Method for reducing influence of frequency drift on phi-OTDR demodulation based on WOA-VMD

Abstract

The invention discloses a method for reducing influence of frequency drift on phi-OTDR demodulation based on WOA-VMD, and relates to the technical field of optical fiber sensing. The invention provides a method for reducing influence of frequency drift on phi-OTDR demodulation based on WOA-VMD. The invention discloses a method for reducing influence of frequency drift on phi-OTDR demodulation based on WOA-VMD. The method specifically comprises the following steps: acquiring an RBS signal of a phi-OTDR system; demodulating the RBS signal through an IQ demodulation method; obtaining a secondary difference signal by using a secondary difference method; obtaining a VMD parameter combination according to the WOA-VMD; carrying out VMD on the secondary differential signal according to the obtained parameter combination, and decomposing the secondary differential signal into different IMFs; and according to the obtained IMF, removing a low-frequency component in the secondary differential signal to obtain a noise reduction demodulation signal. On the basis of not changing an original phi-OTDR system hardware structure, the influence of frequency drift on phi-OTDR demodulation is reduced, and the problem of phi-OTDR demodulation waveform distortion caused by laser frequency drift is solved; and the application range of the phi-OTDR in the field with high requirements on high sensitivity and low frequency response is expanded.

Description

Technical Field

[0001] The invention relates to the field of optical fiber technology, and in particular to a method for reducing the influence of frequency drift on Φ-OTDR demodulation based on WOA-VMD. Background Art

[0002] Phase-Sensitive Optical Time-Domain Reflectometry (Φ-OTDR) is an optical measurement technology for detecting and analyzing changes in the structure or environment around the sensing fiber. This technology can measure the intensity of backscattered light to demodulate the phase information and has the ability to detect subtle changes in the structure or environment. Φ-OTDR has been widely used in peripheral security, pipeline monitoring, geological detection, and power equipment monitoring due to its high sensitivity, long sensing distance, and good dynamic detection capabilities. Compared with point sensors, Φ-OTDR can obtain continuously distributed vibration information on the sensing fiber, which is more suitable for long-distance and large-scale monitoring.

[0003] In fact, the Φ-OTDR system is affected by many kinds of interference in application. Laser frequency drift is a very important interference factor, and its influence cannot be ignored. Affected by laser frequency drift, the demodulated waveform of Φ-OTDR will be distorted, thus limiting the application scope of Φ-OTDR in fields with high requirements for high sensitivity and low frequency response. Summary of the invention

[0004] In order to solve the technical problems raised by the above-mentioned background technology, the present invention aims to provide a method for reducing the influence of frequency drift on Φ-OTDR demodulation based on Whale Optimization Algorithm and Variational Mode Decomposition (WOA-VMD), so as to solve the problem of Φ-OTDR demodulation waveform distortion caused by laser frequency drift and expand the application scope of Φ-OTDR.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is: a method for reducing the influence of frequency drift on Φ-OTDR demodulation based on WOA-VMD, comprising the following steps:

[0006] S1. Obtain the Rayleigh Backscattering (RBS) signal of the Φ-OTDR system by acquisition;

[0007] S2. demodulate the RBS signal by an IQ demodulation method;

[0008] S3, using a quadratic difference method to obtain a quadratic difference signal;

[0009] S4. Obtain VMD parameter combination according to WOA-VMD;

[0010] S5. Perform VMD on the secondary differential signal according to the obtained parameter combination to decompose it into different intrinsic mode functions (IMFs);

[0011] S6. According to the obtained IMF, the low-frequency component in the secondary differential signal is removed to obtain a noise-reduced demodulated signal.

[0012] As described above, a method for reducing the influence of frequency drift on Φ-OTDR demodulation based on WOA-VMD is characterized in that the RBS signal in S1 is:

[0013] V(t)=RE(t)cos[2πΔft+Φ(t)],

[0014] Where t represents time, R represents a constant, E(t) represents the amplitude, Δf represents the frequency shift added by the AOM, and Φ(t) represents the demodulated signal to be demodulated.

