Optical fiber shape sensing method based on fiber core combination optimization adaptive algorithm

Abstract

The invention belongs to the technical field of optical fiber shape sensors, and discloses a shape sensing method based on a fiber core combination optimization adaptive algorithm, which is used for shape sensing of a multi-core optical fiber optical frequency domain reflectometer. The shape sensing method comprises the following steps: calculating the signal-to-noise ratio of a peripheral fiber core of the multi-core optical fiber, removing the fiber core of which the signal-to-noise ratio is lower than a set threshold value, and taking the left fiber core as a candidate fiber core; demodulating the strain of each candidate fiber core, then normalizing the demodulated strain of the fiber core, and performing difference and square accumulation to obtain the strain jump degree of each fiber core with the signal-to-noise ratio higher than a set threshold value; the candidate fiber cores form different triangular combinations, the total jump degree of the fiber core strain of each combination is calculated, and then the fiber core combination with the minimum total strain jump degree is selected for optical fiber shape reconstruction. The measurement precision of the optical fiber shape sensor is greatly improved, and the optical fiber shape sensor is worthy of popularization and application.

Description

Technical Field

[0001] The present invention belongs to the technical field of fiber optic shape sensors, and particularly relates to a high-precision shape sensing method for measuring the shape of optical fibers. Background Art

[0002] Fiber optic shape sensing reconstructs the shape of an optical fiber by measuring the strains of different cores in a multi-core optical fiber or an optical fiber bundle and establishing the relationship between the strains of different cores and the bending direction angle and bending curvature of the optical fiber. The distributed fiber optic shape sensing technology based on an optical frequency domain reflectometer has advantages such as no need for visual assistance, anti-electromagnetic interference, small volume, and high spatial resolution, so it shows extensive application value in many fields such as civil engineering, aerospace, and in-vivo navigation of medical devices. However, for the practical application of this technology, how to improve the shape sensing accuracy is a key problem to be solved.

[0003] First of all, traditional fiber optic shape reconstruction methods all use three cores forming an equilateral triangle in a multi-core optical fiber for shape reconstruction. For example, for a seven-core optical fiber, there are only two combinations of cores forming an equilateral triangle. Once the signal-to-noise ratio of a certain core in these two combinations is low or the bending curvature is large, resulting in a large error in strain measurement, it will seriously affect the accuracy of fiber optic shape sensing. At present, although some scholars have proposed using other combinations of cores in a seven-core optical fiber for fiber optic shape reconstruction, they have not analyzed how to select an optimal combination of cores for shape reconstruction. There are a total of 20 combinations of triangular core combinations that can be used for shape sensing in a seven-core optical fiber, which can greatly improve the fault tolerance rate of shape sensing compared to only using the two combinations of equilateral triangles. Subsequently, by analyzing the strain measurement accuracy of the cores in each combination and selecting an optimal combination, the accuracy of shape sensing of multi-core optical fibers in extreme cases such as large bending curvature, low core signal-to-noise ratio, and even core breakage can be greatly improved.

[0004] Secondly, when a multi-core optical fiber has large bending curvature and low core signal-to-noise ratio, using an optical frequency domain reflectometer based on cross-correlation demodulation to measure the frequency drift of the cores in the multi-core optical fiber will result in large jumps in the demodulated frequency drift. These jumps in frequency drift will reduce the accuracy of strain measurement and thus reduce the accuracy of fiber optic shape reconstruction. To address this problem, most methods use denoising algorithms to remove the jumps in frequency drift, such as smoothing denoising. However, this will cause the frequency drift caused by actual strain to not be correctly demodulated, thus affecting the accuracy of fiber optic shape reconstruction. Therefore, a method for accurately removing the jumps in frequency drift needs to be developed to solve this problem.

[0005] In addition, compared with cross-correlation demodulation, phase demodulation is more vulnerable to coherent fading, polarization fading, large bending curvature of multi-core optical fiber, and low signal-to-noise ratio of the core, resulting in more phase jump points in the demodulated phase. Therefore, it is also necessary to develop a method to remove phase jumps to overcome this defect of the existing technology. Summary of the Invention

[0006] The technical problem to be solved by the present invention is to propose a shape sensing method based on a core combination optimization adaptive algorithm, aiming to improve the shape sensing accuracy.

