Shaft interlayer fluid channeling restraining device of coal bed gas multi-layer commingling production well

Abstract

The invention discloses a shaft interlayer fluid channeling restraining device for a coal bed gas multi-layer commingling production well, and relates to the technical field of petroleum and natural gas extraction. According to the coalbed methane multi-layer commingling production well shaft interlayer fluid channeling restraining device, a layering state sensing unit is used for obtaining hydrodynamic data and structural response data of a plurality of shaft layers in real time and analyzing corresponding multiphase flow stability indexes and shaft structural disturbance response indexes; the sound wave identification and analysis unit is used for acquiring sound wave signal data and analyzing a sound wave disturbance identification index in combination with a pre-trained sound wave identification model; and the fluid channeling risk analysis unit is used for carrying out weighted analysis on the multiphase flow stability index, the wellbore structure disturbance response index and the acoustic disturbance identification index to obtain a fluid channeling risk index. According to the fluid channeling risk index in the packing feedback adjustment unit, intelligent inhibition is carried out on a packer of each wellbore layer of a set coal seam; therefore, the recognition accuracy and the dynamic sealing control capability of the complex interlayer fluid channeling path are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of oil and gas exploitation, and particularly to a device for suppressing interlayer crossflow in a wellbore of a multi-layer combined production well for coalbed methane. Background Art

[0002] During the exploitation of coalbed methane, the multi-layer combined production technology is used to simultaneously exploit multiple coal seams in a single well, which can significantly reduce the input cost and increase the gas production of a single well. However, due to the heterogeneous characteristics of the coal-bearing reservoir, there are large differences in physical properties such as reservoir pressure and permeability grade among different coal seams, which easily leads to interlayer crossflow. For example, the fluid in the high-pressure reservoir layer invades the low-pressure reservoir layer, resulting in a decrease in gas production efficiency or even wellbore blockage. Based on the existing technology, the monitoring of multi-phase flow in the wellbore mainly relies on stage artificial logging or single-parameter sensors, and it is difficult to achieve real-time dynamic feedback; the suppression of interlayer crossflow mainly uses mechanical packers or chemical plugging, but lacks the adaptive adjustment ability linked with real-time monitoring, resulting in unstable plugging effects.

[0003] The limitations of the existing technology at least include the following problems. In the existing technology, relying on the data of a single type of sensor is difficult to reflect the coupling mechanism of flow disturbance and structural response. As a result, during the risk identification of interlayer crossflow in the wellbore, the linkage characteristics between the dynamic evolution of multi-phase flow and the strain of the wellbore structure are often ignored, and it is difficult to effectively capture the formation process of potential abnormal fluid channels. At the same time, there is a lack of an adaptive packer adjustment mechanism for real-time response based on monitoring results, and it is difficult to achieve hierarchical differential plugging control for different wellbore layers and different disturbance types. Especially in the multi-layer combined production working condition, problems such as response lag and regulation failure are likely to occur, thus seriously restricting the accurate identification and dynamic sealing control ability of the interlayer crossflow path, with limited overall suppression effect and difficult to meet the operation requirements of high stability and high tightness. Summary of the Invention

[0004] Aiming at the deficiencies of the existing technology, the present invention provides a device for suppressing interlayer crossflow in a wellbore of a multi-layer combined production well for coalbed methane, which solves the problems that it is difficult for the existing technology to accurately identify the crossflow risk and achieve real-time differential sealing and suppression.

[0005] To achieve the above objectives, the present invention is realized through the following technical solutions: A device for suppressing interlayer crossflow in the wellbore of a multi-layer coalbed methane production well, comprising the following steps: a stratified state perception unit for real-time obtaining hydrodynamic data and structural response data of several wellbore layers of a set coal seam, and respectively analyzing the multiphase flow stability index and the wellbore structure disturbance response index of each wellbore layer of the set coal seam; an acoustic wave identification and analysis unit for obtaining acoustic wave signal data of each wellbore layer of the set coal seam and performing comprehensive analysis in combination with a pre-trained acoustic wave identification model to obtain the acoustic wave disturbance identification index of each wellbore layer of the set coal seam; a crossflow risk analysis unit for performing weighted analysis on the multiphase flow stability index, the wellbore structure disturbance response index, and the acoustic wave disturbance identification index of each wellbore layer of the set coal seam to obtain the crossflow risk index of each wellbore layer of the set coal seam; a packer feedback adjustment unit for intelligently suppressing the packers of each wellbore layer of the set coal seam based on the crossflow risk index.

