Oil casing leakage detection system and method

Abstract

The invention discloses an oil casing pipe leakage detection system and method, the system comprises a detection device, a data processing device and a connection device, and the connection device comprises a data transmission part, a fixing part and a clamping part; the detection equipment is used for detecting a sound wave signal in a shaft annulus and sending the sound wave signal to the data transmission part of the connection equipment; the data transmission part of the connection device is in communication connection with the data processing device, and the data transmission part of the connection device is used for receiving the sound wave signal sent by the detection device and transmitting the sound wave signal to the data processing device; and the data processing equipment is used for receiving the sound wave signal transmitted by the data transmission part, performing denoising processing on the sound wave signal, inputting the denoised sound wave signal into a pre-trained oil casing leakage detection model, and outputting an oil casing leakage detection result. According to the method, the leakage state of the oil casing can be analyzed and judged efficiently and accurately in real time, and powerful support is provided for safety production of subsequent gas wells.

Description

Technical Field

[0001] The present application relates to the technical field of oil casing detection, and in particular to an oil casing leakage detection system and method. Background Art

[0002] In the process of oil and gas production, leakage of downhole wellbore oil casing (hereinafter referred to as oil casing) has always been a common problem in oil and gas production. Oil casing leakage will affect oil and gas production and endanger the safe production of oil and gas wells. In severe cases, blowouts and well kicks may even occur. Therefore, oil casing leakage detection is crucial. At present, in the detection and diagnosis of oil casing leakage, holographic diagnosis is mainly achieved by combining real-time monitoring of annular pressure, determining the source and rate of leakage during the pressure relief process, and locating the leakage point with the help of logging. Noise logging uses an acoustic wave sensor to receive the noise signal generated when a pipeline leaks, and judges the leakage of the pipeline by analyzing the acoustic signal. When a pipeline leaks, the logging curve of the acoustic signal will show an obvious peak. Based on the analysis of the peak, the leakage point can be located.

[0003] The current noise logging collects acoustic signals through acoustic sensors inserted into the wellbore annulus, and then connects to a PC after passing through a signal conditioner and a data acquisition card. The acoustic signals are further processed and analyzed through the analysis and judgment of the detection personnel. This method is relatively expensive and requires detection personnel to be on site in real time. Many production sites have harsh environments and the operability of oil casing detection is low. In addition, this method has poor real-time performance and cannot achieve normal wellbore acoustic signal monitoring, making it difficult to analyze leakage in the first place.

[0004] It can be seen that the existing noise logging method for monitoring oil casing leakage cannot obtain real-time data related to oil casing, has a long action cycle, and cannot detect oil casing leakage in time. In addition, the existing equipment is complicated to operate and requires stopping the well and moving the pipe string, which has high labor and equipment costs. Summary of the invention

[0005] The purpose of the embodiments of the present application is to provide an oil casing leakage detection system and method to solve the problem that oil casing leakage cannot be discovered in time and the detection cost is high.

[0006] In order to solve the above technical problems, the first aspect of this specification provides an oil casing leakage detection system, including:

[0007] A detection device, a data processing device and a connection device, wherein the connection device comprises a data transmission part, a fixing part and a clamping part;

[0008] The detection device is fixed to the connection device through the clamping part, and the connection device and the detection device are fixed through the fixing part so that the detection device extends into the wellbore annulus;

[0009] The detection device is communicatively connected to the data transmission part of the connection device, and the detection device is used to detect the acoustic wave signal in the wellbore annulus and send the acoustic wave signal to the data transmission part of the connection device;

[0010] The data transmission part of the connecting device is communicatively connected to the data processing device, and the data transmission part of the connecting device is used to receive the sound wave signal sent by the detection device and transmit the sound wave signal to the data processing device;

[0011] The data processing device is used to receive the sound wave signal transmitted by the data transmission unit, perform denoising on the sound wave signal, input the denoised sound wave signal into a pre-trained oil casing leakage detection model, and output the oil casing leakage detection result.

[0012] In some embodiments, the fixing portion is provided with a first channel, the clamping portion is provided with a second channel, the first channel and the second channel are connected, and the data transmission portion passes through the first channel and the second channel;

[0013] An accommodating cavity is provided at one end of the clamping portion away from the fixing portion, the accommodating cavity is communicated with the second channel, the detection device is built in the accommodating cavity and fixed to the clamping portion through a threaded hole of the accommodating cavity.

[0014] In some embodiments, the fixing portion is provided with a pressure relief hole, and the pressure relief hole is connected to the wellbore annulus through a pressure relief pipeline, so as to release the pressure in the wellbore annulus.

[0015] In some embodiments, the connection device includes a device protection portion, which is disposed in the accommodating cavity of the clamping portion and between the inner wall of the accommodating cavity and the detection device, and the device protection portion surrounds the circumference of the detection device.

[0016] In some embodiments, the data processing device performs denoising on the sound wave signal, including:

[0017] Decomposing the sound wave signal based on the mode of the sound wave signal to obtain multiple sound wave signal components;

[0018] Removing the sound wave signal components exceeding a preset signal threshold from the multiple sound wave signal components to obtain at least one sound wave signal component after noise reduction;

[0019] A noise-reduced sound wave signal is generated based on the at least one noise-reduced sound wave signal component.

[0020] In some embodiments, the sound wave signal is decomposed based on the mode of the sound wave signal to obtain multiple sound wave signal components, including:

[0021] Performing feature extraction on the acoustic wave signal to determine the mode number, iteration penalty coefficient, and iteration convergence tolerance of the acoustic wave signal, wherein each mode corresponds to an acoustic wave signal component;

[0022] Based on preset constraints and iteration parameters, an iteration constraint model of the acoustic wave signal is determined, wherein the preset constraints include: the sum of the bandwidths of the center frequencies of all modes is minimum, and the sum of the acoustic wave signal components corresponding to all modes matches the acoustic wave signal;

[0023] Iteratively searching for an optimal solution of the iterative constraint model in a current mode based on the iterative constraint model, the iterative penalty coefficient, and the iterative convergence tolerance, and determining a center frequency and a bandwidth of the current mode based on the optimal solution;

[0024] The acoustic wave signal is decomposed based on the center frequency and bandwidth of each mode to obtain the multiple acoustic wave signal components.

