A method and system for reducing noise of low-displacement and low-pressure-difference pulse signal while drilling

Abstract

The present invention provides a method and system for reducing noise of a low-displacement and low-pressure differential downhole pulse signal, which relates to the technical field of downhole pulse signal processing. Downhole pulse signal sequences, displacement information sequences and pressure differential information sequences are obtained under low-displacement and low-pressure differential working conditions. According to the displacement information sequence and the pressure differential information sequence, displacement and pressure differential change significance analysis is performed to obtain a change significance sequence, and downhole pulse signals at each moment are predicted respectively. According to multiple change significances in the change significance sequence, the downhole pulse signal sequence and the predicted downhole pulse signal sequence are fused, and the fused downhole pulse signal sequence is filtered to obtain a reduced noise downhole pulse signal sequence. The technical problem of weak downhole pulse signal strength and large noise interference under low-displacement and low-pressure differential conditions in the prior art is solved. The technical effect of effectively reducing noise of low-displacement and low-pressure differential downhole pulse signals and improving signal quality is achieved.

Description

Technical Field

[0001] The invention relates to the technical field of drilling pulse signal processing, and in particular to a low-displacement and low-pressure difference drilling pulse signal noise reduction method and system. Background Art

[0002] The pulse signal while drilling is a key link in the measurement while drilling technology. It uses a mud pulse generator to encode and modulate the data obtained by the downhole measuring instrument into a mud pressure wave, which is transmitted to the ground through the mud medium inside the drill pipe, and then received and decoded by the ground equipment to restore it to downhole parameter information. This technology is widely used in oil and gas exploration and development. The pulse signal transmission technology while drilling is the core of realizing wireless measurement while drilling. However, in actual processes, due to the low displacement and pressure difference operating conditions, the mud flow rate and pressure are reduced, resulting in limited propagation capacity of mud pulses. The intensity of the generated pulse signal is usually weak, which affects the transmission distance and reception quality of the signal. In addition, under low displacement and low pressure difference conditions, the pulse signal while drilling is easily interfered by various noises, such as drill pipe vibration, mud flow noise, etc., which further increases the difficulty of signal processing and affects the real-time and accuracy of drilling operations.

[0003] In the prior art, there are technical problems that the drilling pulse signal strength is weak and the noise interference is large under the conditions of low displacement and low pressure difference, resulting in low real-time performance and accuracy of drilling operations. Summary of the invention

[0004] The present application provides a method and system for reducing noise of a low-displacement and low-pressure differential downhole pulse signal, which is used to solve the technical problem in the prior art that downhole pulse signal strength is weak and noise interference is large under low-displacement and low-pressure differential conditions, resulting in low real-time performance and accuracy of drilling operations.

[0005] In view of the above problems, the present application provides a low-displacement and low-pressure difference while-drilling pulse signal noise reduction method and system.

[0006] In a first aspect of the present application, a method for reducing noise of a low-displacement and low-pressure-difference downhole pulse signal is provided, the method comprising: under a low-displacement and low-pressure-difference operating condition, collecting continuous downhole pulse signals to obtain a downhole pulse signal sequence, and collecting continuous displacement information and pressure-difference information to obtain a displacement information sequence and a pressure-difference information sequence; performing displacement change significance analysis and pressure-difference change significance analysis according to the displacement information sequence and the pressure-difference information sequence to obtain a displacement change significance sequence and a pressure-difference change significance sequence, and processing to obtain a change significance sequence; predicting the downhole pulse signal at each moment according to the downhole pulse signal sequence to obtain a predicted downhole pulse signal sequence; fusing the downhole pulse signal sequence and the predicted downhole pulse signal sequence according to multiple change significances in the change significance sequence to obtain a fused downhole pulse signal sequence, and filtering the fused downhole pulse signal sequence to obtain a reduced-noise downhole pulse signal sequence.

