Oil and gas well underwater sand production detection device and method

Abstract

The invention discloses an oil and gas well underwater sand production detection device and method.The device comprises a sand production signal sensing module, a sand production signal conditioning module, a sand production signal collecting module and a sand production signal analyzing and processing module.The sand production signal sensing module is used for sensing and measuring high-frequency vibration signals generated by sand grains; the sand production signal conditioning module is used for conditioning the high-frequency vibration signal; the sand production signal acquisition module is used for acquiring the high-frequency vibration signal conditioned by the sand production signal conditioning module in real time; and the sand production signal analyzing and processing module is used for obtaining the high-frequency vibration signals collected by the sand production signal collecting module and carrying out analysis, noise reduction, feature extraction and sand production information inversion on the high-frequency vibration signals on the basis of a self-adaptive mode decomposition method to obtain sand production signals excited when sand grains collide with the pipe wall of the outlet pipe of the subsea Christmas tree. Therefore, accurate identification of the underground sand production information of the oil and gas well is realized by analyzing the sand-containing signal characteristics of the outlet pipeline of the subsea Christmas tree.

Description

TECHNICAL FIELD

[0001] The present application relates to a kind of oil and gas well underwater sand production detection device and method, and it is related to oil and gas well sand production detection field. BACKGROUND

[0002] It is one of the key processes to realize reservoir loose safe and efficient development of deepwater oil and gas field to detect the sand production condition of underwater Christmas tree outlet pipe. At the outlet pipe of Christmas tree, the continuous impact of sand particles carried by fluid on the pipe can easily cause pipe blockage and erosion, greatly limiting the efficiency and safety of pipeline transportation. Therefore, real-time detection of sand production condition of underwater Christmas tree outlet pipe is of great significance to ensure efficient and safe pipeline transportation.

[0003] The prior art discloses a kind of detection system and method for sand concentration and particle size distribution of solid-liquid two-phase flow based on mixed frequency acoustics, which includes mixed frequency signal generating unit, ultrasonic piezoelectric drive control system, ultrasonic transducer and operation unit etc. The advantage of the system / method is that the sand particle information in the liquid-solid two-phase flow system can be accurately detected. For gas-liquid two-phase flow system, the ultrasonic detection accuracy is greatly limited due to the influence of strong gas-liquid coupling, which is not suitable for detecting the sand content in gas-liquid-solid flow system.

[0004] The prior art also discloses a kind of detection method for sand particle velocity of gas-solid two-phase flow based on ESMD, including electrostatic signal acquisition module based on electrostatic sensor and electrostatic signal decomposition and characterization module based on ESMD. The advantage of this method is that it can effectively detect the sand particles in the gas-solid two-phase flow system, but due to the limitation of electrostatic sensor detection principle, it cannot accurately detect the sand particles in the liquid-containing flow system. SUMMARY

[0005] The present application aims to solve at least one of the technical problems existing in the prior art. To this end, in view of the above problems, the purpose of the present application is to provide an oil and gas well underwater sand production detection device and method capable of accurately identifying the sand production information in the oil and gas well.

[0006] In order to achieve the above-mentioned application purpose, the technical scheme adopted by the present application is:

[0007] In a first aspect, the present application provides an oil and gas well underwater sand production detection device, which comprises a sand production signal sensing module, a sand production signal conditioning module, a sand production signal acquisition module and a sand production signal analysis and processing module, wherein:

[0008] The sand production signal sensing module is used to sense and measure the high-frequency vibration signal generated by sand particles;

[0009] The sand production signal conditioning module is used to condition the high-frequency vibration signal;

[0010] The sand production signal acquisition module is configured to acquire the high-frequency vibration signal processed by the sand production signal conditioning module in real time.

[0011] The sand production signal analysis processing module is configured to acquire the high-frequency vibration signal collected by the sand production signal acquisition module, and analyze, denoise, extract features, and invert sand production information of the high-frequency vibration signal based on an adaptive modal decomposition method, to obtain a sand production signal excited by sand particles colliding with a pipe wall at an outlet pipe of a subsea Christmas tree.

[0012] In some possible implementation manners, the sand production signal sensing module adopts a vibration sensor; or / and, the sand production signal conditioning module adopts a low-impedance signal conditioner, which can directly convert the measured sand production signal into a voltage signal and perform amplification and filtering processing; or / and, the sand production signal acquisition module adopts an acquisition card that provides multi-channel differential analog input and has flexible channel conversion range.

[0013] In some possible implementation manners, the sand production signal analysis processing module includes a preprocessing module, a main frequency response analysis module, an adaptive decomposition module, a multi-scale feature analysis module, and a sand production information feature extraction and inversion module, wherein:

[0014] The preprocessing module is configured to pre-process the collected high-frequency vibration signal, and determine whether a frequency response range of the collected high-frequency vibration signal meets a preset requirement.

[0015] The main frequency response analysis module is configured to perform time domain analysis, frequency domain analysis, and power spectrum analysis on the high-frequency vibration signal meeting the preset requirement, to determine a main frequency response range of the vibration signal generated by sand particles at the outlet pipe of the subsea Christmas tree.