[0025] The second aspect of the specification provides an oil casing leakage detection method, which is applied to the oil casing leakage detection system described in the first aspect, comprising:

[0026] The detection device detects the acoustic wave signal in the wellbore annulus and sends the acoustic wave signal to the data transmission part of the connection device;

[0027] The data transmission part of the connection device receives the sound wave signal sent by the detection device and transmits the sound wave signal to the data processing device;

[0028] The data processing device receives the acoustic wave signal transmitted by the data transmission unit, performs denoising on the acoustic wave signal, inputs the denoised acoustic wave signal into a pre-trained oil casing leakage detection model, and outputs an oil casing leakage detection result.

[0029] In some embodiments, the data processing device performs denoising on the sound wave signal, including:

[0030] The data processing device decomposes the sound wave signal based on the mode of the sound wave signal to obtain multiple sound wave signal components;

[0031] The data processing device removes the sound wave signal components exceeding a preset signal threshold from the multiple sound wave signal components to obtain at least one sound wave signal component after noise reduction;

[0032] The data processing device generates a noise-reduced sound wave signal based on at least one noise-reduced sound wave signal component.

[0033] In some embodiments, the data processing device decomposes the sound wave signal based on the mode of the sound wave signal to obtain multiple sound wave signal components, including:

[0034] The data processing device performs feature extraction on the acoustic wave signal to determine the mode number, iteration penalty coefficient and iteration convergence tolerance of the acoustic wave signal, wherein each mode corresponds to an acoustic wave signal component;

[0035] The data processing device determines an iterative constraint model of the acoustic wave signal based on preset constraints and iteration parameters, wherein the preset constraints include: the sum of the bandwidths of the center frequencies of all modes is minimum, and the sum of the acoustic wave signal components corresponding to all modes matches the acoustic wave signal;

[0036] The data processing device iteratively searches for an optimal solution of the iterative constraint model in a current mode based on the iterative constraint model, the iterative penalty coefficient, and the iterative convergence tolerance, and determines a center frequency and a bandwidth of the current mode based on the optimal solution;

[0037] The data processing device decomposes the acoustic wave signal based on the central frequency and bandwidth of each mode to obtain the multiple acoustic wave signal components.

[0038] In some embodiments, the oil casing leakage detection model is trained by:

[0039] The data processing device acquires historical acoustic wave data and oil casing leakage detection data corresponding to the historical acoustic wave data as a training data set;

[0040] The data processing device determines an initial deep learning model based on the training data set, wherein the initial deep learning model includes an acoustic wave mapping function, and the acoustic wave mapping function is used to map the historical acoustic wave data to a target feature space;

[0041] The data processing device inputs the training data set into the initial deep learning model, trains the initial deep learning model, and obtains the oil casing leakage detection model.

[0042] In some embodiments, the data processing device determines an initial deep learning model based on the training data set, including:

[0043] The data processing device determines feature data of historical sound wave data in the training data set;

[0044] The data processing device determines the acoustic wave mapping function of the initial deep learning model based on the distribution characteristics of the feature data, and constructs the initial deep learning model based on the acoustic wave mapping function.

[0045] A third aspect of the present specification provides a computer storage medium, wherein the computer storage medium stores computer program instructions, and when the computer program instructions are executed, the steps of the method described in the second aspect are implemented.

[0046] The oil casing leakage detection system provided in the embodiment of this specification extends the detection equipment into the wellbore annulus to detect the acoustic wave signal in the wellbore annulus, and sends the acoustic wave signal to the data transmission part of the connection device, and then receives the acoustic wave signal sent by the detection equipment through the data transmission part of the connection device that is connected to the data processing device for communication, and transmits the acoustic wave signal to the data processing device. Then, the data processing device can receive the acoustic wave signal transmitted by the data transmission part, denoise the acoustic wave signal, and input the denoised acoustic wave signal into a pre-trained oil casing leakage detection model to output the oil casing leakage detection result. The present application detects the wellbore acoustic wave signal in real time through the detection equipment, and quickly judges the oil casing leakage status through the data processing equipment, which can realize the analysis and judgment of the oil casing leakage status in real time, efficiently and accurately, and provide strong support for the safe production of subsequent gas wells. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the implementation methods of the present application or the technical solutions in the prior art, the drawings required for use in the implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some implementation methods recorded in this application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0048] Figure 1 Shown is a schematic diagram of an oil casing leakage detection system provided by an embodiment of the present application;

[0049] Figure 2 Shown is a structural schematic diagram of a connection device provided in an embodiment of the present application;

[0050] Figure 3 Shown is a structural schematic diagram of a connection device provided in an embodiment of the present application;

[0051] Figure 4 The figure shows a structural schematic diagram of a data processing device provided in an embodiment of the present application;

[0052] Figure 5 The figure is a schematic flow chart of a method for detecting oil casing leakage provided in an embodiment of the present application. DETAILED DESCRIPTION

[0053] In order to enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this application.

[0054] As mentioned above, the existing noise logging method for monitoring oil casing leakage cannot obtain relevant data of oil casing in real time, has a long working cycle, and cannot detect oil casing leakage in time. In addition, the existing equipment is complicated to operate and requires stopping the well and moving the pipe string, which has high labor and equipment costs.

[0055] In order to solve the above problems, an embodiment of the present application provides an oil casing leakage detection system and method, wherein the system includes a detection device, a data processing device and a connecting device, the connecting device includes a data transmission part, a fixing part and a clamping part; the detection device is fixed to the connecting device by the clamping part, and the connecting device and the detection device are fixed by the fixing part so that the detection device extends into the wellbore annulus; the detection device is communicatively connected to the data transmission part of the connecting device, the detection device is used to detect the acoustic wave signal in the wellbore annulus, and send the acoustic wave signal to the data transmission part of the connecting device; the data transmission part of the connecting device is communicatively connected to the data processing device, the data transmission part of the connecting device is used to receive the acoustic wave signal sent by the detection device, and transmit the acoustic wave signal to the data processing device; the data processing device is used to receive the acoustic wave signal transmitted by the data transmission part, denoise the acoustic wave signal, and input the denoised acoustic wave signal into a pre-trained oil casing leakage detection model to output the oil casing leakage detection result.

[0056] The present application detects wellbore acoustic wave signals in real time through detection equipment, and quickly determines the leakage status of oil casing through data processing equipment. It can analyze and determine the leakage status of oil casing in real time, efficiently and accurately, and provide strong support for the safe production of subsequent gas wells.