[0007] In a second aspect of the present application, a low-displacement and low-pressure differential drilling pulse signal noise reduction system is provided, the system comprising: a data acquisition module, the data acquisition module is used to collect continuous drilling pulse signals under low-displacement and low-pressure differential operating conditions to obtain a drilling pulse signal sequence, and collect continuous displacement information and pressure differential information to obtain a displacement information sequence and a pressure differential information sequence; a data analysis module, the data analysis module is used to perform displacement change significance analysis and pressure differential change significance analysis according to the displacement information sequence and the pressure differential information sequence to obtain a displacement change significance sequence and a pressure differential information sequence. a data prediction module, wherein the data prediction module is used to predict the drilling pulse signal at each moment according to the drilling pulse signal sequence, and obtain a predicted drilling pulse signal sequence; a data fusion module, wherein the data fusion module is used to fuse the drilling pulse signal sequence and the predicted drilling pulse signal sequence according to multiple change sensitivities in the change significance sequence, and obtain a fused drilling pulse signal sequence, and filter the fused drilling pulse signal sequence to obtain a noise-reduced drilling pulse signal sequence.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0009] The method provided in the embodiment of the present application is to collect continuous downhole pulse signals under low displacement and low pressure difference working conditions to obtain downhole pulse signal sequences, and collect continuous displacement information and pressure difference information to obtain displacement information sequences and pressure difference information sequences; according to the displacement information sequence and pressure difference information sequence, displacement change significance analysis and pressure difference change significance analysis are performed to obtain displacement change significance sequence and pressure difference change significance sequence, and the change significance sequence is processed; according to the downhole pulse signal sequence, the downhole pulse signal at each moment is predicted to obtain a predicted downhole pulse signal sequence; according to multiple change significances in the change significance sequence, the downhole pulse signal sequence and the predicted downhole pulse signal sequence are fused to obtain a fused downhole pulse signal sequence, and the fused downhole pulse signal sequence is filtered to obtain a de-noised downhole pulse signal sequence. The technical effect of effectively reducing the noise of downhole pulse signals with low displacement and low pressure difference, improving signal quality, and ensuring the real-time and accuracy of drilling operations is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0011] Figure 1 A schematic flow chart of a low-displacement and low-pressure difference pulse signal noise reduction method while drilling provided in this application;

[0012] Figure 2 A schematic diagram of the structure of a low-displacement and low-pressure difference while-drilling pulse signal noise reduction system provided in this application.

[0013] Explanation of the reference numerals: data collection module 11 , data analysis module 12 , data prediction module 13 , data fusion module 14 . DETAILED DESCRIPTION

[0014] The present application provides a method and system for reducing noise of a low-displacement and low-pressure differential while-drilling pulse signal, which is used to solve the technical problem in the prior art that the strength of the while-drilling pulse signal is weak and the noise interference is large under the condition of low-displacement and low-pressure differential, resulting in low real-time and accuracy of drilling operations. The method achieves the technical effect of effectively reducing noise of low-displacement and low-pressure differential while-drilling pulse signals, improving signal quality, and ensuring the real-time and accuracy of drilling operations.

[0015] Below, the technical solutions in the present invention will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments of the present invention. It should be understood that the present invention is not limited to the example embodiments described herein. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. It should also be noted that, for the convenience of description, only the parts related to the present invention are shown in the accompanying drawings, rather than all of them.

[0016] Embodiment 1, as Figure 1 As shown, the present application provides a method for reducing noise of a low-displacement and low-pressure-difference pulse signal while drilling, the method comprising:

[0017] Under the working condition of low displacement and low pressure difference, continuous while drilling pulse signals are collected to obtain a while drilling pulse signal sequence, and continuous displacement information and pressure difference information are collected to obtain a displacement information sequence and a pressure difference information sequence.

[0018] Specifically, the low displacement and low pressure difference operating condition refers to the state in which the mud displacement is low and the pressure difference inside and outside the drill pipe is small during the drilling process. Under this condition, the strength of the mud pulse signal is relatively weak, which may affect the transmission distance and reception quality of the signal. In addition, under the condition of low displacement and low pressure difference, the signal is easily interfered by the background environmental noise, resulting in increased difficulty in signal processing. Under the low displacement and low pressure difference operating condition, the continuous drilling fluid (mud) pressure fluctuation is monitored in real time by a high-sensitivity pressure sensor, and the continuous analog signal is converted into a digital signal to obtain a continuous while-drilling pulse signal sequence. The while-drilling pulse information refers to the signal information generated by the interaction between the drill bit and the formation during the drilling process, which can reflect the change in the drilling state. At the same time, the displacement information and the pressure difference information are continuously collected using the displacement sensor and the pressure difference sensor. The displacement information reflects the flow change of the mud, and the pressure difference information reflects the difference in the pressure inside and outside the drill pipe. According to the collected continuous displacement information and pressure difference information, a displacement information sequence and a pressure difference information sequence are formed. By collecting the actual pulse signal sequence, as well as the corresponding displacement information sequence and pressure difference information sequence, reliable data support is provided for the subsequent change significance analysis and noise reduction processing.

[0019] Furthermore, under the operating condition of low displacement and low pressure difference, continuous downhole pulse signals are collected to obtain a downhole pulse signal sequence, and continuous displacement information and pressure difference information are collected to obtain a displacement information sequence and a pressure difference information sequence, including: under the operating condition of low displacement and low pressure difference, continuous downhole pulse signals are collected to obtain a downhole pulse signal sequence, and a signal timestamp sequence is extracted; according to the signal timestamp sequence, the displacement information and pressure difference information at each signal timestamp are collected to obtain a displacement information sequence and a pressure difference information sequence.

