Submarine cable detection and inspection method and system based on optomagnetic fusion

Abstract

The invention discloses a submarine cable detection and inspection method and system based on optomagnetic fusion. The method comprises the following steps: S1, respectively acquiring magnetic detection data and optical detection data through a magnetic sensing module and a camera module; s2, a data resolving transmission module establishes an identification framework for the preprocessed detection data and generates a result set; the data resolving and transmitting module performs probability assignment on the two identification frameworks according to a D-S evidence theory method, and confidence intervals are obtained respectively; the data resolving and transmitting module calculates the similarity between the two pieces of evidence, and if the similarity is lower than a preset threshold value, the result set is updated; s3, the data resolving and transmitting module fuzzifies the confidence interval and the result set according to a preset fuzzy rule to obtain a fuzzy set and confidence coefficients of the two kinds of detection data; and S4, a data solution transmission module carries out data fusion according to the confidence coefficients of the two kinds of detection data to obtain a final detection result. According to the invention, seamless integration of two kinds of data is realized, and the detection precision and efficiency are significantly improved.

Description

Technical Field

[0001] The present invention relates to the field of submarine cable detection technology, and in particular to a submarine cable detection and inspection method and system based on optical-magnetic fusion. Background Art

[0002] With the vigorous promotion of marine engineering and the booming development of offshore wind power industry, the total length of submarine power cables laid between my country's coastal islands and the mainland has shown a rapid growth trend. However, this trend has also brought about a significant increase in safety risks, making the safety inspection and maintenance of submarine cables particularly urgent. Submarine cables are extremely susceptible to damage in the complex and changing marine environment, so their working status must be regularly inspected and potential dangers must be detected in time to effectively avoid accidents and ensure the reliability and continuous and stable operation of the submarine cable system.

[0003] However, the current detection methods have obvious limitations and insufficient detection accuracy. At different working stages, only a single detection method can be relied upon, and the switching of working modes depends entirely on manual judgment. This traditional method not only leads to a long inspection cycle, but also makes the cost investment too high, and it is difficult to meet the needs of efficient detection in complex marine environments. Even if a variety of detection methods are used to find cables, it is difficult to achieve real-time fusion of data. Therefore, it is urgent to develop a submarine cable detection method that integrates multiple detection methods to effectively solve the above-mentioned technical problems, improve the accuracy and efficiency of submarine cable detection, and provide strong guarantees for the sustainable development of marine engineering and offshore wind power industries. Summary of the invention

[0004] In view of the above-mentioned defects of the prior art, the present invention provides a submarine cable detection and inspection method and system based on optical-magnetic fusion, which organically combines optical detection data with magnetic detection data and uses advanced multi-source data fusion technology to achieve seamless integration of the two data, which not only omits the cumbersome manual data analysis steps in traditional methods, but also can automatically switch working states according to real-time changes in the underwater environment, thereby significantly improving the accuracy and efficiency of detection. .

[0005] In order to achieve the above object, the technical solution adopted by the present invention is:

[0006] In a first aspect, a submarine cable detection and inspection method based on optical-magnetic fusion comprises the following steps:

[0007] S1. Collect and pre-process detection data:

[0008] The detection data includes magnetic detection data and optical detection data;

[0009] Acquire the magnetic detection data through a magnetic sensor module; acquire the optical detection data through a camera module;

[0010] S2, D-S Evidence Processing:

[0011] The data calculation and transmission module first establishes an identification framework for the preprocessed detection data respectively and generates a result set;

[0012] The data calculation and transmission module assigns probabilities to the two identification frameworks respectively according to the D-S evidence theory method and obtains confidence intervals respectively;

[0013] The data calculation and transmission module calculates the similarity between the two pieces of evidence. If the similarity is lower than the preset threshold, the result set is updated;

[0014] S3, Fuzzification Processing:

[0015] The data calculation and transmission module performs fuzzification processing on the confidence interval and the result set according to the preset fuzzy rules to obtain a fuzzy set and the confidence degrees of the two types of detection data;

[0016] S4, Generating Location Information:

[0017] The data calculation and transmission module performs data fusion according to the confidence degrees of the two types of detection data to obtain the final detection result.

