Three-Dimensional Localization Method for Leakage of Subsea Production System Manifold Based on Underwater Acoustic Signals

Abstract

The present invention belongs to the technical field of petroleum engineering, and particularly relates to a three-dimensional positioning method for leaks in subsea production system manifolds based on underwater acoustic signals. By establishing the correlation of underwater acoustic signals between multiple channels of different hydrophones and setting a double-threshold detection method, the positioning and detection result of leaks in subsea production system manifolds is effectively improved in terms of accuracy and reliability, providing technical support for quickly determining the three-dimensional coordinate position of the leak sound source. The present invention provides a three-dimensional positioning method for leaks in subsea production system manifolds based on underwater acoustic signals, including the following steps: Step P1: Preprocess the underwater acoustic signals; Step P2: Calculate the time difference of arrival between the acoustic signals when the reference hydrophone and other hydrophones in multiple channels first detect a leak; Step P3: Construct a three-dimensional positioning model for the leak sound source and calculate the three-dimensional coordinates of the leak in the subsea production system manifold.

Description

Technical Field

[0001] The present invention belongs to the technical field of petroleum engineering, and particularly relates to a three-dimensional positioning method for leaks in a subsea production system pipeline network based on underwater acoustic signals. Background Art

[0002] The subsea production system pipeline network is composed of several pipeline modules and manifold modules, and is mainly used for collecting various fluids required in the oil and gas reservoir production process through subsea wellheads and Christmas trees, and then completing the gathering and transportation operations via subsea pipeline terminal equipment. However, due to the complex and changeable working environment of the subsea production system pipeline network, once a leakage failure occurs, it will not only cause significant economic losses, but may also cause serious damage to the marine ecological environment.

[0003] In order to detect and identify possible leakage phenomena in the subsea production system pipeline network, hydrophone devices are usually used in the prior art to collect underwater acoustic signals generated during the operation of the subsea production system pipeline network, and to screen and analyze the above signals. This technical method is based on the sound waves caused by leakage, and realizes the detection and identification of leakage phenomena based on its vibration characteristics, thereby providing a reliable basis for taking timely countermeasures.

[0004] However, in further research, the inventors found that due to the influence of multiple factors such as the complex pipeline structure of the subsea production system pipeline network and different types of noise pollution in the underwater environment, there are large errors in the existing leakage detection and positioning results based on underwater acoustic signals, and false alarms and missed alarms are likely to occur. The above situation greatly increases the difficulty of leakage detection in the subsea production system pipeline network. Therefore, it is urgent for those skilled in the art to design and provide a new three-dimensional positioning method for leaks in the subsea production system pipeline network, so as to be able to quickly respond to leakage events and provide help for ensuring the efficient and safe operation of the subsea production system pipeline network. Summary of the Invention

[0005] The present invention provides a three-dimensional positioning method for leaks in a subsea production system pipeline network based on underwater acoustic signals. This positioning method effectively improves the accuracy and reliability of the leakage positioning detection results of the subsea production system pipeline network by establishing the correlation of underwater acoustic signals between multiple channels of different hydrophones and setting a double threshold detection method, and provides technical support for quickly determining the three-dimensional coordinate position of the leakage sound source.

[0006] To solve the above technical problems, the present invention adopts the following technical solutions:

[0007] The present invention provides a three-dimensional positioning method for leaks in a subsea production system pipeline network based on underwater acoustic signals, including the following steps:

[0008] Step P1: Preprocess the underwater acoustic signals;

[0009] Step P2: Calculate the time difference of arrival between the underwater acoustic signals when the reference hydrophone and other hydrophones in the multi-channel hydrophone first detect a leak.

[0010] Step P3: Construct a three-dimensional localization model of the leak sound source and calculate the three-dimensional coordinates of the leak in the underwater production system manifold.

[0011] More preferably, the process of preprocessing the underwater acoustic signal in step P1 is specifically described as follows:

[0012] Step P101: Calculate the time-domain average value of the underwater acoustic signal; eliminate the DC component in the underwater acoustic signal to correct its baseline deviation.

[0013] Step P102: Implement time-domain segmentation of the underwater acoustic signal by using frame-by-frame windowing processing.

