receive the reference input noise signal (B1, B2 magnetic sensors receive noise signals), and the other end is used to receive the mixed signal with the useful signal (A1, A2 electrodes receive electric field signals), and then use Figure 2 The adaptive noise cancellation method is used for filtering.

[0084] Least Mean Squares Algorithm:

[0085] The adaptive least mean square algorithm, namely the Least Mean Square (LMS) algorithm, is a simple and widely used adaptive filtering algorithm. Based on the Wiener filtering theory, it is optimized and extended by the steepest descent method. The main features and principles of the LMS algorithm are as follows:

[0086] The basic idea of ​​the LMS algorithm is to minimize the mean square error (MSE) between the filter output and the expected output by continuously adjusting the filter weight coefficients. In specific implementation, the algorithm will update the weight coefficients of the "next moment" based on the error signal of the "current moment" (i.e., the difference between the expected signal and the actual output of the filter). This update process is completed by adding a proportional term of the negative mean square error gradient to the weight coefficients of the "previous moment".

[0087] The general process of the LMS algorithm includes the following steps:

[0088] (1) Initialization: Assign a set of small random non-zero values ​​to the weight coefficients of the filter and set the global step size parameter (also called the learning rate) and the order (or number of taps) of the filter.

[0089] (2) Filter output: Calculate the filter output based on the current input signal and the filter weight coefficients.

[0090] (3) Error calculation: Calculate the error between the expected signal and the filter output.

[0091] (4) Weight coefficient update: According to the error signal and the input signal, the weight coefficient of the filter is adjusted according to the update rule of the LMS algorithm.

[0092] (5) Iteration: Repeat (2) to (4) until a certain convergence condition is met or the preset number of iterations is reached.

[0093] The main features of this algorithm are:

[0094] (1) Low computational complexity: The LMS algorithm has a simple structure and only requires the input vector and expected response for each iteration, thus avoiding complex matrix operations and reducing computational complexity.

[0095] (2) Good convergence: In an environment where the signal is a stationary signal, the LMS algorithm has good convergence, and its expected value can converge to the Wiener solution unbiasedly.

[0096] (3) High stability: The LMS algorithm exhibits good stability when implemented with finite precision and is one of the most stable and widely used adaptive algorithms.

[0097] (4) Easy to implement: The principle of the LMS algorithm is simple, with few parameters, and it is easy to implement in hardware and software.

[0098] Recursive least squares algorithm:

[0099] The adaptive recursive least squares algorithm (RLS) is an algorithm widely used in the field of adaptive filtering. It aims to minimize the weighted square error between the filter output and the expected output in a recursive way. Compared with the traditional least mean square (LMS) algorithm, the RLS algorithm has a faster convergence speed and better tracking performance, but the computational complexity is also relatively high.

[0100] The basic idea of ​​the RLS algorithm is to minimize the weighted square error sum (i.e., cost function) by continuously adjusting the weight coefficients of the filter. Unlike the LMS algorithm, the RLS algorithm considers the error information of all historical moments when updating the weight coefficients, and introduces a forgetting factor to balance the impact of new and old data on the weight coefficient update.

[0101] The general process of the RLS algorithm includes the following steps:

[0102] (1) Initialization: Assign initial values ​​to the weight coefficients and inverse correlation matrix of the filter. Usually, the weight coefficients are initialized to a zero vector or a small random vector, and the inverse correlation matrix is ​​initialized to a large positive definite matrix (for example, a multiple of the unit matrix).

[0103] (2) Calculate the intermediate quantity: Calculate the intermediate quantity used to update the weight coefficient based on the input signal at the current moment and the value of the inverse correlation matrix at the previous moment.

[0104] (3) Calculate the gain vector: Use the intermediate quantity and the input signal to calculate the gain vector. The gain vector plays a key role in the weight coefficient update process, and it determines the direction and size of the weight coefficient update.

[0105] (4) Calculate the prior estimation error: Calculate the prior estimation error based on the expected signal at the current moment and the output of the filter at the previous moment.

[0106] (5) Update filter coefficients: Use the gain vector and prior estimation error to update the filter weight coefficients.

[0107] (6) Update the inverse correlation matrix: Update the inverse correlation matrix according to the gain vector, input signal and the previous value of the inverse correlation matrix. The inverse correlation matrix plays the role of accumulating historical information in the algorithm.

[0108] (7) Iteration: Repeat (2) to (6) until a certain convergence condition is met or a preset number of iterations is reached.