[0018] Preferably, the step S1 includes:

[0019] Performing distance calculation based on the recognition position of the camera module to obtain preprocessed optical detection data;

[0020] Calculating the light intensity based on the preprocessed optical detection data;

[0021] Performing denoising and de-geomagnetic interference processing on the magnetic detection data to obtain preprocessed magnetic detection data;

[0022] Calculating the magnetic field change rates of three components for the preprocessed magnetic detection data.

[0023] Preferably, the step S2 includes:

[0024] Performing boundary division on the light intensity and the magnetic field change rates respectively, and establishing an optical detection identification framework and a magnetic detection identification framework accordingly.

[0025] Preferably, the step S4 includes:

[0026] Comparing the confidence degrees of the two types of detection data and defining the primary and secondary relationships of the two detection means according to the comparison results.

[0027] Preferably, the magnetic sensing module includes three fluxgate sensors for detecting the spatial position of the target submarine cable relative to the origin; the three fluxgate sensors are distributed in an isosceles triangle.

[0028] Preferably, the optical detection data includes the spatial position of the target submarine cable relative to the camera module and the environmental gray value.

[0029] Preferably, the camera module includes at least two cameras.

[0030] In a second aspect, an optical-magnetic fusion-based submarine cable detection and inspection system for performing the optical-magnetic fusion-based submarine cable detection and inspection method includes a magnetic sensing module, a camera module, a data calculation and transmission module, and a host computer;

[0031] The magnetic sensing module acquires the magnetic detection data of the target submarine cable and sends it to the data calculation and transmission module;

[0032] The camera module acquires the optical detection data of the target submarine cable and sends it to the data calculation and transmission module;

[0033] The data calculation and transmission module performs D-S evidence processing and fuzzy processing on the magnetic detection data and the optical detection data to obtain the confidence levels of the two types of detection data and the final detection result.

[0034] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0035] 1. Based on the D-S evidence theory, the present invention introduces a fuzzy set for data category division and deeply fuses the optical and magnetic detection data, effectively making up for the deficiencies of a single detection method and significantly improving the accuracy and reliability of the detection result.

[0036] 2. The present invention uses common sensors and combines a multi-modal data fusion algorithm to optimize the detection data, enabling it to achieve a detection level equivalent to that of high-precision sensors without increasing the cost of high-precision sensors, thereby greatly reducing the equipment cost while ensuring the detection accuracy.

[0037] 3. The invention abandons the cumbersome steps of manually comparing and analyzing the data of various sensors in the traditional method, instead using a processor to automatically complete data fusion according to a preset algorithm and output accurate and reliable detection results, which not only improves the detection accuracy but also saves the time of manual data analysis and significantly improves the detection efficiency.

[0038] 4. The multi-modal data fusion algorithm of the present invention can automatically update the fuzzy set according to the real-time changes of the underwater environment, thereby autonomously switching the detection working state. This adaptive ability effectively reduces the detection error, enables the present invention to maintain high detection accuracy and reliability in a variety of complex underwater environments, and has good environmental adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 is a schematic diagram of the method of Embodiment 1;

[0040] Figure 2 is a schematic flowchart of Embodiment 1;

[0041] Figure 3 is a layout diagram of the triaxial fluxgate sensors of Embodiment 1;

[0042] Figure 4 is the synthetic basic probability assignment of Embodiment 1. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0043] In order to make the technical means, creative features, achieved purposes and effects of the invention easy to understand, the present invention will be further described below with reference to specific drawings. However, the present invention is not limited to the following embodiments.

[0044] It should be noted that the structures, ratios, sizes, etc. shown in the drawings of this specification are only used to cooperate with the content disclosed in the specification for those skilled in the art to understand and read, and are not used to limit the limited conditions under which the present invention can be implemented. Therefore, they do not have a substantial technical meaning. Any modification of the structure, change of the proportional relationship or adjustment of the size should still fall within the scope covered by the technical content disclosed in the present invention without affecting the effects that the present invention can produce and the purposes that can be achieved.

