Anchor damage detection methods, devices, processors, and electronic equipment

Abstract

This application discloses a method, apparatus, processor, and electronic device for anchor damage detection. The method includes: acquiring n phase signals to be detected; extracting features from the n phase signals to obtain p phase signal values ​​corresponding to each phase signal; constructing an initially empty target matrix of size m x n; sequentially filling the target matrix with the p phase signal values ​​corresponding to each phase signal; performing convolution processing on the target matrix in a convolution detection state to obtain a first convolution result; determining a first anchor damage detection result for the n phase signals under first time information based on the first convolution result; removing the phase signal values ​​already filled in the first row of the target matrix; moving the phase signal values ​​already filled in the second to m rows of the target matrix to the first to (m-1)th rows; and restoring the target state of the target matrix from the convolution detection state to the signal filling state. This application solves the problem of low accuracy in anchor damage detection in related technologies.

Description

Technical Field

[0001] This application relates to the field of optical cable inspection technology, and more specifically, to a method, apparatus, processor, and electronic equipment for detecting anchor damage. Background Technology

[0002] Anchor intrusion refers to the accidental contact between a submarine anchor chain or other anchoring equipment and an optical cable, which may cause damage, breakage, or other problems to the cable. This situation not only incurs expensive repair and replacement costs but may also lead to data transmission interruptions, affecting the availability and reliability of communication networks.

[0003] Currently, some existing technologies exist for monitoring and preventing submarine fiber optic cable anchoring intrusion, such as using underwater sensors and cameras to monitor activity around the cables. However, this approach can be limited by underwater visibility and harsh marine conditions, leading to issues with the accuracy of detection results.

[0004] In other words, existing technologies suffer from the technical problem of low accuracy in detecting anchor damage. Summary of the Invention

[0005] The main objective of this application is to provide a method, apparatus, processor, and electronic device for detecting anchor damage, in order to solve the problem of low accuracy in clustering in related technologies.

[0006] To achieve the above objectives, according to one aspect of this application, a method for detecting anchor damage is provided. The method includes: acquiring n phase signals to be detected, where n is a positive integer; performing feature extraction on the n phase signals to obtain p phase signal values ​​corresponding to each phase signal; constructing an initially empty target matrix of size m multiplied by n, where m indicates the first time information selected from all time information corresponding to the p phase signal values, m and p are positive integers and m is less than p; sequentially filling the p phase signal values ​​corresponding to each phase signal into the target matrix, wherein if there is no region in the target matrix where no phase signal values ​​have been filled, the target state of the target matrix is ​​determined as a convolution detection state; performing convolution processing on the target matrix in the convolution detection state to obtain a first convolution result, and determining the first anchor fault detection result of the n phase signals under the first time information based on the first convolution result; removing the phase signal values ​​already filled in the first row of the target matrix, moving the phase signal values ​​already filled in the second to m rows of the target matrix to the first to m-1 rows, and restoring the target state of the target matrix from the convolution detection state to the signal filling state.

[0007] To achieve the above objectives, according to another aspect of this application, an anchor damage detection device is provided. The device includes: an acquisition unit for acquiring n phase signals to be detected, where n is a positive integer; an extraction unit for performing feature extraction on the n phase signals to obtain p phase signal values ​​corresponding to each phase signal; a construction unit for constructing an initially empty target matrix of size m multiplied by n, where m indicates first time information selected from all time information corresponding to the p phase signal values, m and p are positive integers and m is less than p; and a filling unit for sequentially filling the p phase signal values ​​corresponding to each phase signal into the target matrix, wherein there are no unfilled values ​​in the target matrix. In the case of a region of phase signal values, the target state of the target matrix is ​​determined as a convolution detection state; a convolution unit is used to perform convolution processing on the target matrix in the convolution detection state to obtain a first convolution result, and based on the first convolution result, determine the first anchoring fault detection result of the n phase signals under the first time information; a removal unit is used to remove the phase signal values ​​already filled in the first row of the target matrix, move the phase signal values ​​already filled in the second row to the m-th row of the target matrix to the first row to the (m-1)-th row, and restore the target state of the target matrix from the convolution detection state to the signal filling state.

[0008] To achieve the above objectives, according to another aspect of this application, an anchor damage detection processor is provided, which is used to run a program, wherein the program executes the above-described anchor damage detection method when it runs.

[0009] To achieve the above objectives, according to another aspect of this application, an anchor damage detection electronic device is provided, the electronic device including a processor, a memory and a program stored in the memory and executable on the processor, wherein the processor executes the program to implement the above anchor damage detection method.

[0010] Through the embodiments provided in this application, by constructing a target matrix and adopting a first-in-first-out detection order, multiple convolution detection processes are performed on the target number of phase signal values ​​out of p phase signal values. After each convolution detection process to obtain the anchor damage detection result of the phase signal of the current time information, the earliest row of signal value data filled in the target matrix is ​​removed, and the signal value data of the subsequent rows is shifted forward so that a new row of signal value data can be filled in. Then, convolution detection processing is performed on the new overall signal value data to obtain the anchor damage detection result of the phase signal of the new time information. This not only achieves the purpose of autonomously controlling the recognition range and customizing the duration of the anchor damage event, but also enables multiple, global detections of anchor damage events under different time information based on certain signal data, thereby achieving the technical effect of improving the accuracy of anchor damage detection. Attached Figure Description

[0011] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0012] Figure 1 This is a flowchart of the anchor damage detection method provided according to the embodiments of this application;

[0013] Figure 2 This is a schematic diagram of the anchor damage detection method provided according to the embodiments of this application;

[0014] Figure 3 This is a schematic diagram of the original phase characteristics of the optical cable as a whole according to the embodiments of this application;