[0016] The adaptive decomposition module is configured to decompose the high-frequency vibration signal meeting the main frequency response range requirement by using an adaptive decomposition algorithm to obtain a plurality of IMF subsequences.

[0017] The multi-scale feature analysis module calculates pulse factors and integral energies of the obtained IMF sequences, preferably selects a sequence with a high pulse factor and a high integral energy as an optimal sequence, performs spatial reconstruction on the optimal IMF sequence, and obtains a data set that can be used to represent a sand production condition.

[0018] The sand production information feature extraction and inversion module extracts two-dimensional response feature information of the sand production signal based on the data set representing the sand production condition, gives different weights to the extracted feature information of different spatial positions to perform sand production information feature fusion, and obtains pipeline sand production information.

[0019] In some possible implementation manners, the sand production signal analysis processing module further has a watchdog principle-based system fault self-repairing module built in, to avoid the problem that the system cannot be restarted after a fault.

[0020] In some possible implementation manners, the device further has a communication module, a power supply module and a remote control module.

[0021] The communication module uses a multi-core water-tight communication cable to output sand production information in real time through an RS485 serial port.

[0022] The power supply module uses a dual-voltage transformer power supply group, which shares a micro water-tight cable with the communication module, to supply power to the sand production signal conditioning module and the sand production signal analysis processing module.

[0023] The remote control module uses an Ethernet ultra-small water-tight cable, and the remote control module is used for communication between the platform central control and the sand production signal analysis processing module.

[0024] In some possible implementation manners, the device further has a shell made of cylindrical titanium alloy, and each module is fixedly assembled in the shell through a small-sized board card. The sand production signal sensing module is arranged in rigid contact with the pipe wall of the underwater Christmas tree outlet pipe at the bottom of the shell, and the shell and the underwater Christmas tree outlet pipe are fixedly installed through a clamp.

[0025] In some possible implementation manners, an aluminum heat sink is further arranged in the shell, and the aluminum heat sink is in contact with the shell, so as to improve the heat dissipation capacity of the underwater sand production detector. Alternatively or additionally, a handle is designed at the top of the shell, to facilitate underwater installation and maintenance of the underwater sand production detection device of the oil and gas well by an ROV.

[0026] In a second aspect, the present application further provides an underwater sand production detection method of the underwater sand production detection device of the oil and gas well, comprising:

[0027] The underwater sand production detection device of the oil and gas well is fixed on the underwater Christmas tree outlet pipe by an ROV, to ensure that the lower end of the sand production signal sensing module is in rigid contact with the pipe wall.

[0028] Fluids carrying sand particles in the underwater Christmas tree outlet pipe impact the pipe wall to generate a vibration signal, which is sensed by the sand production signal sensing module in contact with the outer wall of the pipe.

[0029] The weak vibration signal sensed and measured by the sand production signal sensing module is transmitted to the signal conditioning module through a high-frequency low-noise communication cable for adaptive filtering and amplification.

[0030] The high-frequency vibration signal output by the sand production signal conditioning module is transmitted to the sand production signal acquisition module through a high-frequency low-noise communication cable.

[0031] The high-frequency sand production vibration signal collected by the sand production signal collection module is analyzed, processed and inverted in real time by the sand production signal analysis and processing module to obtain a sand production signal excited by sand particles colliding with the pipe wall at the outlet pipe of the underwater Christmas tree;

[0032] The obtained sand production information is sequentially transmitted to the platform end through RS485, multi-core water-tight communication cable, SCM and module umbilical cable to complete the detection of sand production information.

[0033] In some possible embodiments, the high-frequency sand production vibration signal collected by the sand production signal collection module is analyzed, processed and inverted in real time by the sand production signal analysis and processing module to obtain a sand production signal excited by sand particles colliding with the pipe wall at the outlet pipe of the underwater Christmas tree, including:

[0034] The pre-processing module pre-processes the collected high-frequency vibration signal, compares the frequency response range of the collected sand production signal, and if the main frequency response range is higher than the set value, the next step is performed, otherwise it is indicated that the collected sand production signal is severely distorted, and the self-checking of each electronic module of the detection device is stopped;

[0035] The main frequency response analysis module analyzes the main frequency response of the pre-processed sand production signal, calculates the time domain, frequency domain and power spectrum response distribution characteristics of the sand production signal, and determines the main frequency response range of the sand production signal based on the amplitude-frequency response characteristics of the calculated sand production signal frequency domain and power spectrum;

[0036] The self-adaptive decomposition module uses the CEEMDAN algorithm to perform self-adaptive decomposition on the sand production signal in the main frequency response range based on the band-pass filtered signal, to obtain an IMF component representing sand particle information;

[0037] The multi-scale feature analysis module calculates the pulse factor and integral energy of each IMF sequence, and optimally selects a sequence with high pulse factor and high integral energy as the optimal sequence;

[0038] Statistical feature analysis is performed on the selected IMF sequence, and the sample entropy and Hurst index of each sub-sequence are calculated; the selected IMF sub-sequence is divided into three frequency scales according to the "high sample entropy-low Hurst", "medium sample entropy-medium Hurst" and "low sample entropy-high Hurst" division methods, and the IMF sequence of low frequency scale is removed;

[0039] The IMF sequence of low frequency scale is removed and spatially reconstructed to obtain a data set for representing the sand production condition;

[0040] The data set characterizing the sand production condition is input into the sand production information feature extraction and inversion module, the input data set is simultaneously input into the CNN module and the GRU module, so as to simultaneously extract sand production signal double-dimension response feature information, and through the built-in self-attention mechanism algorithm, different weights are given to the response features of different spatial positions to be fused, and the pipeline sand production information is obtained.