[0020] Specifically, the downhole pulse signal is generated by the pressure change in the mud. Continuous downhole pulse signals are collected under low displacement and low pressure difference working conditions to obtain a downhole pulse signal sequence, which includes the amplitude information and timestamp information of the signal. Then the timestamp information in the downhole pulse signal sequence is extracted to obtain a signal timestamp sequence, which includes the specific time when each pulse occurs. According to the extracted signal timestamp sequence, the displacement information and pressure difference information under each signal timestamp are synchronously collected. Through synchronous collection, it is ensured that at the moment corresponding to each timestamp, the corresponding displacement information and pressure difference information are obtained from the sensor, and the displacement information sequence and pressure difference information sequence are obtained by integrating the collected displacement information and pressure difference information. According to the timestamp of the downhole pulse signal, the corresponding displacement information sequence and pressure difference information sequence are obtained, which ensures the synchronization of different types of data and improves the reliability of subsequent signal analysis and noise reduction.

[0021] According to the displacement information sequence and the pressure difference information sequence, displacement change significance analysis and pressure difference change significance analysis are performed to obtain a displacement change significance sequence and a pressure difference change significance sequence, and then the change significance sequence is obtained by processing.

[0022] Furthermore, based on the displacement information sequence and the pressure difference information sequence, a displacement change significance analysis and a pressure difference change significance analysis are performed, including: based on the displacement information sequence, performing a displacement change significance analysis to obtain a displacement change significance sequence; based on the pressure difference information sequence, performing a pressure difference change significance analysis to obtain a pressure difference change significance sequence; and performing weighted calculation on multiple displacement change significances and multiple pressure difference change significances in the displacement change significance sequence and the pressure difference change significance sequence to obtain a change significance sequence.

[0023] Specifically, according to the obtained displacement information sequence, the variation range between the mean of adjacent displacement information and the mean of randomly selected displacement information is calculated, the variation significance of each displacement information is analyzed, and according to the analysis result, the displacement variation significance sequence is obtained. Similarly, the pressure difference variation significance analysis is performed according to the pressure difference information sequence, and the variation range between the mean of adjacent pressure difference information and the mean of randomly selected pressure difference information is used to obtain the pressure difference variation significance sequence. The displacement variation significance sequence and the pressure difference variation significance sequence respectively reflect the fluctuation of displacement and pressure difference at different timestamps.

[0024] Then, multiple displacement change significances and multiple pressure difference change significances in the displacement change significance sequence and the pressure difference change significance sequence are weightedly calculated. The weight coefficient can be determined based on expert experience or historical data analysis results. The weight coefficient reflects the relative size of the influence of displacement and pressure difference on the pulse signal under specific working conditions, and then a weighted calculation is performed to obtain a change significance sequence. For example, based on the analysis of historical data, it is known that the influence of displacement on the pulse signal and the influence of pressure difference on the pulse signal are equal, and the weight coefficients can be set to 0.5 and 0.5. Then, when calculating the change significance sequence, the weight coefficients of displacement change significance and pressure difference change significance are 0.5 respectively. Assuming that the displacement change significance is A and the pressure difference change significance is B, the change significance C can be calculated by the following formula: = + ,in, and are the weight coefficients of displacement and pressure difference, =0.5, =0.5. If it is concluded from the analysis of historical data that the change in displacement has a greater impact on the pulse signal, the weight coefficient can be set to 0.6 and 0.4, so that the impact of displacement will be more prominent when calculating the change significance. By flexibly adjusting the weight coefficient, the combined impact of various factors on the signal can be more accurately reflected. The change significance sequence comprehensively considers the possible impact of changes in displacement and pressure difference on the downhole pulse signal. The greater the change significance, the greater the change in displacement and pressure difference, and the greater the probability of change in the pulse signal at this timestamp and the pulse signal at other timestamps. By analyzing the change significance of displacement information and pressure difference information, the dynamic characteristics of the drilling process can be better understood, and the comprehensiveness and reliability of subsequent downhole pulse signal processing can be improved.

[0025] Furthermore, according to the displacement information sequence, a displacement change significance analysis is performed, including: selecting first displacement information in the displacement information sequence, extracting M adjacent displacement information adjacent to the first displacement information, and randomly extracting M comparative displacement information in the displacement information sequence, where M is a positive integer; calculating the difference between the mean of the M adjacent displacement information and the mean of the M comparative displacement information as the first displacement change significance; continuing to perform displacement change significance analysis on other displacement information in the displacement information sequence, obtaining multiple displacement change significances, and generating a displacement change significance sequence.