[0015] Figure 4 This is a schematic diagram of the 01 matrix of vibration positions when the convolution kernel is (3,3) according to an embodiment of this application;

[0016] Figure 5 This is a schematic diagram of the 01 matrix of vibration positions when the convolution kernel is (5,5) according to the embodiments of this application;

[0017] Figure 6 This is a schematic diagram of the data after performing two-dimensional convolution on the original phase features according to the embodiments of this application;

[0018] Figure 7 This is a schematic diagram of the original phase characteristics of an optical cable subjected to a single-point anchor damage test according to an embodiment of this application;

[0019] Figure 8 This is a schematic diagram of an anchor damage detection device provided according to an embodiment of this application;

[0020] Figure 9 This is a schematic diagram of an anchor damage detection electronic device provided according to an embodiment of this application. Detailed Implementation

[0021] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0022] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0023] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate for the embodiments of this application described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0024] It should be noted that all information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for display, data used for analysis, etc.) involved in this disclosure are information and data authorized by the user or fully authorized by all parties. For example, this system has an interface with relevant users or organizations. Before obtaining relevant information, it is necessary to send an acquisition request to the aforementioned user or organization through the interface, and obtain the relevant information after receiving consent information from the aforementioned user or organization.

[0025] The present invention will now be described in conjunction with preferred implementation steps. Figure 1 This is a flowchart of the anchor damage detection method provided according to the embodiments of this application, such as... Figure 1 As shown, the method includes the following steps:

[0026] Step S101: Obtain n phase signals to be detected, where n is a positive integer;

[0027] Step S102: Perform feature extraction on the n phase signals to obtain p phase signal values ​​corresponding to each phase signal;

[0028] Step S103: Construct an initial empty target matrix with a size of m multiplied by n, where m is used to indicate the first time information selected from all time information corresponding to p phase signal values, and m and p are positive integers with m less than p;

[0029] Step S104: P phase signal values ​​corresponding to each phase signal are sequentially filled into the target matrix. In the case that there is no region in the target matrix that has not been filled with phase signal values, the target state of the target matrix is ​​determined as the convolution detection state.

[0030] Step S105: Perform convolution processing on the target matrix of the convolution detection state to obtain the first convolution result, and determine the first anchor damage detection result of n phase signals under the first time information based on the first convolution result;

[0031] Step S106 involves removing the phase signal values ​​already filled in the first row of the target matrix, moving the phase signal values ​​already filled in the second to m rows of the target matrix to the first to m-1 rows, and restoring the target state of the target matrix from the convolution detection state to the signal filling state.

[0032] Optionally, in this embodiment, the above-described anchor damage detection method can be applied, but is not limited to, online assessment of seabed anchor damage. With the rapid development of the Internet, submarine optical cable networks have become one of the main infrastructures for global information transmission. These optical cable networks play a crucial role in connecting international data centers, communication systems, and cloud computing services. However, the laying and maintenance of submarine optical cables in the marine environment faces a series of challenges, one of which is the threat of seabed anchor intrusion.

[0033] Traditionally, anchor intrusion refers to the accidental contact between a submarine anchor chain or other anchoring equipment and an optical cable, potentially causing damage, breakage, or other problems. This not only incurs costly repair and replacement costs but can also lead to data transmission interruptions, impacting the availability and reliability of communication networks.

[0034] Currently, several existing technologies exist for monitoring and preventing anchor damage to submarine optical cables, including but not limited to: 1) Anchor damage detection systems: Some methods use underwater sensors and cameras to monitor activity around the cable. However, these systems may be limited by underwater visibility and harsh marine conditions. 2) Physical protection: Other methods employ physical protection measures, such as armor and protective layers, to improve the cable's resistance to anchor damage. However, this increases cost and complexity.

[0035] At the same time, due to the scarcity of anchor damage data, it is estimated that an anchor damage event occurs only once every 30,000 hours on average, making it difficult to determine anchor damage in the absence of data.

[0036] To address the aforementioned problem, the above-mentioned anchor damage detection method utilizes a constructed target matrix and employs a first-in-first-out detection order to sequentially perform multiple convolution detection processes on the target number of phase signal values ​​out of p phase signal values. After each convolution detection process yields the anchor damage detection result for the phase signal at the current time, the earliest row of signal values ​​in the target matrix is ​​removed, and the signal values ​​of subsequent rows are shifted forward to allow for the addition of a new row of signal values. This new overall signal value convolution detection process is then performed to obtain the anchor damage detection result for the phase signal at the new time. This not only achieves autonomous control of the recognition range and customization of the anchor damage event duration but also enables multiple, global detections of anchor damage events at different time information based on specific signal data, thereby improving the accuracy of anchor damage detection.

[0037] Optionally, in this embodiment, acquiring the n phase signals to be detected can be, but is not limited to, by using a distributed optical fiber disturbance monitoring system. The DAS uses a laser source to emit laser light into a coupler. The coupler inputs 90% of the laser intensity detection light into an acousto-optic modulator. The acousto-optic modulator converts the laser into a pulse signal, which is amplified by an erbium-doped fiber amplifier and then input into an optical fiber. Rayleigh scattering from the optical fiber and the other 10% intrinsic light split off from the coupler enter the coupler, outputting two coherent beams that enter a balanced detector, converting the optical signal into an electrical signal. Finally, the phase signal is input into the acquisition card.

