Self-decoupling method for multi-characteristic spectrum inversion and ribbon-parallel fiber bragg grating sensing optical cable

Abstract

The invention provides a self-decoupling method for multi-characteristic spectrum inversion and a ribbon-parallel fiber grating sensing optical cable, and relates to the technical field of optical fiber sensing, and it needs to be explained that a grating reflection spectrum of the ribbon-parallel fiber grating sensing optical cable is obtained by determining a reference quantity according to the embodiment of each method; the method comprises the following steps: placing a ribbon-parallel fiber bragg grating sensing optical cable in a preset characteristic calibration state, collecting a reflection spectrum of the ribbon-parallel fiber bragg grating sensing optical cable, and traversing the reflection spectrum through a main peak global extremum search algorithm to obtain a central wavelength of the ribbon-parallel fiber bragg grating sensing optical cable; extracting a first-order left side lobe intensity and a first-order right side lobe intensity in symmetrical wavelength windows at two sides of the central wavelength by adopting a peak detection algorithm, and respectively calculating a wavelength reference variation, a left side lobe reference variation and a right side lobe reference variation according to the calibration reference quantity; and constructing a response characteristic calibration model so as to respectively solve the temperature variation and the strain variation of the ribbon fiber grating sensing optical cable. According to the invention, the decoupling precision and stability can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of optical fiber sensing, in particular to a self-decoupling method for multi-feature spectrum inversion and a band-type fiber grating sensing cable. BACKGROUND

[0002] Optical fiber grating sensors are widely used in structural health monitoring due to their advantages such as electromagnetic interference resistance and small size. In actual engineering, temperature and strain often exist simultaneously and are coupled with each other, and it is necessary to accurately separate the independent changes of the two. For example, the interference signal caused by temperature needs to be excluded for structural strain evaluation, and the interference signal of mechanical strain also needs to be avoided for temperature measurement. The traditional single FBG sensor only responds to environmental changes through a single wavelength shift, and cannot distinguish between temperature and strain effects, so it is usually necessary to additionally set up a separate temperature compensation unit, which increases the complexity of the system and makes it difficult to ensure the spatial consistency of the temperature / strain measurement points.

[0003] A multi-physical field strain sensing signal decoupling calibration method is disclosed in Chinese Patent No. CN117109465B, which establishes a characteristic response model between the measured pressure, temperature, strain and fiber grating sensor wavelength under the influence of multiple physical parameters by jointly arranging multiple sensors on a substrate, realizing fast response, high sensitivity and decoupling calibration of multiple physical parameters of fiber grating. According to the measured physical quantity, the influence of temperature, pressure or strain on the relationship model is considered respectively, and corresponding strain correction parameters, temperature correction parameters and pressure correction parameters are introduced, which improves the measurement and decoupling accuracy of the sensor. However, the above-mentioned scheme only relies on the multi-parameter decoupling of wavelength characteristics, which is prone to underdetermination or ill-conditioning when noise and drift exist, and the robustness is insufficient. Therefore, it is necessary to provide a self-decoupling method for multi-feature spectrum inversion and a band-type fiber grating sensing cable to improve the decoupling accuracy and stability. SUMMARY

[0004] Therefore, the present application provides a self-decoupling method for multi-feature spectrum inversion and a band-type fiber grating sensing cable. By utilizing the different sensitivity of the center wavelength and the left and right side lobes to different physical quantities, a multi-feature response model is constructed to suppress the cross-sensitivity under single feature, improve the distinguishability and separability of temperature and strain, and significantly reduce the decoupling error and drift.

[0005] The present application provides a self-decoupling method for multi-feature spectrum inversion, which comprises: Obtaining the grating reflection spectrum of the band-type fiber grating sensing cable and the calibration reference quantity; placing the co-axial fiber grating sensing cable in a preset characteristic calibration state, collecting a reflection spectrum corresponding to the co-axial fiber grating sensing cable, and traversing the reflection spectrum by using a global extremum search algorithm of a main peak to obtain a center wavelength of the co-axial fiber grating sensing cable; extracting first-order left and right side lobe intensities in a symmetric wavelength window on both sides of the center wavelength by using a peak detection algorithm, and calculating wavelength reference variation, left side lobe reference variation and right side lobe reference variation corresponding to the center wavelength, the first-order left side lobe intensity and the first-order right side lobe intensity respectively according to the calibration reference quantity; constructing a response characteristic calibration model according to the wavelength reference variation, the left side lobe reference variation and the right side lobe reference variation to solve temperature variation and strain variation of the co-axial fiber grating sensing cable respectively.

