CN-121980910-A - Sampling optimization method of high-spatial-resolution OTDR optical fiber temperature measurement system

Abstract

The invention provides a sampling optimization method of a high-spatial-resolution OTDR optical fiber temperature measurement system, which is based on a variable-frequency sampling frequency selection optimization method of a multi-target particle swarm algorithm (MOPSO), an optimal frequency combination is found under a given frequency range and spatial resolution, data under different sampling frequencies in the frequency combination are acquired, a sampling point closest to the spatial distance of an ideal sampling point is found by utilizing a reconstruction algorithm, and a temperature value of the sampling point is calculated and obtained. The invention can automatically select the sampling frequency combination and the corresponding sampling point, thereby realizing higher-efficiency variable-frequency sampling.

Inventors

• ZHANG HAO
• SHI ZHIMIN
• YANG YONG
• LI FENG
• LI JIJUN
• Zhong Yuehai
• ZHU ZHAOHUAN
• ZHENG YUBIN

Assignees

• 浙江浙能温州发电有限公司
• 浙江大学湖州研究院

Dates

Publication Date

20260505

Application Date

20251215

Claims (5)

1. 1. A sampling optimization method of a high-spatial-resolution OTDR optical fiber temperature measurement system is characterized in that an optimal frequency combination is found under a given frequency range and spatial resolution based on a variable-frequency sampling frequency selection optimization method of a multi-target particle swarm algorithm (MOPSO), data under different sampling frequencies in the frequency combination are collected, a sampling point closest to an ideal sampling point in spatial distance is found by utilizing a reconstruction algorithm, and a temperature value of the sampling point is calculated and obtained.
2. 2. A method of optimizing the sampling of a high spatial resolution OTDR optical fiber thermometry system according to claim 1, characterized in that the temperature measurement at the ideal sampling point location is predicted using trend extrapolation or interpolation using the temperature information of the adjacent sampling points.
3. 3. A method for optimizing sampling of a high spatial resolution OTDR optical fiber temperature measurement system according to claim 1, wherein the method for optimizing variable frequency sampling frequency selection based on a multi-objective particle swarm algorithm (MOPSO) comprises the steps of: Step 1, determining a sampling rate frequency conversion range supported by hardware equipment, inputting a selectable frequency range f max ～f min and ideal spatial resolution deltax into an algorithm, randomly generating a group of particle swarm optimization variables X (T) = (X 1 ,x 2 ,…x i ,…,x m ) according to a multi-objective particle swarm optimization algorithm, wherein each optimized particle X i (i=1, 2,) corresponds to a frequency combination, algorithm parameters comprise a maximum iteration number T max , the number m of particles, an inertia factor maximum omega max , an inertia factor minimum w min , an acceleration constant c 1 、c 2 and a speed maximum v max , setting the current optimization algebra as t=1, T is less than or equal to T max , randomly generating m particles X 1 ,x 2 ,…,x i ,…,x m in a space with a dimension range to form a population X (T), randomly generating initial speeds v 1 ,v 2 ,…,v i ,…,v m of each particle, wherein the position of the i-th particle is X i ＝(x i,1 ,x i,2 ,…,x i,j , the speed is v i ＝(v i,1 ,v i,2 ,…,v i,j ), wherein j=f2, the value of each element in the particles X i (1, 2,) is 0 or 1; Step 2, calculating the distribution situation of sampling points under each sampling frequency according to the frequency combination represented by the optimized particle x i , re-screening two groups of sampling points according to the ideal spatial resolution delta x according to a reconstruction algorithm, and designing to obtain two groups of high-precision reconstruction data distribution situations p 1 ,p 2 ,…,p n and q 1 ,q 2 …,q n ; step 3, establishing a mathematical model to obtain an objective function of the accuracy and sampling times of the measurement reconstruction data: Value 2 (x i )=count Wherein Value 1 (x i ) and Value 2 (x i ) are two objective functions of the optimization algorithm, Δd i is the deviation of the reconstructed data from the position of the point at the ideal spatial resolution, ad i ＝|i·Δx-q i |+|i·Δx-p i |, and count is the number of samples. Calculating an objective function under each optimized particle according to the model, wherein the objective function is used for selecting an individual historical optimal position Prest and a population global historical optimal position Gbest; And 4, generating a group of new particle swarm optimization variables X (t+1) according to the transformation rule of the multi-target particle swarm algorithm, returning to the step 2, continuing the next optimization until the particle swarm algorithm reaches the maximum iteration number, stopping calculation, and outputting the frequency combination corresponding to the optimal particle.