[0007] To solve the above technical problem, the present invention proposes a shape sensing method based on a core combination optimization adaptive algorithm. First, calculate the signal-to-noise ratio of the measurement signals of the outer cores in the multi-core optical fiber; then sort the signal-to-noise ratios of the outer cores from small to large, remove the cores with signal-to-noise ratios lower than the set threshold, and the remaining outer cores are used as candidate cores for fiber shape reconstruction.

[0008] Preferably, the specific method for calculating the signal-to-noise ratio of the measurement signals of the outer cores in the multi-core optical fiber is as follows: Perform Fourier transform on the reference signal and the measurement signal of the multi-core in the fiber shape sensor to obtain the distance domain signal; Calculate the signal-to-noise ratio of the measurement signal of each outer core using the following formula SNR i , i is the number of each outer core: In the formula P si is the i th signal power of the distance domain signal of the outer core, P ni is the i th noise power of the distance domain signal of the outer core.

[0009] Furthermore, the shape sensing method based on the core combination optimization adaptive algorithm includes the steps of: After obtaining the candidate cores, demodulate the strain of each candidate core; Normalize, differentiate, and square-accumulate the strain of the demodulated cores, and then obtain the strain jump degree of each candidate core; Form different triangular combinations of the candidate cores, and calculate the total jump degree of the strain of each combination of cores; Select the core combination with the smallest total strain jump degree for fiber shape reconstruction.

[0010] Optionally, for the shape sensing method based on the core combination optimization adaptive algorithm, the demodulation of the strain of each candidate core is cross-correlation demodulation, specifically including the following steps: Step A-1: Measure the signal of a certain candidate core in the natural state as the reference signal I ri ; Measure the signal of the same core after applying the shape as the measurement signal I mi , and perform Fourier transform on I ri and I mi to obtain the distance-domain signals and ; Step A-2: Divide the distance-domain signals and into n segments respectively, namely and ; Perform inverse Fourier transform on and to obtain I R1 and I M1 , perform cross-correlation on I R1 and I M1 to obtain the frequency drift curve of the measurement signal relative to the reference signal C 1; Locate the peak of the frequency drift curve to obtain the frequency corresponding to the peak as the frequency drift magnitude of the first segment of this core fs 1; Step A-3: Perform inverse Fourier transform on and to obtain I R2 and I M2 , and perform cross-correlation on I R2 and I M2 to obtain the frequency drift curve of the measurement signal relative to the reference signal C 2; Locate within a range near the frequency drift magnitude fs 1 corresponding to the previous peak. This range is , w is a set positive integer,[[]] f is the horizontal axis frequency interval of the frequency drift curve C 2. Locate the frequency corresponding to the peak within this range as the frequency drift magnitude of the second segment of this core fs 2.

[0011] Step A-4: Repeatedly process subsequent core segments using the method of Step A-3 to obtain the frequency drift magnitude of each core segment, and finally obtain FS ={ fs 1, fs 2… fs n}}, that is, the frequency drift magnitude at each position of the core along the distance domain; Step A-5: Repeat Steps A-1 to A-4 to calculate the frequency drift magnitude of each candidate core along the distance domain, that is FS i , i where is the number of the candidate core; FS i divide by the constant e P, where P is the photoelastic coefficient of the optical fiber and its value is the constant 0.78, v 0 is the center frequency of the tuning laser, to obtain the strain magnitude of each candidate core along the distance domain .

[0012] Optionally, for the shape sensing method based on the core combination optimization adaptive algorithm, the demodulation of the strain of each candidate core is phase demodulation, which specifically includes the steps of: Step B-1: Calculate the phase of the distance domain signal after Fourier transform of the candidate core reference signal, calculate the phase of the distance domain signal after Fourier transform of the candidate core measurement signal, and then subtract the two and perform unwrapping to obtain the unwrapped differential phase , i where is the number of the candidate core; Step B-2: Assume that there are N data in q . Calculate the mean value W 1 of the first data before the M -th data, and the mean value q 2 of the last W data after the -th data. If M is greater than or equal to the set threshold , then subtract from the q -th to the N -th data; ; Step B-3: Loop through q= 1 to q = N , so as to complete the compensation of All discontinuous segments in, and then for the compensated Perform SG filtering; Step B-4, obtain the for each candidate core obtained after step B-3 by the above steps, and perform differencing to obtain , and obtain the strain magnitude of each candidate core along the distance domain through the following formula : where is the theoretical spatial resolution, is the initial wavelength of the tuning laser, is the effective refractive index of the core.