[0045] Embodiment 1:

[0046] Such as Figure 1 , Figure 2 shown, a submarine cable detection and inspection method based on optical and magnetic fusion includes the following steps:

[0047] S1. Collect and preprocess the detection data,

[0048] Obtain magnetic detection data and optical detection data through the magnetic sensing module and the camera module respectively. In this embodiment, the magnetic sensing module is preferably a triaxial fluxgate sensor array, and the camera module is a binocular camera module with two cameras.

[0049] The binocular camera module is used to optically detect the spatial position (x1, y1, z1) of the target submarine cable relative to the camera module and the gray value γ of the environment. The binocular camera module is placed in a transparent cabin and located at the coordinate origin. The underwater ambient light intensity l can be calculated based on the gray value:

[0050]

[0051] The arrangement of the three-axis fluxgate sensor array is as Figure 3 shown, in an isosceles triangle. The coordinates of fluxgate sensor 1 are (d1 / 2, 0, 0), the coordinates of fluxgate sensor 2 are (0, d2, 0), and the coordinates of fluxgate sensor 3 are (-d1 / 2, 0, 0), which are used to detect the spatial position (x2, y2, z2) of the target submarine cable relative to the origin. According to the magnetic gradient tensor B ij (i = 1, 2, 3; j = x, y, z) output by the three sensors, the magnetic gradient tensor matrix G of the origin can be calculated:

[0052]

[0053] From equation (3), the relative spatial position vector r of the target submarine cable from the origin O can be inferred:

[0054] r = -3G -1 B (3)

[0055] S2, D-S evidence processing,

[0056] The data calculation and transmission module first establishes an identification framework for the preprocessed detection data respectively and generates a result set; the data calculation and transmission module assigns probabilities to the two identification frameworks respectively according to the D-S evidence theory method and obtains the confidence intervals respectively; the data calculation and transmission module calculates the similarity between the two pieces of evidence, and if the similarity is lower than the preset threshold, the result set is updated.

[0057] The optical detection data and magnetic detection data are demarcated, and the identification frameworks Θ1 and Θ2 for optical detection and magnetic detection are established respectively according to the demarcation results:

[0058] Θ1 = {θ 11 , θ 21 , …, θ 2i} (4)

[0059] Θ2 = {θ 21 , θ 22 ,..., θ 2j} (5)

[0060] The elements in the sets of the two identification frameworks satisfy mutual exclusivity respectively, that is, the intersection of any two elements is an empty set.

[0061] Power set of the identification framework and are respectively defined as the set systems composed of all subsets of Θ1 and Θ2, including the empty set and the universal set itself.

[0062] Taking Θ1 as an example, when constructing the initial trust assignment, the basic probability assignment function m needs to satisfy the mapping condition: And the sum of the probability assignments of all subsets is 1, as shown in Equation (6). Based on this, the established trust function Bel(A) represents the lowest confidence in the optical detection data A, and its expression is as shown in Equation (7).

[0063]

[0064] To comprehensively evaluate the confidence of optical detection data, it is necessary to introduce the likelihood function Pl(A) to represent the highest confidence in not denying A. Its calculation method is as shown in Equation (8). This function expands the evaluation range of the trust function by summing up the BPA values of all optical detection data B that intersect with A. A* is the subset that has an intersection with A. Among them, BPA can also be called the mass function, which represents the relationship between the basic trust degree m(A) of each subset A. The process of synthesizing the probability assignment is as Figure 4 . Satisfying Bel(A) ≤ Pl(A), they jointly constitute the confidence interval [Bel(A), Pl(A)] of the proposition.

[0065]

[0066] For the magnetic detection data B, the confidence interval is also established following the above steps.

[0067] For the synthesis of the trust function Bel1 of optical detection data and the trust function Bel2 of magnetic detection data, it is necessary to handle the intersection conflict of their corresponding BPA functions m1 and m2.