[0038] Optionally, in this embodiment, feature extraction may, but is not limited to, taking into account the characteristics of the acquisition card, outputting a phase signal every 0.1 seconds. Therefore, the variance of the phase signal every 0.1 seconds is calculated, as shown in the following formula:

[0039]

[0040] Where N is the number of phase signals in 0.1 seconds, x i Let μ be the value of each phase signal, and let μ be the average phase value over 0.1 seconds. The phase signals over 0.1 seconds are accumulated to 1 second and summed to obtain the phase characteristic signal value over 1 second. This 1-second phase characteristic signal value is then determined as the corresponding phase signal value for the aforementioned phase signals.

[0041] Optionally, in this embodiment, different phase signal values ​​among the p phase signal values ​​correspond to different time information. The time information corresponding to the phase signal values ​​obtained first by feature extraction / variance calculation is earlier, and the time information is filled into the target matrix earlier.

[0042] Optionally, in this embodiment, if the target matrix has already been filled with m times n phase signal values, the target state of the target matrix is ​​determined to be the convolution detection state. It can be understood that if there are regions in the target matrix that have not been filled with phase signal values, the target state of the target matrix is ​​determined to be the signal filling state.

[0043] Optionally, in this embodiment, when the target state of the target matrix is ​​a convolution detection state, the target matrix is ​​subjected to convolution processing to obtain a first convolution result for indicating the first anchor damage detection result of n phase signals under the first time information.

[0044] It should be noted that if the phase signal values ​​included in the target matrix are different, the time information indicated by the anchor damage detection result obtained by subsequent convolution processing will also be different. The time information indicated by the anchor damage detection result corresponds to the phase signal values ​​included in the target matrix.

[0045] Optionally, in this embodiment, after obtaining the first anchor hazard detection result under the first time information, the phase signal values ​​already filled in the first row of the target matrix are removed, the phase signal values ​​already filled in the second row to the m-th row of the target matrix are moved to the first row to the (m-1)-th row, and the target state of the target matrix is ​​restored from the convolution detection state to the signal filling state.

[0046] It should be noted that after restoring the target state of the target matrix from the convolution detection state to the signal filling state, n phase signal values ​​are sequentially determined from the p phase signal values ​​that have not been filled into the target matrix; the n phase signal values ​​are sequentially filled into the m-th row region of the target matrix; after all n phase signal values ​​are filled into the target matrix, the target state is adjusted from the signal filling state to the convolution detection state; the target matrix in the convolution detection state is subjected to a second convolution process to obtain the second convolution result, and the second anchor damage detection result of the n phase signals under the second time information is determined based on the second convolution result.

[0047] To illustrate further, when the phase signal values ​​are sequentially input into an empty matrix A(m,n), if the number of phase feature signals exceeds m and becomes m+1, the original first signal is deleted from matrix A(m,n), and the original second to mth data are padded forward to become the first to m-1th data. The subsequent m+1th data becomes the mth data, and so on for subsequent new signals.

[0048] Through the embodiments provided in this application, by constructing a target matrix and adopting a first-in-first-out detection order, multiple convolution detection processes are performed on the target number of phase signal values ​​out of p phase signal values. After each convolution detection process to obtain the anchor damage detection result of the phase signal of the current time information, the earliest row of signal value data filled in the target matrix is ​​removed, and the signal value data of the subsequent rows is shifted forward so that a new row of signal value data can be filled in. Then, convolution detection processing is performed on the new overall signal value data to obtain the anchor damage detection result of the phase signal of the new time information. This not only achieves the purpose of autonomously controlling the recognition range and customizing the duration of the anchor damage event, but also enables multiple, global detections of anchor damage events under different time information based on certain signal data, thereby achieving the technical effect of improving the accuracy of anchor damage detection.

[0049] As an optional approach, the target matrix of the convolution detection state is convolved to obtain the first convolution result. Based on the first convolution result, the anchor damage detection results of n phase signals are determined, including:

[0050] S1, use a two-dimensional convolution kernel to perform two-dimensional convolution calculation on the target matrix to obtain the first calculation matrix with a size of i multiplied by j;

[0051] S2, elements in the first calculation matrix whose values ​​are greater than the first preset threshold are recorded as 1, and elements in the convolution matrix whose values ​​are not greater than the first preset threshold are recorded as 0, to obtain the second calculation matrix;

[0052] S3, sum the elements in the second calculation matrix column by column to obtain a third calculation matrix with a size of 1 multiplied by j;

[0053] S4. Based on the values ​​of the j elements included in the third calculation matrix and the second preset threshold, the values ​​of the j elements are judged respectively to obtain the first anchor damage detection result.

[0054] Optionally, in this embodiment, the target matrix may be, but is not limited to, A(m,n), and the two-dimensional convolution kernel may be, but is not limited to, K(p,q). Whenever a new phase signal value enters a row of matrix A(m,n), a two-dimensional convolution operation is performed on matrix A(m,n) to obtain the first computation matrix O(i,j). The operation process is as follows:

[0055]

[0056] Where O(i,j) is the first computational matrix of the output, and A(i+p,j+q)K(p,q) represents the multiplication of an element A(i+p,j+q) of the input matrix with an element K(p,q) of the kernel matrix. and The summation is performed horizontally and vertically along the matrix. During convolution, matrix A(m,n) needs to be padded with P. h Line and P w Columns are used to ensure that inputs and outputs have the same height and width. Generally, P is set... h =p-1,P w =q-1, the convolution kernel uses an odd number of heights and widths to ensure that the number of padding elements at both ends is equal.

[0057] Optionally, in this embodiment, for the first computation matrix O(i,j) obtained after convolution, a threshold a is set and compared with the threshold. Elements in the first computation matrix that are greater than the threshold a are recorded as 1, and elements that are less than the threshold a are recorded as 0, and stored to obtain a new second computation matrix O2(i,j).