[0013] Preferably, the normalization, differencing, and square summation of the strain of the demodulated core are performed to obtain the strain jump degree of each candidate core, including: For perform normalization processing with the following formula: Perform differencing processing on the normalized strain with the following formula, that is: For , square each value in it and sum them to obtain S i , S i which is used to measure the strain jump degree of the core.

[0014] Further preferably, the candidate cores are combined into different triangular combinations, and the total strain jump degree of each combination of cores is calculated, including: Combine the candidate cores into different triangular core combinations. Assume there are g different triangular core combinations in total, and calculate the magnitude of the total strain jump degree of each combination of cores GS m ( m = 1, 2 …g), that is: S m1 , S m2 , S m3 are the magnitudes of the strain jump degrees of the three cores forming the triangular core combination, that is, the set { S m1 , S m2 , S m3} is the set {S i a subset of

[0015] The present invention has the following beneficial effects: 1. The present invention proposes an adaptive algorithm for shape reconstruction of multi-core optical fiber. This algorithm calculates the signal-to-noise ratio of each peripheral core of the multi-core optical fiber, eliminates the cores with low signal-to-noise ratio and does not use them for shape reconstruction. Then, the strain of the candidate cores is demodulated, and the demodulated core strain is normalized, differentiated, and squared and accumulated to obtain the strain jump degree of each candidate core. Finally, the candidate cores are combined into different triangular combinations, and the combination with the smallest strain jump degree among all combinations is selected for shape reconstruction. Compared with using only the traditional equilateral triangular core combination for shape reconstruction, this algorithm has a high fault tolerance rate and greatly improves the accuracy of shape reconstruction in the case of large bending curvature of multi-core optical fiber, low core signal-to-noise ratio, or even core fracture.

[0016] 2. The present invention improves the cross-correlation peak-seeking algorithm to solve the problem that when the multi-core optical fiber shape sensor has a large bending curvature of multi-core optical fiber and low core signal-to-noise ratio, there will be many jump points when using the traditional cross-correlation algorithm to demodulate frequency drift. Compared with using denoising algorithms such as smoothing denoising to solve the frequency jump, the improved cross-correlation peak-seeking algorithm can correctly demodulate the frequency drift caused by actual strain, thereby improving the accuracy of shape reconstruction.

[0017] 3. The present invention proposes an algorithm for removing phase jumps in phase demodulation, which removes phase jump points caused by coherent fading, polarization fading, large bending curvature of multi-core optical fiber, and low core signal-to-noise ratio, thereby improving the demodulation accuracy and the accuracy of shape reconstruction. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The technical solutions of the present invention will be further specifically described below in conjunction with the drawings and specific embodiments.

[0019] Figure 1 is a diagram of an optical frequency domain reflectometer shape sensing device; Figure 2 is a schematic diagram of 20 triangular core combinations in a seven-core optical fiber; Figure 3 is a flowchart of the adaptive algorithm of the present invention; Figure 4 is a comparison diagram of the frequency drift curve demodulated by using the traditional cross-correlation algorithm and the frequency drift curve demodulated by using the improved cross-correlation peak-seeking algorithm; Figure 5 is a comparison diagram of the differential phase diagrams before and after removing phase jump points; Figure 6 is a schematic diagram of the positional relationship between two adjacent sensing points in the optical fiber. Detailed implementation mode

[0020] In order to enable those skilled in the art to better understand the technical solutions in the embodiments of the present invention and make the above objects, features, and advantages of the embodiments of the present invention more obvious and understandable, the technical solutions in the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0021] Without loss of generality, the optical frequency domain reflectometer system used in the present invention adopts a seven-core optical fiber. See Figure 1 . The measuring device includes: a tunable laser 1; a 1:99 optical fiber coupler 2; an auxiliary interferometer part, including: a 1:1 optical fiber coupler 3, a 60-meter delay optical fiber 4, a 1:1 optical fiber coupler 5, and a balanced detector 6; a main interferometer part, including: a 1:99 optical fiber coupler 7, a polarization controller 8, an optical fiber circulator 9, a 1:1 polarization-maintaining optical fiber coupler 10, polarization diversity reception 11, a data acquisition card 12, an optical switch 13, a fan-in / fan-out 14, and a seven-core optical fiber 15. <X