[0041] The present application has the following characteristics due to the above technical scheme:

[0042] 1. The present application realizes accurate identification of the sand production information of the oil and gas well by analyzing the sand signal characteristics of the underwater Christmas tree outlet pipeline.

[0043] 2. The present application can realize real-time detection of the sand particle information of the underwater Christmas tree outlet pipeline by integrating the sand production signal sensing module, the sand production signal conditioning module, the sand production signal acquisition module and the sand production signal analysis and processing module.

[0044] 3. The present application proposes a multi-scale statistical feature analysis method to analyze the collected sand production signal, and compared with the existing method, the advantages are that: by calculating the pulse factor and integral energy of each IMF component, the optimal IMF sequence representing the sand particle vibration information can be preliminarily determined; by calculating the sample entropy and Hurst index of each IMF component, the fractal characteristics of each IMF component are determined, and the vibration signal of the sand particle excitation is divided in three scales through the division mode of "high sample entropy-low Hurst", "medium sample entropy-medium Hurst" and "low sample entropy-high Hurst", wherein the low sample entropy-high Hurst scale signal is the fluid noise signal, and the fluid noise signal is automatically removed, so that the noise reduction of the sand production signal of the underwater Christmas tree outlet pipeline can be realized.

[0045] In summary, the present application can be widely applied to the detection of underwater sand production of oil and gas wells. BRIEF DESCRIPTION OF DRAWINGS

[0046] Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The accompanying drawings are included to provide a description of the preferred embodiments and are not intended to limit the scope of the application. Throughout the drawings, same reference numerals are used for same components. In the drawings:

[0047] Figure 1 The figure is a schematic diagram of the principle of the oil and gas well underwater sand production detection device of the embodiment of the present application;

[0048] Figure 2 The figure is a schematic diagram of the structure of the oil and gas well underwater sand production detection device of the embodiment of the present application;

[0049] Figure 3A schematic diagram of a sand production signal analysis processing module according to an embodiment of the present application;

[0050] Figure 4 A flow chart of an operation of the oil and gas well underwater sand production detection device according to an embodiment of the present application;

[0051] Figure 5 A flow chart of an operation of the oil and gas well underwater sand production detection method according to an embodiment of the present application. DETAILED DESCRIPTION

[0052] It is to be understood that the terminology used herein is for the purpose of describing particular example embodiments only and is not intended to be limiting. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. The terms "comprises", "comprising", "includes", "including" and "has" are inclusive and therefore specify the presence of stated features, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, steps, operations, elements, components, and / or groups thereof. The method steps, processes, and operations described herein are not to be construed as necessarily requiring their performance in the particular order in which they are described, unless specifically identified as an order dependent step. It is also to be understood that additional or alternative steps can be employed.

[0053] Although the terms first, second, third, and the like can be used herein to describe various elements, components, regions, layers and / or sections, these elements, components, regions, layers and / or sections should not be limited by these terms. These terms can be only used to distinguish one element, component, region, layer or section from another region, layer or section. Terms such as "first", "second", and other numerical terms when used herein do not imply a sequence or order unless clearly indicated by the context. Thus, a first element, component, region, layer or section discussed below could be termed a second element, component, region, layer or section without departing from the teachings of the example embodiments.

[0054] Spatially relative terms, such as "inner", "outer", "beneath", "below", "lower", "above", "upper", and the like, can be used herein for ease of description to describe one element or feature's relationship to another element(s) or feature(s) as illustrated in the figures. The spatially relative terms are intended to encompass different orientations of the device in use or operation in addition to the orientations depicted in the figures.

[0055] It is of great significance to realize real-time detection of sand production of the underwater Christmas tree outlet pipe for guaranteeing efficient and safe transportation of the pipeline. The present application provides an oil and gas well underwater sand production detection device and method, which comprises a sand production signal sensing module, a sand production signal conditioning module, a sand production signal collecting module and a sand production signal analysis and processing module. The sand production signal sensing module is used for sensing and measuring the high-frequency vibration signal generated by sand particles. The sand production signal conditioning module is used for conditioning the high-frequency vibration signal. The sand production signal collecting module is used for collecting the high-frequency vibration signal conditioned by the sand production signal conditioning module in real time. The sand production signal analysis and processing module is used for obtaining the high-frequency vibration signal collected by the sand production signal collecting module, and analyzing, denoising, feature extracting and sand production information inversion of the high-frequency vibration signal based on an adaptive modal decomposition method, so as to obtain the sand production signal excited by the sand particle collision with the pipe wall of the underwater Christmas tree outlet pipe. Therefore, the present application can realize real-time and accurate detection of the underwater sand production condition of the oil and gas well.

[0056] Exemplary embodiments of the present application will be described in greater detail below with reference to the accompanying drawings. While exemplary embodiments of the present application are shown in the drawings, it is understood that the present application can be embodied in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the application to those skilled in the art.

