Method for predicting detectable distance of phi-OTDR (Optical Time Domain Reflectometer)

Abstract

The invention discloses a method for predicting detectable distance of # imgabs0 #, and belongs to the technical field of measurement. According to the method, the distortion rate of a demodulation signal is taken as a reference index, after a # imgabs 1 # system is successfully constructed and disturbance within a certain range is applied, the standard deviation of a Rayleigh Backscattered (RBS) signal and the standard deviation of thermal noise of a photoelectric detector (BPD) are obtained through measurement, then the relation between the ratio of the two signals and the signal distortion rate is established, a critical ratio is selected, and the distortion rate of the demodulation signal is calculated. The value meets the requirement that the RBS demodulation waveform does not have difference, and the D value is small enough. And if not, adjusting the incident light power by adjusting the EDFA driving current, and repeating the steps. Finally, the relation between the ratio and the optical fiber position is established, and the detection distance can be determined. According to the method, the detection distance can be predicted only by obtaining the standard deviation of the RBS signal generated by the first section of the optical fiber and the standard deviation of the thermal noise output by the BPD under the dark condition.

Description

Technical Field

[0001] This invention patent relates to a method for predicting the detectable distance, belonging to the field of measurement technology. Background Art

[0002] Phase-Sensitive Optical Time-Domain Reflectometry ( ) is one of the technologies in the Distributed Optical Fiber Sensor System (DOFS). This technology uses optical fiber as the sensing element and transmission medium, and has applications in marine geophysics, intelligent transportation systems, underwater monitoring, perimeter security and other fields. In addition, as the mainstream technology in , the characteristics of high signal-to-noise ratio and high sensitivity of coherence have attracted the attention and research of many scholars at home and abroad. At present, most literature points out that the detection distance is one of the key performance indicators of . If you want to increase the detection distance of Summary of the Invention

[0003] The purpose of this invention is to provide a method for predicting the detectable distance to solve the problem of reasonable allocation of system resources such as intelligent monitoring based on .

[0004] To achieve the above purpose, the technical solution adopted by this invention is as follows:

[0005] This invention takes the distortion rate of the demodulated signal as the reference index, based on analyzing the composition of the RBS signal noise, and determines the detectable distance according to the ratio of the standard deviation of the RBS signal to the standard deviation of the BPD thermal noise obtained from the measured signal. The specific operation steps are as follows:

[0006] Step 1: After successfully constructing the system, apply a certain vibration signal to the end of the optical fiber.

[0007] Step 2: According to the actual measurement signal, obtain the standard deviation σ f of the noise-free RBS signal generated by the optical fiber and the standard deviation σ of the thermal noise of the Balanced Photodetector (BPD) under the condition of no lightther 。

[0008] Step 3: We establish the relationship between the ratio ρ of the above two and the complete distortion rate D through a coordinate diagram, where: σ f (x max ) / σ ther = ρ (1) In the above two formulas: σ f (x max ) is the standard deviation of the noiseless RBS signal at the maximum detection distance; σ ther is the standard deviation of the BPD thermal noise; n is the number of points of complete distortion in the optical fiber within a certain range; n0 is the total number of detection points of the optical fiber in the same range.

[0009] Step 4: Determine the critical value of the ratio ρ, denoted as ρ crit , which is called the threshold; at this time, the RBS demodulation waveform is not bad, and the D value is small enough; if not satisfied, adjust the driving current of the Erbium-Doped Fiber Amplifier (EDFA) to adjust the optical power of the incident light, and repeat the above steps.

[0010] Step 5: The prediction of the detection distance can be realized from Equation (1), and the maximum detectable distance x max is:

[0011] where: ρ crit represents the threshold; α represents the optical fiber attenuation coefficient; σ ther represents the standard deviation of the thermal noise; σ U (x1) represents the standard deviation of the noisy RBS signal at the first section of the optical fiber.

[0012] In addition, the expressions for the standard deviation σ U (x1) of the noisy RBS signal at the first section of the optical fiber and the attenuation coefficient α in the above formula are respectively:

[0013] That is: σ U (x1) is obtained by Gaussian distribution fitting of the standard deviation of the noiseless RBS signal at the first section and the standard deviation of the thermal noise of the BPD output signal without optical input.

[0014] The present invention designs a method for predicting the detectable distance. Compared with the prior art, the advantages and beneficial effects of the present invention are:

[0015] Based on For intelligent monitoring systems and others, it provides a more reasonable resource allocation; when the fiber optic length is no more than 35.7 km, the maximum error between the measured distance and the predicted distance is only 1.0 km, which can better achieve the prediction of the detectable distance. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The present invention will be further described in detail below with reference to the drawings.