[0026] Specifically, firstly, the first displacement information is selected according to the displacement information sequence as the basic point for analysis. Then, starting from the first displacement information, M adjacent displacement information is extracted, and the adjacent displacement information refers to the sorted data that is relatively close to the first displacement information in the time series. At the same time, M comparative displacement information is randomly extracted from the displacement information sequence, and the comparative displacement information comes from different positions in the displacement information sequence, and M is a positive integer. Then, the mean information of the two data groups is calculated: first, the mean of the M adjacent displacement information is calculated, and the mean of the adjacent displacement information reflects the fluctuation of the displacement during this period of time. Then, the mean of the M comparative displacement information is calculated, and the mean of the comparative displacement information reflects the fluctuation of the overall displacement sequence. By calculating the difference between the mean of the adjacent displacement information and the mean of the comparative displacement information, the first displacement change significance is obtained, and the first displacement change significance is used to reflect the degree of fluctuation of the first displacement information in the displacement information sequence. Continue this process, perform the same change significance analysis on other displacement information in the displacement information sequence, and repeat this method in a loop to finally obtain multiple displacement change significances and form a displacement change significance sequence. The sorted change significance sequence provides key data support for subsequent analysis and noise reduction, and can accurately reflect the fluctuation of displacement information under low displacement and low pressure difference conditions. Similarly, based on the above method, pressure difference change significance analysis can be performed to obtain a pressure difference change significance sequence. Since the method is similar, it will not be elaborated here.

[0027] According to the drilling-while-pulse signal sequence, the drilling-while-pulse signal at each moment is predicted respectively to obtain a predicted drilling-while-pulse signal sequence.

[0028] Specifically, a time series prediction method is used to predict the downhole pulse signal at each moment in the downhole pulse signal sequence to obtain the predicted downhole pulse signal value at the corresponding moment. By traversing the entire downhole pulse signal sequence, the downhole pulse signal at each timestamp is predicted point by point, and finally a predicted downhole pulse signal sequence corresponding to the downhole pulse signal sequence is formed. Under low-displacement and low-pressure differential operating conditions, since the signal quality may be unstable, the obtained predicted downhole pulse signal sequence can be used as a supplement to the actual signal. By subsequently comparing the predicted downhole pulse signal sequence with the actual signal, the signal fusion is achieved, and the overall quality and stability of the signal are further improved, providing more reliable basic data for subsequent noise reduction processing, achieving the effect of auxiliary noise reduction, and making the downhole pulse signal more realistically and accurately reflect the drilling status.

[0029] Furthermore, according to the drilling pulse signal sequence, the drilling pulse signal at each moment is predicted respectively to obtain the predicted drilling pulse signal sequence, including: according to the historical data of the low-displacement and low-pressure difference drilling pulse signal, a sample preamble signal group set is collected, and the drilling pulse signal after each sample preamble signal group is collected to obtain a sample prediction signal set; the sample preamble signal group set and the sample prediction signal set are used to train a drilling pulse signal predictor; a plurality of preamble signal groups are selected in the drilling pulse signal sequence, and the preamble signal groups are input into the drilling pulse signal predictor to predict the predicted signals of the signal timestamps after the plurality of preamble signal groups to obtain the predicted drilling pulse signal sequence.

[0030] Specifically, the historical data of low-displacement and low-pressure difference while-drilling pulse signals are extracted from the drilling operation data collection library. Then, based on the historical data of the low-displacement and low-pressure difference while-drilling pulse signals, a sample preamble signal group set is collected, and the sample preamble signal group set includes multiple while-drilling pulse signals before the current time point, which are used to reflect the dynamic characteristics of the signal. Then, for each sample preamble signal group, the subsequent while-drilling pulse signal is collected to obtain a sample prediction signal set. By collecting the preamble signal and the subsequent signal, the foundation for subsequent model training is laid.