[0057] Embodiment one: as shown in the figure, the present embodiment provides an oil and gas well underwater sand production detection device, which comprises a shell 1, and a sand production signal sensing module 2, a sand production signal conditioning module 3, a sand production signal collecting module 4 and a sand production signal analysis and processing module 5 arranged in the shell 1, wherein: Figure 1

[0058] The sand production signal sensing module 2 is arranged in rigid contact with the pipe wall of the underwater Christmas tree outlet pipe at the bottom of the shell 1, and is used for sensing and measuring the high-frequency vibration signal generated by the sand particles of the underwater Christmas tree outlet pipe;

[0059] The sand production signal conditioning module 3 is used for conditioning the high-frequency vibration signal;

[0060] The sand production signal collecting module 4 is used for collecting the high-frequency vibration signal outputted after conditioning by the sand production signal conditioning module 3 in real time;

[0061] The sand production signal analysis and processing module 5 is used for obtaining the high-frequency vibration signal collected by the sand production signal collecting module 4, and analyzing, denoising, feature extracting and sand production information inversion of the high-frequency vibration signal based on an adaptive modal decomposition method, so as to obtain the sand production signal excited by the sand particle collision with the pipe wall of the underwater Christmas tree outlet pipe.

[0062] ​In a preferred embodiment of the present application, the sand production signal sensing module 2 can adopt a vibration sensor without external power supply, which has the characteristics of ultra-low noise, wide dynamic frequency response range, wide working temperature range, high sensitivity, high signal-to-noise ratio, strong anti-interference, etc.

[0063] Further, the sand production signal sensing module 2 selected in the embodiment is a piezoelectric vibration sensor based on M5 top end connection mode, which has wide frequency response, ultra-low noise and high sensitivity, and the like, and is not limited to this.

[0064] In a preferred embodiment of the present application, the sand production signal conditioning module 3 can adopt a low-impedance signal conditioner with a built-in dual-gain scheme, which can directly convert the measured sand production signal into a voltage signal and amplify and filter the electrical signal.

[0065] Further, the sand production signal conditioning module 3 adopted in the embodiment is a low-impedance signal conditioner with integrated high-pass filtering, voltage amplification and bias compensation functions.

[0066] In a preferred embodiment of the present application, the sand production signal acquisition module 4 can adopt a high-precision acquisition card that provides multi-channel differential analog input and has flexible channel conversion range.

[0067] Further, the sand production signal acquisition module 4 includes an ADC module, a self-calibration module, an FPGA control module and a USB controller, and can realize synchronous acquisition of multi-channel differential signals and buffer the collected data through SDRAM.

[0068] In a preferred embodiment of the present application, as shown in Figure 3 The sand production signal analysis and processing module 5 includes a preprocessing module 51, a main frequency response analysis module 52, an adaptive decomposition module 53, a multi-scale feature analysis module 54 and a sand production information feature extraction and inversion module 55, wherein:

[0069] The preprocessing module 51 is used for preprocessing the collected high-frequency vibration signal and judging whether the frequency response range of the collected high-frequency vibration signal meets the preset requirement;

[0070] The main frequency response analysis module 52 is used for adopting time domain analysis, frequency domain analysis and power spectrum analysis on the high-frequency vibration signal meeting the preset requirement, determining the main frequency response range of the high-frequency vibration signal generated by the sand particles at the outlet pipe of the underwater Christmas tree, and providing a reference frequency band for sand production signal noise reduction;

[0071] The adaptive decomposition module 53 is used for decomposing the high-frequency vibration signal meeting the main frequency response range requirement by using an adaptive decomposition algorithm and obtaining a plurality of IMF sequences;

[0072] The multi-scale feature analysis module 54 calculates the pulse factor and integral energy of each IMF sequence, preferably takes the sequence with high pulse factor and high integral energy as the optimal sequence, and reconstructs the optimal IMF sequence in space to obtain a data set for representing the sand production condition;

[0073] The sand production information feature extraction and inversion module 55 is used for extracting sand production signal two-dimensional response feature information from the data set representing the sand production condition, and fusing the feature information of different spatial positions by giving different weights to obtain the pipeline sand production information.

[0074] Further, the sand production signal analysis processing module 5 also has a watchdog principle-based detection system fault self-repairing module, wherein the watchdog (Watchdog Timer, WDT) is a hardware or software level fault tolerance mechanism, which realizes fault repair by monitoring the system running state in a timing manner and triggering system reset when detecting an exception. In the system, the main program needs to periodically send a signal to the watchdog when the system is running normally, which indicates that the system is running normally. The watchdog has an independent timer, and if the signal is not received within the preset time, it is determined that the system is faulty. When the timer overflows, the watchdog immediately restarts the CPU for fault repair, so as to avoid the problem that the system cannot be restarted after a fault.

[0075] In a preferred embodiment of the present application, the oil and gas well underwater sand production detection device further comprises a communication module 6, a power supply module 7 and a remote control module 8, wherein:

[0076] The communication module 6 adopts a multi-core water-tight communication cable processed by a high-temperature and high-pressure vulcanization process, and outputs the sand production information to the platform central control 9 in real time through an RS485 serial port. The RS485 interface can be combined with a balanced driver and a differential receiver, which has strong anti-noise interference. The RS485 is a sand production information transmitter based on the Modbus communication protocol, and the transmission rate of the sand production signal can reach 10Mbps in the half-duplex working mode of the sending module.