[0006] Further, the hydrodynamic data includes fluid pressure value, temperature value, flow velocity value, gas-liquid ratio value, and flow angle value. The specific steps for analyzing the multiphase flow stability index of each wellbore layer of the set coal seam are as follows: Read the hydrodynamic data of several wellbore layers of the set coal seam and analyze the fluid evaluation set of each wellbore layer of the set coal seam, including the flow disturbance intensity index and the gas-liquid structure equilibrium index; perform weighted analysis on the fluid evaluation set of each wellbore layer of the set coal seam to obtain the multiphase flow stability index of each wellbore layer of the set coal seam.

[0007] Further, the specific steps for analyzing the fluid evaluation set of each wellbore layer of the set coal seam are as follows: Perform weighted analysis on the flow velocity value and the flow angle value of each wellbore layer of the set coal seam based on Z-Score standardization to obtain the flow disturbance intensity index of each wellbore layer of the set coal seam; perform weighted analysis on the temperature value, the flow velocity value, and the gas-liquid ratio value of each wellbore layer of the set coal seam based on Z-Score standardization to obtain the gas-liquid structure equilibrium index of each wellbore layer of the set coal seam.

[0008] Further, the structural response data includes vibration acceleration value, axial stress value, radial strain value, and attitude deflection angle value. The specific steps for analyzing the wellbore structure disturbance response index of each wellbore layer of the set coal seam are as follows: Read the vibration acceleration value, axial stress value, radial strain value, and attitude deflection angle value of each wellbore layer of the set coal seam and perform standardization processing; perform weighted analysis on the vibration acceleration value, axial stress value, radial strain value, and attitude deflection angle value of each wellbore layer of the set coal seam after standardization processing to obtain the wellbore structure disturbance response index of each wellbore layer of the set coal seam.

[0009] Further, the acoustic wave signal data is specifically the sound pressure amplitude at each time point, and the acoustic wave recognition model is specifically a convolutional neural network, including an input layer, an encoding layer, a joint modeling layer, and an output mapping layer.

[0010] Further, the specific steps for obtaining the acoustic wave disturbance recognition index of each wellbore layer of the set coal seam are as follows: Input the acoustic wave signal data of each wellbore layer of the set coal seam into a pre-trained convolutional neural network for feature analysis to obtain the acoustic wave feature set of each wellbore layer of the set coal seam, including an abnormal waveform offset index, a recognition confidence index, and a spectral energy centroid offset index; perform weighted analysis on the acoustic wave feature set of each wellbore layer of the set coal seam to obtain the acoustic wave disturbance recognition index of each wellbore layer of the set coal seam.

[0011] Further, the specific steps for obtaining the acoustic wave feature set of each wellbore layer of the set coal seam are as follows: In the input layer of the convolutional neural network, receive the acoustic wave signal data of each wellbore layer of the set coal seam and perform preprocessing; in the encoding layer of the convolutional neural network, perform convolutional extraction processing on the preprocessed acoustic wave signal data of each wellbore layer of the set coal seam to obtain the disturbance feature tensor of each wellbore layer of the set coal seam; in the joint modeling layer of the convolutional neural network, perform waveform and spectrum joint modeling processing on the disturbance feature tensor of each wellbore layer of the set coal seam to obtain the fusion feature tensor of each wellbore layer of the set coal seam; in the output mapping layer of the convolutional neural network, perform extraction processing on the fusion feature tensor of each wellbore layer of the set coal seam to obtain the abnormal waveform offset index, the recognition confidence index, and the spectral energy centroid offset index of each wellbore layer of the set coal seam.

[0012] Further, the specific steps for intelligently suppressing the packer of each wellbore layer of the set coal seam based on the crossflow risk index are as follows: Compare and analyze the crossflow risk index of each wellbore layer of the set coal seam with a preset crossflow risk index threshold respectively; if the crossflow risk index of each wellbore layer of the set coal seam is lower than or equal to the preset crossflow risk index threshold, do not perform suppression processing on the packer of the corresponding wellbore layer of the set coal seam; if the crossflow risk index of each wellbore layer of the set coal seam is higher than the preset crossflow risk index threshold, perform suppression processing on the packer of the corresponding wellbore layer of the set coal seam.