[0057] The following first introduces in detail the oil casing leakage detection system provided by the embodiment of the present application in conjunction with the accompanying drawings.

[0058] Figure 1 FIG. 1 is a schematic diagram of an oil casing leakage detection system provided in an embodiment of the present application. Figure 1 As shown, the system may include:

[0059] A detection device 101 , a connection device 102 and a data processing device 103 , wherein the connection device 102 includes a data transmission portion 1021 , a fixing portion 1022 and a clamping portion 1023 .

[0060] The detection device 101 is fixed to the connection device 102 via the clamping portion 1023 , and the connection device 102 and the detection device 101 can be fixed via the fixing portion 1022 , so that the detection device 101 extends into the wellbore annulus.

[0061] It can be understood that the fixing portion 1022 and the clamping portion 1023 can be used as hardware components of the connecting device 102. The clamping portion 1023 is used to fix the detection device 101 on the connecting device 102 to fixedly connect the detection device 101 to the connecting device 102. The fixing portion 1022 is used to fix the detection device 101 at a predetermined position to install the detection device 101 in an environment where detection is required, that is, the detection device 101 can be extended into a position in the wellbore annulus.

[0062] The detection device 101 is communicatively connected to the data transmission unit 1021 of the connection device 102, and the detection device 101 is used to detect the acoustic wave signal in the wellbore annulus and send the acoustic wave signal to the data transmission unit 1021 of the connection device 102. It can be understood that the detection device 101 can detect the acoustic wave signal in the wellbore annulus in real time and transmit it, which can then be received by the data processing device 103 to determine the pipeline leakage situation.

[0063] The data transmission unit 1021 of the connection device 102 is communicatively connected to the data processing device 103 , and is used to receive the sound wave signal sent by the detection device 101 , and transmit the sound wave signal to the data processing device 103 .

[0064] The data processing device 103 is used to receive the acoustic wave signal transmitted by the data transmission unit 1, perform denoising on the acoustic wave signal, input the denoised acoustic wave signal into a pre-trained oil casing leakage detection model, and output the oil casing leakage detection result.

[0065] It can be understood that the data processing device 103 can be an electronic device with data calculation, processing and storage capabilities. The electronic device can be a terminal such as a PC (Personal Computer), a tablet computer, a smart phone, a wearable device, an intelligent robot, etc.; it can also be a server; it can also be a system on chip (System on Chip, SOC), such as PYNQ (Python Productivity for Zynq, i.e., Python productivity of Zynq), etc. Among them, the server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud computing services. This application does not limit this.

[0066] In some embodiments, the detection device 101 may include an acoustic wave sensor for detecting acoustic wave signals in the wellbore annulus, and the acoustic wave sensor may be communicatively connected to the data transmission unit 1021 of the connection device 102 .

[0067] In some embodiments, the communication connection between the data transmission unit 1021 and the detection device 101 and the data processing device 103 can be a wired electrical connection, and the data transmission unit 1021 can be a signal line, such as a data transmission line, a coaxial line, etc., and then the detection device 101 and the data processing device 103 can be provided with corresponding hardware interfaces, and the data transmission unit 1021 can be respectively communicated with the detection device and the data processing device 103 through the corresponding hardware interface to receive the acoustic wave signal in the wellbore annulus detected by the detection device 101, and transmit the received acoustic wave signal to the data processing device 103.

[0068] In some embodiments, the communication connection between the data transmission unit 1021 and the detection device 101 and the data processing device 103 can be a wireless communication connection. The data transmission unit 1021 can be a device that can realize wireless reception and wireless transmission of signals, such as a signal relay device, etc., and then the detection device 101 and the data processing device 103 can include corresponding wireless communication units. After the detection device 101 detects the acoustic wave signal in the wellbore annulus, the acoustic wave signal can be wirelessly transmitted to the data transmission unit 1021 through its wireless communication unit. After receiving the acoustic wave signal, the data transmission unit 1021 can amplify and restore the acoustic wave signal, and wirelessly transmit the amplified and restored acoustic wave signal to the data processing device 103, or can directly transmit the received acoustic wave signal wirelessly to the data processing device 103.

[0069] Figure 2 Shown is a structural schematic diagram of a connection device provided in an embodiment of the present application.

[0070] like Figure 2As shown, in some embodiments, the fixing portion 1022 is provided with a first channel 201, the clamping portion 1023 is provided with a second channel 202, the first channel 201 and the second channel 202 are connected, and the data transmission portion 1021 passes through the first channel 201 and the second channel 202;

[0071] An accommodating cavity 203 is provided at one end of the clamping portion 1023 away from the fixing portion 1022 . The accommodating cavity 203 is communicated with the second channel 202 . The detection device 101 is built in the accommodating cavity 203 and fixed to the clamping portion 1023 through a threaded hole 2031 of the accommodating cavity 203 .

[0072] It can be understood that the detection device 101 can be fixed to the clamping portion 1023 by means of screws and threaded holes 2031 to withstand higher vibration and pressure in the installation environment of the detection device 101 and the connection device 102 .

[0073] In some embodiments, the detection device 101 can be fixed to the clamping portion 1023 by means of a latch through the threaded hole 2031 of the clamping portion 1023 to withstand higher vibration and pressure in the installation environment of the detection device 101 and the connection device 102 .

[0074] In some embodiments, the connection device 102 includes a device protection portion, which is disposed in the accommodating cavity 203 of the clamping portion and between the inner wall of the accommodating cavity 203 and the detection device 101 , and surrounds the circumference of the detection device 101 .

[0075] In some embodiments, the equipment protection part can be composed of a corrosion-resistant plastic film and a fine filter mesh. The equipment protection part is used to wrap the detection device 101, especially the probe part of the detection device 1011, to protect it from the oil casing annulus impurities that may return from the oil casing, such as the annulus protection fluid, so as to obtain more accurate sound wave signal detection results and ensure the service life of the detection device 101.