[0058] For the second computation matrix O2(i,j) containing 0 and 1, it is added column by column to become the third computation matrix O2'(1,j), where each value of j in the third computation matrix O2'(1,j) is the sum of each column in the 0 and 1 matrix O2(i,j).

[0059] It should be noted that when each row of phase signal values ​​is input into matrix A(m,n), a one-dimensional convolution kernel K(1,q) can be used to convolve each single phase signal value to obtain O'(1,j). Then, each convolved phase feature signal O'(1,j) is input into the matrix to obtain matrix O(i,j). This achieves the same effect as directly performing a two-dimensional convolution on the entire matrix using a two-dimensional convolution kernel K(p,q). The one-dimensional convolution formula is as follows:

[0060]

[0061] The embodiments provided in this application use a two-dimensional convolution method to determine seabed anchor damage events. This method enables autonomous control of the identification range, customization of the duration and vibration intensity of anchor damage events, and multiple, global detections of anchor damage events at different time points based on certain signal data. This achieves the technical effect of improving the accuracy of anchor damage detection.

[0062] As an optional approach, based on the values ​​of the j elements included in the third calculation matrix and the second preset threshold, the values ​​of the j elements are determined separately, including:

[0063] S1, determine the current element from j elements in turn;

[0064] S2, if the value of the current element is greater than the second preset threshold multiplied by i, determine that there is a risk of anchor damage in the optical fiber region of the phase signal corresponding to the current element.

[0065] Optionally, in this embodiment, for the third calculation matrix O2'(1,j), the number of j is the number of monitoring points, and its value is the sum of each column in the second calculation matrix O2(i,j). A second preset threshold b∈(0,1) is set, and each j value is judged. When j>b*i, it means that the signal at that point exceeds the threshold by b*100%, and it is judged as an anchor damage alarm signal. Then, the index signal of that point is restored to the true length, so as to achieve the effect of anchor damage alarm.

[0066] Through the embodiments provided in this application, the values ​​of the j elements included in the third calculation matrix and the second preset threshold are used to determine the values ​​of the j elements respectively, and the first anchor damage detection result is obtained. This enables the autonomous control of the identification range and the customization of the duration and vibration intensity of the anchor damage event. Even in the absence of actual anchor damage data, the identification and location of the anchor damage event can be achieved, thereby improving the technical effect of improving the accuracy of anchor damage detection.

[0067] As an optional approach, two-dimensional convolutional kernels are used to perform two-dimensional convolution calculations on the target matrix, including:

[0068] S1, obtain a two-dimensional convolution kernel of size k multiplied by k, where k is an odd number and k is not less than m and not less than n;

[0069] S2, perform row and column filling on the target matrix to obtain a filled target matrix of size k multiplied by k;

[0070] S3 uses a two-dimensional convolution kernel to perform multiplication on the padded target matrix.

[0071] Optionally, in this embodiment, the two-dimensional convolution kernel can be, but is not limited to, a matrix of size k multiplied by k, where k is an odd number and k is not less than m and not less than n.

[0072] Optionally, in this embodiment, when performing convolution, row padding is required for matrix A(m,n) to achieve the desired filling effect for P. h The purpose of the row is to perform column filling to fill P. w The purpose of this column is to ensure that the input and output have the same height and width.

[0073] As an optional approach, after restoring the target state of the target matrix from the convolutional detection state to the signal filling state, the method further includes:

[0074] S1, determine n phase signal values ​​sequentially from the p phase signal values ​​that have not been filled into the target matrix;

[0075] S2, fill the m-th row region of the target matrix with n phase signal values ​​in sequence;

[0076] S3, after all n phase signal values ​​are filled into the target matrix, the target state is adjusted from the signal filling state to the convolution detection state;

[0077] S4, perform a second convolution process on the target matrix of the convolution detection state to obtain the second convolution result, and determine the second anchor damage detection result of n phase signals under the second time information based on the second convolution result.

[0078] Optionally, in this embodiment, if the phase signal values ​​included in the target matrix are different, the time information indicated by the anchor damage detection result obtained by subsequent convolution processing will also be different, wherein the time information indicated by the anchor damage detection result corresponds to the phase signal values ​​included in the target matrix.

[0079] Optionally, in this embodiment, after obtaining the anchor damage detection result corresponding to the previous time information, the phase signal values ​​of the first row included in the target matrix are removed, and the phase signal values ​​of the other rows are shifted forward by one row to obtain the region of the last row to be filled. The unfilled phase signal values ​​are then sequentially filled into the region of the last row to be filled, and after filling is completed, the target matrix is ​​determined to enter the convolution detection state. The same processing method is performed as described above for the convolution process to obtain the anchor damage detection result of the new time information corresponding to the new target matrix.

[0080] Through the embodiments provided in this application, by constructing a target matrix and adopting a first-in-first-out detection order, multiple convolution detection processes are performed on the target number of phase signal values ​​out of p phase signal values. After each convolution detection process to obtain the anchor damage detection result of the phase signal of the current time information, the earliest row of signal value data filled in the target matrix is ​​removed, and the signal value data of the subsequent rows is shifted forward so that a new row of signal value data can be filled in. Then, convolution detection processing is performed on the new overall signal value data to obtain the anchor damage detection result of the phase signal of the new time information, thereby achieving the technical effect of improving the accuracy of anchor damage detection.

[0081] As an optional approach, feature extraction is performed on n phase signals to obtain p phase signal values ​​corresponding to each phase signal, including:

[0082] S1, according to the first cycle, perform feature variance extraction processing on n phase signals respectively to obtain multiple phase feature values ​​corresponding to the first cycle;

[0083] S2, according to the second period, accumulate the phase characteristic values ​​corresponding to multiple first periods to obtain p phase signal values ​​corresponding to each phase signal, wherein a second period includes at least two first periods and different second periods do not overlap with each other.