[0015] As described above, a method for reducing the influence of frequency drift on Φ-OTDR demodulation based on WOA-VMD is characterized in that the specific steps of IQ demodulation in S2 are:

[0016] S21, multiply V(t) by cos(2πΔft) and sin(2πΔft) respectively and pass through a low-pass filter to obtain two orthogonal signals I(t) and Q(t);

[0017] S22. Demodulated signal Φ(t)=arctan[Q(t) / I(t)], arctan() is an inverse tangent function.

[0018] As described above, a method for reducing the influence of frequency drift on Φ-OTDR demodulation based on WOA-VMD is characterized in that the specific steps of the secondary difference method in S3 include:

[0019] S31, setting the differential length L, where L is greater than the sum of the vibration duration length and the system spatial resolution;

[0020] S32, select points X2 and X1 before and after the optical fiber vibration area, respectively, with X1>X2, and the distance between the two points is L, and differentiate the demodulated signal at point S1 from the demodulated signal at point S2 to obtain the demodulated signal between the two points, recorded as (X1, X2);

[0021] S33. Select two points X4 and X3 with an interval of L in the non-vibration area, and X3>X4, and obtain (X3, X4). Difference of (X1, X2) and (X3, X4) is obtained to obtain a secondary differential signal.

[0022] As described above, a method for reducing the influence of frequency drift on Φ-OTDR demodulation based on WOA-VMD is characterized in that the specific steps of S4 are:

[0023] S41, constructing a fitness function;

[0024] Further, the fitness function is minimum sample entropy;

[0025] S42, initialize the number of whale populations N, the individual whale positions Z = [k, α], and the maximum number of iterations M;

[0026] Where k and α represent the number of IMFs and penalty factors that need to be set in VMD, respectively;

[0027] S43, establishing a mathematical model of shrinking and surrounding, random search, and spiral shrinkage, specifically:

[0028] Generate a random number p in [0,1], and perform shrinkage when p<0.5 and |A|<1. Shrinkage model:

[0029] D=|CZ * (m)-Z(m)|

[0030] Z(m+1)=Z * (m)-AD

[0031] Among them, D is the distance between the best individual whale position and the average individual whale position in the whale population, and Z * (m) is the best individual whale position in the mth iteration. Z(m) is the individual whale position in the mth iteration. A and C are control coefficients:

[0032] A=2βr-β

[0033] C=2r

[0034] Where β = 2 – (2m / M), β decreases linearly from 2 to 0 during the iteration process. r is a random value in [0,1].

[0035] When p<0.5 and |A|≥1, a random search is performed. Random search model:

[0036] D r =|CZ r (m)-Z(m)|

[0037] Z(m+1)=Z r (m)-AD r

[0038] Among them, D ris the distance between the best individual whale position and the average individual whale position in the whale population. r (m) represents the position of a random whale individual in the mth iteration.

[0039] When p≥0.5, spiral contraction is performed. Spiral contraction model:

[0040] D p =|Z * (m)-Z(m)|

[0041] Z(m+1)=e bl D p cos(2πl)+Z * (m)

[0042] Among them, D p is the distance between the best individual whale position and the average individual whale position in the whale population. b is the logarithmic spiral coefficient, and l is a random number in the interval [-1,1].

[0043] S44, performing VMD on the secondary differential signal according to each whale individual position Z, calculating the fitness of each whale individual position, selecting the minimum fitness value, and obtaining the best whale individual position in the current whale population;

[0044] S45, updating the individual positions of whales in the current whale population;

[0045] S46, repeat steps S33 and S45 to complete the maximum number of iterations M;

[0046] S47. Obtain the best individual whale position Z0=[k0,α0] and output it. [k0,α0] is the VMD parameter combination for this time.

[0047] The advantages and beneficial effects of the present invention are:

[0048] The present invention reduces the influence of frequency drift on Φ-OTDR demodulation without changing the original Φ-OTDR system hardware structure, and solves the problem of Φ-OTDR demodulation waveform distortion caused by laser frequency drift.