[0076] In some embodiments, the connection device 102 may be a tree plug structure, and the same material as the tree plug is used. The detection device 101 may be installed at the tree plug position of the connection device 102 through a plug design, and the detection device may be extended into the wellbore annulus through the clamping portion of the connection device to receive the acoustic wave signal in real time. Specifically, the connection device 102 may have a connection channel running through its first end face 204 and second end face 205, and the connection channel may include a first channel 201 provided at a fixing portion 1022, a second channel 202 provided at a clamping portion 1023, and a receiving cavity 203 provided at the clamping portion 1023, the clamping portion 1023 is located at the first end face 204 of the connection device 102, and the fixing portion 1022 is located at the second end face 205 of the connection device 102. Furthermore, the accommodating cavity 203 of the clamping portion 1023 is located at one end of the first end face 203 of the connecting channel close to the connecting device 102, the first channel 201 of the fixing portion 1022 is located at one end of the second end face 205 of the connecting channel close to the connecting device 102, and the second channel 202 is located between the first channel 201 of the connecting channel and the accommodating cavity 203.

[0077] In some embodiments, an opening is provided around the fixing portion 1022 , and a connecting channel penetrating the opening and the first end surface may be provided inside the connecting device. In this case, the second channel 202 may be an L-shaped channel.

[0078] Figure 3 Shown is a structural schematic diagram of a connection device provided in an embodiment of the present application.

[0079] like Figure 3 As shown, in some embodiments, the clamping portion 1023 may be a hollow tubular structure, and a threaded hole 301 is provided on the pipe column wall of the clamping portion 1023. The accommodating cavity of the clamping portion 1023 may include a first chamber 302 and a second chamber 303. The diameter of the first chamber 302 may be different from the diameter of the second chamber 303. For example, Figure 3 The diameter of the first chamber 302 is greater than the diameter of the second chamber 303 , and the threaded hole 301 can be obtained by opening a hole above the second chamber 303 . The threaded hole 301 can connect the outside of the clamping part 1023 with the second chamber 303 .

[0080] Continue to refer Figure 3As shown, in some embodiments, the fixing portion 1022 may include a cylindrical first portion 304, a hexagonal second portion 305, and a cylindrical third portion 306, and the first channel 201 of the fixing portion 1022 may be disposed inside the first portion 304, the second portion 305, and the third portion 306. The first portion 304 may be fixedly connected to the clamping portion 1023, one end of the first channel 2011 located at the first portion 304 is communicated with the second channel 202 of the clamping portion 1023, the third portion 306 is disposed at one end of the connecting device 102 close to the second end surface 205, and the second portion 305 may be disposed between the first portion 304 and the third portion 306.

[0081] In some embodiments, a clamp (eg, Figure 3 307), the clamp can be used to fix the connection device 102 of the clamping detection device 102 at a predetermined position, so as to install the detection device 101 in an environment where detection is required, that is, the detection device 101 can be extended into the position in the wellbore annulus.

[0082] In some embodiments, the fixing portion is provided with a pressure relief hole 308, and the pressure relief hole 308 is connected to the wellbore annulus through a pressure relief pipeline to release the pressure in the wellbore annulus.

[0083] Continue to refer Figure 3 As shown, in some embodiments, the pressure relief hole 308 can be provided at the periphery of the hexagonal tube of the second part 305 of the fixing part 1022, and the fixing part 1022 can include a third channel connecting the second part 305 and the first part 306 thereof, and a pressure relief pipeline 309 can be provided in the third channel, and the pressure relief pipeline 309 passes through the third channel and is connected in the wellbore annulus, and the annular pressure accumulated in the wellbore annulus during the oil casing leakage detection process can be released through the pressure relief pipeline 309 and the pressure relief hole 308. Among them, the pressure relief pipeline 309 can be a high-pressure sealed electrical connection line.

[0084] In some embodiments, the pressure relief hole 308 and the pressure relief line 309 may be connected via a pressure relief valve. When the pressure relief valve detects that the pressure in the wellbore annulus exceeds a preset pressure, the pressure relief valve is in an open state and the pressure in the wellbore annulus may be released to the outside of the wellbore through the pressure relief line 309, the pressure relief valve and the pressure relief hole 308.

[0085] In some embodiments, the data processing device 103 performs denoising on the sound wave signal, which may include the following steps:

[0086] Decomposing the sound wave signal based on the mode of the sound wave signal to obtain multiple sound wave signal components;

[0087] Removing the sound wave signal components exceeding a preset signal threshold from the multiple sound wave signal components to obtain at least one sound wave signal component after noise reduction;

[0088] A noise-reduced sound wave signal is generated based on the at least one noise-reduced sound wave signal component.

[0089] In some embodiments, the sound wave signal is decomposed based on the mode of the sound wave signal to obtain multiple sound wave signal components, including:

[0090] Performing feature extraction on the acoustic wave signal to determine the mode number, iteration penalty coefficient, and iteration convergence tolerance of the acoustic wave signal, wherein each mode corresponds to an acoustic wave signal component;

[0091] Based on preset constraints and iteration parameters, an iteration constraint model of the acoustic wave signal is determined, wherein the preset constraints include: the sum of the bandwidths of the center frequencies of all modes is minimum, and the sum of the acoustic wave signal components corresponding to all modes matches the acoustic wave signal;

[0092] Iteratively searching for an optimal solution of the iterative constraint model in a current mode based on the iterative constraint model, the iterative penalty coefficient, and the iterative convergence tolerance, and determining a center frequency and a bandwidth of the current mode based on the optimal solution;

[0093] The acoustic wave signal is decomposed based on the center frequency and bandwidth of each mode to obtain the multiple acoustic wave signal components.

[0094] It can be understood that the acoustic signals detected in the wellbore annulus are mostly high-frequency noise of field production noise. The collected acoustic signals can be solved as a variational problem, and the center frequency and bandwidth of each mode can be determined by iteratively searching the optimal solution of the iterative constraint model, and the effective components corresponding to each center frequency in the frequency domain are found to obtain the modal function; and, by extracting features from the acoustic signal, determining the number of modes, iterative penalty coefficients, and iterative convergence tolerance, the signal can be adaptively and effectively separated, and the decomposed acoustic signal components have independent center frequencies and exhibit sparse characteristics in the frequency domain. They have the characteristics of sparse research and can effectively avoid modal aliasing. The leakage acoustic signal processed by this method is more robust. After modal decomposition of the acoustic signal, the main acoustic signal components can be selected by energy, that is, multiple acoustic signal components whose weight does not exceed the preset signal threshold are selected, and then the original acoustic signal can be restored based on the selected main acoustic signal components. Compared with the traditional empirical mode decomposition denoising algorithm, the above denoising method will not discard small-order components, can restore the true characteristics of the acoustic signal, and can use the efficient parallel computing power of data processing equipment to denoise a large number of acoustic signals, retain the effective characteristics of the acoustic signal, and lay a data foundation for oil casing leakage detection.