[0014] More preferably, the time-domain average value of the underwater acoustic signal and the underwater acoustic signal after eliminating the DC component in step P101 are specifically described as follows:

[0015] Express the underwater acoustic signal as x(t). The time-domain average value of the underwater acoustic signal satisfies:

[0016] (1);

[0017] In formula (1), N is the sample length of the underwater acoustic signal;

[0018] The underwater acoustic signal after eliminating the DC component satisfies:

[0019] (2).

[0020] More preferably, step P102 is specifically described as follows:

[0021] For an underwater acoustic signal with a sample length of N, take each frame length as l, the displacement of the latter frame relative to the former frame as m, and the overlapping part as o = l - m, and a total of frames are obtained;

[0022] Among them, Satisfies: (3);

[0023] Then the underwater acoustic signal after frame-by-frame windowing processing satisfies:

[0024] (⑷);

[0025] In formula (4), w(i) is the Hamming window function, and c is the windowing position;

[0026] Further, set the time-domain waveform of the underwater acoustic signal as s(n). Then the i-th frame of the underwater acoustic signal s obtained after frame-by-frame windowing processingi (n), satisfying:

[0027] (5);

[0028] Among them, the i-th frame underwater acoustic signal s i (n) short-term energy, satisfying:

[0029] (6);

[0030] In formula (5) and (6), .

[0031] Preferably, the process of calculating the arrival time difference between the water acoustic signals of the reference hydrophone and other hydrophones in the multi-channel hydrophone when the leak is first detected in step P2 is specifically described as follows:

[0032] Step P201: establishing a correlation relationship between the reference hydrophone and the underwater acoustic signals of other hydrophones in the multi-channel hydrophone;

[0033] Step P202: Statistically obtain the time point when each hydrophone in the multi-channel hydrophone detects a leak for the first time;

[0034] Step P203: Based on the time point when each hydrophone in the multi-channel hydrophone first detects a leak calculated in step P202, calculate the arrival time difference between the water acoustic signal when the reference hydrophone and other hydrophones in the multi-channel hydrophone first detect a leak.

[0035] Preferably, the process of establishing the correlation relationship between the water acoustic signals of the reference hydrophone and other hydrophones in the multi-channel hydrophone in step P201 is specifically described as follows:

[0036] Construct a multivariate regression model that integrates multi-channel hydrophone characteristics, frequency response parameters, and measured energy values;

[0037] Among them, the constructed multiple regression model satisfies:

[0038] (7);

[0039] In formula (7), S1 is the short-time energy value of the reference hydrophone in the multi-channel hydrophone; S i is the observation value of other hydrophones in the multi-channel hydrophone, ; β0, β1, ..., β n is the regression coefficient of the multiple regression model, S1 2 、S1 3 ,…,S1 n is the high-order power of the short-time energy value S1; ε is an independent distribution error term with constant variance;

[0040] Among them, the coefficient combinations β0, β1, …, β that best fit the predicted values and the observed values n , satisfy:

[0041] (8);

[0042] Construct matrix X, and according to the normal equations, solve to obtain the regression coefficients, which satisfy:

[0043] (9).

[0044] More preferably, the process of statistically obtaining the time nodes when each hydrophone in the multi-channel hydrophone first detects a leak in step P202 is specifically described as:

[0045] Set the double threshold values of the reference hydrophone in the multi-channel hydrophone;

[0046] Among them, the double threshold values of the reference hydrophone in the multi-channel hydrophone satisfy:

[0047] (10);

[0048] (11);

[0049] In equations (10) and (11), , ;

[0050] Based on the regression coefficients obtained by analyzing the multiple regression model, determine the double threshold values of the other hydrophones in the multi-channel hydrophone;

[0051] Among them, the double threshold values of the other hydrophones in the multi-channel hydrophone satisfy:

[0052] (12);

[0053] (13);

[0054] In equation (13), ;

[0055] Among them, the first threshold value in the double threshold values is used to initially screen the hydrophone signal when the multi-channel hydrophone first detects a leak, so as to determine the interval where the hydrophone signal is located when the leak is first detected; the second threshold value in the double threshold values is used to detect and determine the starting position of the hydrophone signal when the multi-channel hydrophone first detects a leak.