[0049] The present invention improves the performance of the Φ-OTDR system and expands the application range of the Φ-OTDR in fields with high requirements for high sensitivity and low frequency response. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 This is the structure diagram of the Φ-OTDR system;

[0051] Figure 2 It is a flow chart of a method for reducing the influence of frequency drift on Φ-OTDR demodulation based on WOA-VMD;

[0052] Figure 3 Flow chart of obtaining VMD parameter combination according to WOA-VMD;

[0053] Figure 4 It is the demodulated signal waveform of the Φ-OTDR system.

[0054] Figure 5 The demodulated signal waveform of the Φ-OTDR system processed by the present invention. DETAILED DESCRIPTION

[0055] The present invention will be further described in detail below with reference to the accompanying drawings and by way of examples. The following examples are intended to explain the present invention but the present invention is not limited to the following examples.

[0056] The present invention provides a method for reducing the influence of frequency drift on Φ-OTDR demodulation based on WOA-VMD, and its flow chart is as follows: Figure 2 As shown, the specific implementation steps are as follows:

[0057] Figure 2 As shown in S1, the Rayleigh Backscattering (RBS) signal of the Φ-OTDR system is obtained by acquisition.

[0058] Figure 1 The structure diagram of the Φ-OTDR system is shown. In this embodiment, the narrow linewidth laser (NLL) with a linewidth of 3kHz emits laser light and is divided into two beams of light through coupler 1 (OC1). One beam of light is modulated into pulse light by an acousto-optic modulator (AOM), and the other beam of light is used as a local reference light. The frequency shift of the acousto-optic modulator is 80MHz. The pulse light is amplified by an optical amplifier (EDFA) and injected into a single-mode optical fiber (SMF) through a circulator (CIR). After the RBS light is generated, it is mixed with the local reference light through coupler 2 (OC2), detected by a photoelectric balanced detector (BPD), filtered by a bandpass filter (BPF), and collected by an acquisition card (DAQ), and then the RBS signal is obtained on a personal computer (PC). Piezoelectric ceramics (PZT) are placed somewhere in the optical fiber to generate vibrations. The total length of the single-mode optical fiber is 3.84km, and the optical fiber in the range of about 3.6 to 3.64km is wound around the PZT. The detection pulse width was set to 80ns and the frequency was set to 400Hz, and a 1V, 0.2Hz sine wave was applied to the PZT. The sampling rate of the DAQ was set to 125MSPS, and 400 RBS signals were sampled.

[0059] V(t)=RE(t)cos[2πΔft+Φ(t)],

[0060] Where t represents time, R represents a constant, E(t) represents the amplitude, Δf represents the frequency shift added by the AOM, and Φ(t) represents the demodulated signal to be demodulated.

[0061] Figure 2 As shown in S2, the RBS signal is demodulated by an IQ demodulation method, and the specific steps are:

[0062] S21. Multiply V(t) by cos(2πΔft) and sin(2πΔft) respectively and pass them through a low-pass filter to obtain two orthogonal signals I(t) and Q(t).

[0063] S22. Demodulated signal Φ(t)=arctan[Q(t) / I(t)], arctan() is an inverse tangent function.

[0064] In this embodiment, the demodulated signal waveform of the Φ-OTDR system is as follows: Figure 4 shown.

[0065] Figure 2 As shown in S3, a quadratic differential signal is obtained by using a quadratic differential method. The specific steps of the quadratic differential method include:

[0066] S31, setting the differential length L, where L is greater than the sum of the vibration duration length and the system spatial resolution. In this embodiment, L=60m.

[0067] S32. Select points X2 and X1 before and after the optical fiber vibration area, respectively, with X1>X2 and a distance L between the two points. Differ the demodulation signal at point S1 from the demodulation signal at point S2 to obtain a demodulation signal between the two points, recorded as (X1, X2). In this embodiment, X1=3650, X2=3590. (3650, 3590) is the demodulation signal at the two points.

[0068] S33, select two points X4 and X3 with an interval of L in the non-vibration area, and X3>X4, and obtain (X3, X4), (X1, X2) and (X3, X4) to obtain a secondary differential signal. In this embodiment, X3 = 3590, X4 = 3530. The secondary differential signal is the difference between (3650, 3590) and (3590, 3520).