[0006] In the above technical solution, preferably, the placing of the co-axial fiber grating sensing cable in a preset characteristic calibration state specifically includes: placing the co-axial fiber grating sensing cable in a constant stress clamp, and grading the ambient temperature of the co-axial fiber grating sensing cable, changing the ambient temperature of the co-axial fiber grating sensing cable at a predetermined gradient in a state of keeping the strain of the co-axial fiber grating sensing cable as zero, and collecting the reflection spectrum of the co-axial fiber grating sensing cable.

[0007] In the above technical solution, preferably, the placing of the co-axial fiber grating sensing cable in a preset characteristic calibration state further includes: fixing the co-axial fiber grating sensing cable on a strain loading platform with constant temperature, applying an axial strain load to the co-axial fiber grating sensing cable, changing the strain value borne by the co-axial fiber grating sensing cable at a preset step, and collecting the reflection spectrum of the co-axial fiber grating sensing cable.

[0008] Further preferably, the extracting of the first-order left and right side lobe intensities specifically includes: performing a peak detection algorithm in a symmetric region on both sides of the center wavelength according to a wavelength offset reference of initial calibration, extracting the first-order left side lobe intensity in a first predefined wavelength interval based on the peak detection algorithm, and extracting the first-order right side lobe intensity in a second predefined wavelength interval based on the peak detection algorithm, and a range of the first predefined wavelength interval is smaller than a range of the second predefined wavelength interval.

[0009] Further preferably, the constructing the response characteristic calibration model according to the wavelength reference variation, the left side lobe reference variation and the right side lobe reference variation specifically comprises: constructing a central wavelength linear representation function based on the wavelength reference variation, the temperature sensitivity coefficient and the strain sensitivity coefficient; constructing a side lobe intensity response function based on the left side lobe reference variation and the right side lobe reference variation; constructing a response characteristic calibration model according to the central wavelength linear representation function and the side lobe intensity response function.

[0010] Further preferably, the calibration reference quantity comprises an initial central wavelength, an initial first-order left side lobe intensity and an initial first-order right side lobe intensity of the FBG sensing cable.

[0011] In a second aspect of the present application, a FBG sensing cable is provided for the self-decoupling method of the multi-feature spectrum inversion, the FBG sensing cable comprises a plurality of single-mode optical fibers arranged side by side, each single-mode optical fiber is coated with a coloring layer and a bonding layer in sequence, the bonding layer is used to form a strip structure on the surface of the single-mode optical fiber, and at least one single-mode optical fiber in the FBG sensing cable is engraved with a uniform FBG with a high-intensity side lobe, and the bonding layer is used to suppress fiber micro-bending and vibration.

[0012] Further preferably, the uniform FBG is a uniform refractive index modulation structure without apodization processing, and the reflection spectrum of the uniform FBG comprises a main peak reflection spectrum, a first-order left side lobe reflection spectrum and a first-order right side lobe reflection spectrum.

[0013] Further preferably, the bonding layer and the coloring layer are both UV-cured resin layers, the curing process of the single-mode optical fiber is carried out in an inert atmosphere, and the geometric size and resin distribution uniformity of the FBG sensing cable are controlled through a precision eye mold.

[0014] Further preferably, the FBG sensing cable is matched with a multi-channel FBG demodulator, and a plurality of gratings are arranged along the length direction of the FBG sensing cable to form a high-density sensing network.