[0017] Figure 1 It is the system operation principle of the entire embodiment.

[0018] Figure 2 For system schematic diagram.

[0019] Figure 3 It is a schematic diagram of the sliding window method.

[0020] Figure 4 It is the output voltage distribution of BPD under light and dark conditions. Among them: Figure 4 a is the output voltage distribution of BPD under dark conditions, Figure 4 b is the output voltage distribution of BPD under light conditions.

[0021] Figure 5 It is the waterfall diagram of the demodulation results of the RBS signal at different signal-to-noise ratios under a specific standard deviation σ U where: Figure 5 aσ U = 0.435 V, Figure 5 bσ U = 0.294 V, Figure 5 cσ U = 0.216 V, Figure 5 dσ U = 0.107 V, Figure 5 eσ U = 0.074 V, Figure 5 fσ U = 0.034 V.

[0022] Figure 6 It is the relationship between D and σ f / σ ther where the x-axis is in logarithmic coordinates.

[0023] Figure 7 It is the relationship between σ f / σ ther of the RBS signal and the distance for 3 groups of different signal-to-noise ratios. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0024] The present invention will be further described in detail below in conjunction with the accompanying drawings and through embodiments. The following embodiments are explanations of the present invention, and the present invention is not limited to the following embodiments.

[0025] Through calculation, it can be obtained that compared with other DOFS technologies, the method for predicting the detectable distance reduces the maximum detectable distance error to 1.0 km.

[0026] Figure 1 is the basic flowchart of this embodiment. After successfully constructing the system, a vibration signal is applied in the area about 1 km from the end of the optical fiber; the standard deviation σ of the ideal RBS signal is measured at the head of the optical fiber f and the standard deviation σ of the thermal noise at the output end of the BPD ther ; then, based on the above two data, the standard deviation σ of the noisy RBS signal is obtained U ; then, the relationship between the different ratios ρ of the two signals and the signal distortion rate D is established through Equation (12); finally, after multiple experimental measurements, the threshold ρ crit is obtained. At this time, the RBS demodulation waveform is not bad, and the D value is small enough; if not satisfied, the driving current of the EDFA is adjusted to adjust the optical power injected into the optical fiber, and the above steps are repeated. After obtaining ρ crit , the maximum detectable distance x max can be obtained by combining Equation (14). The details of each item in the basic process of this embodiment will be described below:

[0027] Figure 2 is the system principle block diagram built in this embodiment , where the length of the sensing optical fiber is 1 km. After the system works normally, the highly coherent laser emitted by the narrow linewidth laser (NLL) is divided into two branches. The upper branch is called the RBS light, and the lower branch is called the local light. The two beams of light are finally detected to obtain the RBS signal through coupling. The ideal RBS signal f(t) can be simply expressed as:

[0028] where: t is time, A(t) is the signal amplitude, and f p is the phase shift generated by the light passing through the acousto-optic modulator, is the phase shift generated by the laser at this time. Since generating RBS light requires the interference of backward Rayleigh scattered light from a large number of scattering centers, f(t) is random at different times t, and f(t) satisfies a Gaussian distribution, which is also verified in the subsequent experiments. However, due to the attenuation of the optical fiber, the intensity of the RBS signal at a relatively far detection distance is low, and the detection noise generated by the BPD will affect the quality of the RBS signal at this time, resulting in abnormal demodulation. Considering the thermal noise and shot noise brought by the BPD, the noisy RBS signal U(t) can be expressed as: U(t) = f(t) + u shot (t) + u ther (t) (8)

[0029] where: u shot (t) and u ther (t) refer to the shot noise and thermal noise of the BPD respectively, and both noises follow a Gaussian distribution. The standard deviation of the thermal noise is independent of the input optical power and is mainly related to the device's own parameters and temperature. The standard deviation of the shot noise is related to the input optical power at this time, that is, the greater the optical power, the greater the standard deviation of this noise. The shot noise at different fiber positions does not affect the signal-to-noise ratio, that is, the shot noise has no effect on the detection distance; so the shot noise is ignored in this embodiment, and therefore the noisy RBS signal can be simplified as: U(t) = f(t) + u ther (t) (9)

[0030] It can be seen from the above formula that the maximum detection distance is related to the amplitude of the RBS signal and the intensity of the thermal noise of the BPD. Since both f(t) and u ther (t) follow a Gaussian distribution and they are independent of each other, U(t) also follows a Gaussian distribution. Assume that the standard deviations of U(t), f(t), and u ther (t) are σ U , σ f , and σ ther respectively. According to the properties of the Gaussian distribution:

[0031] Therefore, after measuring the noisy RBS signal in a certain position interval and the output voltage of the BPD without light input, and fitting the two with a Gaussian distribution, σ U and σ ther can be obtained; then the standard deviation σ f of the noise-free RBS signal can be obtained according to the above formula. According to the above theoretical analysis, when the environment changes little, σ ther can be approximated as a constant. However, due to the attenuation of the optical fiber, when the sensing optical fiber is long, σ U , σ fIt will vary according to the different interval positions. Therefore, according to the actual situation, the above formula can be changed to:

[0032] Where: x is the starting point of the selected position interval, with the unit of m. σ U (x) represents the standard deviation of the noisy RBS signal U(t) within the range from x to x + z, where z represents the interval size, and this value should not be too large or too small. In this embodiment, z = 2000m.

[0033] Figure 3 To improve the resolution of the prediction results, the sliding window method is adopted, and each value of x needs to be set, and x n+1 and x n The distance s between them is the interval of each selected position interval. In this embodiment, s is 300m.

[0034] Let x max be the maximum detectable distance. By calculating the critical ratio of the ratio of the standard deviation of the noise-free RBS signal to the standard deviation of the BPD thermal noise of this section of the signal, we have: σ f (x max ) / σ ther = ρ (12)

[0035] Where: ρ is a constant value in the experiment; the fiber attenuation α refers to the attenuation coefficient per unit length of the optical fiber, and its unit is dB / m; in this embodiment, α can be defined as

[0036] Where: x1 is the starting position of the fiber head section position interval, that is, 0m, and x j is the starting position of any position interval. In summary, since α can be obtained from the optical fiber manufacturer or measured in advance, therefore, when ρ crit and σ ther are known and σ U (x1) is measured through a shorter optical fiber, the maximum detectable distance x max is:

[0037] Figure 4 is the BPD output voltage distribution under light and no-light conditions. In this embodiment, a 1kHz sinusoidal vibration is applied within about 40m at the end of the optical fiber. When the NLL does not generate laser, the BPD output voltage is measured, and the voltage distribution with a standard deviation of 0.009V as shown in Figure 4 (a) is obtained. That is, u ther (t) follows a normal distribution with a mean of 0V and a standard deviation of σ therIt is a Gaussian distribution with a mean of 0.009V. Then, NLL is enabled to make the system work properly. The pulse repetition frequency is set to 20 kHz, and the pulse width is 80 ns. 200 RBS signals are collected, and their distribution is as Figure 4 shown in Fig. b. It can be seen that the RBS signals follow a Gaussian distribution with a mean of 0V and a standard deviation of 0.351V. In summary, the distribution results of the measured signals are consistent with the corresponding theories.

[0038] Figure 5 To simulate different demodulation qualities, based on the above settings, while keeping other experimental parameter settings unchanged, the EDFA drive current is adjusted to make the optical power incident on the optical fiber different, and different groups of RBS signals are obtained. Each group of RBS signals follows a Gaussian distribution. The demodulation results of each group of RBS signals are as shown.

[0039] From Figure 5 it can be observed that as σ U decreases, the quality of the demodulation waveform deteriorates, specifically manifested by an increase in the demodulation error positions (dark regions). This is because as σ U decreases, the signal-to-noise ratio of the RBS signals decreases, resulting in a deterioration of the demodulation waveform quality.

[0040] Figure 6 Fig. shows the relationship between D and σ f / σ ther . Among them, D is defined as the complete distortion rate, which represents the ratio of the number of completely distorted points within 1 km of the optical fiber to the total number of detection points, and can be used to evaluate the quality of each group of demodulation waveforms. In this paper, when the maximum value of the absolute value of the demodulation waveform error at a certain distance is greater than 1.5 rad, that point is considered a completely distorted point, and the judgment criterion can also be appropriately adjusted according to the actual situation during application. Figure 6 The D values of the subfigures in Fig. are 0.5%, 1.2%, 1.6%, 4.2%, 10.1%, and 40.2% respectively.

[0041] From Figure 6 it can be found that as σ f / σ ther [[ID=3?]]increases, the mean value, maximum value, and minimum value of D all decrease. Therefore, D can effectively reflect the negative impact of BPD thermal noise on the demodulated signal. In this embodiment, considering that when the critical values of the D mean are 16%, 6%, 3%, and 1%, the corresponding ρ values are 6.6, 12.0, 15.8, and 33.6 respectively. At this time, the maximum sensing distance can be calculated according to Equation (14). Note that appropriate D and ρ values can be selected as the critical values according to actual needs.