[0022] When collecting signals, the light output by the tunable laser 1 is divided into two optical paths by the optical fiber coupler 2. The light with an energy ratio of 1% enters the auxiliary interferometer part, and the beat signal of the auxiliary interferometer is detected by the balanced detector 6 and then collected by the data acquisition card 12. The other optical path of the optical fiber coupler 2 with an energy ratio of 99% enters the main interferometer part. The optical fiber coupler 7 divides the input light into two paths, and 1% of the light enters the polarization controller 8 and then enters the polarization-maintaining optical fiber coupler 10. The polarization controller is used to control the equal intensity of the P-polarized light and the S-polarized light. 99% of the light passes through the optical fiber circulator 9, and then passes through the optical switch 13 and the fan-in / fan-out 14 and enters the core of the seven-core optical fiber 15. The backward Rayleigh scattered light of the core returns along the original path to the optical fiber circulator 9 and enters the polarization-maintaining optical fiber coupler 10. The polarization diversity reception 11 is used to detect the beat signals of the P-polarized light and the S-polarized light, and the beat signals of the P-polarized light and the S-polarized light are collected by the data acquisition card 12.

[0023] The seven-core optical fiber 15 is composed of the middle core 7 and the outer cores 1 to 6. The distances from the six outer cores to the center are equal, and the angle between adjacent cores is 60°. See Figure 2 . In the experiment, one of the 20 triangular core combinations shown in Figure 2 will be selected through an adaptive algorithm for shape reconstruction. The measured signals and reference signals in the experiment both include the auxiliary interferometer data and the main interferometer data from the seven cores.

[0024] The shape sensing method based on the core combination optimization adaptive algorithm proposed by the present invention has the following overall technical solution: First, calculate the signal-to-noise ratio of the measurement signals of the outer cores in a multi-core optical fiber; sort the signal-to-noise ratios of the outer cores from small to large, eliminate the cores with signal-to-noise ratios lower than the set threshold, and the remaining outer cores are used as candidate cores for fiber shape reconstruction; demodulate the strain of each candidate core; normalize, differentiate, and square and accumulate the strains of the demodulated cores to obtain the strain jump degree of each candidate core; form different triangular combinations of the candidate cores, and calculate the total strain jump degree of each combination of cores; select the core combination with the smallest total strain jump degree for fiber shape reconstruction.

[0025] Next, a more specific and detailed description of the technical solution of the present invention will be given.

[0026] Process 1: Conduct two signal acquisition experiments. In the first experiment, the optical fiber is in a natural elongation state, and the signals of each core collected in the first experiment are used as reference signals; in the second experiment, the optical fiber is bent into a specific shape, and the signals of each core collected in the second experiment are used as measurement signals. First, use the beat frequency signals of the auxiliary interferometer of the reference signal and the measurement signal of each core to obtain the instantaneous optical frequency information of the tunable laser in the two experiments, and then perform synchronous equal-frequency resampling on the beat frequency signals of the main interferometer of the reference signal and the measurement signal of each core to compensate for the nonlinear tuning of the tunable laser and the inconsistent frequency scanning ranges of the tunable lasers in the two experiments.

[0027] Process 2: Use an adaptive algorithm to reconstruct the fiber shape, which includes sub-processes (1) to (4) as shown in Figure 3 : (1) Perform Fourier transform on the main interferometer signals in the measurement signals of the seven cores after compensating for the nonlinear tuning of the tunable laser and the inconsistent frequency scanning range to obtain the distance domain signals I mi , and calculate the signal-to-noise ratio of the measurement signals of the six outer cores in the seven-core optical fiber through the following expression : SNR i ( i is the number of each outer core, that is i = 1, 2…6): In the formula P si is the signal power of the distance domain signal of the i th outer core, and P ni is the noise power of the distance domain signal of the i th outer core. Sort the signal-to-noise ratios of the six cores from small to large, and calculate the 10th percentile T, and use T as the set threshold, and eliminate the optical fiber cores with signal-to-noise ratios lower than the threshold T corresponding thereto.