[0015] The multi-feature spectrum inversion self-decoupling method and the FBG sensing cable provided by the present application have the following beneficial effects compared with the prior art: (1) By utilizing the sensitivity difference of the center wavelength and the left and right side lobes to different physical quantities, a multi-feature response model is constructed to suppress the cross-sensitivity under single feature, improve the distinguishability and separability of temperature and strain, compared with the method using only Bragg wavelength, the condition number is more optimal, the inversion is more stable, the decoupling error and drift are significantly reduced, and the global extremum search of the main peak is adopted to avoid local extremum misjudgment, enhance the main peak tracking stability under noise, spectral distortion and side mode interference, set a symmetric window on both sides of the center wavelength and perform peak detection to effectively suppress the influence of spectral asymmetry, spectral bottom fluctuation and optical path disturbance on side lobe extraction, and based on the feature construction and solution of the reference change quantity, the error accumulation caused by common mode intensity drift such as light source power fluctuation and insertion loss change is weakened, the response characteristic calibration model is established by calibrating the reference quantity, which can compensate for individual differences and ensure long-term online monitoring stability, the main peak search and side lobe extraction algorithm has low computational complexity and is easy to realize in real time in edge devices, which meets the dynamic working condition monitoring and is suitable for multi-point / multi-channel application scenarios of the optical fiber grating sensing cable, and has better tolerance and migration to spectral shape changes.

[0016] (2) The center wavelength change and the left and right side lobe intensity change are jointly included in the model to form a dual-feature constraint of frequency domain plus amplitude, which can significantly reduce the deviation of a single parameter caused by noise, spectral peak jitter or local fitting error, the center wavelength linear representation function is used to characterize the dominant linear response of temperature and strain, and the side lobe intensity response function is used to compensate the nonlinear effect of spectral shape, which improves the overall range accuracy and robustness, the side lobe intensity feature is more sensitive to spectral energy distribution changes, and it can be used as an indicator of zero point drift and coupling state change after being introduced, and it can form a self-checking constraint with the initial reference quantity to reduce the influence of long-term environmental drift on the results. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained without creative labor on the basis of these drawings.

[0018] Figure 1 The flowchart of the self-decoupling method of multi-feature spectrum inversion provided by the present application; Figure 2 The side lobe grating spectrum provided by the present application; Figure 3 The cross-sectional view of the band-pass optical fiber grating sensing cable provided by the present application; Figure 4 The grating center wavelength-temperature response diagram provided by the present application; Figure 5A grating center wavelength-strain response graph provided by the present application; Figure 6 A grating sidelobe intensity-temperature response graph provided by the present application; Figure 7 A grating sidelobe intensity-strain response graph provided by the present application. DETAILED DESCRIPTION

[0019] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0020] The present application discloses a method for predicting the excitation of a valve train of an internal combustion engine based on digital twinning, referring to Figure 1 The steps of the method include S1-S4.

[0021] In step S1, the grating reflection spectrum of the banded fiber grating sensing cable and the calibration reference quantity are obtained, and the calibration reference quantity includes the initial center wavelength, the initial first-order left sidelobe intensity and the initial first-order right sidelobe intensity of the banded fiber grating sensing cable.

[0022] In this step, the banded fiber grating sensing cable is a plurality of side-by-side single-mode optical fibers. Considering that a general fiber grating demodulator generally includes 16 channels, the number of ordinary optical fibers can be any number between 2 and 16, and a plurality of sidelobe gratings are written on any single-mode optical fiber. The outer layer of the optical fiber can be wrapped by a colored layer, and the outermost layer is a cured UV glue. The design of the temperature-strain self-decoupling banded fiber grating sensing cable is realized through four steps of sidelobe grating writing, fiber banded cable production, temperature-strain characteristic calibration and test verification.

[0023] In step S2, the banded fiber grating sensing cable is placed in a preset characteristic calibration state, the corresponding reflection spectrum of the banded fiber grating sensing cable is collected, and the reflection spectrum is traversed through a main peak global extreme value search algorithm to obtain the center wavelength of the banded fiber grating sensing cable.

[0024] In this step, steps S21-S22 are further included.