4. 4. A method for optimizing the sampling of a high spatial resolution OTDR optical fiber temperature measurement system according to claim 1, wherein said step 2 comprises the steps of, in order: Step 201, setting the position set of sampling points on the sensing optical fiber as { d i |i=1, 2,.. N }, wherein d i is an arithmetic series with interval deltax from 0, determining frequency combination according to particle x i , firstly calculating the deviation between each ideal sampling point and two nearest actual sampling points in the combination under each single sampling frequency, setting the actual sampling point set at each frequency to form a data set A, Wherein a f,n is the position of the nth sampling point from the initial point at the frequency f; Step 202, screening out the sampling points closest to the actual sampling points under each frequency and the sampling points under the ideal spatial resolution according to the calculated sampling point distribution matrix; For any n E [1, N ], m E { f 1 ,f 2 ,…,f m } satisfies Wherein i is E [1, N ] Step 203, excluding the sample points screened in the step C 1 , screening again, and screening out the sample points of which the actual sample points are second closest to the sample points under the ideal spatial resolution at each frequency to obtain a sample point C 2 ; Step 204, under different sampling frequencies, finding the actual sampling point closest to each ideal sampling point according to C 1 ,C 2 , to obtain the final index matrix: For any m E { f 1 ,f 2 ,…,f m }, all satisfy i=1,2,...,N In the above index matrix F, the first row F 1,i represents the sampling frequency selected by the i-th point of the reconstructed data curve, and the second row F 2,i represents the F 2,i -th sampling point at the frequency F 1,i ; step 205, excluding the sample points screened in the F in the C 1 ,C 2 , and calculating and finding the actual sample point closest to each ideal sample point again to obtain a second index matrix F': Step 205, for each ideal sampling point, obtaining the position and temperature data of two actual sampling points closest to the ideal sampling point from an index matrix F and a second index matrix F', assuming that the positions are D 1 and D 2 respectively, the temperature measurement results are T 1 and T 2 , predicting the temperature of the ideal sampling point by using a trend extrapolation method or an interpolation method according to the proximity degree of the positions, predicting by using a trend extrapolation method if the two points are on the same side of the ideal sampling point, predicting by using an interpolation method if the two points are on different sides of the ideal sampling point, and using a linear regression model.
5. 5. The method for optimizing sampling of high spatial resolution OTDR fiber temperature measurement system according to claim 1, wherein step 4 comprises the steps of: Step 401, using the objective function obtained in step 3 as the fitness Value to evaluate the quality of each particle, embedding Pareto dominant relations into PSO to determine the optimal position pbest and the global optimal position gbest in the whole population, and updating the local optimal pbest according to the Pareto dominant principle based on the objective functions Value 1 (x i ) and Value 2 (x i obtained in step 3), and if the current Value and the original pbest are not dominant, updating randomly. Selecting global optimum gbest in an Archive (Archive set) by calculating crowding in the Archive set, and updating the speed and position of particles to generate a new set of particle swarm optimization variables; step 402, according to the formula: v i,j (t+1)＝ωv i,j (t)+c 1 r 1 [P i,j -x i,j (t)]+c 2 r 2 [P g,j -x i,j (t)] Updating the velocity and formula of particles Updating the position of the particles to generate a new population X (t+1), v i,j being the current speed of the j parameter of the i-th particle, c 1 、c 2 representing a positive acceleration coefficient, r 1 、r 2 being a random number between 0 and 1, p i,j representing the best position pbest found so far for the i-th particle, p g,j representing the best position gbest found for the whole particle population, X i,j (t) being the current position of the j parameter of the i-th particle; Step 403, updating the inertia factor of the optimization algorithm according to the formula ω=ω min +r 3 ·(ω max -ω min ), wherein r 3 is a random number between 0 and 1; Step 404, judging whether T is equal to T max , if yes, outputting the frequency combination corresponding to the optimal particle, and outputting the high-precision data index matrix under the frequency combination, otherwise, t=t+1, and returning to step 402 to continue searching.