[0077] The power supply module 7 adopts a dual-voltage power supply group, which shares the multi-core water-tight communication cable with the communication module 6. The external power supply module 10 receives the instructions of the platform central control 9 and supplies power to the sand production signal conditioning module 3 and the sand production signal analysis processing module 5 through the dual-voltage power supply group.

[0078] The remote control module 8 adopts an Ethernet super-small water-tight cable, which receives and transmits the instructions of the platform central control 9 to adjust and control the parameters of each module.

[0079] In a preferred embodiment of the present application, as shown in Figure 2As shown, each device of the oil and gas well underwater sand production detection device is processed through miniaturization and low power consumption, and is fixed and assembled through a small-sized board card. The shell 1 is made of cylindrical titanium alloy, each small-sized board card is armored with the cylindrical titanium alloy shell 1, and the shell 1 is fixed and installed with the outlet pipe of the underwater Christmas tree through a clamp. Specifically, each component is fixed and assembled in the shell 1 through a board card in the form of a bolt, and is armored through the cylindrical titanium alloy shell 1 with a pressure resistance of 5000 psi, and the armored body is fixed with the outer wall of the outlet pipe of the underwater Christmas tree through a clamp. The obtained sand production information is transmitted to the SCM underwater Christmas tree control module through a water-tight communication cable and an underwater electrical connector, and is finally transmitted to the platform central control 9 through a umbilical cable.

[0080] Further, an aluminum heat sink 11 is also provided, which is the same size as the small-sized sand production signal processor board card of the built-in sand production signal analysis processing module 5, and is in contact with the shell 1 to improve the heat dissipation capacity of the underwater sand production detector.

[0081] Further, as shown in Figure 2 The top of the shell 1 is designed with a D-shaped handle 12, which facilitates underwater installation and maintenance of the oil and gas well underwater sand production detection device through an ROV.

[0082] Embodiment two: as shown in Figures 3-5 The application also provides a method for underwater sand production detection, which comprises:

[0083] S1, initializing and self-checking the oil and gas well underwater sand production detection device on the ground, if each module can work normally, then executing the next step, otherwise, repeating the step;

[0084] S2, fixing the oil and gas well underwater sand production detection device at the elbow position of the outlet pipe of the underwater Christmas tree through a clamp by an ROV, and the lower end of the sand production signal sensing module 2 is in rigid contact with the pipe wall;

[0085] S3, the platform central control 9 controls the external power supply module 10 to supply power to the internal power supply module 23 of the underwater sand production detection device through a water-tight power supply cable, and each electronic module in the underwater sand production detection device is powered on and started;

[0086] S4, the platform central control 9 sends instructions to the sand production signal analysis processing module 5 through the remote control module 8 to set the sand production signal sampling frequency and other key detection parameters, and the underwater sand production detection device formally enters the working mode;

[0087] S5, the fluid carrying sand particles in the outlet pipe of the underwater Christmas tree impact the pipe wall to generate a high-frequency vibration signal, which is sensed by the sand production signal sensing module 2 in contact with the outer wall of the pipe;

[0088] S6, the weak vibration signal measured by the sand-out signal sensing module 2 is transmitted to the sand-out signal conditioning module 3 through a high-frequency low-noise communication cable for adaptive filtering, amplification and output;

[0089] S7, the high-frequency vibration signal output by the sand-out signal conditioning module 3 is transmitted to the sand-out signal acquisition module 4 through a high-frequency low-noise communication cable;

[0090] S8, the high-frequency sand-out vibration signal collected by the sand-out signal acquisition module 4 is analyzed, processed and inverted in real time by the sand-out signal analysis processing module 5 to obtain sand-out information.

[0091] In this embodiment, the high-frequency sand-out vibration signal collected by the sand-out signal acquisition module 4 is analyzed, processed and inverted in real time by the sand-out signal analysis processing module to obtain sand-out information, including:

[0092] S81, the collected high-frequency vibration signal is preprocessed by the preprocessing module 51 to determine the frequency response range of the collected sand-out signal. If the main frequency response range is higher than the set value, for example, 20 kHz, the next step is performed, otherwise it means that the collected sand-out signal is severely distorted, and the self-checking of each electronic module of the detection device is stopped;

[0093] S82, the main frequency response analysis module 52 analyzes the main frequency response of the high-frequency vibration signal within the frequency response range, and calculates the time domain, frequency domain and power spectrum response distribution characteristics of the high-frequency vibration signal.

[0094] In this embodiment, the frequency domain response distribution is calculated as follows:

[0095]

[0096] In the formula, N is the length of the sand-out signal, F k is the frequency spectrum function after Fourier transform, f n is the nth sample value of the time domain signal, k is the frequency domain index, is the DFT kernel function.

[0097] S83, based on the amplitude-frequency response characteristics of the sand-out signal frequency domain and power spectrum calculated, the main frequency response range of the sand-out signal is determined, wherein the purpose of determining the main frequency response range is to suppress the background noise interference of the sand-out signal and separate and extract the characteristic signal component representing the sand particle impact event.