[0025] In step S21, the banded fiber grating sensing cable is placed in a constant stress clamp, and the ambient temperature of the banded fiber grating sensing cable is controlled in stages. The ambient temperature of the banded fiber grating sensing cable is changed at a predetermined gradient while keeping the strain of the banded fiber grating sensing cable at zero, and the reflection spectrum of the banded fiber grating sensing cable is collected.

[0026] In this step, the fiber grating is placed in a constant stress clamp, and the environmental temperature is controlled by a high-precision temperature control device. In the state of keeping the strain as zero, the temperature condition is changed according to a predetermined gradient, and the reflected spectrum is collected in real time. After each temperature stage is stable, the spectrum data is recorded, and the mapping relationship data set of the temperature variable and the spectrum characteristics is established. In the constant stress clamp, the strain is kept as zero, and only the environmental temperature is changed, so that the functional relationship between the grating center wavelength drift and the temperature change can be identified alone, and the temperature sensitivity coefficient can be accurately obtained. Through the step-by-step control and the predetermined gradient change, the full range of the working temperature zone can be covered, the linearity, hysteresis and repeatability can be tested, and if necessary, a quadratic or segmented model can be fitted to improve the fitting accuracy of the full range. After the pure temperature response is obtained, the temperature compensation for the strain measurement can be performed in the subsequent calculation, the pseudo-strain caused by the temperature drift is reduced, the robustness of the temperature-strain decoupling is improved, the main peak form of the grating reflection spectrum is more stable in the zero-strain state, the side lobe and spectrum width change can be quantified and modeled, which is beneficial to the optimization of the center wavelength extraction algorithm and the reduction of the measurement noise.

[0027] In step S22, the FBG sensing cable is fixed on the strain loading platform with constant temperature, and an axial strain load is applied to the FBG sensing cable. The strain value borne by the FBG sensing cable is changed in a preset step, and the reflected spectrum of the FBG sensing cable is collected.

[0028] In this step, the fiber grating is fixed on the strain loading platform with constant temperature, and an axial strain load is applied by a precision mechanical device. The strain value is changed in a preset step, and the reflected spectrum is collected after the system is stable. This process constructs a corresponding relationship data set of the strain variable and the spectrum characteristics. On the constant temperature platform, the axial strain is loaded in a preset step, a quantitative relationship between the center wavelength drift and the strain is established, the strain sensitivity coefficient is accurately obtained, and through the step-by-step loading / unloading curve, the linearity zone, hysteresis characteristics and time-related effects of the strain response are evaluated to provide parameters for historical compensation and filtering in the sensing algorithm. Under the stress condition, the reflection spectrum may be broadened, tilted or multi-peak. The calibration can establish the correlation characteristics of the spectrum shape and the strain to improve the calculation stability under the conditions of noise, micro-bending or multi-mode coupling. The different grating regions in the FBG sensing cable are loaded and calibrated respectively to compensate for the inconsistent responses caused by manufacturing differences and uneven bonding, and the measurement consistency of the whole link is improved.

[0029] Further, the reflected spectrum S t ( λ ) is the change of the light signal intensity with the light wavelength. The center wavelength and the side lobe intensity can be obtained by the following methods: performing a global extremum search to determine the center wavelength λ B,t, the peak value position of the reflection spectrum is determined by a global extreme value search algorithm, and the wavelength value corresponding to the point is the center wavelength. In the implementation process, an adaptive threshold is used to identify the main lobe range, and the sidelobe interference step is excluded.

[0030] S3, in the symmetric wavelength window on both sides of the center wavelength, the first order left sidelobe intensity and the first order right sidelobe intensity are extracted by using the peak value detection algorithm, and the wavelength reference variation, the left sidelobe reference variation and the right sidelobe reference variation corresponding to the center wavelength, the first order left sidelobe intensity and the first order right sidelobe intensity are calculated according to the calibration reference quantity.

[0031] In this step, according to the wavelength offset reference of initial calibration, the peak value detection algorithm is carried out in the symmetric region on both sides of the center wavelength, the first order left sidelobe intensity is extracted in the first predefined wavelength interval based on the peak value detection algorithm, and the first order right sidelobe intensity is extracted in the second predefined wavelength interval based on the peak value detection algorithm, and the range of the first predefined wavelength interval is less than the range of the second predefined wavelength interval.