[0056] More preferably, the three-dimensional positioning model of the leak sound source constructed in step P3 is specifically described as:

[0057] The three-dimensional positioning model of the leakage sound source satisfies:

[0058] (14);

[0059] (15);

[0060] In formulas (14) and (15), the spatial coordinates of each hydrophone in the multi-channel hydrophone are (x i , y i , z i ), and the spatial coordinates of the leakage sound source to be determined are (x, y, z);

[0061] Formula (14) is further expressed as:

[0062] (16);

[0063] In formula (16), ;

[0064] And, according to formula (15), the following relationship can be further obtained:

[0065] (17);

[0066] Substitute formula (17) into formula (16), then we get:

[0067] (18);

[0068] In formula (18), , , .

[0069] The present invention provides a three-dimensional positioning method for the leakage of the pipeline manifold of an underwater production system based on underwater acoustic signals, which includes the following steps: Step P1: Preprocess the underwater acoustic signals; Step P2: Calculate the time difference of arrival between the acoustic signals detected for the first time by the reference hydrophone and other hydrophones in the multi-channel hydrophone; Step P3: Construct a three-dimensional positioning model of the leakage sound source and calculate the three-dimensional coordinates of the leakage of the pipeline manifold of the underwater production system.

[0070] The three-dimensional positioning method for the leakage of the pipeline manifold of an underwater production system based on underwater acoustic signals with the above step characteristics has at least the following technical advantages compared with the prior art:

[0071] (1). The three-dimensional positioning method for the leakage of the underwater production system manifold based on underwater acoustic signals provided by the present invention establishes the correlation relationship between the reference hydrophone and other hydrophones in the multi-channel hydrophone, achieving the technical effect of synchronous detection of the multi-channel hydrophone array. Compared with the traditional single-hydrophone leakage positioning method, this three-dimensional positioning method has higher positioning accuracy and more reliable positioning results;

[0072] (2). The three-dimensional positioning method for the leakage of the underwater production system manifold based on underwater acoustic signals provided by the present invention constructs a multiple regression model that incorporates the characteristics of multi-channel hydrophones, frequency response parameters, and measured energy values, achieving comprehensive response processing of environmental noise and acoustic wave propagation characteristics, greatly overcoming the influence of environmental noise on the underwater acoustic signals of leakage, and providing assistance for optimizing the accuracy of the sound source positioning results. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] The accompanying drawings are used to provide a further understanding of the present invention, and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention, but do not constitute a limitation to the present invention. In the following drawings:

[0074] Figure 1 It is a schematic flow chart of the three-dimensional positioning method for the leakage of the underwater production system manifold based on underwater acoustic signals provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0075] The present invention provides a three-dimensional positioning method for the leakage of the underwater production system manifold based on underwater acoustic signals. This positioning method effectively improves the accuracy and reliability of the leakage positioning detection results of the underwater production system manifold by establishing the correlation of underwater acoustic signals between different multi-channel hydrophones and setting a double threshold detection method, providing technical support for quickly determining the three-dimensional coordinate position of the leakage sound source.

[0076] As Figure 1 shown, the present invention provides a three-dimensional positioning method for the leakage of the underwater production system manifold based on underwater acoustic signals, including the following steps:

[0077] Step P1: Preprocess the underwater acoustic signals.

[0078] It should be noted that the underwater acoustic signals are interfered by factors such as environmental noise and the hydrophone for collecting underwater acoustic signals (specific frequencies introduced by its DC power supply), resulting in spectral aliasing. Therefore, when implementing the three-dimensional positioning of the leakage of the underwater production system manifold, it is necessary to first preprocess the underwater acoustic signals.

[0079] As a relatively preferred embodiment of the present invention, the process of preprocessing the underwater acoustic signals in step P1 is specifically described as:

[0080] Step P101: Calculate the time-domain average value of the underwater acoustic signal; eliminate the DC component in the underwater acoustic signal to correct its baseline deviation.

[0081] Step P102: Perform frame segmentation and windowing processing to achieve time-domain segmentation of the underwater acoustic signal.

[0082] Among them, the time-domain average value of the underwater acoustic signal and the underwater acoustic signal after eliminating the DC component in Step P101 are preferably specifically described as:

[0083] Express the underwater acoustic signal as x(t), and the time-domain average value of the underwater acoustic signal satisfies:

[0084] (1);

[0085] In formula (1), N is the sample length of the underwater acoustic signal.