[0098] In this embodiment, the collected sand particle vibration signal is analyzed by FFT frequency domain analysis method and power spectrum analysis method to determine the main frequency response reference range. Specifically, the sensor is fixed to one end of the metal pipe wall, a standard sand particle impact load is applied to the other end of the metal pipe wall, the vibration response signal is synchronously collected and the Fourier spectrum and power spectrum density distribution are calculated. By comparing the baseline spectrum characteristics under the non-impact state, the characteristic frequency band with significant energy increment is identified, and the main frequency response domain of the sand particle impact event is determined accordingly.

[0099] S84, band-pass filtering the sand-out signal in the non-main frequency response frequency band, wherein the band-pass filtering of the sand-out signal in the non-main frequency response frequency band is to eliminate noise and only keep the signal in the main frequency response frequency band, and the signal in the main frequency response reference range is processed in subsequent processing.

[0100] In this embodiment, the formula for band-pass filtering the sand-out signal in the non-main frequency response frequency band is:

[0101]

[0102] In the above formula, f L and f H respectively represent the lower and upper cut-off frequencies.

[0103] S85, the adaptive decomposition module 12 performs adaptive decomposition on the band-pass filtered signal using the CEEMDAN algorithm to obtain the IMF components representing the sand particle information.

[0104] In this embodiment, the sand-out signal is decomposed as follows:

[0105]

[0106] In the formula, a Gaussian white noise sequence is added to the original sand-out signal f(t), δ0 is the signal-to-noise ratio, and ω i is the white noise sequence added for the i-th time.

[0107] The decomposition is stopped until the residual error shows a monotonic trend, and the original sand-out signal is decomposed into m modal components and a residual term R(t), which is calculated as follows:

[0108]

[0109] S86, the multi-scale feature analysis module 54 calculates the pulse factor and integral energy of each IMF sequence obtained, and preferably selects the sequence with high pulse factor and high integral energy as the optimal sequence. The minimum energy threshold is set to 0.85 times the sum of the integral energies of all components, and the components with energy greater than the average energy are selected until the cumulative energy exceeds the minimum energy threshold, ensuring that the integral energy of the selected sub-sequence accounts for no less than 85% of the total integral energy.

[0110] S87, performing statistical feature analysis on the preferred IMF sequence, calculating sample entropy and Hurst index of each IMF sequence;

[0111] In this embodiment, the calculation process of sample entropy calculation is as follows:

[0112] By phase space reconstruction of each IMF sequence X, a space matrix Y is obtained:

[0113]

[0114] Based on this, the absolute value of the maximum difference of corresponding elements in vectors Y(i) and Y(j) is defined as their distance d(ij), and the number of d(ij)<r is B i Thus, the sample entropy calculation formula is as follows:

[0115]

[0116] In the formula, N is the length of the IMF sequence, and m is the modal dimension. In addition, the Hurst index is calculated by the mature formula, which can be directly calculated according to the literature formula, and will not be repeated here.

[0117] In this embodiment, the preferred IMF subsequence is divided into three frequency scales according to the "high sample entropy-low Hurst", "medium sample entropy-medium Hurst" and "low sample entropy-high Hurst" division methods. The larger the sample entropy is, the higher the complexity and uncertainty of the signal are, and the lower the signal is. The larger the Hurst value is, the stronger the signal persistence is, the lower the complexity is, and the lower the Hurst value is. High sample entropy-low Hurst value represents high complexity and uncertainty of the signal, representing the process of sand particles impacting the pipe wall in the flow process; medium sample entropy-medium Hurst value represents higher complexity and uncertainty of the signal, representing the signal generated by the sliding of sand particles at the bottom of the pipeline in the flow process; low sample entropy-high Hurst represents strong signal persistence, low complexity and regular signal, representing low-frequency flow noise in the signal. Corresponding to microscale, mesoscale and macroscale in turn, the macroscale signal is the fluid noise signal, the macroscale sequence is removed, the IMF sequence of low frequency scale is removed, and the noise of the sand out signal of the underwater Christmas tree outlet pipe can be realized. The high-frequency and sub-high-frequency scale IMF sequences are reconstructed, the reconstructed signal is the denoising signal of the original sand out signal, and the denoising signal of the original sand out signal of the underwater Christmas tree outlet pipe is obtained.

[0118] S88, the optimal IMF sequence of the aforementioned denoised sand production signal is reconstructed in space, and the specific process is as follows: first, the depth, height and width of the three-dimensional matrix are set, each position (i, j, k) of the three-dimensional matrix is traversed according to the depth-first principle, and the one-dimensional signal index corresponding to each coordinate point is calculated, finally the denoised one-dimensional sand production signal is filled into the position of the three-dimensional matrix according to the index, and a structured three-dimensional space matrix is obtained, that is, a data set for representing the sand production condition can be obtained.

[0119] S89, the data set for representing the sand production condition is input into the sand production information feature extraction and inversion module 55, and the input data set is simultaneously input into the CNN module and the GRU module to simultaneously extract the sand production signal two-dimensional response feature information.

[0120] In this embodiment, the sand production signal two-dimensional response feature information includes sand production signal spatial feature and sand production signal time delay feature.