[0032] In one example, according to the wavelength offset reference of initial calibration, local extreme value detection is carried out in the symmetric region on both sides of the center wavelength, the left sidelobe intensity is extracted in the predefined wavelength interval λ B,t - a, λ B,t - b ] based on the peak value detection algorithm I L,t , the right sidelobe intensity is extracted in the predefined wavelength interval λ B,t + a, λ B,t + b ] based on the peak value detection algorithm I R,t , the extraction process uses the peak value detection algorithm, and a dynamic threshold is set to exclude false peak interference.

[0033] Step S4, according to the wavelength reference variation, the left sidelobe reference variation and the right sidelobe reference variation, a response characteristic calibration model is constructed to solve the temperature variation and strain variation of the FBG sensing cable.

[0034] In this step, steps S41-S43 are also included.

[0035] Step S41, based on the wavelength reference variation, the temperature sensitivity coefficient and the strain sensitivity coefficient, a center wavelength linear representation function is constructed.

[0036] In this step, a center wavelength linear representation model is established, that is, a quantitative relationship between the center wavelength offset and the physical quantity change is established: Δ λB = α· Δ T + β· Δ ε wherein, α denotes the temperature sensitivity coefficient, β denotes the strain sensitivity coefficient, both of which are scalar constants. The center wavelength linear representation model follows the linear superposition principle and represents the coupling effect of the two physical quantities on the center wavelength.

[0037] In step S42, a sidelobe intensity response function is constructed based on the left sidelobe reference change amount and the right sidelobe reference change amount.

[0038] In this step, a sidelobe intensity response function model is established, that is, a nonlinear response function of the bilateral sidelobe intensity is constructed:

[0039] wherein, f L denotes a left sidelobe intensity continuous function with temperature and strain as independent variables, f R denotes a right sidelobe intensity continuous function with temperature and strain as independent variables, both of which represent the nonlinear response characteristics of the sidelobe intensity to the combined environmental changes.

[0040] In step S43, a sidelobe intensity response function is constructed based on the left sidelobe reference change amount and the right sidelobe reference change amount.

[0041] In this embodiment, the center wavelength change and the left and right sidelobe intensity changes are jointly included in the model to form a dual-feature constraint of frequency domain plus amplitude, which can significantly reduce the deviation of a single parameter caused by noise, spectral peak jitter or local fitting error. The center wavelength linear representation function is used to depict the dominant linear response of temperature and strain, and the sidelobe intensity response function is used to compensate the nonlinear effect of the spectral shape, thereby improving the overall accuracy and robustness. The sidelobe intensity feature is more sensitive to changes in spectral energy distribution, and can be introduced as an indicator of zero-point drift and coupled state change, which, together with the initial reference quantity, forms a self-checking constraint to reduce the impact of long-term environmental drift on the results. The linear response of the center wavelength to temperature and strain provides the main solution, and the left and right sidelobes are sensitive to non-uniform stress fields and bending. Joint modeling of the two can distinguish between center drift caused by uniform load and spectral shape changes caused by non-uniform load, thereby improving the temperature / strain decoupling accuracy. The differential characteristics of the left and right sidelobes can sensitively reflect abnormal working conditions such as eccentric stress, local bonding mismatch, and bending radius change, providing online health monitoring and data credibility rating capabilities. When the center peak is affected by noise, shielding or multi-peak, the sidelobe intensity response function can provide alternative or correction information to reduce the risk of single-point failure.

[0042] Further, in the calibration experiment, the change value Δ T of the strain ε jointly affects the change value Δ λ of the center wavelength B and the side lobe intensity Δ I L , Δ I R , the corresponding response calibration data sets are formed by using the test data of each group of experiments.

[0043] The center wavelength coefficient calibration is based on the temperature response calibration data set {Δ T i , Δ λ B,i} and the strain response calibration data set {Δ ε j , Δ λ B,j}, and the temperature sensitivity coefficient and the strain sensitivity coefficient are solved by multivariate least squares method:

[0044] wherein, X represents a design matrix, X T represents the transpose of the design matrix, Y represents an observation vector, Y = Δ λ B .