[0028] (2) Since there are mainly two modes, i.e., cross-correlation demodulation and phase demodulation, for signal demodulation during the measurement of existing optical fiber sensors. The adaptive demodulation algorithms for these two demodulation modes will be separately described below.

[0029] A. Cross-correlation demodulation. As shown in the left dotted box Figure 3 , the specific steps are as follows: Step A-1: Take the main interferometer signal i in the reference signal of a certain candidate optical fiber core I ri ; measure the signal of the same optical fiber core after applying the shape as the measurement signal I mi , and perform Fourier transform on I ri , I mi to obtain the distance-domain signals , .

[0030] Step A-2: Divide and into n segments, i.e., and . Perform inverse Fourier transform on and to obtain I R1 , I M1 , and perform cross-correlation on I R1 , I M1 to obtain the frequency drift curve of the measurement signal relative to the reference signal C 1. Locate the peak of the frequency drift curve, and obtain the frequency corresponding to the peak as the frequency drift magnitude fs 1 of the first segment of this optical fiber core.

[0031] Step A-3: Perform inverse Fourier transform on and to obtain I R2 , I M2 , and perform cross-correlation on I R2 , I M2 to obtain the frequency drift curve of the measurement signal relative to the reference signal C2. Since the change in strain on the fiber core should be continuous and there will be no sudden jump in the strain of adjacent two fiber core segments, when locating the peak of the frequency drift curve C 2, it can be located within a range near the frequency drift magnitude fs 1 corresponding to the previous peak. This range is , w is a set positive integer, f is the horizontal axis frequency interval of the frequency drift curve C 2. The frequency corresponding to the peak located within this range is used as the frequency drift magnitude fs 2 of the second segment of this fiber core.

[0032] Step A-4, repeat Step A-3 to process subsequent fiber core segments in sequence, obtain the frequency drift magnitude of each fiber core segment, and finally obtain fs 1, fs 2… fs n , that is, the frequency drift magnitude FS at each position of this fiber core along the distance domain.

[0033] Figure 4 shows the frequency drift curve of a certain fiber core demodulated using the traditional cross-correlation algorithm and the frequency drift curve of this fiber core demodulated using the improved cross-correlation peak search algorithm described in Steps 1 to 4 of the present invention. The bending radius of the seven-core optical fiber is 2.5 cm, the demodulation length is 0.21 m, and the fiber core is divided into n = 101 segments, w = 50, f = 0.54 GHz. It can be seen that when the bending curvature is large, using the improved cross-correlation peak search algorithm of the present invention can well suppress the frequency drift jump, so as to correctly reflect the frequency drift caused by strain.

[0034] Step A-5, repeat Steps A-2 to A-4 to obtain the frequency drift magnitude FS i , i at each position of each candidate fiber core along the distance domain,

[0035] Step A-6, divide FS i by the constant , P e is the photoelastic coefficient of the optical fiber, and its value is a constant 0.78, v 0 is the central frequency of the tuning laser, to obtain the strain magnitude T along the distance domain of each fiber core with a signal-to-noise ratio higher than the threshold .

[0036] B. Phase demodulation. Combining with Figure 3 as shown in the right dashed box, the specific steps are as follows: Step B-1: Calculate the phase of the candidate core distance domain signal and calculate the phase of the candidate core distance domain signal. Then subtract the two and perform unwrapping to obtain the unwrapped differential phase , i where is the number of the candidate core; Step B-2: Since the phase jump points will cause to be divided into many discontinuous segments, it is necessary to compensate for the discontinuous parts in . Assume that there are N data in q . Calculate the mean value 1 of the first 4 M data before the q -th data and the mean value 2 of the last 4 M data after the -th data. If is greater than or equal to the set threshold q , then subtract N from the -th to the -th data; Step B-3: Loop through q= 1 to q = N ; thus compensating for all the discontinuous segments in . Finally, perform SG filtering on the compensated ; Figure 5 shows the differential phase result diagram of a certain core before and after using the algorithm of the present invention. It can be seen that the phase jump can be well removed by the algorithm of the present invention.

[0037] Step B-4: Obtain the of each candidate core through the above steps, and perform differencing on it to obtain . Obtain the strain magnitude along the distance domain of each core with a signal-to-noise ratio higher than the threshold T through the following formula : where is the theoretical spatial resolution, is the initial wavelength of the tunable laser, is the effective refractive index of the core.