[0084] Optionally, in this embodiment, the first period may be, but is not limited to, every 0.1 seconds, and the second period may be, but is not limited to, every 1 second. It is understood that the first period and the second period may also be, but are not limited to, other time values, and this embodiment does not impose specific restrictions on the specific time values.

[0085] Optionally, in this embodiment, considering the characteristics of the acquisition card, a phase signal is output once every 0.1 seconds. Therefore, the first period is set to 0.1 seconds, and the variance of the phase signal every 0.1 seconds is calculated using the following formula:

[0086]

[0087] Where N is the number of phase signals in 0.1 seconds, x iis the value of each phase signal, and μ is the average phase value over 0.1 seconds.

[0088] Furthermore, the second period is set to 1 second, the phase signal of 0.1 seconds is accumulated to 1 second and summed to obtain the phase characteristic signal value of 1 second, and the phase characteristic signal value of 1 second is determined as a phase signal value corresponding to the above phase signal.

[0089] Through the embodiments provided in this application, by preprocessing such as variance calculation, a target object that can cover global signal information over a period of time is obtained, thereby achieving the goal of improving the accuracy of subsequent data processing from the source of data processing, and thus realizing the technical effect of improving the accuracy of anchor damage detection.

[0090] As an optional approach, n phase signals to be detected are acquired, including:

[0091] S1, a laser light source is used to emit a laser light into the coupler. The coupler inputs the first proportional intensity detection light of the laser light into the acousto-optic modulator to obtain a pulse signal. The acousto-optic modulator is used to convert the laser light into a pulse signal.

[0092] S2, the pulse signal is amplified by an erbium-doped fiber amplifier and then input into the fiber to obtain Rayleigh scattering of the fiber. Based on the Rayleigh scattering and the second proportion of intrinsic light separated by the coupler, two coherent lights are determined.

[0093] S3, input the two coherent light sources into the balanced detector to obtain the target electrical signal, wherein the smoothing detector is used to convert the optical signal into an electrical signal;

[0094] S4 inputs the target electrical signal to the acquisition card to obtain n phase signals output by the acquisition card.

[0095] Optionally, in this embodiment, acquiring the n phase signals to be detected can be, but is not limited to, by using a distributed optical fiber disturbance monitoring system. The DAS uses a laser source to emit laser light into a coupler. The coupler inputs 90% of the laser intensity detection light into an acousto-optic modulator. The acousto-optic modulator converts the laser into a pulse signal, which is amplified by an erbium-doped fiber amplifier and then input into an optical fiber. Rayleigh scattering from the optical fiber and the other 10% intrinsic light split off from the coupler enter the coupler, outputting two coherent beams that enter a balanced detector, converting the optical signal into an electrical signal. Finally, the phase signal is input into the acquisition card.

[0096] As an alternative, the above-mentioned anchor damage detection method is applied to an online anchor damage determination scenario based on a distributed fiber optic disturbance monitoring system. Based on the long-distance and long-term characteristics of anchor damage events, the method identifies anchor damage events and their locations, and provides accurate and effective alarms.

[0097] Specifically, a distributed fiber optic disturbance monitoring system uses a laser source to emit a laser beam that enters a coupler. The coupler inputs 90% of the laser's intensity detection light into an acousto-optic modulator, which converts the laser beam into a pulse signal. After being amplified by an erbium-doped fiber amplifier, the pulse signal is input into an optical fiber. Rayleigh scattering from the optical fiber and the remaining 10% of the intrinsic light split off from the coupler enter the coupler, outputting two coherent beams that enter a balanced detector. This converts the optical signal into an electrical signal, which is then input into a data acquisition card to obtain the phase and envelope signals.

[0098] Furthermore, the specific steps for extracting and processing features from the phase signals obtained by the acquisition card to achieve anchor damage intrusion event identification are as follows: Figure 2 As shown, the complete steps are as follows:

[0099] Step 1: Due to the characteristics of the data acquisition card, a phase signal is output every 0.1 seconds. Therefore, the variance of the phase signal every 0.1 seconds is calculated using the following formula:

[0100]

[0101] Where N is the number of phase signals in 0.1 seconds, x i is the value of each phase signal, and μ is the average phase value over 0.1 seconds.

[0102] Step 2: Accumulate the 0.1-second phase signal to 1 second and sum them to obtain the 1-second phase characteristic signal.

[0103] Step 3: Create an empty matrix A(m,n), where m is the number of rows, is a user-defined parameter representing the selection of m seconds of data, and n is the number of fiber sampling points in the DAS.

[0104] Step 4: Input the 1-second phase feature signals from Step 2 into the empty matrix A(m,n) sequentially. When the number of phase feature signals exceeds m and becomes m+1, the original first signal is deleted from matrix A(m,n), and the original second to m-th data are padded forward to become the first to m-1 data. The subsequent m+1 data becomes the m-th data, and so on for subsequent new signals.

[0105] Step 5: Determine the two-dimensional convolution kernel K(p,q). Whenever matrix A(m,n) enters a new phase feature, perform a two-dimensional convolution operation on matrix A(m,n). The operation process is as follows:

[0106]

[0107] Where O(i,j) is the output matrix, and A(i+p,j+q)K(p,q) represents the multiplication of an element A(i+p,j+q) of the input matrix with an element K(p,q) of the kernel matrix. and The summation is performed horizontally and vertically along the matrix. During convolution, matrix A(m,n) needs to be padded with P. h Line and P w Columns are used to ensure that inputs and outputs have the same height and width. Generally, P is set... h =p-1,P w =q-1, the convolution kernel uses an odd number of heights and widths to ensure that the number of padding elements at both ends is equal.