[0045] The side lobe response calibration is based on the temperature-strain joint calibration data set {Δ T k , Δ ε k , Δ I L,k , Δ I R,k} and is constructed by the following steps: The temperature and strain of the calibration data are combined to define a two-dimensional independent variable space coordinate point P k = (Δ T k , Δ ε k ); based on P k and the side lobe intensity change value Δ I L,k and Δ I R,kThe scattered data points set is constructed; the triangular interpolation algorithm is used to generate continuous surface in two-dimensional space; and the response value prediction of any coordinate point is realized through segmented polynomial function.

[0046] In one example, the current reflection spectrum of the grating in the optical cable is acquired by the optical fiber demodulator S t ( λ ), and the initial moment λ B,0 、 I L,0 and I R,0 are recorded in the manner mentioned above λ B,t 、 I L,t and I R,t are read in real time at the subsequent moment t. The characteristic change amount is calculated relative to the initial reference state as follows: Δ λ B = λ B,t - λ B,0 Δ I L = I L,t - I L,0 Δ I R = I R,t - I R,0 A constraint system containing three equations is established, the central wavelength constraint is a linearly coupled equation, and the expression is as follows: Δ λ B = α· Δ T + β· Δ ε The left side lobe response constraint is a nonlinear equation, and the expression is as follows:

[0047] Wherein, f L ( ) represents the binary interpolation function calibrated and constructed.

[0048] The right side lobe response constraint is a nonlinear equation, and the expression is as follows:

[0049] in, f R ( ) represents the binary interpolation function constructed by calibration.

[0050] Construct a residual function and define a vector-valued function F(x) to characterize the deviation of the equation:

[0051] The parameter optimization solution framework includes initial estimation settings, iterative solution mechanisms, and convergence criteria. The initial estimation setup includes setting the vector of parameters to be solved. initial value x (0) ; The iterative solution mechanism package employs a numerical optimization method based on the derivative of the objective function, approximating the optimal solution through an iterative update strategy:

[0052] in, Φ Indicates the optimization operator, H represents the gradient of the objective function, and H represents the Hessian matrix or its approximation. The convergence criteria include terminating the iterative process when any of the following sets of conditions are met: Condition group I: ; Condition group II: ; Condition group III: ; in, δ 1 indicates the first preset precision threshold. δ 2 indicates the second preset precision threshold. δ 3 indicates the third preset precision threshold.

[0053] The derivative information construction mechanism includes selectively constructing the following mathematical features based on the type of optimization method: Gradient vector: objective function The first derivative; Curvature information: including but not limited to: Jacobian matrix Hessian matrix and quasi-Newton approximation matrix B k .

[0054] The parameter update rule implements parameter updates through a numerical optimization algorithm, and its general form is expressed as:

[0055] Wherein, Γ represents an update operator defined by an optimization algorithm, and represents a parameter space update operation.

[0056] When the convergence condition is met, the temperature change amount is Δ T out =x 1, the strain change amount is Δ ε out =x 2.

[0057] The decoupling accuracy depends on the stabilization effect of the sideband structure on the side lobe signal. If a single fiber is independently packaged, the side lobe intensity is easily affected by micro-bending disturbance, resulting in deviation of the decoupling result from the true value.

[0058] In this embodiment, by utilizing the different sensitivities of the center wavelength and left and right side lobes to different physical quantities, a multi-feature response model is constructed to suppress cross-sensitivity under single feature and improve the distinguishability and separability of temperature and strain. Compared with the method of using only the Bragg wavelength, the condition number is more optimal and the inversion is more stable, significantly reducing the decoupling error and drift. The main peak global extreme value search is adopted to avoid local extreme value misjudgment and enhance the stability of main peak tracking under noise, spectral distortion and side mode interference. The symmetric window is set on both sides of the center wavelength and the peak value is detected to effectively suppress the influence of spectral asymmetry, spectral bottom fluctuation and optical path disturbance on side lobe extraction. At the same time, based on the feature construction and solution of the reference change amount, the error accumulation caused by common mode intensity drift such as light source power fluctuation and insertion loss change is weakened. The response characteristic calibration model is established by calibrating the reference quantity, which can compensate for individual differences and ensure long-term online monitoring stability. The main peak search and side lobe extraction algorithm has low computational complexity and is easy to realize in real time in edge devices, meeting the dynamic working condition monitoring and being suitable for multi-point / multi-channel application scenarios of the sideband type fiber grating sensing optical cable. It has better tolerance and migratability to spectral shape changes.