[0031] The sample preamble signal group set and the sample prediction signal set are used, the sample preamble signal group set is used as the model input data, and the sample prediction signal set is used as the target value of the model prediction, and the drilling pulse signal predictor is trained. The drilling pulse signal predictor is constructed based on machine learning or deep learning algorithms, and can learn and capture how to infer the change pattern of subsequent signals from preamble signals, and has the ability to predict future signals based on historical signals. After the model training is completed, multiple preamble signal groups are selected from the drilling pulse signal sequence as input, and for each signal timestamp, the drilling pulse information of multiple signal timestamps before the timestamp is selected to construct a preamble signal group, which contains the historical signal data related to the current timestamp in the time series, and can reflect the dynamic changes and trends of the signal. Then, these preamble signal groups are input into the trained drilling pulse signal predictor. The drilling pulse signal predictor predicts multiple predicted drilling pulse signals of multiple signal timestamps based on the input preamble signal group characteristics and the patterns and laws learned internally, and finally forms a predicted drilling pulse signal sequence. By using the information in historical data to train the prediction model, real-time prediction of the drilling pulse signal under low displacement and low pressure difference conditions is achieved, improving the accuracy and reliability of signal prediction.

[0032] According to multiple change significances in the change significance sequence, the drilling-while-drilling pulse signal sequence and the predicted drilling-while-drilling pulse signal sequence are fused to obtain a fused drilling-while-drilling pulse signal sequence, and the fused drilling-while-drilling pulse signal sequence is filtered to obtain a noise-reduced drilling-while-drilling pulse signal sequence.

[0033] Specifically, the signal quality is unstable under low displacement and low pressure difference. According to the multiple change significances in the change significance sequence, the actually collected downhole pulse signal sequence and the predicted downhole pulse signal sequence obtained by prediction are fused to achieve the purpose of auxiliary noise reduction. In the fusion process, different weights are assigned to different signals to ensure the accuracy and stability of the fusion results and improve the reliability of the overall signal sequence. After the fusion is completed, a fused downhole pulse signal sequence is obtained. The signal quality of the fused downhole pulse signal sequence is improved, and the signal-to-noise ratio is improved. The higher the signal-to-noise ratio, the less noise components in the signal and the better the signal quality. Due to the low displacement and low pressure difference, the signal is easily interfered by the background environmental noise. There may be a certain amount of noise in the fused downhole pulse signal sequence. In order to further improve the signal quality in the fused downhole pulse signal sequence and remove noise and unnecessary interference, filtering algorithms such as low-pass filtering, median filtering, Kalman filtering and other technologies are used to filter the fused downhole pulse signal sequence. Through signal fusion and filtering and noise removal processing, the signal quality of the fused while-drilling pulse signal sequence can be improved, and finally the noise reduction of the low-displacement and low-pressure difference while-drilling pulse signal can be achieved, and the quality and reliability of the low-displacement and low-pressure difference while-drilling pulse signal can be improved, thus ensuring the real-time and accuracy of the drilling operation.

[0034] Furthermore, according to multiple change significances in the change significance sequence, the drilling-while-drilling pulse signal sequence and the predicted drilling-while-drilling pulse signal sequence are fused to obtain a fused drilling-while-drilling pulse signal sequence, including: obtaining the change significances of the signal timestamps of the multiple preceding signal groups in the change significance sequence, and calculating the mean to obtain multiple signal group change significances; according to the multiple signal group change significances, performing correction calculation on the preset prediction signal weights to obtain multiple corrected prediction signal weights, wherein the preset prediction signal weight is 0.4, and performing correction calculation on the preset prediction signal weights according to the ratio of the mean of the change significance sequence and the change significance of each signal group; according to the multiple corrected prediction signal weights, multiple original signal weights are calculated, and the signals in the drilling-while-drilling pulse signal sequence and the predicted drilling-while-drilling pulse signal sequence are weightedly calculated and fused to obtain multiple fused drilling-while-drilling pulse signals of multiple signal timestamps to obtain a fused drilling-while-drilling pulse signal sequence.

[0035] Specifically, in the process of fusing the drilling pulse signal sequence and the predicted drilling pulse signal sequence, the change significance of the signal timestamps of multiple preceding signal groups in the change significance sequence is obtained, the change significance of all timestamps in each preceding signal group is added, divided by the data of the timestamp, and the mean of the change significance of the multiple preceding signal groups is calculated. According to the mean value of the change significance of multiple timestamps before the timestamp of each predicted drilling pulse signal, the weight of the predicted drilling pulse signal in the fusion is allocated. The larger the mean value, the poorer the signal quality of the previous timestamps, and the smaller the weight of the predicted drilling pulse signal in the fusion. According to the calculation results, the change significance of multiple signal groups is obtained. The signal group change significance is a quantitative indicator to measure the degree of fluctuation of the signal over a period of time. The initial preset predicted signal weight value is set to 0.4. The preset predicted signal weight value indicates the contribution of the predicted signal in the fusion when there is no change significance information. Then, using the change degree of multiple signal groups, the ratio of the mean of the change significance sequence and the change degree of each signal group is multiplied by the preset prediction signal weight to complete the correction calculation of the preset prediction signal weight. The signal predicted at the time stamp with more displacement and pressure difference changes has a lower prediction signal quality and a smaller weight, thereby improving the quality of the fused signal. After obtaining the corrected multiple corrected prediction signal weights, the corresponding corrected prediction signal weight is subtracted from 1 to calculate and obtain multiple original signal weights. Finally, the signals in the drilling pulse signal sequence and the predicted drilling pulse signal sequence are respectively multiplied by the corresponding original signal weights, and the multiple calculation results are added to obtain the fused multiple fused drilling pulse information. This process is repeated until the fusion of all time stamp signals in the drilling pulse signal sequence and the predicted drilling pulse signal sequence is completed, and finally the fused drilling pulse signal sequence is obtained. By using the multiple change significances in the change significance sequence to perform dynamic weight fusion processing on the drilling pulse signal sequence and the predicted drilling pulse signal sequence, the accuracy and reliability of the fused signal are further improved, and the quality of the drilling pulse signal with low displacement and low pressure difference is improved.