[0013] The present invention has the following beneficial effects: (1). The device for suppressing interlayer crossflow in the wellbore of a multi-layer coalbed methane commingled production well obtains hydrodynamic data, structural response data, and acoustic signal data through the collaborative work of a layered state sensing unit and an acoustic wave identification and analysis unit, constructs a triple evaluation system including a multiphase flow stability index, a wellbore structure disturbance response index, and an acoustic wave disturbance identification index, and generates a crossflow risk index weighted by a crossflow risk analysis unit, so that the packer feedback adjustment unit can implement layered identification and differential suppression for the disturbance intensity of different wellbore layers, and then can achieve rapid response and precise intervention for abnormal fluid channels, avoid problems of regulation lag and suppression failure, and then improve the identification accuracy and dynamic sealing control ability for complex interlayer crossflow paths, so as to meet the wellbore operation requirements of high stability and high tightness.

[0014] (2). The device for suppressing interlayer crossflow in the wellbore of a multi-layer coalbed methane commingled production well, through the constructed acoustic wave disturbance identification model and by integrating a mesoscale convolution and a dilated convolution module, respectively extracts the waveform change characteristics and spectral distribution characteristics in the acoustic wave signal, and thus generates corresponding indices through a multi-channel fully connected structure in the output mapping layer, so as to achieve refined and multi-dimensional discrimination of disturbance anomalies, and use it as a core input parameter to participate in the evaluation of the crossflow risk index, jointly determine the packing strategy of the wellbore layer with multiphase flow and structural disturbance information, and immediately physically block potential crossflow channels when the risk index is higher than the threshold, and then realizes the full-process closed-loop regulation from intelligent identification of acoustic wave disturbance to physical sealing control execution, and then effectively improves the intelligent level of the system and the timeliness and accuracy of risk suppression.

[0015] Of course, it is not necessary for any product implementing the present invention to achieve all the above-mentioned advantages simultaneously. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a block diagram of a device for suppressing interlayer crossflow in the wellbore of a multi-layer coalbed methane commingled production well according to the present invention.

[0017] Figure 2 It is a specific step flowchart for obtaining the acoustic wave disturbance identification index of each wellbore layer of a set coal seam in a device for suppressing interlayer crossflow in the wellbore of a multi-layer coalbed methane commingled production well according to the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0018] Please refer to Figure 1, an embodiment of the present invention provides a technical solution: a device for suppressing interlayer channeling in the wellbore of a multi-layer coalbed methane production well, including the following steps: a stratification state perception unit, configured to obtain in real time (where "in real time" here refers to a preset extremely short time window, such as 0.3 s) the hydrodynamic data and structural response data of several wellbore layers of a set coal seam, and analyze the multiphase flow stability index and wellbore structure disturbance response index of each wellbore layer of the set coal seam respectively; an acoustic wave identification and analysis unit, configured to obtain the acoustic wave signal data of each wellbore layer of the set coal seam, and perform comprehensive analysis in combination with a pre-trained acoustic wave identification model to obtain the acoustic wave disturbance identification index of each wellbore layer of the set coal seam; a channeling risk analysis unit, configured to perform weighted analysis on the multiphase flow stability index, wellbore structure disturbance response index, and acoustic wave disturbance identification index of each wellbore layer of the set coal seam to obtain the channeling risk index of each wellbore layer of the set coal seam; a packer feedback adjustment unit, configured to perform intelligent suppression on the packers of each wellbore layer of the set coal seam based on the channeling risk index.