[0086] The underwater acoustic signal after eliminating the DC component satisfies:

[0087] (2).

[0088] Step P102 is preferably specifically described as:

[0089] For the underwater acoustic signal with a sample length of N, take each frame length as l, the displacement of the latter frame relative to the former frame as m, and the overlapping part as o = l - m, and a total of frames are obtained;

[0090] Among them, satisfies: (3).

[0091] Then the underwater acoustic signal after frame segmentation and windowing processing satisfies:

[0092] (4);

[0093] In formula (4), w(i) is the Hamming window function, and c is the windowing position.

[0094] Further, set the time-domain waveform of the underwater acoustic signal as s(n), then the i-th frame of the underwater acoustic signal s i (n) obtained after frame segmentation and windowing processing satisfies:

[0095] (5).

[0096] Among them, the short-time energy of the i-th frame of the underwater acoustic signal s i (n) satisfies:

[0097] (6);

[0098] In formulas (5) and (6), 。

[0099] Step P2: Calculate the time difference of arrival between the underwater acoustic signals when the reference hydrophone and other hydrophones in the multi-channel hydrophone first detect a leak.

[0100] Based on the completion of Step P1, further calculate the time difference of arrival between the underwater acoustic signals when the reference hydrophone and other hydrophones first detect a leak.

[0101] As a relatively preferred embodiment of the present invention, the process of calculating the time difference of arrival between the underwater acoustic signals when the reference hydrophone and other hydrophones in the multi-channel hydrophone first detect a leak in Step P2 is specifically described as follows:

[0102] Step P201: Establish the correlation relationship between the underwater acoustic signals of the reference hydrophone and other hydrophones in the multi-channel hydrophone.

[0103] Step P202: Statistically obtain the time nodes when each hydrophone in the multi-channel hydrophone first detects a leak.

[0104] Step P203: According to the time nodes when each hydrophone in the multi-channel hydrophone first detects a leak calculated in Step P202, calculate the time difference of arrival between the underwater acoustic signals when the reference hydrophone and other hydrophones first detect a leak.

[0105] It should be noted that there are many composite factors affecting the underwater acoustic signals of the reference hydrophone and other hydrophones in the multi-channel hydrophone, such as: the inherent characteristics of different hydrophone sensors, the spatial arrangement of the multi-channel hydrophone array, environmental noise, acoustic wave propagation characteristics, etc. Therefore, it is necessary to determine the correlation between the underwater acoustic signals of the reference hydrophone and other hydrophones and use an appropriate threshold to improve the accuracy of screening underwater acoustic signals.

[0106] Preferably, the process of establishing the correlation relationship between the underwater acoustic signals of the reference hydrophone and other hydrophones in the multi-channel hydrophone in Step P201 is specifically described as follows:

[0107] Construct a multiple regression model that incorporates the characteristics of the multi-channel hydrophone, frequency response parameters, and measured energy values.

[0108] Among them, the constructed multiple regression model satisfies:

[0109] (7);

[0110] In Equation (7), S1 is the short-time energy value of the reference hydrophone in the multi-channel hydrophone; S i is the observed value of other hydrophones in the multi-channel hydrophone, ; β0, β1,..., βn is the regression coefficient of the multiple regression model, S1 2 、S1 3 、…、S1 n are the high powers of the short-time energy value S1; ε is an independent distribution error term with a constant variance.

[0111] It should be noted that this multiple regression model is specifically used to compensate for the sensitivity differences between hydrophones and simultaneously achieve noise reduction for the spatial heterogeneity differences of environmental noise.

[0112] Furthermore, the coefficient combinations β0, β1, …, β n that best fit the predicted values and the observed values satisfy:

[0113] (8);

[0114] Construct matrix X, and according to the normal equations, solve to obtain the regression coefficients, satisfying:

[0115] (9).

[0116] Additionally, to confirm that the above multiple regression model is truly valid. Here, the underwater acoustic signal dataset can be selected and substituted into the multiple regression model for solution. Among them, the underwater acoustic signal dataset is obtained by multi-channel hydrophones collecting multi-environment sound waves after standard tapping at fixed positions, so different working conditions can be contained in its dataset. After substituting the underwater acoustic signal dataset into the multiple regression model, the least squares method is used for parameter estimation; then, the sum of squared residuals is minimized through iterative optimization. It should be noted that as the data volume of the dataset continues to increase and the number of calculations accumulates, the regression coefficients gradually converge to a stable solution, and thus the coefficient combination that best fits the predicted values and the observed values is obtained.