[0068] For example, Θ1 = {P1, Q1}, Θ2 = {P2, Q2}, P represents that the measured object is on the left, and Q represents that the measured object is on the right. The basic probability assignments (BPA) of the two evidence sources are as follows:

[0069] Evidence source 1: m1(P1) = 0.9, m1(Q1) = 0.1

[0070] Evidence source 2: m2(P2) = 0.1, m2(Q2) = 0.9

[0071] Both pieces of evidence assign high confidence to P and Q, but the directions are completely opposite, and the conflict is significant.

[0072] The conflict coefficient k = m1(P1)*m2(Q2) + m1(P2)*m2(Q1) = 0.9×0.9 + 0.1×0.1 = 0.82, indicating that there is a great conflict between the evidences.

[0073] When , the empty set will obtain part of the BPA value. At this time, it is necessary to introduce a normalization factor K for correction. The definition of the conflict coefficient k is as shown in Eqs. (9) and (10). Under the condition of k < 1, the combined result C is finally expressed as Eq. (11). When conducting underwater detection, it is set that when both have detection results, optical detection is the main method. Therefore, there is no situation where k = 1, that is, a complete conflict.

[0074]

[0075] Since the D-S evidence theory may produce the result of evidence conflict, it is necessary to first quantify the degree of conflict and then judge whether the two evidences belong to highly conflicting evidences:

[0076]

[0077] This judgment adopts the cosine similarity calculation method. Let the angle threshold of the relative position vectors of the two detection methods be 30°. Then the preset threshold of the similarity S is √3 / 2. If the similarity S is lower than the preset threshold, it means that the two belong to highly conflicting evidences, indicating that the detection system is in an underwater environment with rapid changes in light or magnetic field, resulting in misjudgment of one of the detection methods.

[0078] For example, after processing the optical detection and magnetic detection data in the previous period, the following two evidence sources are obtained respectively:

[0079] Evidence source 1: m1(P1) = 0.2, m1(Q1) = 0.8

[0080] Evidence source 2: m2(P2) = 0.1, m2(Q2) = 0.9

[0081] Then the conflict coefficient k = m1(P1)*m2(Q2) + m1(P2)*m2(Q1) = 0.2×0.9 + 0.8×0.1 = 0.26, and the similarity S = [m1(P1)*m2(Q1) + m1(P2)*m2(Q2)] / √[m1(P1) 2 + m1(Q1) 2 ·√[m2(P2) 2 + m2(Q2) 2 = (0.2*0.1 + 0.8*0.9) / [√(0.2 2 + 0.8 2 )·√(0.1 2 + 0.9 2)] = 0.99, indicating that the detection data of both are relatively accurate

[0082] After processing the detection data in the next cycle, the following two evidence sources are obtained respectively:

[0083] Evidence source 1: m1(P1) = 0.2, m1(Q1) = 0.8

[0084] Evidence source 2: m2(P2) = 0.6, m2(Q2) = 0.4

[0085] Then the conflict coefficient k = m1(P1) * m2(Q2) + m1(P2) * m2(Q1) = 0.2×0.4 + 0.8×0.6 = 0.56, and the similarity S = [m1(P1) * m2(Q1) + m1(P2) * m2(Q2)] / √[m1(P1) 2 + m1(Q1) 2 ·√[m2(P2) 2 + m2(Q2) 2 = (0.2 * 0.6 + 0.8 * 0.4) / [√(0.2 2 + 0.8 2 )·√(0.4 2 + 0.6 2 )] = 0.739 < √3 / 2, the two have high conflict. Combining the evidence source of the previous cycle, it can be seen that the magnetic field has changed greatly, and this change can be reflected by the magnetic tensor norm ||G||, that is, the magnetic field change rate.

[0086] Therefore, it is necessary to modify the normalized belief function to obtain the updated result set C:

[0087]

[0088] Among them, α is the conflict factor attenuation coefficient, used to reduce the probability of evidence failure in high-conflict scenarios; A i 、B j are respectively the identification framework elements of the light intensity l and the magnetic gradient change norm ||G||.