[0069] Figure 2 As shown in S4, a VMD parameter combination is obtained according to WOA-VMD. The flowchart of step S4 is as follows Figure 3 As shown, the specific steps are:

[0070] S41. Construct a fitness function.

[0071] Furthermore, the fitness function is minimum sample entropy.

[0072] S42. Initialize the number of whale populations N, the individual whale positions Z = [k, α], and the maximum number of iterations M.

[0073] In this embodiment, N=20, the range of k is [2,9], the range of α is [100,5000], and M=30. k and α represent the number of IMFs and the penalty factor required to be set in VMD, respectively.

[0074] S43, establish a shrinking surround, random search, and spiral shrinking model. Specifically:

[0075] Generate a random number p in [0,1], and perform shrinkage when p<0.5 and |A|<1. Shrinkage model:

[0076] D=|CZ * (m)-Z(m)|

[0077] Z(m+1)=Z * (m)-AD

[0078] Among them, D is the distance between the best individual whale position and the average individual whale position in the whale population, and Z * (m) is the best individual whale position in the mth iteration. Z(m) is the individual whale position in the mth iteration. A and C are control coefficients:

[0079] A=2βr-β

[0080] C=2r

[0081] Where β = 2 – (2m / M), β decreases linearly from 2 to 0 during the iteration process. r is a random value in [0,1].

[0082] When p<0.5 and |A|≥1, a random search is performed. Random search model:

[0083] D r =|CZ r (m)-Z(m)|

[0084] Z(m+1)=Z r (m)-AD r

[0085] Among them, D r is the distance between the best individual whale position and the average individual whale position in the whale population. r (m) represents the position of a random whale individual in the mth iteration.

[0086] When p≥0.5, spiral contraction is performed. Spiral contraction model:

[0087] D p =|Z * (m)-Z(m)|

[0088] Z(m+1)=e bl Dp cos(2πl)+Z * (m)

[0089] Among them, D p is the distance between the best individual whale position and the average individual whale position in the whale population. b is the logarithmic spiral coefficient, and l is a random number in the interval [-1,1].

[0090] S44, performing VMD on the secondary differential signal according to each whale individual position Z, calculating the fitness of each whale individual position, selecting the minimum fitness value, and obtaining the best whale individual position in the current whale population.

[0091] S45. Update the individual positions of whales in the current whale population according to step S43.

[0092] S46. Repeat steps S44 and S45 to complete the maximum number of iterations M.

[0093] S47. Obtain the best individual whale position Z0=[k0,α0] and output it. [k0,α0] is the VMD parameter combination for this time.

[0094] Figure 2 As shown in S5, VMD is performed on the secondary differential signal according to the obtained parameter combination to decompose it into different IMFs.

[0095] Figure 2 As shown in S6, according to the obtained IMF, the low-frequency component in the secondary differential signal is removed to obtain a noise-reduced demodulated signal.

[0096] In this embodiment, the waveform of the demodulated signal of the Φ-OTDR system processed by the present invention is as follows: Figure 5 As shown. Figure 4 By comparing the original demodulated signal waveforms, it can be found that the present invention solves the problem of Φ-OTDR demodulated waveform distortion caused by laser frequency drift.

[0097] In this embodiment, the Φ-OTDR system has a simple structure, and the influence of frequency drift on Φ-OTDR demodulation is reduced without changing its hardware structure, thereby expanding the application scope of Φ-OTDR in fields with high requirements for high sensitivity and low frequency response.

[0098] The above is only one embodiment of the present invention, not all or the only embodiment. Any equivalent changes made to the technical solution of the present invention by ordinary technicians in this field after reading the specification of the present invention are covered by the claims of the present invention.

Claims

1. A method for reducing the influence of frequency drift on phase-sensitive optical time-domain reflectometry (Φ-OTDR) demodulation based on Whale Optimization Algorithm and Variational Mode Decomposition (WOA-VMD), characterized in that: The specific steps are as follows: S1. Obtain the Rayleigh Backscattering (RBS) signal of the Φ-OTDR system by acquisition; S2. demodulate the RBS signal by an IQ demodulation method; S3, using a quadratic difference method to obtain a quadratic difference signal; S4. Obtain VMD parameter combination according to WOA-VMD; S5. Perform VMD on the secondary differential signal according to the obtained parameter combination to decompose it into different intrinsic mode functions (IMF); S6. According to the obtained IMF, the low-frequency component in the secondary differential signal is removed to obtain a noise-reduced demodulated signal.