[0019] The specific steps for performing intelligent suppression on the packers of each wellbore layer of the set coal seam based on the channeling risk index are as follows: Compare and analyze the channeling risk index of each wellbore layer of the set coal seam with a preset channeling risk index threshold respectively; If the channeling risk index of each wellbore layer of the set coal seam is lower than or equal to the preset channeling risk index threshold, do not suppress the packer of the corresponding wellbore layer of the set coal seam; If the channeling risk index of each wellbore layer of the set coal seam is higher than the preset channeling risk index threshold, perform suppression processing on the packer of the corresponding wellbore layer of the set coal seam, that is, the intelligent packer corresponding to this wellbore layer performs an expansion operation to selectively block possible channeling channels. The shape memory alloy ring inside the intelligent packer and the hydraulic drive unit cooperate to make the packer expand radially and fit the inner wall of the wellbore at the target layer position, realizing physical isolation of abnormal fluid channels and suppressing the channeling path.

[0020] Specifically, the hydrodynamic data includes fluid pressure value, temperature value, flow velocity value, gas-liquid ratio, and flow angle value. The specific steps for analyzing the multiphase flow stability index of each wellbore layer of the set coal seam are as follows: Read the hydrodynamic data of several wellbore layers of the set coal seam, and analyze the fluid evaluation set of each wellbore layer of the set coal seam, including flow disturbance intensity index and gas-liquid structure equilibrium index; Perform weighted analysis on the fluid evaluation set of each wellbore layer of the set coal seam to obtain the multiphase flow stability index of each wellbore layer of the set coal seam.

[0021] Among them, the fluid pressure value is obtained through an integrated pressure sensor (using a piezoresistive MEMS sensor with a measurement range covering 0 - 50 MPa, a temperature resistance of 150 °C, the shell is treated with 316L stainless steel for corrosion protection, and an internal temperature compensation module is built in).

[0022] The temperature value is obtained through a temperature sensor (platinum resistance probe, accuracy ±0.1 °C, in direct contact with the wellbore fluid through a heat conduction module).

[0023] The flow velocity value is obtained through an ultrasonic flowmeter (dual-probe opposed design, emission frequency 1 - 5 MHz, measurement range 0.1 - 10 m / s, supports bidirectional flow detection, installed on the inner wall of the casing and equipped with an anti-scaling coating).

[0024] The gas-liquid ratio is obtained through a capacitive gas holdup meter (ring-shaped capacitive electrode design, detects the gas holdup of gas-liquid two-phase flow through the difference in dielectric constant, resolution ±2%, built-in self-cleaning electrodes to prevent coal dust adhesion).

[0025] The flow angle value is the spatial angle of the wellbore fluid flow vector relative to the direction of the wellbore main axis, and is directly obtained through a three-dimensional ultrasonic Doppler flow direction sensor arranged in the wellbore. The sensor consists of at least three non-coplanar ultrasonic transducers. By receiving the ultrasonic signal frequency shift reflected by suspended particles in the fluid, the fluid velocity vector in the wellbore is calculated, and further the angle between this vector and the wellbore axis is solved as a real-time numerical index of the deviation degree of the local flow direction in the wellbore.

[0026] The specific steps for analyzing the fluid evaluation set of each wellbore layer of the set coal seam are as follows: Based on Z-Score standardization, weighted analysis is performed on the flow velocity value and flow angle value of each wellbore layer of the set coal seam (that is, based on Z-Score standardization, the unit of the flow velocity value and flow angle value of each wellbore layer of the set coal seam is removed, and weighted processing is performed based on the result after removing the unit), to obtain the flow disturbance intensity index of each wellbore layer of the set coal seam; Based on Z-Score standardization, weighted analysis is performed on the temperature value, flow velocity value, and gas-liquid ratio of each wellbore layer of the set coal seam (that is, based on Z-Score standardization, the unit of the temperature value, flow velocity value, and gas-liquid ratio of each wellbore layer of the set coal seam is removed, and weighted processing is performed based on the result after removing the unit), to obtain the gas-liquid structure equilibrium index of each wellbore layer of the set coal seam.

[0027] In this implementation plan, multi-source fluid dynamic data is introduced for the wellbore multiphase flow state, and unit removal processing is performed through the Z-Score standardization method, so as to be able to eliminate the influence of the dimension and value differences of different physical quantities on the calculation results, and does not improve the comparability between indicators and the accuracy of fusion analysis. Secondly, the flow disturbance intensity index is obtained through the combined analysis of flow velocity and flow angle data, which can accurately depict the fluid disturbance intensity and deviation trend in the local wellbore; the gas-liquid structure equilibrium index is based on the collaborative modeling of temperature, flow velocity, and gas-liquid ratio, effectively characterizing the stability of the two-phase flow structure and the energy exchange characteristics, thus improving the identification sensitivity of the crossflow inducement, and also providing high-resolution and traceable basic parameter support for the subsequent construction of risk indexes.