[0089] S3. Fuzzification processing,

[0090] The data calculation and transmission module performs fuzzification processing on the confidence interval and the result set according to the preset fuzzy rules to obtain the fuzzy set and the confidence degrees of the two detection data.

[0091] The fuzzy sets are introduced, and the two detection confidence levels of the underwater environment are divided into fuzzy intervals according to the synthesis result after D-S evidence processing and the BPA value. The confidence level is obtained through the fuzzy rule table. For example, if the synthesis result is likely to fall into the ZO interval and the similarity is relatively high, the optical detection confidence level can be taken as 0.5 - 0.6. Since the magnetic gradient change norm is always positive, representing the magnetic field stability of the underwater environment, ZO represents the most stable interval, and there are no three cases of NB, NM, and NS. Based on this, the conventional fuzzy sets are modified, and the modified fuzzy interval division is shown in Table 1. Therefore, Θ1 = {NB1, NM1, NS1, ZO1, PS1, PM1, PB1}, and Θ2 = {ZO2, PS2, PM2, PB2}. Combining with Equation (10), we can get A i ∈Θ1, B j ∈Θ2. The confidence level ζ a of the optical detection data ∈ {NB, NM, NS, ZO, PS, PM, PB}, and the confidence level ζ b of the magnetic detection data = 1 - ζ a .

[0092] Table 1 Optical-magnetic fusion fuzzy algorithm rules

[0093]

[0094] Among them, in the subset elements of the light intensity l, ZO1 is 18% of the average light intensity of the object, and it gradually increases from NB to PB; in the subset elements of the magnetic gradient change norm, ZO2 is when the magnetic detection data is within the credible range, and the result is from NB to PB, ζ a increases and ζ b decreases. The optical-magnetic fusion detection mainly relies on optical detection and supplemented by magnetic detection. Therefore, when ζ a = ZO, ζ a > ζ b . When the light intensity is close to 0 and the magnetic gradient change is small, only the magnetic detection data is relied on for positioning. At this time, NB in the result set C represents ζ a = 0, ζ b = 1; when the light intensity is much higher than the average level and the magnetic gradient change is extremely large, only the optical detection data is relied on for positioning. At this time, PB represents ζ a = 1, ζ b = 0.

[0095] S4. Generate position information,

[0096] The data calculation and transmission module performs data fusion according to the confidence levels of the two detection data to obtain the final detection result.

[0097] Based on the optical detection coordinates [x1, y1, z1] output after detecting the target submarine cable and the calculated magnetic detection coordinates [x2, y2, z2], the finally resolved position information can be expressed as

[0098] [X, Y, Z] = ζ a [x1, y1, z1] + ζ b [x2, y2, z2] (14)

[0099] To sum up, the data calculation and transmission module is based on the optical detection data and magnetic detection data. First, it establishes an identification framework for the two, then assigns probabilities through the D-S evidence theory, then fuzzifies through fuzzy rules, and determines whether there is a conflict between the two detection methods. If there is a conflict, the trust degrees of the two detection methods are recalculated after conflict processing and the fuzzy set is updated, so as to obtain the confidence degrees of the two detection data. The final detection result is obtained through data integration using the confidence degrees of the two detection data.

[0100] Embodiment 2:

[0101] A submarine cable detection and inspection system based on optical-magnetic fusion, which is used to execute the submarine cable detection and inspection method based on optical-magnetic fusion, includes a magnetic sensing module, a camera module, a data calculation and transmission module, and a host computer;

[0102] The magnetic sensing module acquires the magnetic detection data of the target submarine cable and sends it to the data calculation and transmission module;

[0103] The camera module acquires the optical detection data of the target submarine cable and sends it to the data calculation and transmission module;

[0104] The data calculation and transmission module performs D-S evidence processing and fuzzy processing according to the magnetic detection data and the optical detection data to obtain the confidence degrees of the two detection data and the final detection result.