[0121] In this embodiment, the CNN module is a convolutional neural network optimized based on an EluRelu activation function, which can extract sand production signal spatial dimension response information, avoid the interference of underwater Christmas tree outlet pipe gas-liquid fluid noise, and maximize the spatial feature extraction capability of the CNN.

[0122] In this embodiment, the GRU module is a gate-controlled recurrent unit optimized based on a regularization technique, which can effectively capture the time dimension feature information of the sand production signal. It is a gate-controlled recurrent unit optimized based on a regularization technique. This module has fewer parameter quantities and a simpler network structure, and significantly improves the fitting and extraction precision of the model for the time dimension feature information of the sand production signal.

[0123] S810, the sand production information feature extraction and inversion module 55 is built-in self-attention mechanism algorithm, and different weights are given to the time delay features of different spatial positions extracted in S89 to obtain the pipeline sand production information.

[0124] In this embodiment, the sand production information feature extraction and inversion model based on deep learning algorithm is provided, which first introduces a self-attention mechanism to adaptively fuse the sand production information extracted by the CNN and GRU algorithms, that is, the CNN-GRU-SATT algorithm architecture; the CNN model is a convolutional neural network optimized based on an EluRelu activation function, which can extract sand production signal spatial dimension response information; the GRU model is a gate-controlled recurrent unit optimized based on a regularization technique, which can effectively capture the time dimension feature information of the sand production signal; the SATT module is a sand production information feature fusion algorithm based on a self-attention mechanism, which gives different weights to the time delay features of the sand production signal time series at different spatial positions, and improves the recognition and inversion precision of the model for the sand production information.

[0125] Further, the self-attention mechanism core algorithm is as follows:

[0126] s t =tanh(W h ·h t +b h )

[0127] a ι =softmax(exp(s ι ) / ∑exp(s ι ))

[0128] s=∑a t ·h t

[0129] In the formula, s t is the dot product attention weight matrix between feature elements; W h is the weight matrix of the self-attention mechanism, and a t is the normalized dot product attention weight.

[0130] S9, the obtained sand information is sequentially transmitted to the platform central control 9 through the RS485, multi-core water-tight communication cable, SCM and umbilical cable 13, and the sand information detection and output are completed.

[0131] Each embodiment in the specification is described in a progressive manner, and the same or similar parts between each embodiment can be referred to each other. Each embodiment focuses on the difference from other embodiments. In the description of the specification, the description of the reference terms "one preferred embodiment", "further", "specifically", "in the embodiment", and the like means that the specific features, structures, materials or characteristics described in combination with the embodiment or example are contained in at least one embodiment or example of the embodiments of the specification. In the specification, the illustrative description of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any appropriate manner in any one or more embodiments or examples. In addition, the person skilled in the art can combine and combine the different embodiments or examples described in the specification and the features of the different embodiments or examples without contradiction.

[0132] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. An underwater sand detection device for oil and gas wells, characterized in that: The device includes a sand signal sensing module, a sand signal conditioning module, a sand signal acquisition module and a sand signal analysis and processing module, wherein: The sand signal sensing module is used to sense and measure the high-frequency vibration signals generated by sand particles; The sand signal conditioning module is used to condition the high-frequency vibration signal; The sand signal acquisition module is used to collect the high-frequency vibration signal conditioned by the sand signal conditioning module in real time; The sand production signal analysis and processing module is used to obtain the high-frequency vibration signal collected by the sand production signal acquisition module, and analyze, reduce noise, extract features, and invert sand production information on the high-frequency vibration signal based on an adaptive modal decomposition method to obtain a sand production signal excited by sand particles colliding with the pipe wall of the underwater oil tree outlet pipe.

2. The underwater sand detection device for oil and gas wells according to claim 1, characterized in that: The sand signal sensing module adopts a vibration sensor; or / and, the sand signal conditioning module adopts a low-impedance signal conditioner, which can directly convert the measured sand signal into a voltage signal and amplify and filter it; or / and, the sand signal acquisition module adopts an acquisition card that provides multiple differential analog inputs and has a flexible channel conversion range.

3. The underwater sand detection device for oil and gas wells according to claim 1, characterized in that: The sand production signal analysis and processing module includes a preprocessing module, a main frequency response analysis module, an adaptive decomposition module, a multi-scale feature analysis module, and a sand production information feature extraction and inversion module, wherein: The preprocessing module is used to preprocess the collected high-frequency vibration signal to determine whether the frequency response range of the collected high-frequency vibration signal meets the preset requirements; The main frequency response analysis module is used to use time domain analysis, frequency domain analysis and power spectrum analysis on the high-frequency vibration signal that meets the preset requirements to determine the main frequency response range of the vibration signal generated by the sand particles in the outlet pipe of the underwater oil tree; The adaptive decomposition module is used to decompose the high-frequency vibration signal that meets the main frequency response range requirements using an adaptive decomposition algorithm to obtain a plurality of IMF subsequences; The multi-scale feature analysis module calculates the pulse factor and integrated energy of each obtained IMF sequence, selects a sequence with a high pulse factor and a high integrated energy as the optimal sequence, and spatially reconstructs the optimal IMF sequence to obtain a data set that can be used to characterize sand production conditions; The sand production information feature extraction and inversion module extracts two-dimensional response feature information of the sand production signal based on a data set representing the sand production status, assigns different weights to the extracted feature information at different spatial locations, and performs sand production feature fusion to obtain pipeline sand production information.