[0028] Specifically, the structural response data includes vibration acceleration values, axial stress values, radial strain values, and attitude deflection angle values. The specific steps for analyzing the wellbore structure disturbance response index of each wellbore layer of the set coal seam are as follows: Read the vibration acceleration values, axial stress values, radial strain values, and attitude deflection angle values of each wellbore layer of the set coal seam, and perform standardization processing; Perform weighted analysis on the vibration acceleration values, axial stress values, radial strain values, and attitude deflection angle values of each wellbore layer of the set coal seam after standardization processing to obtain the wellbore structure disturbance response index of each wellbore layer of the set coal seam.

[0029] Among them, the vibration acceleration value is obtained through a piezoelectric acceleration sensor.

[0030] The axial stress value is the stress magnitude per unit area borne by the wellbore structure along the wellbore axis direction (usually the vertical direction), and it can be obtained through a fiber optic strain gauge.

[0031] The radial strain value is the relative deformation amount of the wellbore structure in the radial direction (perpendicular to the wellbore axis), and it can be obtained through an FBG fiber optic strain gauge.

[0032] The attitude deflection angle value is the spatial attitude deflection angle of the packer, which are the tilt angles along the three coordinate axes (pitch, roll, yaw, all obtained through a three-axis MEMS inclinometer sensor). Obtain the historical tilt angles (historical pitch, historical roll, historical yaw) at several historical time points, and perform mean processing to obtain the tilt reference angles (pitch reference, roll reference, yaw reference), and perform ratio processing with the tilt angles (pitch, roll, yaw) respectively. Based on the ratio results, perform weighted processing, and the obtained result is the attitude deflection angle value.

[0033] In this implementation plan, by obtaining multi-dimensional structural response parameters such as vibration acceleration, axial stress, radial strain, and attitude deflection angle, and uniformly performing standardization and weighted analysis, the high-precision quantification of the wellbore structure disturbance state can be realized, so as to comprehensively reflect the stability state of the wellbore structure during the multi-layer combined mining process. Especially, the attitude deflection angle value is modeled based on the historical mean ratio, which can effectively identify the attitude deviation and abnormal deformation trend of the packer, improve the early identification ability of the structural instability risk, and then enhance the sensitivity and robustness of the risk assessment, providing a precise decision-making basis for the subsequent intelligent packer response.

[0034] Specifically, the acoustic signal data is specifically the sound pressure amplitude at each time point (the time point within a preset extremely short time window), and the acoustic recognition model is specifically a convolutional neural network, including an input layer, an encoding layer, a joint modeling layer, and an output mapping layer.

[0035] Such as Figure 2As shown below, the specific steps to obtain the acoustic wave disturbance recognition index for each shaft layer of the set coal seam are as follows: Input the acoustic wave signal data of each shaft layer of the set coal seam into a pre-trained convolutional neural network for feature analysis to obtain the acoustic wave feature set for each shaft layer of the set coal seam, including the abnormal waveform offset index, recognition confidence index, and spectral energy centroid offset index; perform weighted analysis on the acoustic wave feature set for each shaft layer of the set coal seam to obtain the acoustic wave disturbance recognition index for each shaft layer of the set coal seam.