[0059] In one example, a side lobe grating is written on a fiber according to the side lobe grating writing requirement, and the spectrum of the written side lobe is read by a spectrometer. The result is shown in Figure 2 . Wherein, 1 is the reflection center peak, 2 is the first left side lobe, and 3 is the first right side lobe. The fiber with the written side lobe grating and the remaining 15 ordinary fibers are completed according to the production process of the fiber ribbon cable to complete the production of the fiber sideband. The result is shown in Figure 3 . Figure 3 , wherein 4 is a single mode fiber, and 5 is a fixed mold.

[0060] The calibration test is completed on the built platform. In the case of ensuring that the strain is 0, the temperature is set from 30℃ to 100℃ with a gradient of every 5℃, a total of 15 groups of tests. In the case of ensuring that the temperature is 30°, the strain is set from 0 to 1000με with a gradient of every 50με, a total of 21 groups of tests. The initial reference state is: λ 0 = 1558.748 nm; I L,0 = -44.871 dBm; I R,0 = -49.419 dBm.

[0061] The relationship between the center wavelength and temperature and strain was studied using the read spectrum data, and the response curve was obtained, as shown in Figure 4 , Figure 5 The wavelength temperature response formula is: Δ λ B = 0.010 x Δ T The wavelength strain response formula is: Δ λ B = 0.00118 x Δ ε The relationship between the reflection intensity of the two side lobes and temperature and strain was studied using the read spectrum data, and the response curve was obtained, as shown in Figure 6 , Figure 7 The wavelength temperature response formula is:

[0062] The wavelength temperature response formula is:

[0063] T = 70℃, and ε= 200 when the micro-strain is measured: λ B,t = 1559.385 nm; I L,t = -47.643 dBm; I R,t = -47.903 dBm.

[0064] The relative change amount was calculated: Δ λ B = λ B,t - λ B,0 = 0.637 nm Δ I L = I L,t - I L,0 = -2.772 dBm Δ IR = I R,t - I R,0 =1.516dBm The decoupling equation set is constructed by using the relative variation amount and the response of the center wavelength and the sidelobe intensity.

[0065] The center wavelength constraint is 0.010 x Δ T+ 0.00118 x Δ ε= 0.637.

[0066] The left sidelobe constraint is:

[0067] The right sidelobe constraint is:

[0068] The residual function is further constructed,

[0069] and the initial guess is set: Δ T 0 = 0℃, Δ ε 0 = 0 με .

[0070] The Levenberg-Marquardt (L-M) optimization algorithm is adopted, and the parameter update rule is:

[0071] wherein, λ k is an adaptive damping parameter, λ 0=0.001 When the residual decreases, the adjustment strategy is: λ k+1 =0.5 λ k , and the Gauss-Newton characteristic is enhanced; when the residual increases, λ k+1 =2 λ k , the gradient descent characteristic is enhanced.

[0072] The convergence criterion is: the parameter variation amount threshold is =0.0001, the residual norm threshold is =0.001, and the maximum iteration number is k max = 50.

[0073] Please refer to Figure 6 ,Figure 6 The response relationship of the side lobe intensity of the fiber Bragg grating to the temperature change under the condition of zero strain (ε = 0) is shown. The right side lobe tends to gradually increase in intensity from about -49 dBm to about -47 dBm as the temperature rises. The left side lobe tends to gradually decrease in intensity from about -45.5 dBm to about -49 dBm as the temperature rises. The left and right side lobe intensities have significant and opposite linear response characteristics to temperature, which can be used for temperature-sensitive spectral feature extraction and model decoupling.