[0095] In some embodiments, the number of modes, the iteration penalty coefficient, and the iteration convergence tolerance can be specified. The iteration penalty coefficient can be set to 3000, and the iteration convergence tolerance can be set to 0.000001.

[0096] In some embodiments, after performing modal decomposition on the acoustic wave signal, a plurality of intrinsic mode functions (IMFs) of the acoustic wave signal may be obtained. The plurality of intrinsic mode functions may be used as a plurality of acoustic wave signal components obtained after decomposing the acoustic wave signal.

[0097] In some embodiments, the preset signal threshold may include a preset energy threshold and / or a preset frequency threshold. It is understood that in some embodiments, by removing the sound wave signal components exceeding the preset energy threshold from multiple sound wave signal components, at least one sound wave signal component after noise reduction can be obtained. In other embodiments, by removing the sound wave signal components exceeding the preset frequency threshold from multiple sound wave signal components, at least one sound wave signal component after noise reduction can be obtained. In other embodiments, by removing the sound wave signal components exceeding the preset energy threshold and the preset frequency threshold from multiple sound wave signal components, at least one sound wave signal component after noise reduction can be obtained. In other embodiments, the preset signal threshold can also be set based on other characteristics of the noise signal, not limited to energy and frequency. The above examples are only partial embodiments of the present application, and the present application is not limited to this.

[0098] In some embodiments, the data processing device 103 may store a pre-trained oil casing leakage detection model, and after the data processing device 103 performs denoising on the received acoustic wave signal, the denoised acoustic wave signal may be input into the stored oil casing leakage detection model to perform oil casing leakage detection. The oil casing leakage detection may specifically be oil casing annulus leakage detection.

[0099] In some embodiments, the oil casing leakage detection model stored in the data processing device 103 can be trained in the following way:

[0100] Acquire historical acoustic wave data and oil casing leakage detection data corresponding to the historical acoustic wave data as a training data set;

[0101] Determining an initial deep learning model based on the training data set, the initial deep learning model includes an acoustic wave mapping function, and the acoustic wave mapping function is used to map the historical acoustic wave data to a target feature space;

[0102] The training data set is input into the initial deep learning model, and the initial deep learning model is trained to obtain the oil casing leakage detection model.

[0103] It can be understood that the historical acoustic wave data can be obtained by measuring the acoustic wave signals in the wellbore annulus, and the acoustic wave mapping function can map the historical acoustic wave data from a low-dimensional feature space to a high-dimensional feature space to classify the historical acoustic wave data.

[0104] In some embodiments, the oil casing leakage detection data may include data on whether the oil casing is leaking when corresponding historical acoustic wave data is detected, so as to obtain an oil casing leakage detection model for detecting whether the oil casing is leaking by using the oil casing leakage detection data including whether the oil casing is leaking and the historical acoustic wave data as training data sets.

[0105] In some embodiments, the oil casing leakage detection data may include whether the oil casing leaks or not and the corresponding leakage degree data when the corresponding historical acoustic wave data is detected. That is, by using the oil casing leakage detection data including the data of whether the oil casing leaks or not and the leakage degree and the historical acoustic wave data as the training data set, an oil casing leakage detection model for detecting whether the oil casing leaks or not and detecting the leakage degree of the oil casing is obtained. Among them, the leakage degree data can be the oil casing leakage level obtained by dividing based on the state of the oil casing, and different levels can adopt different treatment measures for the oil casing.

[0106] In other embodiments, the oil casing detection data may also include other data that can characterize the judgment of oil casing leakage, such as leakage location, etc., so that when oil casing leakage detection is performed based on the oil casing leakage detection model, the corresponding oil casing detection result can be output. This application does not limit this.

[0107] In some embodiments, the training data set may also include simulated acoustic wave data obtained through experimental simulation analysis, and oil casing leakage detection data corresponding to the simulated acoustic wave data. Through measured data and field measured data, a large number of training data sets that meet actual application scenarios can be obtained to train a more accurate oil casing leakage detection model.

[0108] In some embodiments, the initial deep learning model can be a support vector machine model in machine learning. It can be understood that the support vector machine (SVM) is a generalized classifier that performs binary classification on the training data set in a supervised learning manner, and its decision boundary is the maximum margin hyperplane solved for the learning samples composed of the training data set. The classification problem can be transformed into a problem of solving convex quadratic programming. The oil casing leakage detection model obtained by training the experimental and field measured oil casing leakage data as the training data set is deployed to the data processing device 103, and the noise-reduced acoustic wave signal is input in real time and the classification result is given to determine whether there is a leakage. The powerful parallel capability and low power consumption of the data processing device 103 are used to accelerate the calculation and improve the efficiency, so as to realize the actual field application of intelligent detection.

[0109] In some embodiments, during the training process of the oil casing leakage detection model, determining the initial deep learning model based on the training data set may include:

[0110] Determine feature data of historical sound wave data in a training data set;

[0111] Based on the distribution characteristics of the feature data, an acoustic wave mapping function of an initial deep learning model is determined, and the initial deep learning model is constructed based on the acoustic wave mapping function.

[0112] In some embodiments, data processing, such as feature extraction, can be performed on historical acoustic wave data when oil casing leakage occurs (hereinafter referred to as leakage acoustic wave data) to obtain feature data of the historical acoustic wave data, and then the acoustic wave mapping function of the initial deep learning model can be determined by analyzing and judging the feature data and combining the application scenarios of multiple acoustic wave mapping functions.

[0113] In some embodiments, since leakage acoustic wave data has many features, it is difficult for traditional acoustic wave mapping functions to classify them well. The mapping and classification effects of multiple acoustic wave mapping functions can be tested to select leakage acoustic wave data with better effects that are more suitable for leakage acoustic wave data.

[0114] In some embodiments, the acoustic wave mapping function may be a kernel function, which may map the historical acoustic wave data from a low-dimensional feature space to a high-dimensional feature space to classify the historical acoustic wave data.

[0115] In some embodiments, the determined acoustic wave mapping function may be a radial basis function (RBF) kernel function, which may map characteristic data of acoustic wave signals that are difficult to classify to a high dimension, thereby better realizing the classification of acoustic wave signals and the detection of oil casing leakage.