[0105] After sampling by the magnetic sensing module and the camera module, the data calculation and transmission module needs to preprocess the optical detection and magnetic detection results respectively: perform target recognition on the optical detection data, and calculate the distance according to the recognition position of the camera of the camera module to obtain the preprocessed optical detection distance data; perform noise removal and geomagnetic interference removal processing on the magnetic detection data, and then calculate the magnetic field change rate of the three components for the processed result to obtain the processed magnetic detection distance data. At the same time, the light intensity is deduced from the gray value in the original data of the optical detection.

[0106] The data calculation and transmission module divides the data boundaries by combining the preprocessed data with the calculated light intensity, constructs an identification framework, performs D-S evidence processing, obtains the confidence intervals and belief functions of the two, and determines whether there is a high conflict between the two. If there is, a conflict factor attenuation coefficient is introduced to recalculate the results and obtain the belief degrees of the two.

[0107] The data calculation and transmission module defines a fuzzy set according to the belief functions of the two and obtains the interval to which the synthesis result belongs, thereby obtaining the final confidence degrees of the two.

[0108] Finally, the data calculation and transmission module compares the obtained confidence degrees, determines the confidence levels of the optical detection and magnetic detection results, thereby defining the primary and secondary relationships of the two detection methods, calculates the final distance data, and sends it to the host computer. If multiple samplings are required, repeat all the above operations.

Claims

1. A submarine cable detection and inspection method based on optical and magnetic fusion, characterized in that: The following steps are involved: S1. Collect and pre-process detection data: The detection data includes magnetic detection data and optical detection data; Acquire the magnetic detection data through a magnetic sensor module; acquire the optical detection data through a camera module; S2. DS evidence processing: The data solution transmission module first establishes an identification framework for the pre-processed detection data and generates a result set; The data solution transmission module respectively assigns probability to the two identification frames according to the DS evidence theory method, and obtains confidence intervals respectively; The data solution transmission module calculates the similarity between two pieces of evidence, and updates the result set if the similarity is lower than a preset threshold; S3, fuzzy processing: The data solution transmission module performs fuzzification processing on the confidence interval and the result set according to preset fuzzy rules to obtain a fuzzy set and two confidence levels of the detection data; S4. Generate location information: The data calculation and transmission module performs data fusion according to the confidence of the two detection data to obtain a final detection result.

2. The method according to claim 1, characterized in that The step S1 comprises: Performing distance calculation according to the recognition position of the camera module to obtain pre-processed optical detection data; Calculating the light intensity according to the pre-processed optical detection data; De-noising and de-geomagnetic interference processing are performed on the magnetic detection data to obtain pre-processed magnetic detection data; The pre-processed magnetic detection data is calculated to obtain the magnetic field change rates of the three components.

3. The method according to claim 2, characterized in that The step S2 comprises: The light intensity and the magnetic field change rate are respectively divided into boundaries, and an optical detection and identification framework and a magnetic detection and identification framework are respectively established accordingly.

4. The method according to claim 1, characterized in that: The step S4 comprises: The confidence levels of the two detection data are compared, and the primary and secondary relationship of the two detection methods is defined according to the comparison result.

5. The method according to claim 1, characterized in that The magnetic sensing module includes three fluxgate sensors for detecting the spatial position of the target submarine cable relative to the origin; the three fluxgate sensors are distributed in an isosceles triangle.

6. The method according to claim 1, characterized in that The optical detection data includes the spatial position of the target submarine cable relative to the camera module and the environmental grayscale value.

7. The method according to claim 6, characterized in that The camera module includes at least two cameras.

8. A submarine cable detection and inspection system based on optical and magnetic fusion, characterized in that: Used to execute the method according to any one of claims 1 to 7, comprising a magnetic sensor module, a camera module, a data solution transmission module and a host computer; The magnetic sensing module acquires magnetic detection data of the target submarine cable and sends it to the data solution transmission module; The camera module acquires the optical detection data of the target submarine cable and sends it to the data solution transmission module; The data solution transmission module performs DS evidence processing and fuzzification processing according to the magnetic detection data and the optical detection data to obtain the confidence of the two detection data and the final detection result.

Images