[0038] (III) Normalize the strain magnitude : Differentiate the strain of each normalized candidate core, i.e.: For each value in, square them and accumulate them to get S i , that is, the degree of strain jump of core i .

[0039] (4) Combine the remaining cores into different triangular core combinations as shown in Figure 2 . Assume there are g different triangular core combinations , and calculate the total degree of strain jump of each combined core GS m ( m = 1, 2 …g), that is: S m1 , S m2 , S m3 are the degrees of strain jump of the three cores forming the triangular core combination, that is, the set { S m1 , S m2 , S m3} is a subset of the set { S i}, and select the triangular core combination corresponding to the smallest GS m value to reconstruct the shape of the seven-core optical fiber.

[0040] Process 3. Optical fiber shape reconstruction, including the steps: Step 3-1. Assume that the strain magnitudes of the three cores of the optimal triangular core combination along the distance domain are . a, b, and c represent the three cores in the optimal triangular core combination. The angle between the straight line passing through core a and the middle core 7 and the straight line passing through core b and the middle core 7 is α 1, and the angle between the straight line passing through core b and the middle core 7 and the straight line passing through core c and the middle core 7 is α 2, that is, as shown in Figure 6 .

[0041] Step 3-2. Solve the following system of equations to obtain the bending direction angle, bending curvature, and torsion magnitude of the optical fiber: where is the bending curvature, is the bending direction angle, is the torsional strain, r is the distance between the outer core and the middle core. When 、 the solutions of the above equations are: When 、 the solutions of the above equations are: When 、 the solutions of the above equations are: Since the torsional arc length of the outer core rotating on the fiber cross-section l , that is, as shown in Figure 6 , according to the modulus constant G between the shear strain and the torsional angle of the fiber, the torsional angle can be obtained according to the following expression: Step 3-3, fiber shape reconstruction based on the secondary transformation matrix. As shown in Figure 6 , assume P j and P j+1 are the coordinates of two adjacent sensing points on the fiber, j represents the j th sensing point, that is j = 1, 2, 3…, the distance between two adjacent sensing points on the fiber central axis is the sensing resolution z of the system. The coordinates j + 1 of the P j+1 th sensing point can be obtained by substituting the coordinates j of the P j th sensing point into the following secondary transformation matrix framework: where is the bending direction angle of each sensing point after compensating for the torsional angle, , is the bending curvature of each sensing point obtained by solving, is the bending radius at each sensing point. Therefore, the coordinates of each sensing point on the optical fiber can be obtained successively according to the above homogeneous transformation matrix framework, so as to reconstruct the shape of the entire optical fiber.

[0042] Those skilled in the art will readily conceive of other embodiments of the present invention after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present invention that follow the general principles of the present invention and include known common knowledge or conventional technical means in the technical field not disclosed by the present invention. The specification and examples are only illustrative, and the true scope and spirit of the present invention are pointed out by the following claims.

[0043] It should be understood that the present invention is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present invention is only regulated by the appended claims.

[0044] Finally, it should be noted that the above specific embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered within the scope of the claims of the present invention.

Claims

1. A shape sensing method based on a core combination optimization adaptive algorithm, characterized in that, First, calculate the signal-to-noise ratio of the measurement signals of the outer cores in the multi-core optical fiber; then, sort the signal-to-noise ratios of the outer cores from small to large, eliminate the cores with signal-to-noise ratios lower than the set threshold, and the remaining outer cores are used as candidate cores for fiber shape reconstruction; After obtaining the candidate cores, demodulate the strain of each candidate core; Normalize, differentiate, and square-accumulate the demodulated strain of the cores to obtain the strain jump degree of each candidate core; Form different triangular combinations of the candidate cores and calculate the total strain jump degree of the cores in each combination; Select the core combination with the smallest total strain jump degree for fiber shape reconstruction.

2. The shape sensing method based on the core combination optimization adaptive algorithm according to claim 1 is characterized in that: The specific steps for calculating the signal-to-noise ratio of the measurement signals of the outer cores in the multi-core optical fiber include: Perform Fourier transform on the reference signal and the measurement signal of the multi-core in the fiber shape sensor to obtain the distance-domain signal; The signal-to-noise ratio of the measurement signal of each peripheral fiber core is calculated using the following formula: SNR i , i Number each outer fiber core: In the formula P si For the i The signal power of the signal in the root outer core distance domain, P ni For the i The noise power of the signal in the distance domain of the outer fiber core.