[0116] In some embodiments, the feature data of the historical sound wave data may include at least one of the following dimensions:

[0117] The peak-to-peak value of the historical sound wave data, the absolute mean of the historical sound wave data, the variance of the historical sound wave data, the root amplitude of the historical sound wave data, the root mean square of the historical sound wave data, the average value of the historical sound wave data, the kurtosis factor of the historical sound wave data, the margin factor of the historical sound wave data, and the margin factor of the autocorrelation coefficient of the historical sound wave data.

[0118] In some embodiments, the oil casing leakage detection model can be trained on the data processing device 103. In other embodiments, the oil casing leakage detection model can be trained on other devices or servers, and transplanted to the data processing device 103 after the training is completed. This application does not limit this.

[0119] In some embodiments, after obtaining the oil casing leakage detection result, the data processing device 103 can upload the calculated and processed oil casing leakage detection result and the acoustic signal after noise reduction to the cloud, and can view the oil casing annulus leakage and the corresponding acoustic signal data in real time through the mobile phone or computer software. In other embodiments, after obtaining the oil casing leakage detection result, the data processing device 103 can transmit the calculated and processed oil casing leakage detection result and the acoustic signal after noise reduction to the mobile phone or computer, and then can view the oil casing annulus leakage and the corresponding acoustic signal data in real time through the mobile phone or computer. Specifically, the data upload and / or transmission can be realized through the wireless communication unit (such as a wifi module) of the data processing device 103, and the acoustic signal and the leakage detection result can be uploaded to the cloud database or transmitted to the computer in real time through the wireless network, and can be monitored and viewed in real time through the computer or mobile phone software. In some embodiments, the computer or mobile phone software can visualize the received acoustic signal and leakage detection result on the display interface so as to clearly and vividly display the detection data of the wellbore status.

[0120] In some embodiments, the data processing device 103 may be a SOC, and the SOC may be used to implement noise reduction processing on the acoustic wave signal and detection of oil casing leakage.

[0121] The following takes the data processing device 103 as an SOC as an example to further introduce the data processing device.

[0122] Figure 4 Shown is a structural schematic diagram of a data processing device provided in an embodiment of the present application.

[0123] like Figure 4 As shown, in some embodiments, the data processing device 103 may include a signal processing unit 401. The signal processing unit 401 may perform denoising on the acoustic wave signal, input the denoised acoustic wave signal into a pre-trained oil casing leakage detection model, and output the oil casing leakage detection result.

[0124] In some embodiments, the signal processing unit 401 may collect received sound wave signals in real time at a specified sampling frequency to reduce the amount of data processing.

[0125] In some embodiments, the signal processing unit 401 may be a PYNQ module. The PYNQ module may be based on a hardware reconfigurable architecture. The function of each logic unit is determined during reprogramming. The operation is accelerated at the system level, and high-speed processing of a large number of sound wave signals can be achieved.

[0126] In some embodiments, the PYNQ module may include a wireless communication sub-module, such as a wifi sub-module. The wireless communication sub-module can be used to communicate with the data transmission unit 1021 of the connection device 102, and / or to upload the acoustic wave signal and leakage detection results to a cloud database or transmit them to a computer in real time via a wireless network, so that real-time monitoring and viewing can be performed through computer or mobile phone software.

[0127] In some embodiments, the PYNQ module may include a high-speed communication interface, which can be directly or indirectly electrically connected to the data transmission unit 1021, and then through the design of the high-speed communication interface, high-speed signal processing can be achieved to further process the received sound wave signal, such as extraction and filtering, so that the signal can be easily processed, transmitted and stored.

[0128] In some embodiments, the PYNQ module can include two data processing modes, concurrent and pipeline, to have greater parallelism. The PYNQ module can build a corresponding pipeline according to the combing process of the acoustic signal (i.e., the number of data packet steps), thereby achieving data parallelism and pipeline parallelism. The concurrency of the PYNQ module can be performed between different logical functions, and is not limited to executing the same function at the same time, thereby improving the processing speed of the received acoustic signal, and obtaining the noise-reduced acoustic signal and / or the oil casing leakage detection result more quickly.

[0129] In some embodiments, since the PYNQ module has good scalability, the data transmission unit can use different interfaces to connect to the PYNQ module to realize the collection and real-time monitoring of various signals in the wellbore. The PYNQ module can use Python language to write instructions, which is effectively integrated with the mainstream artificial intelligence algorithms, making the deployment and operation of the oil casing leakage detection model more concise, efficient and scalable.

[0130] In some embodiments, the data processing device 103 may further include a signal conditioning unit 402. The signal conditioning unit 402 may be communicatively connected between the signal processing unit 401 and the data transmission unit 1021 of the connection device 102. The signal conditioning unit 402 may be used to receive the acoustic wave signal transmitted by the data transmission unit 1021 of the connection device 102, and perform signal conditioning on the received acoustic wave signal, such as preliminary amplification and filtering operations, and then transmit it to the signal processing unit 401 communicatively connected thereto. The signal processing unit 401 may perform denoising on the acoustic wave signal after signal conditioning by the signal conditioning unit 402, and input the denoised acoustic wave signal into a pre-trained oil casing leakage detection model, and output the oil casing leakage detection result.

[0131] In some embodiments, the data processing device 103 may further include a signal acquisition unit 403. The signal acquisition unit 403 may be communicatively connected between the signal conditioning unit 402 and the signal processing unit 401. The signal conditioning unit 402 receives the acoustic wave signal transmitted by the data transmission unit 1021 of the connection device 102, and performs signal conditioning on the received acoustic wave signal, such as preliminary amplification and filtering operations, and then transmits the conditioned acoustic wave signal to the signal acquisition unit 403 communicatively connected thereto. The signal acquisition unit 403 may be used to convert the received acoustic wave signal characterized as an analog signal processed by the signal conditioning unit 402 into a digital signal that can be processed by a computer, and transmit the digital signal to the signal processing unit 401. The signal processing unit 401 may perform denoising on the acoustic wave signal conditioned by the signal conditioning unit 402, and input the denoised acoustic wave signal into a pre-trained oil casing leakage detection model, and output the oil casing leakage detection result.