Description

Sampling optimization method of high-spatial-resolution OTDR optical fiber temperature measurement system Technical Field The invention relates to a temperature measuring method of an OTDR optical fiber temperature measuring system. Background A distributed temperature measurement system (DTS) is a real-time, online, continuous temperature monitoring system. The system is designed based on the temperature-sensitive effect of Raman scattering and the Optical Time Domain Reflectometry (OTDR) principle. The sensing optical fiber can be used as a sensing probe, so that the on-line monitoring of the full-length temperature field of the optical fiber can be realized. The system is corrosion-resistant, free of electromagnetic interference and intrinsically safe, has the capability of continuously monitoring long-distance and large-range environmental temperature, and is widely applied to pipe network monitoring in various scenes. A typical OTDR optical fiber temperature measurement system comprises an upper computer, a data transmission line, a pulse light source, a wavelength division multiplexing device, a photoelectric detector, a high-speed acquisition card and a sensing temperature measurement optical fiber, wherein the pulse light source emits ns-level pulse light with the wavelength of 1550nm, the ns-level pulse light is emitted into the temperature measurement optical fiber through the wavelength division multiplexing device, raman scattering occurs in the optical fiber to generate a back scattering signal containing temperature information, and Stokes light are screened through the wavelength division multiplexing device. The APD converts the two paths of optical signals into electric signals, and the electric signals are collected by the collecting card and transmitted into the upper computer for processing. The spatial resolution is used as one of important indexes of the distributed temperature measurement system, represents the characterization capability of the system on the abrupt change of the temperature of a narrow-length unit, and is defined as the distance length between 10% and 90% of the temperature step change value measured by the optical fiber temperature measurement system when the environmental temperature where the sensing optical fiber is located is subjected to step change. The smaller the spatial resolution, the more robust the system is to the characterization of narrow length unit temperature discontinuities. In the distributed temperature measurement system based on the optical time domain reflection technology, the optical speed in the optical fiber is fixed, so that the signal position can be positioned according to the time delay of the received optical signal, and the distributed measurement can be realized. The index is limited by hardware parameters of the temperature measuring system, namely pulse light width emitted by the pulse light source, bandwidth of the photoelectric converter and sampling rate of the acquisition card. The former two determine the waveform shape collected by the collecting card. The sampling rate determines the spacing of the sampling points as known from OTDR techniques. The spatial resolution of the system is determined by the lower limits of the three. The prior literature discloses a plurality of optimization schemes for solving the problem of low spatial resolution of a system caused by too low sampling rate of an acquisition card, wherein a variable frequency sampling technology is utilized, the property that sampling point time intervals can be converted into space intervals in an OTDR principle is utilized, and a method for screening and reconstructing a group of high-precision data by changing sampling frequency of the acquisition card and further changing distribution density of the sampling points is designed. However, the frequency conversion sampling and reconstruction algorithm proposed at present is set subjectively on frequency selection, so that partial sampling redundancy is caused, the utilization rate of sampling points is low, the response time of a system is prolonged, a large amount of storage and calculation resources are occupied, an optimization algorithm aiming at frequency selection is required to be designed, and an evaluation index is designed for different frequency combinations to comprehensively consider the measurement precision and the response time. Disclosure of Invention The invention aims to provide a sampling optimization method of a high-spatial-resolution OTDR optical fiber temperature measurement system, which can automatically select sampling frequency combinations and corresponding sampling points thereof to realize higher-efficiency variable-frequency sampling. For this purpose, the invention adopts the following technical scheme: A sampling optimization method of a high-spatial-resolution OTDR optical fiber temperature measurement system is characterized in that an optimal frequency combination is found