[0036] Furthermore, filtering is performed on the fused while-drilling pulse signal sequence, including: extracting multiple fused while-drilling pulse signals in the fused while-drilling pulse signal sequence; filtering the multiple fused while-drilling pulse signals to obtain a noise-reduced while-drilling pulse signal sequence.

[0037] Specifically, multiple fused Drilling pulse signals are extracted from the fused Drilling pulse signal sequence. The multiple fused Drilling pulse signals are obtained after weighted fusion of the original Drilling pulse signal and the predicted Drilling pulse signal. They have high signal quality and signal-to-noise ratio and can more accurately reflect the drilling status. Then, the multiple fused Drilling pulse signals are filtered to further improve the signal quality. For example, the multiple fused Drilling pulse signals are pre-processed first, including amplification, impedance matching, etc., to ensure the readability and accuracy of the signal. Then, one or more filtering techniques such as low-pass filtering, median filtering or Kalman filtering are used to implement noise reduction filtering of the multiple fused Drilling pulse signals. The low-pass filter can effectively remove high-frequency noise, while the median filter can process burst noise in the pulse signal. The Kalman filter combines signal prediction and measurement to provide more accurate signal estimation. High-frequency noise and burst interference are effectively removed by noise reduction, thereby improving the clarity and reliability of the signal. Finally, a noise-reduced while-drilling pulse signal sequence is obtained based on multiple fused while-drilling pulse signals after noise reduction, and the noise-reduced while-drilling pulse signal sequence includes the while-drilling pulse signal data after filtering and noise reduction processing. Through the fusion noise reduction processing, the noise in the while-drilling pulse signal is effectively reduced under the working conditions of low displacement and low pressure difference, and the quality and reliability of the while-drilling pulse signal are improved, thereby ensuring the safety and effectiveness of the drilling operation.

[0038] Embodiment 2 is based on the same inventive concept as the method for reducing noise of a low-displacement and low-pressure-difference pulse signal while drilling in the above embodiment. Figure 2 As shown, the present application provides a low-displacement and low-pressure difference while drilling pulse signal noise reduction system, wherein the system comprises:

[0039] The data acquisition module 11 is used to collect continuous while-drilling pulse signals under low-displacement and low-pressure differential operating conditions to obtain a while-drilling pulse signal sequence, and to collect continuous displacement information and pressure differential information to obtain a displacement information sequence and a pressure differential information sequence;

[0040] A data analysis module 12, the data analysis module 12 is used to perform displacement change significance analysis and pressure difference change significance analysis according to the displacement information sequence and the pressure difference information sequence, obtain a displacement change significance sequence and a pressure difference change significance sequence, and process to obtain a change significance sequence;

[0041] A data prediction module 13, the data prediction module 13 is used to predict the drilling pulse signal at each moment according to the drilling pulse signal sequence, and obtain a predicted drilling pulse signal sequence;

[0042] The data fusion module 14 is used to fuse the drilling-while-drilling pulse signal sequence and the predicted drilling-while-drilling pulse signal sequence according to multiple change significances in the change significance sequence to obtain a fused drilling-while-drilling pulse signal sequence, and filter the fused drilling-while-drilling pulse signal sequence to obtain a noise-reduced drilling-while-drilling pulse signal sequence.

[0043] Furthermore, the data acquisition module 11 is used to perform the following steps: under low displacement and low pressure difference operating conditions, collect continuous downhole pulse signals to obtain a downhole pulse signal sequence, and extract a signal timestamp sequence; according to the signal timestamp sequence, collect the displacement information and pressure difference information at each signal timestamp to obtain a displacement information sequence and a pressure difference information sequence.