[0132] In some embodiments, the data processing device 103 may further include a power supply 404 to provide power to various components of the data processing device 103 to ensure normal operation of the various components.

[0133] In some embodiments, the data processing device 102 may further include a memory. The memory may be used as a non-transitory computer-readable storage medium, and may be used to store non-transitory software programs, non-transitory computer executable programs, and modules, such as program instructions / modules corresponding to the method for processing an acoustic wave signal by the data processing device in an embodiment of the present invention. The signal processing unit 401 may execute various functional applications and data processing of the signal processing unit 401 by running the non-transitory software programs, instructions, and modules stored in the memory.

[0134] The memory may include a program storage area and a data storage area, wherein the program storage area may store an operating system and an application required by at least one function; the data storage area may store data created by the processor, etc. In addition, the memory may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage devices.

[0135] In some embodiments, the memory may optionally include a memory remotely disposed relative to the signal processing unit 401, and these remote memories may be connected to the signal processing unit 401 via a network. Examples of the above network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0136] In some embodiments, the signal conditioning unit 402, the signal acquisition unit 403 and the signal processing unit 401 can be integrated into a SOC as the data processing device 103 in the embodiment of the present application. The integrated SOC can be packaged in a small explosion-proof box, which can be used for explosion-proof, and has a small size, is portable and reliable.

[0137] In some embodiments, the signal conditioning unit 402, the signal acquisition unit 403 and the signal processing unit 401 can be connected via a remote control switch 405. During the process of oil casing leakage detection, the components in the data processing device 103 can be turned on or off through a remote PC or mobile phone or other electronic devices, without the need for manual start-up or shutdown on site, so as to improve operational flexibility.

[0138] The present application also provides a method for detecting oil casing leakage. Figure 5 The oil casing leakage detection method provided in the embodiment of the present application is introduced.

[0139] Figure 5 FIG. 1 is a flow chart of a method for detecting oil casing leakage provided by an embodiment of the present application. Figure 5 As shown, the method may include:

[0140] S501: The detection device detects the acoustic wave signal in the wellbore annulus and sends the acoustic wave signal to the data transmission part of the connection device.

[0141] S502: The data transmission unit of the connection device receives the sound wave signal sent by the detection device, and transmits the sound wave signal to the data processing device.

[0142] S503: The data processing device receives the sound wave signal transmitted by the data transmission unit.

[0143] S504: the data processing device performs denoising on the acoustic wave signal, inputs the denoised acoustic wave signal into a pre-trained oil casing leakage detection model, and outputs an oil casing leakage detection result.

[0144] In some embodiments, the data processing device performs denoising on the sound wave signal, which may include:

[0145] The data processing device decomposes the sound wave signal based on the mode of the sound wave signal to obtain multiple sound wave signal components;

[0146] The data processing device removes the sound wave signal components exceeding a preset signal threshold from the multiple sound wave signal components to obtain at least one sound wave signal component after noise reduction;

[0147] The data processing device generates a noise-reduced sound wave signal based on at least one noise-reduced sound wave signal component.

[0148] In some embodiments, the data processing device decomposes the sound wave signal based on the mode of the sound wave signal to obtain multiple sound wave signal components, including:

[0149] The data processing device performs feature extraction on the acoustic wave signal to determine the mode number, iteration penalty coefficient and iteration convergence tolerance of the acoustic wave signal, wherein each mode corresponds to an acoustic wave signal component;

[0150] The data processing device determines an iterative constraint model of the acoustic wave signal based on preset constraints and iteration parameters, wherein the preset constraints include: the sum of the bandwidths of the center frequencies of all modes is minimum, and the sum of the acoustic wave signal components corresponding to all modes matches the acoustic wave signal;

[0151] The data processing device iteratively searches for an optimal solution of the iterative constraint model in a current mode based on the iterative constraint model, the iterative penalty coefficient, and the iterative convergence tolerance, and determines a center frequency and a bandwidth of the current mode based on the optimal solution;

[0152] The data processing device decomposes the acoustic wave signal based on the central frequency and bandwidth of each mode to obtain the multiple acoustic wave signal components.

[0153] In some embodiments, the oil casing leakage detection model is trained by:

[0154] The data processing device acquires historical acoustic wave data and oil casing leakage detection data corresponding to the historical acoustic wave data as a training data set;

[0155] The data processing device determines an initial deep learning model based on the training data set, wherein the initial deep learning model includes an acoustic wave mapping function, and the acoustic wave mapping function is used to map the historical acoustic wave data to a target feature space;

[0156] The data processing device inputs the training data set into the initial deep learning model, trains the initial deep learning model, and obtains the oil casing leakage detection model.

[0157] In some embodiments, the data processing device determines an initial deep learning model based on the training data set, including:

[0158] The data processing device determines feature data of historical sound wave data in the training data set;

[0159] The data processing device determines the acoustic wave mapping function of the initial deep learning model based on the distribution characteristics of the feature data, and constructs the initial deep learning model based on the acoustic wave mapping function.

[0160] The description and functions of the above steps can be understood by referring to the contents of the oil casing leakage detection system section, which will not be repeated here.

[0161] The present specification also provides a computer storage medium, wherein the computer storage medium stores computer program instructions, and when the computer program instructions are executed, the steps of the above-mentioned oil casing leakage detection method are implemented.

[0162] The present specification also provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the steps of the above-mentioned oil casing leakage detection method are implemented.

[0163] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the storage medium can be a disk, an optical disk, a read-only memory (ROM), a random access memory (RAM), a flash memory, a hard disk drive (HDD) or a solid-state drive (SSD), etc.; the storage medium can also include a combination of the above-mentioned types of memory.

[0164] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0165] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions.

[0166] For the convenience of description, the above device is described in various units according to their functions. Of course, when implementing the present application, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0167] It can be known from the description of the above implementation methods that those skilled in the art can clearly understand that the present application can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present application can be essentially or partly contributed to the prior art in the form of a software product, which can be stored in a storage medium such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute certain parts of the methods of each implementation method of the present application.

[0168] The present application can be used in many general or special computer system environments or configurations, such as personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronic devices, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc.

[0169] The present application may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.

[0170] Although the present application has been described through embodiments, those skilled in the art will appreciate that there are many modifications and variations to the present application without departing from the spirit of the present application, and it is intended that the appended claims include these modifications and variations without departing from the spirit of the present application.