[0117] Further preferably, the process of statistically obtaining the time nodes when each hydrophone in the multi-channel hydrophones first detects leakage in step P202 is specifically described as:

[0118] Set the double threshold values of the reference hydrophone in the multi-channel hydrophones;

[0119] Among them, the double threshold values of the reference hydrophone in the multi-channel hydrophones satisfy:

[0120] (10);

[0121] (11);

[0122] In equations (10) and (11), , ;

[0123] Determine the double threshold values of other hydrophones in the multi-channel hydrophone based on the regression coefficients obtained by parsing the multiple regression model;

[0124] Among them, the double threshold values of other hydrophones in the multi-channel hydrophone satisfy:

[0125] (12);

[0126] (13);

[0127] In formula (13), ;

[0128] Among them, the first threshold value in the double threshold value is used to preliminarily screen the hydrophone signal when the multi-channel hydrophone first detects a leak, so as to determine the interval where the hydrophone signal is located when the leak is first detected; the second threshold value in the double threshold value is used to detect and determine the starting position of the hydrophone signal when the multi-channel hydrophone first detects a leak.

[0129] It should be added that by setting the double threshold of each hydrophone as the leak criterion, the time nodes when each hydrophone first detects the leak signal can be accurately calibrated, and they are respectively marked as t1, t i , among them, . Then, the time difference of arrival between the reference hydrophone and other hydrophones in the multi-channel hydrophone when the underwater acoustic signal is first detected for leakage satisfies:

[0130] (19).

[0131] As a basic parameter for underwater acoustic propagation research, the accurate determination of the underwater acoustic propagation speed is directly related to the accuracy of sound source localization. In a fluid medium, sound waves propagate in the form of elastic longitudinal waves, and its propagation speed can be expressed by formula (20):

[0132] (20);

[0133] In formula (20), ρ is the medium density; κ is the adiabatic compression coefficient of the medium. Further research finds that both ρ and κ in seawater are functions of temperature, salinity, and pressure. Therefore, the sound speed in seawater is also a function of temperature, salinity, and pressure.

[0134] Step P3: Construct a three-dimensional localization model of the leakage sound source, and calculate the three-dimensional coordinates of the leakage of the underwater production system manifold.

[0135] On the basis of completing step P2, further implement step P3 to construct a three-dimensional localization model of the leakage sound source. On this basis, based on the three-dimensional localization model of the leakage sound source, combined with the known spatial position information of the multi-channel hydrophone array and the time difference data calculated in the above steps, the three-dimensional coordinates of the leakage sound source can be accurately estimated.

[0136] As a more preferred embodiment of the present invention, the three-dimensional localization model of the leakage sound source constructed in step P3 is specifically described as follows:

[0137] The three-dimensional localization model of the leakage sound source satisfies:

[0138] (14);

[0139] (15);

[0140] In formulas (14) and (15), the spatial coordinates of each hydrophone in the multi-channel hydrophone are (x i , y i , z i ), and the spatial coordinates of the leakage sound source to be determined are (x, y, z).

[0141] Formula (14) is further expressed as:

[0142] (16);

[0143] In formula (16), .

[0144] Moreover, according to formula (15), the following relationship can be further obtained:

[0145] (17).

[0146] Substitute formula (17) into formula (16), then we get:

[0147] (18);

[0148] In formula (18), , , .

[0149] It can be found that formula (18) eliminates the quadratic term of the unknowns and only retains the linear term, obtaining a system of linear equations. After obtaining the calculation result of formula (18), the following derivation can be further carried out to complete the three-dimensional localization of the leakage based on the arrival time of the underwater acoustic signals between the reference hydrophone and other hydrophones in the multi-channel hydrophone when the leakage is first detected.

[0150] Specifically, for example, when i = 2, 3, 4, there are the following expressions:

[0151] (21);

[0152] In formula (21), r i1 , K i , x i1 are known, and the unknown terms are r1, x, y, z. If r1 is assumed to be known, then formula (21) can be regarded as a system of linear equations for solution, and by using elimination, we get:

[0153] (22).