[0044] Furthermore, the data analysis module 12 is used to perform the following steps: perform displacement change significance analysis based on the displacement information sequence to obtain a displacement change significance sequence; perform pressure difference change significance analysis based on the pressure difference information sequence to obtain a pressure difference change significance sequence; perform weighted calculation on multiple displacement change significances and multiple pressure difference change significances within the displacement change significance sequence and the pressure difference change significance sequence to obtain a change significance sequence.

[0045] Furthermore, the data analysis module 12 is also used to perform the following steps: select the first displacement information in the displacement information sequence, extract M adjacent displacement information adjacent to the first displacement information, and randomly extract M comparative displacement information in the displacement information sequence, where M is a positive integer; calculate the difference between the mean of the M adjacent displacement information and the mean of the M comparative displacement information as the first displacement change significance; continue to perform displacement change significance analysis on other displacement information in the displacement information sequence, obtain multiple displacement change significances, and generate a displacement change significance sequence.

[0046] Furthermore, the data prediction module 13 is used to perform the following steps: based on the historical data of low-displacement and low-pressure difference downhole pulse signals, a set of sample preamble signal groups is collected, and the downhole pulse signals after each sample preamble signal group are collected to obtain a sample prediction signal set; the sample preamble signal group set and the sample prediction signal set are used to train a downhole pulse signal predictor; multiple preamble signal groups are selected in the downhole pulse signal sequence, and input into the downhole pulse signal predictor to predict the predicted signals of the signal timestamps after the multiple preamble signal groups to obtain a predicted downhole pulse signal sequence.

[0047] Furthermore, the data fusion module 14 is used to perform the following steps: obtain the change significance of the signal timestamps of the multiple preceding signal groups within the change significance sequence, and calculate the mean to obtain multiple signal group change significances; perform correction calculation on the preset prediction signal weights according to the multiple signal group change significances to obtain multiple corrected prediction signal weights, wherein the preset prediction signal weight is 0.4, and perform correction calculation on the preset prediction signal weights according to the ratio of the mean of the change significance sequence and the change significance of each signal group; perform calculation to obtain multiple original signal weights according to the multiple corrected prediction signal weights, perform weighted calculation and fusion on the signals within the downhole pulse signal sequence and the predicted downhole pulse signal sequence, obtain multiple fused downhole pulse signals of multiple signal timestamps, and obtain a fused downhole pulse signal sequence.

[0048] Furthermore, the data fusion module 14 is further configured to perform the following steps: extracting multiple fused while drilling pulse signals in the fused while drilling pulse signal sequence; and filtering the multiple fused while drilling pulse signals to obtain a noise-reduced while drilling pulse signal sequence.

[0049] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0050] This specification and the drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A method for reducing noise of low-displacement and low-pressure-difference pulse signals while drilling, characterized in that: The method comprises: Under the working condition of low displacement and low pressure difference, continuous while-drilling pulse signals are collected to obtain a while-drilling pulse signal sequence, and continuous displacement information and pressure difference information are collected to obtain a displacement information sequence and a pressure difference information sequence; According to the displacement information sequence and the pressure difference information sequence, a displacement change significance analysis and a pressure difference change significance analysis are performed to obtain a displacement change significance sequence and a pressure difference change significance sequence, and the change significance sequence is obtained by processing; According to the drilling pulse signal sequence, respectively predicting the drilling pulse signal at each moment to obtain a predicted drilling pulse signal sequence; According to multiple change significances in the change significance sequence, the drilling-while-drilling pulse signal sequence and the predicted drilling-while-drilling pulse signal sequence are fused to obtain a fused drilling-while-drilling pulse signal sequence, and the fused drilling-while-drilling pulse signal sequence is filtered to obtain a noise-reduced drilling-while-drilling pulse signal sequence.

2. The low-displacement and low-pressure-difference pulse signal noise reduction method while drilling according to claim 1, characterized in that: Under the working condition of low displacement and low pressure difference, continuous pulse signals while drilling are collected to obtain pulse signal sequences while drilling, and continuous displacement information and pressure difference information are collected to obtain displacement information sequences and pressure difference information sequences, including: Under the working conditions of low displacement and low pressure difference, continuous while-drilling pulse signals are collected to obtain the while-drilling pulse signal sequence, and the signal time stamp sequence is extracted; According to the signal timestamp sequence, the displacement information and the pressure difference information at each signal timestamp are collected to obtain a displacement information sequence and a pressure difference information sequence.