[0108] Step 6: For the convolutional matrix O(i,j), set a threshold a and compare it with the threshold. If it is greater than the threshold a, the value of matrix O(i,j) is recorded as 1, and if it is less than the threshold a, it is recorded as 0 and stored in a new matrix O2(i,j).

[0109] Step 7: For the 0-1 matrix O2(i,j), sum it column by column to get O2'(1,j). Since matrix A(m,n) and matrix O2(i,j) have the same dimension, O2'(1,j) can also be written as O2'(1,n), where each value of n in O2'(1,n) is the sum of each column in the 0-1 matrix O2(i,j).

[0110] Step 8: For matrix O2'(1,n), the number of n is the number of monitoring points, and its value is the sum of each column in the 01 matrix O2(i,j). Set the ratio b∈(0,1), and judge each value of n. When n>b*i, it means that the signal at that point exceeds the threshold by b*100%, and it is judged as an anchor damage alarm signal. Then restore the index signal of that point to the real length to achieve the effect of anchor damage alarm.

[0111] To further illustrate, after the acquisition card obtains the phase signal, the characteristic variance of the phase signal is extracted every 0.1 seconds, and then summed after accumulating to 1 second. In this embodiment, for example... Figure 3 As shown, there are a total of 1689 sampling points, so 1689 phase feature signals will be obtained per second.

[0112] An empty matrix A(m,n) is constructed, where m is a user-defined parameter, defined as 10 in this embodiment, representing the selection of 10 seconds of data, and n is the number of fiber sampling points in the distributed fiber optic disturbance monitoring system, which is 1689 in this embodiment. Therefore, the constructed matrix A(m,n) is specifically A(10,1689).

[0113] The phase characteristic signals for each second are sequentially input into the empty matrix A(10,1689). When the cumulative time reaches 11 seconds, the number of phase characteristic signals exceeds 10 and becomes 11. The original first signal is deleted from matrix A(10,1689), and the original second to tenth data are padded forward to become the first to ninth data. The subsequent eleventh data becomes the tenth data, and so on for subsequent new signals.

[0114] The two-dimensional convolution kernel K(p,q) is determined. In this embodiment, to ensure that the number of padding elements at both ends of the matrix is ​​equal during the convolution operation, an odd number is selected as the height and width of the convolution kernel. The two-dimensional convolution kernels are determined to be (3,3) and (5,5) respectively. Whenever matrix A(10,1689) enters a new phase feature, a two-dimensional convolution operation is performed on matrix A(10,1689). The operation process is as follows:

[0115]

[0116] For the convolutional matrix O(i,j), which is O(10,1689) in this example, a threshold 'a' is set, and the matrix is ​​compared with this threshold. If the value is greater than the threshold 'a', the value of matrix O(10,1689) is recorded as 1; if it is less than the threshold 'a', it is recorded as 0, and stored in a new matrix O2(10,1689). In this example, the matrix threshold 'a' is set to 500. The 0-1 matrices of the oscillation points in matrix O2(10,1689) with convolution kernels of (3,3) and (5,5) are as follows: Figure 4 and Figure 5 As shown.

[0117] For a 0-1 matrix O2(10,1689), summing it column by column gives O2'(1,1689), where each value in O2'(1,1689) is the sum of each column in the 0-1 matrix O2(10,1689).

[0118] For matrix O2'(1,1689), the number of monitoring points in this example is 1689. Its value is the sum of each column in the 0-1 matrix O2(10,1689). Setting the ratio b∈(0,1), each value of the 1689 monitoring points is judged. When n>b*i, it means that the signal at that point exceeds the threshold by b*100%, which is judged as an anchor damage alarm signal. Then, the index signal of that point is restored to its true length to achieve the anchor damage alarm effect. In this example, b is 0.8, meaning that if the signal exceeds the convolutional threshold a=500 for 8 seconds within 10 seconds, an alarm is generated. Figure 4 In the matrix, the sum of the 0-1 matrices at points 1631, 1632, and 1633 is 8, which exceeds the threshold. Figure 5In the data, the sum of the 0 and 1 matrices at points 1630, 1631, 1632, 1633, and 1634 is 8, which exceeds the threshold. This shows that different convolution kernels have different effects on the sensitivity of anchor damage events, thus realizing the control of the sensitivity of anchor damage identification by the convolution kernel.

[0119] It should be noted that the anchor damage detection effect obtained by using the above two-dimensional convolution calculation is as follows: Figure 6 As shown, the anchor damage detection effect obtained without utilizing the above two-dimensional convolution calculation is as follows: Figure 7 As shown, the anchor damage detection effect obtained by using the above two-dimensional convolution calculation is more complete, comprehensive and accurate.

[0120] It should be noted that when performing two-dimensional convolution operations on matrix A(m,n), there is an alternative: when each phase feature signal is input into matrix A(m,n), a one-dimensional convolution kernel K(1,q) can be used to convolve each single phase feature signal to obtain O'(1,j). Then, each convolved phase feature signal O'(1,j) is input into the matrix to obtain matrix O(i,j). The effect is the same as directly performing two-dimensional convolution on the entire matrix using a two-dimensional convolution kernel K(p,q). The one-dimensional convolution formula is as follows:

[0121]

[0122] Through the embodiments provided in this application, a two-dimensional convolution method is used to determine seabed anchor damage events based on the fiber optic disturbance monitoring system. It can realize the functions of autonomously controlling the identification range and customizing the duration and vibration intensity of anchor damage events. Even in the absence of actual anchor damage data, it can identify and locate anchor damage events and achieve a relatively accurate level.