3. The shape sensing method based on the core combination optimization adaptive algorithm according to claim 2, characterized in that: The demodulation of the strain of each candidate core is cross-correlation demodulation, and the specific steps include: Step A-1, measure the signal of a certain candidate fiber core in its natural state as the reference signal I ri ; measure the signal of the same fiber core after applying the shape as the measurement signal I mi , for I ri , I mi perform Fourier transform to obtain the distance-domain signals , ; Step A-2, the distance domain signal , are respectively divided into n segments, namely and ; Perform inverse Fourier transform on and to obtain I R1 , I M1 . Perform cross-correlation on I R1 , I M1 to obtain the frequency drift curve of the measurement signal relative to the reference signal C 1; Locate the peak of the frequency drift curve to obtain the frequency corresponding to the peak as the frequency drift magnitude of the first segment of this core fs 1; Step A-3, and Perform inverse Fourier transform and get I R2 、 I M2 , and I R2 、 I M2 Perform cross-correlation to obtain the frequency drift curve of the measured signal relative to the reference signal C 2. Frequency drift corresponding to the previous peak fs Positioning is performed within a range near 1, which is , w is a set positive integer, f Frequency drift curve C 2 horizontal axis frequency interval, the frequency corresponding to the peak value located within this range is used as the frequency drift of the second section of the fiber core fs 2; Step A-4: Repeat the method of Step A-3 to process the subsequent core segments in sequence, obtain the frequency drift magnitude of each core segment, and finally obtain FS ={ fs 1, fs 2… fs n}, that is, the frequency drift magnitude at each position of the core along the distance domain; Step A-5: Repeat steps A-1 to A-4 to calculate the frequency drift of each candidate fiber core along the distance domain, that is, FS i , i is the number of the candidate fiber core; Step A-6, divide FS i by the constant , where P e is the photoelastic coefficient of the optical fiber, and its value is the constant 0.78, v 0 is the center frequency of the tunable laser, and the strain magnitude of each candidate core along the distance domain is obtained .

4. The shape sensing method based on the core combination optimization adaptive algorithm according to claim 2, characterized in that, The demodulation of the strain of each candidate core is phase demodulation, and the specific steps include: Step B-1: Calculate the phase of the distance-domain signal after Fourier transform of the candidate core reference signal, calculate the phase of the distance-domain signal after Fourier transform of the candidate core measurement signal, then subtract the two and perform unwrapping to obtain the unwrapped differential phase , i is the number of the candidate core; Step B-2: Assume that There are a total of N data, and calculate the mean q of the first W data before the th data M 1, and the mean q of the last W data after the th data M 2. If is greater than or equal to the set threshold , then subtract q from the N th to the th data ; Step B-3, loop through q= 1 to q = N , thus compensating All discontinuous segments in the Perform SG filtering; Step B-4: For each candidate core obtained after Step B-3, perform differencing to obtain , and obtain the strain magnitude of each candidate core along the distance domain through the following formula : wherein is the theoretical spatial resolution, is the initial wavelength of the tunable laser, is the effective refractive index of the fiber core.

5. The shape sensing method based on the fiber core combination optimization adaptive algorithm according to claim 3 or 4, characterized in that The normalization, differentiation, and square-accumulation of the demodulated strain of the cores to further obtain the strain jump degree of each candidate core include: Pairwise Normalize using the following formula: Perform differential processing on the normalized strain using the following formula, that is: For each value in S i , S i is squared and they are accumulated to obtain 6. The shape sensing method based on the core combination optimization adaptive algorithm according to claim 5, characterized in that: The forming of different triangular combinations of the candidate cores and the calculation of the total strain jump degree of the cores in each combination include: Form different triangular core combinations with the candidate cores. Assuming there are a total of g different triangular core combinations, calculate the magnitude of the total jump in core strain for each combination GS m ( m = 1, 2 …g), that is: S m1 , S m2 , S m3 are the magnitudes of the strain jump degrees of the three optical fibers that make up the triangular optical fiber core combination, that is, the set { S m1 , S m2 , S m3} is a subset of the set { S i}.

Images