2. A method for reducing the influence of frequency drift on Φ-OTDR demodulation based on WOA-VMD according to claim 1, characterized in that: The RBS signal in S1 is: V(t)=RE(t)cos[2πΔft+Φ(t)], Where t represents time, R represents a constant, E(t) represents the amplitude, Δf represents the frequency shift added by the AOM, and Φ(t) represents the demodulated signal to be demodulated.

3. A method for reducing the influence of frequency drift on Φ-OTDR demodulation based on WOA-VMD according to claim 1, characterized in that: The specific steps of IQ demodulation in S2 are: S21, multiply V(t) by cos(2πΔft) and sin(2πΔft) respectively and pass through a low-pass filter to obtain two orthogonal signals I(t) and Q(t); S22. Demodulated signal Φ(t)=arctan[Q(t) / I(t)], arctan() is an inverse tangent function.

4. A method for reducing the influence of frequency drift on Φ-OTDR demodulation based on WOA-VMD according to claim 1, characterized in that: The specific steps of the quadratic difference method in S3 include: S31, setting the differential length L, where L is greater than the sum of the vibration duration length and the system spatial resolution; S32, select points X2 and X1 before and after the optical fiber vibration area, respectively, with X1>X2, and the distance between the two points is L, and differentiate the demodulated signal at point S1 from the demodulated signal at point S2 to obtain the demodulated signal between the two points, recorded as (X1, X2); S33. Select two points X4 and X3 with an interval of L in the non-vibration area, and X3>X4, and obtain (X3, X4). Difference between (X1, X2) and (X3, X4) to obtain a secondary differential signal.

5. The method for reducing the influence of frequency drift on Φ-OTDR demodulation based on WOA-VMD according to claim 1, characterized in that: The specific steps of S4 are: S41, constructing a fitness function; Further, the fitness function is minimum sample entropy; S42, initialize the number of whale populations N, the individual whale positions Z = [k, α], and the maximum number of iterations M; Where k and α represent the number of IMFs and penalty factors that need to be set in VMD, respectively; S43, establishing a mathematical model of shrinking and surrounding, random search, and spiral shrinkage, specifically: Generate a random number p in [0,1]. When p<0.5 and |A|<1, perform shrinkage and encirclement. The shrinkage and encirclement model is: D=|CZ * (m)-Z(m)| Z(m+1)=Z * (m)–AD Among them, D is the distance between the best individual whale position and the average individual whale position in the whale population, and Z * (m) is the best individual whale position in the mth iteration, Z(m) is the individual whale position in the mth iteration, and A and C are control coefficients: A＝2βr–β C＝2r Where β = 2 – (2m / M), β decreases linearly from 2 to 0 during the iteration process, and r is a random value in [0,1]; When p<0.5 and |A|≥1, a random search is performed. The random search model is: D r =|CZ r (m)-Z(m)| Z(m+1)=Z r (m)-AD r Among them, D r is the distance between the best individual whale position and the average individual whale position in the whale population, Z r (m) represents the position of a random whale individual in the mth iteration; When p ≥ 0.5, spiral contraction is performed. The spiral contraction model is: D p =|Z * (m)-Z(m)| Z(m+1)=e bl D p cos(2πl)+Z * (m) Among them, D p is the distance between the best individual whale position and the average individual whale position in the whale population, b is the logarithmic spiral coefficient, and l is a random number in the interval [-1,1]; S44, performing VMD on the secondary differential signal according to each whale individual position Z, calculating the fitness of each whale individual position, selecting the minimum fitness value, and obtaining the best whale individual position in the current whale population; S45, updating the individual positions of whales in the current whale population; S46, repeat steps S33 and S45 to complete the maximum number of iterations M; S47. Obtain the best individual whale position Z0=[k0,α0] and output it. [k0,α0] is the VMD parameter combination for this time.

Images