[0042] To verify the prediction effect of this method, an optical fiber with a length of 42 km and an attenuation coefficient of 2.0×10 -4Replace the above optical fiber with an optical fiber of dB / m and keep the experimental environment unchanged, σ ther is about 0.009V. Adjust the drive current of the EDFA to obtain 3 groups of RBS signals with different signal-to-noise ratios.

[0043] Figure 7 The 3 groups of signals σ f / σ ther are the relationships with the sensing distance obtained by the above sliding window method.

[0044] Through actual measurement, the distances corresponding to the 3 groups of signals at different thresholds ρ crit are obtained. The results predicted by Equation (14) are shown in the following table. Among them, "-" indicates that the threshold cannot be reached, which is due to the low power of the RBS signal. For long optical fibers, D in the table represents the D value of the demodulation waveform of the optical fiber within 1 km starting from the predicted distance. The distances corresponding to the 3 groups of RBS signals with different signal-to-noise ratios at different thresholds ρ crit and D

[0045] By comparison Figure 7 with the above table, it can be seen that the maximum error between the measured distance and the predicted distance is only -1.0 km. For a sensing distance of dozens of kilometers, this error is relatively small. In addition, the basic distribution of D satisfies Figure 6 in which its relationship with σ f / σ ther further verifies the reliability of this evaluation method.

[0046] The present invention uses the ratio of the noise-free RBS signal to the standard deviation of the thermal noise output of the BPD for prediction. The experimental results show that when the length of the optical fiber is not more than 35.7 km, the maximum error between the measured distance and the predicted distance is only 1.0 km, and the detectable distance can be effectively predicted, which provides help for the rational allocation of resources in the #imgpt33# intelligent monitoring system.

[0047] The above is only one implementation manner of the present invention, not all or the only implementation manner. Any equivalent transformation of the technical solution of the present invention adopted by those of ordinary skill in the art by reading the specification of the present invention is covered by the claims of the present invention.

Claims

1. A method for predicting the detectable distance, which is mainly characterized by Including the following steps: Step 1: After successfully constructing the system, apply a certain vibration signal to the end of the optical fiber; Step 2: According to the actually measured signal, obtain the noise-free Rayleigh Backscattered (RBS) signal generated by the optical fiber, and calculate its standard deviation σ f ; Measure the standard deviation σ of the thermal noise of the Balanced Photodetector (BPD) under the condition of no light ther ; Step 3: Establish the relationship between the ratio ρ of the above two and the complete distortion rate D; Step 4: Determine the critical value of ρ, denoted as ρ crit , which is called the threshold value. This value satisfies that the RBS demodulation waveform is not bad and the D value is small enough. If it does not meet the requirements, adjust the optical power of the incident light by adjusting the drive current of the Erbium-Doped Fiber Amplifier (EDFA), and repeat the above steps; Step 5: The prediction of the detection distance can be achieved by Equation (1), and the maximum detectable distance x max is as follows: Where: ρ crit is the threshold value, representing the critical ratio of the standard deviation of the noise-free RBS signal to the standard deviation of the BPD thermal noise; α represents the fiber attenuation coefficient; σ ther represents the standard deviation of the thermal noise; σ U (x1) represents the standard deviation of the noisy RBS signal at the first section of the optical fiber.

2. A method for predicting the detectable distance, characterized in that: Taking the distortion rate of the demodulated signal as the reference index, based on the analysis of the RBS signal noise composition, and determining the detectable distance according to the ratio of the standard deviation of the RBS signal to the standard deviation of the BPD thermal noise obtained from the measured signal.

3. A method for predicting the detectable distance, characterized in that: Standard deviation σ of the noisy RBS signal at the first section of the optical fiber U (x1) is obtained by the following formula, that is: As can be seen from the above formula, σ U (x1) is obtained by performing Gaussian distribution fitting on the standard deviation σ f of the noise-free RBS signal in the first segment and the standard deviation σ ther of the thermal noise of the BPD output signal under no optical input.

4. A method for predicting the detectable distance, characterized in that: ρ is the standard deviation σ of the noiseless RBS signal at the maximum detection range f (x max ) and the standard deviation σ of the BPD thermal noise ther The ratio, i.e.: σ f (x max ) / σ ther =ρ (3) where: x max is the starting position of the maximum detectable distance range.

5. A method for predicting the detectable distance, characterized in that: The optical fiber attenuation α at the maximum detection distance is:

6. A method for predicting the detectable distance, characterized in that: The complete distortion rate D represents the probability of the failure of the vibration signal demodulation within a certain range, that is: where n is the number of completely distorted points in the optical fiber within a certain range, and n0 is the total number of detected points of the optical fiber in the same range.

Images