[0123] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0124] This application also provides an anchor damage detection device. It should be noted that the anchor damage detection device of this application can be used to execute the anchor damage detection method provided in this application. The anchor damage detection device provided in this application is described below.

[0125] Figure 8 This is a schematic diagram of an anchor damage detection device according to an embodiment of this application. Figure 8 As shown, the device includes:

[0126] Acquisition unit 802 is used to acquire n phase signals to be detected, where n is a positive integer;

[0127] The extraction unit 804 is used to extract features from n phase signals to obtain p phase signal values ​​corresponding to each phase signal.

[0128] The construction unit 806 is used to construct an initial empty target matrix with a matrix size of m multiplied by n, where m is used to indicate the first time information selected from all time information corresponding to p phase signal values, and m and p are positive integers with m less than p;

[0129] The filling unit 808 is used to fill the target matrix with p phase signal values ​​corresponding to each phase signal in sequence. In the case that there is no region in the target matrix that has not been filled with phase signal values, the target state of the target matrix is ​​determined as the convolution detection state.

[0130] Convolutional unit 810 is used to perform convolution processing on the target matrix of the convolution detection state to obtain the first convolution result, and to determine the first anchor damage detection result of n phase signals under the first time information based on the first convolution result.

[0131] The elimination unit 812 is used to eliminate the phase signal values ​​already filled in the first row of the target matrix, move the phase signal values ​​already filled in the second to m rows of the target matrix to the first to m-1 rows, and restore the target state of the target matrix from the convolution detection state to the signal filling state.

[0132] As an optional solution, the convolutional unit 810 includes:

[0133] The first calculation module is used to perform two-dimensional convolution calculation on the target matrix using a two-dimensional convolution kernel to obtain a first calculation matrix with a size of i multiplied by j.

[0134] The second calculation module is used to record elements in the first calculation matrix whose values ​​are greater than the first preset threshold as 1, and elements in the convolution matrix whose values ​​are not greater than the first preset threshold as 0, to obtain the second calculation matrix.

[0135] The third calculation module is used to add the elements included in the second calculation matrix column by column to obtain a third calculation matrix with a size of 1 multiplied by j.

[0136] The judgment module is used to judge the values ​​of the j elements based on the values ​​of the j elements included in the third calculation matrix and the second preset threshold, so as to obtain the first anchor damage detection result.

[0137] As an optional solution, the judgment module includes:

[0138] The first determining submodule is used to determine the current element from j elements in sequence;

[0139] The second determination submodule is used to determine that there is a risk of anchor damage in the optical fiber region of the phase signal corresponding to the current element when the value of the current element is greater than the second preset threshold multiplied by i.

[0140] As an optional solution, the first computing module includes:

[0141] The acquisition submodule is used to acquire a two-dimensional convolution kernel of size k multiplied by k, where k is an odd number and k is not less than m and not less than n;

[0142] The filling submodule is used to perform row and column filling on the target matrix to obtain a filled target matrix of size k multiplied by k.

[0143] The product submodule is used to perform product processing on the padded target matrix using a two-dimensional convolution kernel.

[0144] As an optional solution, the device also includes:

[0145] The determination module is used to determine n phase signal values ​​sequentially from p phase signal values ​​that have not been filled into the target matrix after restoring the target state of the target matrix from the convolution detection state to the signal filling state.

[0146] The filling module is used to fill the m-th row region of the target matrix with n phase signal values ​​after the target state of the target matrix is ​​restored from the convolution detection state to the signal filling state.

[0147] The adjustment module is used to adjust the target state from the signal filling state to the convolution detection state after the target state of the target matrix is ​​restored from the convolution detection state to the signal filling state and after all n phase signal values ​​are filled into the target matrix.

[0148] The convolution module is used to perform a second convolution process on the target matrix in the convolution detection state after restoring the target state of the target matrix from the convolution detection state to the signal filling state, to obtain the second convolution result, and to determine the second anchor damage detection result of n phase signals under the second time information based on the second convolution result.

[0149] As an optional solution, extraction unit 804 includes:

[0150] The extraction module is used to perform feature variance extraction processing on n phase signals according to the first period to obtain multiple phase feature values ​​corresponding to the first period.

[0151] The accumulation module is used to accumulate the phase feature values ​​corresponding to multiple first cycles according to the second cycle to obtain p phase signal values ​​corresponding to each phase signal. Here, a second cycle includes at least two first cycles and different second cycles do not overlap with each other.

[0152] As an optional solution, the acquisition unit 802 includes:

[0153] The transmitting module is used to emit laser light from a laser source into the coupler. The coupler inputs the first proportional intensity detection light of the laser light into the acousto-optic modulator to obtain a pulse signal. The acousto-optic modulator is used to convert the laser light into a pulse signal.

[0154] The amplification module is used to amplify the pulse signal using an erbium-doped fiber amplifier and input it into the fiber to obtain Rayleigh scattering of the fiber. Based on the Rayleigh scattering and the second proportion of intrinsic light separated by the coupler, two coherent beams are determined.

[0155] The first input module is used to input two coherent light sources into the balanced detector to obtain the target electrical signal. The smoothing detector is used to convert the optical signal into an electrical signal.

[0156] The second input module is used to input the target electrical signal to the acquisition card to obtain n phase signals output by the acquisition card.

[0157] This invention provides a computer-readable storage medium storing a program that, when executed by a processor, implements the above-described anchor damage detection method.

[0158] This invention provides a processor for running a program, wherein the program executes the above-described anchor damage detection method during runtime.

[0159] like Figure 9 As shown, an embodiment of the present invention provides an electronic device, which includes a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, it implements the above-mentioned anchor damage detection method.

[0160] The processor contains a kernel, which retrieves the corresponding program units from memory. One or more kernels can be configured; by adjusting kernel parameters, the target model can be trained or optimized, thereby improving data loading efficiency.