[0154] Formula (22) can be rewritten as:

[0155] (23).

[0156] That is, A x = C; using the inverse of the matrix, i.e., x = A-1C. The following three coordinates can be obtained:

[0157] (24);

[0158] In formula (24), C1 = (K2 - K1) - (r 21 2 + 2r 21 r1).

[0159] Substitute C1, C2, and C3 into formula (24) and simplify to get:

[0160] (25);

[0161] p1, q1 are shown in the following formulas (26), (27). The forms of p2, p3, q2, and q3 are the same as those of p1 and q1, so no more details will be given here. There is only one unknown solution r1 in formula (25); therefore, by finding r1, the specific location where the leakage occurs can be solved.

[0162] (26);

[0163] Substitute formula (25) into formula (15) and let i = 1, then the following formula (28) can be obtained. Formula (28) is a quadratic equation with r1 as the unknown, and two root values can be obtained. According to the prior information, an invalid root is discarded, and the valid r1 is brought back into formula (25) to find the three-dimensional coordinate position of the leakage sound source.

[0164] (27);

[0165] (28).

[0166] So far, through the above calculations, the three-dimensional positioning method for the leakage of the subsea production system pipeline based on underwater acoustic signals provided by the present invention finally realizes the three-dimensional positioning of the leakage of the subsea production system pipeline.

[0167] The present invention provides a three-dimensional positioning method for the leakage of the subsea production system pipeline based on underwater acoustic signals, which includes the following steps: Step P1: Preprocess the underwater acoustic signals; Step P2: Calculate the time difference of arrival between the underwater acoustic signals when the reference hydrophone and other hydrophones in the multi-channel hydrophone first detect the leakage; Step P3: Construct a three-dimensional positioning model of the leakage sound source and calculate the three-dimensional coordinates of the leakage of the subsea production system pipeline.

[0168] The three-dimensional positioning method for the leakage of the subsea production system pipeline based on underwater acoustic signals with the above step characteristics has at least the following technical advantages compared with the prior art:

[0169] (1) The three-dimensional positioning method for the leakage of the subsea production system pipeline based on underwater acoustic signals provided by the present invention establishes the correlation relationship between the underwater acoustic signals of the reference hydrophone and other hydrophones in the multi-channel hydrophone, and realizes the technical effect of synchronous detection of the multi-channel hydrophone array. Compared with the traditional single-hydrophone leakage positioning method, this three-dimensional positioning method has higher positioning accuracy and more reliable positioning results;

[0170] (2) The three-dimensional positioning method for the leakage of the subsea production system pipeline based on underwater acoustic signals provided by the present invention constructs a multiple regression model that integrates the characteristics of multi-channel hydrophones, frequency response parameters and measured energy values, and realizes the comprehensive response processing of environmental noise and acoustic wave propagation characteristics, greatly overcoming the influence of environmental noise on the leakage underwater acoustic signals and helping to optimize the accuracy of the sound source positioning results.