4. The underwater sand detection device for oil and gas wells according to claim 1, characterized in that: The sand signal analysis and processing module also has a built-in detection system fault self-repair module based on the watchdog principle to avoid the problem of being unable to restart after a system failure.

5. The underwater sand detection device for oil and gas wells according to claim 1, characterized in that: The device also includes a communication module, a power supply module and a remote control module, wherein: The communication module uses a multi-core watertight communication cable to output sand production information in real time through the RS485 serial port; The power supply module adopts a dual-voltage transformer power supply group, which shares a micro watertight cable with the communication module to power the sand signal conditioning module and the sand signal analysis and processing module; The remote control module adopts an Ethernet ultra-small watertight cable, and the remote control module is used for communication between the platform central control and the sand signal analysis and processing module.

6. The underwater sand detection device for oil and gas wells according to claim 1, characterized in that: The device also includes a shell, which is made of cylindrical titanium alloy. Each module is fixedly assembled in the shell through a small board. The sand signal sensing module is arranged at the bottom of the shell and is in rigid contact with the wall of the underwater oil tree outlet pipe. The shell and the underwater oil tree outlet pipe are installed and fixed by a clamp.

7. The underwater sand detection device for oil and gas wells according to claim 6, characterized in that: An aluminum heat sink is further provided in the housing, and the aluminum heat sink is in contact with the housing to improve the heat dissipation capacity of the underwater sand detector; or / and a handle is designed on the top of the housing to facilitate underwater installation and maintenance of the oil and gas well underwater sand detection device using an ROV.

8. An underwater sand detection method for oil and gas wells based on the underwater sand detection device according to any one of claims 1 to 7, characterized in that: include: The underwater sand detection device of the oil and gas well is fixed to the outlet pipe of the underwater oil tree by ROV, ensuring that the lower end of the sand signal sensing module is in rigid contact with the pipe wall; The sand particles carried by the fluid in the outlet pipe of the underwater oil tree hit the pipe wall and generate a vibration signal. The vibration signal is sensed by the sand signal sensing module in contact with the outer wall of the pipe. The weak vibration signal sensed and measured by the sand signal sensing module is transmitted to the signal conditioning module through a high-frequency, low-noise communication cable for adaptive filtering and amplification; The high-frequency vibration signal output by the sand signal conditioning module is transmitted to the sand signal acquisition module through a high-frequency low-noise communication cable; The high-frequency sand vibration signal collected by the sand signal acquisition module is analyzed, processed and inverted in real time by the sand signal analysis and processing module to obtain the sand signal excited by the sand particles colliding with the pipe wall at the outlet pipe of the underwater oil tree; The obtained sand production information is finally transmitted to the middle end of the platform through RS485, multi-core watertight communication cable, SCM and module umbilical cable to complete the sand production information detection.

9. The underwater sand detection method of the underwater sand detection device for oil and gas wells according to claim 8, characterized in that: The high-frequency sand vibration signal collected by the sand signal acquisition module is analyzed, processed and inverted in real time by the sand signal analysis and processing module to obtain the sand signal excited by the sand particles colliding with the pipe wall at the outlet pipe of the underwater oil tree, including: The preprocessing module preprocesses the collected high-frequency vibration signal and compares the frequency response range of the collected sand signal. If the main frequency response range is higher than the set value, it will proceed to the next step. Otherwise, it means that the collected sand signal is seriously distorted, and the collection will be stopped and the electronic modules of the detection device will be self-checked. The main frequency response analysis module performs main frequency response analysis on the pre-processed sanding signal, calculates the time domain, frequency domain and power spectrum response distribution characteristics of the sanding signal, and determines the main frequency response range of the sanding signal based on the calculated amplitude-frequency response characteristics of the sanding signal frequency domain and power spectrum; The adaptive decomposition module uses the CEEMDAN algorithm to adaptively decompose the sand production signal in the main frequency response range of the band-pass filtered signal to obtain the IMF component that represents the sand particle information; The multi-scale feature analysis module calculates the impulse factor and integrated energy of each IMF sequence, and selects the sequence with high impulse factor and high integrated energy as the optimal sequence; Perform statistical feature analysis on the selected IMF sequences and calculate the sample entropy and Hurst index of each subsequence. Then, distinguish the selected IMF subsequences at three frequency scales based on the "high sample entropy-low Hurst", "medium sample entropy-medium Hurst", and "low sample entropy-high Hurst" classification methods, and eliminate IMF sequences at low frequency scales. The IMF series with low-frequency scale removed is spatially reconstructed to obtain a data set for characterizing sand production conditions; The data set representing the sand production condition is input into the sand production information feature extraction and inversion module. The input data set will be simultaneously input into the CNN module and the GRU module to simultaneously extract the two-dimensional response feature information of the sand production signal. The built-in self-attention mechanism algorithm assigns different weights to the extracted response features of different spatial positions and fuses them to obtain the pipeline sand production information.

Images