Claims

1. An oil casing leakage detection system, characterized in that: include: A detection device, a data processing device and a connection device, wherein the connection device comprises a data transmission part, a fixing part and a clamping part; The detection device is fixed to the connection device through the clamping part, and the connection device and the detection device are fixed through the fixing part so that the detection device extends into the wellbore annulus; The detection device is communicatively connected to the data transmission part of the connection device, and the detection device is used to detect the acoustic wave signal in the wellbore annulus and send the acoustic wave signal to the data transmission part of the connection device; The data transmission part of the connecting device is communicatively connected to the data processing device, and the data transmission part of the connecting device is used to receive the sound wave signal sent by the detection device and transmit the sound wave signal to the data processing device; The data processing device is used to receive the sound wave signal transmitted by the data transmission unit, perform denoising on the sound wave signal, input the denoised sound wave signal into a pre-trained oil casing leakage detection model, and output the oil casing leakage detection result.

2. The system according to claim 1, characterized in that The fixing portion is provided with a first channel, the clamping portion is provided with a second channel, the first channel and the second channel are communicated, and the data transmission portion passes through the first channel and the second channel; An accommodating cavity is provided at one end of the clamping portion away from the fixing portion, the accommodating cavity is communicated with the second channel, the detection device is built in the accommodating cavity and fixed to the clamping portion through a threaded hole of the accommodating cavity.

3. The system according to claim 1, characterized in that The fixing portion is provided with a pressure relief hole, and the pressure relief hole is connected with the wellbore annulus through a pressure relief pipeline, so as to release the pressure in the wellbore annulus.

4. The system according to claim 1, characterized in that The connecting device comprises a device protection part, which is arranged in the accommodating cavity of the clamping part and located between the inner wall of the accommodating cavity and the detection device, and the device protection part surrounds the periphery of the detection device.

5. The system according to claim 1, characterized in that The data processing device performs denoising processing on the sound wave signal, including: Decomposing the sound wave signal based on the mode of the sound wave signal to obtain multiple sound wave signal components; Removing the sound wave signal components exceeding a preset signal threshold from the multiple sound wave signal components to obtain at least one sound wave signal component after noise reduction; A noise-reduced sound wave signal is generated based on the at least one noise-reduced sound wave signal component.

6. The system according to claim 5, characterized in that Based on the mode of the sound wave signal, the sound wave signal is decomposed to obtain multiple sound wave signal components, including: Performing feature extraction on the acoustic wave signal to determine the mode number, iteration penalty coefficient, and iteration convergence tolerance of the acoustic wave signal, wherein each mode corresponds to an acoustic wave signal component; Based on preset constraints and iteration parameters, an iteration constraint model of the acoustic wave signal is determined, wherein the preset constraints include: the sum of the bandwidths of the center frequencies of all modes is minimum, and the sum of the acoustic wave signal components corresponding to all modes matches the acoustic wave signal; Iteratively searching for an optimal solution of the iterative constraint model in a current mode based on the iterative constraint model, the iterative penalty coefficient, and the iterative convergence tolerance, and determining a center frequency and a bandwidth of the current mode based on the optimal solution; The acoustic wave signal is decomposed based on the center frequency and bandwidth of each mode to obtain the multiple acoustic wave signal components.

7. A method for detecting oil casing leakage, applied to the oil casing leakage detection system according to any one of claims 1 to 6, characterized in that: include: The detection device detects the acoustic wave signal in the wellbore annulus and sends the acoustic wave signal to the data transmission part of the connection device; The data transmission part of the connection device receives the sound wave signal sent by the detection device and transmits the sound wave signal to the data processing device; The data processing device receives the acoustic wave signal transmitted by the data transmission unit, performs denoising on the acoustic wave signal, inputs the denoised acoustic wave signal into a pre-trained oil casing leakage detection model, and outputs an oil casing leakage detection result.

8. The method according to claim 7, characterized in that The data processing device performs denoising processing on the sound wave signal, including: The data processing device decomposes the sound wave signal based on the mode of the sound wave signal to obtain multiple sound wave signal components; The data processing device removes the sound wave signal components exceeding a preset signal threshold from the multiple sound wave signal components to obtain at least one sound wave signal component after noise reduction; The data processing device generates a noise-reduced sound wave signal based on at least one noise-reduced sound wave signal component.

9. The method according to claim 8, characterized in that The data processing device decomposes the sound wave signal based on the mode of the sound wave signal to obtain multiple sound wave signal components, including: The data processing device performs feature extraction on the acoustic wave signal to determine the mode number, iteration penalty coefficient and iteration convergence tolerance of the acoustic wave signal, wherein each mode corresponds to an acoustic wave signal component; The data processing device determines an iterative constraint model of the acoustic wave signal based on preset constraints and iteration parameters, wherein the preset constraints include: the sum of the bandwidths of the center frequencies of all modes is minimum, and the sum of the acoustic wave signal components corresponding to all modes matches the acoustic wave signal; The data processing device iteratively searches for an optimal solution of the iterative constraint model in a current mode based on the iterative constraint model, the iterative penalty coefficient, and the iterative convergence tolerance, and determines a center frequency and a bandwidth of the current mode based on the optimal solution; The data processing device decomposes the acoustic wave signal based on the central frequency and bandwidth of each mode to obtain the multiple acoustic wave signal components.

10. The method according to claim 7, characterized in that The oil casing leakage detection model is trained in the following way: The data processing device acquires historical acoustic wave data and oil casing leakage detection data corresponding to the historical acoustic wave data as a training data set; The data processing device determines an initial deep learning model based on the training data set, wherein the initial deep learning model includes an acoustic wave mapping function, and the acoustic wave mapping function is used to map the historical acoustic wave data to a target feature space; The data processing device inputs the training data set into the initial deep learning model, trains the initial deep learning model, and obtains the oil casing leakage detection model.

11. The method according to claim 10, characterized in that The data processing device determines an initial deep learning model based on the training data set, including: The data processing device determines feature data of historical sound wave data in the training data set; The data processing device determines the acoustic wave mapping function of the initial deep learning model based on the distribution characteristics of the feature data, and constructs the initial deep learning model based on the acoustic wave mapping function.

12. A computer storage medium, characterized in that: The computer storage medium stores computer program instructions, and when the computer program instructions are executed, the steps of the method according to any one of claims 7 to 11 are implemented.

Images