Claims

1. An adaptive cancellation data processing system based on polarization reception of magnetoelectric signals, characterized in that: It comprises a signal acquisition module, an adaptive filtering module, a data processing module and an output module which are electrically connected in sequence; the signal acquisition module comprises a multi-channel receiver and two electrodes and two magnetic sensors which are electrically connected to the multi-channel receiver.

2. The adaptive cancellation data processing system based on polarization reception of magnetoelectric signals according to claim 1, characterized in that: The magnetic sensors are flux negative feedback magnetic sensors, one of which is placed at the well site or near the wellhead, and the other is placed 100 meters away.

3. The adaptive cancellation data processing system based on polarization reception of magnetoelectric signals according to claim 1, characterized in that: The electrodes are copper rods, one electrode is placed at the wellhead and the other is placed 50 meters away.

4. An adaptive cancellation data processing method based on polarization reception of magnetoelectric signals, using the system of claim 1, characterized in that: The signals received by the magnetic sensor and the electrode are synchronously input into the multi-channel receiver. After signal conditioning and amplification, hardware filtering, and synchronous data acquisition, the signals are uploaded to the host computer and demodulated and decoded after being processed by the adaptive algorithm.

5. The method for adaptive cancellation data processing based on polarization reception of magnetoelectric signals according to claim 4, characterized in that: The adaptive filter adopts an adaptive filter based on feedforward control; a high-order low-pass filter is used to accurately extract polarization noise from the amplified signal in real time; the cutoff frequency of the filter is adjusted by an external capacitor; The polarization noise reference end is received by the magnetic sensor; the extracted polarization noise is used as the feedforward quantity and subtracted from the original signal to achieve adaptive cancellation of the polarization noise; Data processing includes signal recovery and filtering and denoising.

6. The method for adaptive cancellation data processing based on polarization reception of magnetoelectric signals according to claim 4, characterized in that: The adaptive algorithm uses an adaptive filter: From e(n)=s(n)+u1(n)-y(n), we can get: e 2 (n)=s 2 (n)+[u1(n)-y(n)] 2 +2s(n)[u1(n)-y(n)] Take the ensemble average on both sides of the above equation, and consider that s(n), u1(n) and y(n) are unrelated, so we have: Yes 2 (n)}=E{s 2 (n)}+E{[u1(n)-y(n)] 2 } In the above formula, E{s 2 (n)} is a deterministic quantity, let E{e 2 (n)} is minimized, which is equivalent to making E{[u1(n)-y(n)] 2 } is minimum. When H(z) converges to the optimal solution, y(n) approaches u1(n), and the error sequence e(n) approaches the signal s(n).

7. The method for adaptive cancellation data processing based on polarization reception of magnetoelectric signals according to claim 4, characterized in that: The adaptive algorithm uses the least mean square algorithm, which includes the following steps: (1) Initialization: Assign a set of small random non-zero values ​​to the filter weight coefficients and set the global step size parameters and the filter order; (2) Filter output: Calculate the filter output based on the current input signal and the filter weight coefficients; (3) Error calculation: Calculate the error between the desired signal and the filter output; (4) Weight coefficient update: According to the error signal and the input signal, the weight coefficient of the filter is adjusted according to the update rule of the LMS algorithm; (5) Iteration: Repeat (2) to (4) until a certain convergence condition is met or the preset number of iterations is reached.

8. The method for adaptive cancellation data processing based on polarization reception of magnetoelectric signals according to claim 4, characterized in that: The adaptive algorithm uses a recursive least squares algorithm, which includes the following steps: (1) Initialization: Assign initial values ​​to the weight coefficients and inverse correlation matrix of the filter. The weight coefficients are initialized to a zero vector or a small random vector, and the inverse correlation matrix is ​​initialized to a large positive definite matrix. (2) Calculating the intermediate quantity: Calculating the intermediate quantity used to update the weight coefficient based on the input signal at the current moment and the value of the inverse correlation matrix at the previous moment; (3) Calculate the gain vector: Calculate the gain vector using the intermediate quantity and the input signal; (4) Calculate the prior estimation error: Calculate the prior estimation error based on the expected signal at the current moment and the output of the filter at the previous moment; (5) Update filter coefficients: Use the gain vector and the prior estimation error to update the filter weight coefficients; (6) Updating the inverse correlation matrix: updating the inverse correlation matrix according to the gain vector, the input signal and the previous moment value of the inverse correlation matrix; (7) Iteration: Repeat (2) to (6) until a certain convergence condition is met or a preset number of iterations is reached.

Images