[0171] The above is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, and all should be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. A three-dimensional positioning method for the leakage of manifolds in an underwater production system based on underwater acoustic signals, characterized in that, It includes the following steps: Step P1: Preprocess the underwater acoustic signal; Step P2: Calculate the time difference of arrival between the underwater acoustic signals when the reference hydrophone and other hydrophones in the multi-channel hydrophone first detect a leak; Step P3: Construct a three-dimensional localization model of the leak sound source and calculate the three-dimensional coordinates of the leak in the underwater production system manifold; The process of calculating the time difference of arrival between the underwater acoustic signals when the reference hydrophone and other hydrophones in the multi-channel hydrophone first detect a leak in Step P2 is specifically described as follows: Step P201: Establish the correlation relationship between the underwater acoustic signals of the reference hydrophone and other hydrophones in the multi-channel hydrophone; Step P202: Statistically obtain the time nodes when each hydrophone in the multi-channel hydrophone first detects a leak; Step P203: Calculate the time difference of arrival between the underwater acoustic signals when the reference hydrophone and other hydrophones in the multi-channel hydrophone first detect a leak according to the time nodes when each hydrophone in the multi-channel hydrophone first detects a leak obtained in Step P202; The process of establishing the correlation relationship between the underwater acoustic signals of the reference hydrophone and other hydrophones in the multi-channel hydrophone in Step P201 is specifically described as follows: Construct a multiple regression model that incorporates the characteristics of the multi-channel hydrophone, frequency response parameters, and measured energy values; Among them, the constructed multiple regression model satisfies: （7）； In Equation (7), S1 is the short-time energy value of the reference hydrophone in the multi-channel hydrophone; S i is the observed value of other hydrophones in the multi-channel hydrophone, ; β0, β1, …, β n are the regression coefficients of the multiple regression model, and S1 2 , S1 3 , …, S1 n are the high-order powers of the short-time energy value S1; ε is an independent distributed error term with a constant variance; Among them, the coefficient combinations β0, β1, …, β that best fit the predicted values to the observed values n , satisfy: （8）； Construct matrix X, and solve for the regression coefficients according to the normal equations, satisfying: （9）； The process of statistically obtaining the time nodes when each hydrophone in the multi-channel hydrophone first detects a leak in Step P202 is specifically described as follows: Set the double threshold values of the reference hydrophone in the multi-channel hydrophone; Among them, the double threshold values of the reference hydrophone in the multi-channel hydrophone satisfy: （10）； （11）； In formulas (10) and (11), , ; Based on the regression coefficients obtained by analyzing the multiple regression model, determine the double threshold values of other hydrophones in the multi-channel hydrophone; Among them, the double threshold values of other hydrophones in the multi-channel hydrophone satisfy: （12）； （13）； In formula (13), ; Among them, the first threshold value in the double threshold values is used to preliminarily screen the hydrophone signals when the multi-channel hydrophone first detects a leak, so as to determine the interval where the hydrophone signals are located when the leak is first detected; the second threshold value in the double threshold values is used to detect and determine the starting position of the hydrophone signals when the multi-channel hydrophone first detects a leak.

2. The three-dimensional positioning method for the leakage of the underwater production system manifold based on the underwater acoustic signal according to claim 1, wherein The process of preprocessing the underwater acoustic signal in Step P1 is specifically described as follows: Step P101: Calculate the time-domain average value of the underwater acoustic signal; eliminate the DC component in the underwater acoustic signal to correct its baseline deviation; Step P102: Perform frame-by-frame windowing processing to achieve time-domain segmentation of the underwater acoustic signal.

3. The three-dimensional positioning method for the leakage of the subsea production system manifold based on underwater acoustic signals according to claim 2, wherein The time-domain average value of the underwater acoustic signal and the underwater acoustic signal after eliminating the DC component in Step P101 are specifically described as follows: Express the underwater acoustic signal as x(t), and the time-domain average value of the underwater acoustic signal satisfies: （1）； In Equation (1), N is the sample length of the underwater acoustic signal; The underwater acoustic signal after eliminating the DC component satisfies: （2）。 4. The three-dimensional positioning method for the leakage of the underwater production system manifold based on underwater acoustic signals according to claim 2, characterized in that, Step P102 is specifically described as follows: For an underwater acoustic signal with a sample length of N, each frame has a length of l, the displacement of the latter frame relative to the former frame is m, and the overlapping part is o = l - m. A total of frames are obtained; Among them, Satisfy: (3); Then the underwater acoustic signal after frame-by-frame windowing processing satisfies: （4）； In Equation (4), w(i) is the Hamming window function, and c is the windowing position; Further, the time-domain waveform of the underwater acoustic signal is set as s(n), and the i-th frame of the underwater acoustic signal s i (n) obtained after frame division and windowing processing satisfies: （5）； Among them, the short-time energy of the i-th frame of underwater acoustic signal s i (n) satisfies: （6）； In formulas (5) and (6), .

5. The three-dimensional positioning method for the leakage of the underwater production system manifold based on underwater acoustic signals according to claim 1, wherein The three-dimensional localization model of the leak sound source constructed in Step P3 is specifically described as follows: The three-dimensional localization model of the leakage sound source satisfies: （14）； （15）； In formulas (14) and (15), the spatial coordinates of each hydrophone in the multi-channel hydrophone are (x i , y i , z i ), and the spatial coordinates of the leakage sound source to be determined are (x, y, z); Equation (14) is further expressed as: （16）； In formula (16), ; And, according to Equation (15), the following relationship can be further obtained: （17）； Substituting Equation (17) into Equation (16), we get: （18）； In formula (18), , , .

Images