[0036] The specific steps to obtain the acoustic wave feature set for each shaft layer of the set coal seam are as follows: In the input layer of the convolutional neural network, receive the acoustic wave signal data of each shaft layer of the set coal seam and perform preprocessing; in the encoding layer of the convolutional neural network, perform convolutional extraction processing on the preprocessed acoustic wave signal data of each shaft layer of the set coal seam (i.e., one-dimensional convolutional operation and non-linear activation processing, extracting local disturbance edge features and initial morphological change features in the acoustic wave signal and generating a disturbance feature tensor), to obtain the disturbance feature tensor for each shaft layer of the set coal seam; in the joint modeling layer of the convolutional neural network, perform waveform and spectrum joint modeling processing on the disturbance feature tensor for each shaft layer of the set coal seam (extracting waveform change features through medium-scale convolution and combining dilated convolution to mine spectral distribution features), to obtain the fused feature tensor for each shaft layer of the set coal seam; in the output mapping layer of the convolutional neural network, perform extraction processing on the fused feature tensor for each shaft layer of the set coal seam (performing a Flatten operation on the fused feature tensor to convert it into a one-dimensional feature vector; then input it into the first fully connected layer to perform linear transformation and ReLU activation operations to extract high-order disturbance patterns; then input it into the second fully connected layer, with three independent output channels respectively corresponding to three disturbance sub-indices; among them, the output of the first channel is used to generate the abnormal waveform offset index by fitting the waveform segment change rate and abnormal residuals; the output of the second channel is the recognition confidence index, indicating the probability that the model judges that this waveform belongs to the disturbance abnormal category; the output of the third channel is the spectral energy centroid offset index, calculated based on the relative offset value between the centroid coordinates of the frequency-domain main component vector and the reference state), to obtain the abnormal waveform offset index, recognition confidence index, and spectral energy centroid offset index for each shaft layer of the set coal seam.

[0037] The input layer is used to receive and preprocess the acoustic wave signal data.

[0038] The encoding layer is used to extract local disturbance features.

[0039] The joint modeling layer is used to construct waveform and spectrum features.

[0040] The output mapping layer is used to output various disturbance indices.

[0041] In this implementation, by constructing a convolutional neural network model with a clear structural division, local perturbation features, mesoscale waveform features, and spectral energy features are extracted in sequence, and finally three types of acoustic perturbation sub-indices with physical meanings are output, so as to achieve accurate identification and classification of abnormal acoustic perturbations in the wellbore layer. Secondly, the model adopts a serial structure, progressing layer by layer and fully integrating information, which can not only identify short-time mutation signals but also capture frequency-domain trend changes, adapting to various types of perturbation forms. Finally, the output mapping layer separates various acoustic features through a multi-channel fully connected structure, effectively improving the discrimination accuracy and stability of the indices, and the overall process can run in real time, thereby enhancing the intelligent processing ability and recognition reliability of acoustic data.

[0042] Although the preferred embodiments of the present invention have been described, those skilled in the art can make additional changes and modifications once they learn the basic creative concept. Therefore, the appended claims are intended to be construed as including the preferred embodiments as well as all changes and modifications falling within the scope of the present invention.

[0043] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention also intends to include these modifications and variations.

Claims

1. A device for suppressing interlayer crossflow in the wellbore of a multi-layer coalbed methane production well, characterized in that, Comprising: A layered state perception unit, configured to obtain in real time hydrodynamic data and structural response data of several wellbore layers of a set coal seam, and respectively analyze the multiphase flow stability index and the wellbore structure disturbance response index of each wellbore layer of the set coal seam; An acoustic wave identification and analysis unit, configured to obtain acoustic wave signal data of each wellbore layer of the set coal seam, and perform comprehensive analysis in combination with a pre-trained acoustic wave identification model to obtain the acoustic wave disturbance identification index of each wellbore layer of the set coal seam; A crossflow risk analysis unit, configured to perform weighted analysis on the multiphase flow stability index, the wellbore structure disturbance response index, and the acoustic wave disturbance identification index of each wellbore layer of the set coal seam to obtain the crossflow risk index of each wellbore layer of the set coal seam; A packer feedback adjustment unit, configured to perform intelligent suppression on the packers of each wellbore layer of the set coal seam based on the crossflow risk index.

2. The device for suppressing interlayer channeling in the wellbore of a coalbed methane multi-layer combined production well according to claim 1, wherein The hydrodynamic data includes fluid pressure value, temperature value, flow velocity value, gas-liquid ratio value, and flow angle value. The specific steps for analyzing the multiphase flow stability index of each wellbore layer of the set coal seam are as follows: Read the hydrodynamic data of several wellbore layers of the set coal seam, and analyze the fluid evaluation set of each wellbore layer of the set coal seam, including the flow disturbance intensity index and the gas-liquid structure equilibrium index; Perform weighted analysis on the fluid evaluation set of each wellbore layer of the set coal seam to obtain the multiphase flow stability index of each wellbore layer of the set coal seam.