[0161] The memory may include non-permanent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0162] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0163] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0164] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0165] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0166] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0167] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0168] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transfer medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0169] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0170] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0171] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for detecting anchor damage, characterized in that, include: Acquire n phase signals to be detected, where n is a positive integer; Feature extraction is performed on the n phase signals to obtain p phase signal values ​​corresponding to each phase signal; Construct an initial empty target matrix of size m multiplied by n, where m indicates the first time information selected from all time information corresponding to the p phase signal values, and m and p are positive integers with m less than p; The p phase signal values ​​corresponding to each phase signal are sequentially filled into the target matrix. Where there is no region in the target matrix that has not been filled with phase signal values, the target state of the target matrix is ​​determined as the convolution detection state. The target matrix is ​​subjected to two-dimensional convolution calculation using a two-dimensional convolution kernel to obtain a first calculation matrix with a size of i multiplied by j. Elements in the first calculation matrix whose values ​​are greater than a first preset threshold are recorded as 1, and elements in the first calculation matrix whose values ​​are not greater than the first preset threshold are recorded as 0, to obtain a second calculation matrix; The elements included in the second calculation matrix are added column by column to obtain a third calculation matrix with a size of 1 multiplied by j; The current element is determined sequentially from the j elements included in the third calculation matrix; If the value of the current element is greater than the second preset threshold multiplied by i, it is determined that there is a risk of anchor damage in the optical fiber region of the phase signal corresponding to the current element, and the first anchor damage detection result is obtained. The phase signal values ​​already filled in the first row of the target matrix are removed, the phase signal values ​​already filled in the second to m rows of the target matrix are moved to the first to m-1 rows, and the target state of the target matrix is ​​restored from the convolution detection state to the signal filling state.

2. The method according to claim 1, characterized in that, The step of performing two-dimensional convolution calculation on the target matrix using a two-dimensional convolution kernel includes: Obtain the two-dimensional convolution kernel of size k multiplied by k, where k is an odd number and k is not less than m and not less than n; The target matrix is ​​subjected to row-filling and column-filling processes to obtain the padded target matrix with a size of k multiplied by k. The two-dimensional convolution kernel is used to perform a product operation on the padded target matrix.

3. The method according to any one of claims 1 to 2, characterized in that, After restoring the target state of the target matrix from the convolution detection state to the signal filling state, the method further includes: From the p phase signal values, n phase signal values ​​are sequentially determined from the multiple phase signal values ​​that were not filled into the target matrix; The n phase signal values ​​are sequentially filled into the m-th row region of the target matrix; After all n phase signal values ​​are filled into the target matrix, the target state is adjusted from the signal filling state to the convolution detection state; The target matrix in the convolution detection state is subjected to a second convolution process to obtain a second convolution result. Based on the second convolution result, the second anchor damage detection result of the n phase signals under the second time information is determined.

4. The method according to any one of claims 1 to 2, characterized in that, The step of extracting features from the n phase signals to obtain p phase signal values ​​corresponding to each phase signal includes: According to the first period, feature variance extraction processing is performed on the n phase signals respectively to obtain multiple phase feature values ​​corresponding to the first period; According to the second period, the phase characteristic values ​​corresponding to multiple first periods are accumulated to obtain p phase signal values ​​corresponding to each phase signal. Among them, a second period includes at least two first periods, and different second periods do not overlap with each other.

5. The method according to any one of claims 1 to 2, characterized in that, The acquisition of the n phase signals to be detected includes: A laser light source emits a laser beam into a coupler, wherein the coupler inputs a first-proportion intensity detection light of the laser beam into an acousto-optic modulator to obtain a pulse signal, and the acousto-optic modulator is used to convert the laser beam into the pulse signal; The pulse signal is amplified by an erbium-doped fiber amplifier and then input into an optical fiber to obtain Rayleigh scattering of the optical fiber. Based on the Rayleigh scattering and the second ratio of intrinsic light separated by the coupler, two coherent beams are determined. The two coherent beams are input into a balanced detector to obtain the target electrical signal, wherein the balanced detector is used to convert the optical signal into an electrical signal; The target electrical signal is input to the acquisition card to obtain the n phase signals output by the acquisition card.

6. An anchor damage detection device, used to perform the method described in any one of claims 1 to 5, characterized in that, include: The acquisition unit is used to acquire n phase signals to be detected, where n is a positive integer; An extraction unit is used to extract features from the n phase signals to obtain p phase signal values ​​corresponding to each phase signal; The construction unit is used to construct an initial empty target matrix with a size of m multiplied by n, where m is used to indicate the first time information selected from all time information corresponding to the p phase signal values, and m and p are positive integers and m is less than p; A filling unit is used to sequentially fill the p phase signal values ​​corresponding to each phase signal into the target matrix, wherein, if there is no region in the target matrix that has not been filled with phase signal values, the target state of the target matrix is ​​determined as a convolution detection state; A convolution unit is used to perform convolution processing on the target matrix of the convolution detection state to obtain a first convolution result, and to determine the first anchor damage detection result of the n phase signals under the first time information based on the first convolution result; The elimination unit is used to eliminate the phase signal values ​​already filled in the first row of the target matrix, move the phase signal values ​​already filled in the second to m rows of the target matrix to the first to m-1 rows, and restore the target state of the target matrix from the convolution detection state to the signal filling state.

7. A processor, characterized in that, The processor is used to run a program, wherein the program executes the method according to any one of claims 1 to 5 when it runs.

8. An electronic device, characterized in that, The method includes one or more processors and a memory for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to implement the method of any one of claims 1 to 5.

Images