3. The low-displacement and low-pressure-difference pulse signal noise reduction method while drilling according to claim 1, characterized in that: According to the displacement information sequence and the pressure difference information sequence, a displacement change significance analysis and a pressure difference change significance analysis are performed, including: According to the displacement information sequence, performing displacement change significance analysis to obtain a displacement change significance sequence; According to the pressure difference information sequence, performing pressure difference change significance analysis to obtain a pressure difference change significance sequence; A weighted calculation is performed on a plurality of displacement change significances and a plurality of pressure difference change significances in the displacement change significance sequence and the pressure difference change significance sequence to obtain a change significance sequence.

4. The low-displacement and low-pressure-difference pulse signal noise reduction method while drilling according to claim 3, characterized in that: According to the displacement information sequence, a displacement change significance analysis is performed, including: Selecting first displacement information in the displacement information sequence, extracting M adjacent displacement information adjacent to the first displacement information, and randomly extracting M comparative displacement information in the displacement information sequence, where M is a positive integer; Calculating the difference between the mean of the M adjacent displacement information and the mean of the M comparative displacement information as the first displacement change significance; Continue to perform displacement change significance analysis on other displacement information in the displacement information sequence to obtain multiple displacement change significances and generate a displacement change significance sequence.

5. The low-displacement and low-pressure-difference pulse signal noise reduction method while drilling according to claim 1, characterized in that: According to the drilling pulse signal sequence, the drilling pulse signal at each moment is predicted respectively to obtain a predicted drilling pulse signal sequence, including: According to the historical data of the low-displacement and low-pressure differential drilling pulse signal, a sample pre-order signal group set is collected, and the drilling pulse signal after each sample pre-order signal group is collected to obtain a sample prediction signal set; Using the sample pre-sequence signal group set and the sample prediction signal set to train a pulse signal predictor while drilling; A plurality of preamble signal groups are selected in the while-drilling pulse signal sequence, and are input into the while-drilling pulse signal predictor to predict the predicted signals of the signal timestamps after the plurality of preamble signal groups to obtain the predicted while-drilling pulse signal sequence.

6. The low-displacement and low-pressure-difference pulse signal noise reduction method while drilling according to claim 5, characterized in that: According to a plurality of change significances in the change significance sequence, the drilling-while-drilling pulse signal sequence and the predicted drilling-while-drilling pulse signal sequence are fused to obtain a fused drilling-while-drilling pulse signal sequence, including: Obtaining the change significance of the signal timestamps of the plurality of preceding signal groups within the change significance sequence, and calculating the mean to obtain the change significance of the plurality of signal groups; According to the change degrees of the multiple signal groups, the preset prediction signal weights are modified and calculated to obtain multiple modified prediction signal weights, wherein the preset prediction signal weight is 0.4, and the preset prediction signal weights are modified and calculated according to the ratio of the mean of the change significance sequence and the change degree of each signal group; According to the multiple corrected predicted signal weights, multiple original signal weights are calculated, and the signals in the drilling pulse signal sequence and the predicted drilling pulse signal sequence are weightedly calculated and fused to obtain multiple fused drilling pulse signals with multiple signal timestamps, thereby obtaining a fused drilling pulse signal sequence.

7. The low-displacement and low-pressure-difference pulse signal noise reduction method while drilling according to claim 5, characterized in that: The fused while-drilling pulse signal sequence is filtered, including: Extracting a plurality of fused while-drilling pulse signals in the fused while-drilling pulse signal sequence; The plurality of fused while-drilling pulse signals are filtered to obtain a noise-reduced while-drilling pulse signal sequence.

8. A low-displacement and low-pressure-difference pulse signal noise reduction system while drilling, characterized in that: The steps for implementing the method according to any one of claims 1 to 7 include: A data acquisition module, wherein the data acquisition module is used to collect continuous while-drilling pulse signals under low-displacement and low-pressure differential operating conditions to obtain a while-drilling pulse signal sequence, and to collect continuous displacement information and pressure differential information to obtain a displacement information sequence and a pressure differential information sequence; A data analysis module, the data analysis module is used to perform displacement change significance analysis and pressure difference change significance analysis according to the displacement information sequence and the pressure difference information sequence, obtain a displacement change significance sequence and a pressure difference change significance sequence, and process to obtain a change significance sequence; A data prediction module, the data prediction module is used to predict the drilling pulse signal at each moment according to the drilling pulse signal sequence, and obtain a predicted drilling pulse signal sequence; A data fusion module is used to fuse the drilling-while-drilling pulse signal sequence and the predicted drilling-while-drilling pulse signal sequence according to multiple change significances in the change significance sequence to obtain a fused drilling-while-drilling pulse signal sequence, and to filter the fused drilling-while-drilling pulse signal sequence to obtain a noise-reduced drilling-while-drilling pulse signal sequence.

Images