3. The device for suppressing interlayer crossflow in the wellbore of a coalbed methane multi-layer combined production well according to claim 2, wherein The specific steps for analyzing the fluid evaluation set of each wellbore layer of the set coal seam are as follows: Perform weighted analysis on the flow velocity value and the flow angle value of each wellbore layer of the set coal seam based on Z-Score standardization to obtain the flow disturbance intensity index of each wellbore layer of the set coal seam; Perform weighted analysis on the temperature value, the flow velocity value, and the gas-liquid ratio value of each wellbore layer of the set coal seam based on Z-Score standardization to obtain the gas-liquid structure equilibrium index of each wellbore layer of the set coal seam.

4. The interlayer crossflow inhibition device for the wellbore of a coalbed methane multi-layer combined production well according to claim 1, wherein The structural response data includes vibration acceleration value, axial stress value, radial strain value, and attitude deflection angle value. The specific steps for analyzing the wellbore structure disturbance response index of each wellbore layer of the set coal seam are as follows: Read the vibration acceleration value, axial stress value, radial strain value, and attitude deflection angle value of each wellbore layer of the set coal seam, and perform standardization processing; Perform weighted analysis on the vibration acceleration value, axial stress value, radial strain value, and attitude deflection angle value of each wellbore layer of the set coal seam after standardization processing to obtain the wellbore structure disturbance response index of each wellbore layer of the set coal seam.

5. The interlayer crossflow suppression device for a coalbed methane multi-layer combined production wellbore according to claim 1, characterized in that The acoustic wave signal data is specifically the sound pressure amplitude at each time point, and the acoustic wave identification model is specifically a convolutional neural network, including an input layer, an encoding layer, a joint modeling layer, and an output mapping layer.

6. The interlayer crossflow inhibition device for the wellbore of a multi-layer coalbed methane commingled production well according to claim 5, characterized in that, The specific steps for obtaining the acoustic wave disturbance identification index of each wellbore layer of the set coal seam are as follows: Input the acoustic wave signal data of each wellbore layer of the set coal seam into the pre-trained convolutional neural network for feature analysis to obtain the acoustic wave feature set of each wellbore layer of the set coal seam, including the abnormal waveform offset index, the recognition confidence index, and the spectral energy centroid offset index; Perform weighted analysis on the acoustic wave feature sets of each shaft layer of the set coal seam to obtain the acoustic wave disturbance identification index of each shaft layer of the set coal seam.

7. The device for suppressing interlayer crossflow in the wellbore of a coalbed methane multi-layer combined production well according to claim 6, wherein The specific steps to obtain the acoustic wave feature sets of each shaft layer of the set coal seam are as follows: In the input layer of the convolutional neural network, receive the acoustic wave signal data of each shaft layer of the set coal seam and perform preprocessing; In the encoding layer of the convolutional neural network, perform convolutional extraction processing on the preprocessed acoustic wave signal data of each shaft layer of the set coal seam to obtain the disturbance feature tensors of each shaft layer of the set coal seam; In the joint modeling layer of the convolutional neural network, perform joint waveform and spectrum modeling processing on the disturbance feature tensors of each shaft layer of the set coal seam to obtain the fusion feature tensors of each shaft layer of the set coal seam; In the output mapping layer of the convolutional neural network, perform extraction processing on the fusion feature tensors of each shaft layer of the set coal seam to obtain the abnormal waveform offset index, recognition confidence index, and spectrum energy centroid offset index of each shaft layer of the set coal seam.

8. The device for suppressing interlayer crossflow in the wellbore of a coalbed methane multi-layer combined production well according to claim 1, characterized in that, The specific steps for intelligent suppression of the packers of each shaft layer of the set coal seam based on the crossflow risk index are as follows: Compare and analyze the crossflow risk indices of each shaft layer of the set coal seam with the preset crossflow risk index thresholds respectively; If the crossflow risk index of each shaft layer of the set coal seam is lower than or equal to the preset crossflow risk index threshold, no suppression treatment is performed on the packers of the corresponding shaft layer of the set coal seam; If the crossflow risk index of each shaft layer of the set coal seam is higher than the preset crossflow risk index threshold, suppression treatment is performed on the packers of the corresponding shaft layer of the set coal seam.

Images