[0074] Please refer to Figure 7 , Figure 7 The response relationship of the left and right side lobe intensities of the fiber Bragg grating to the axial strain change under the condition of constant temperature T = 30℃ is shown. Under the same constant temperature, the right side lobe intensity increases with the increase of strain, and the left side lobe intensity decreases with the increase of strain. Both of the two fitting curves show a slight quadratic effect. The right side lobe intensity is in the range of -46 to -42 dBm, and the left side lobe intensity is in the range of -52 to -48 dBm, which are clearly separated in value. The responses of the left and right side lobes to strain have significant and opposite trends, which can be combined with the linear response of the center wavelength for strain identification and temperature decoupling. In a larger strain range, the quadratic or segmented model can improve the calculation accuracy and robustness better than the simple linear model.

[0075] The L-M algorithm uses the Jacobian matrix as the core curvature information:

[0076] The specific update process of the L-M algorithm is initialization x (0) = [0,0] T , λ= 0.001.

[0077] The residual error calculation is represented as .

[0078] The residual norm is represented as .

[0079] The Jacobian matrix is represented as .

[0080] The search direction is calculated as .

[0081] The new point to be evaluated is represented as .

[0082] The gain ratio is calculated as .

[0083] If the gain satisfies 0.25 ≤ ρ≤0.75, the new measurement point is updated, the current iteration is satisfied, and the subsequent evaluation point is updated.

[0084] If not satisfied, the damping parameter is updated, and when ρ> 0.75, λ k+1 =0.5 λ k , the Gauss-Newton characteristic is enhanced when ρ< 0.25, λ k+1 =2 λ k , the gradient descent characteristic is enhanced, and is recalculated.

[0085] When the convergence condition is satisfied, the temperature change amount Δ T out =39.8645, and the strain change amount Δ ε out =194.6725.

[0086] It is obtained that T out =69.8645, ε out =194.6725, which is close to the true values T =70.0 and ε =200.

[0087] Based on the above method, the application further discloses a ribbon-type fiber grating sensing optical cable. The ribbon-type fiber grating sensing optical cable comprises a plurality of single-mode optical fibers arranged side by side. The single-mode optical fibers are coated with a coloring layer and a bonding layer in sequence. The bonding layer is used to form a ribbon structure on the surface of the single-mode optical fiber. At least one single-mode optical fiber in the ribbon-type fiber grating sensing optical cable is engraved with a uniform fiber grating with reserved high-intensity side lobes. The bonding layer is used to suppress fiber micro-bending and vibration.

[0088] The bonding layer and the coloring layer are both UV-cured resin layers. The curing process of the single-mode optical fiber is carried out in an inert atmosphere. The geometric size and resin distribution uniformity of the ribbon-type fiber grating sensing optical cable are controlled through a precision eye mold.

[0089] The bonding layer and the coloring layer are both UV-cured resin layers. The curing process of the single-mode optical fiber is carried out in an inert atmosphere. The geometric size and resin distribution uniformity of the ribbon-type fiber grating sensing optical cable are controlled through a precision eye mold.

[0090] In this embodiment, the cured UV glue layer of the ribbon fiber grating sensing cable ensures the measurement stability of the grating sidelobe intensity by suppressing fiber micro-bend vibration, which is the key support for decoupling method engineering. The specific steps of the production process of the ribbon fiber grating sensing cable are as follows: Single fiber coloring process: First, the qualified bare fiber is subjected to coloring process, and a UV cured resin layer with a specific color is coated on its surface to form a colored fiber with distinguishable identification and stable mechanical properties. After completion, it is wound and stored in a clean and dry environment.

[0091] Coloring fiber pay-off and tension control: Load a predetermined number of colored fiber discs into the pay-off system, and realize accurate tension control of each fiber through independent pay-off unit and tension adjusting mechanism. After convergence through the guiding device, perform the first electrostatic elimination to remove surface charges and prevent dust adsorption.

[0092] Fiber precise positioning and secondary static elimination: Make the electrostatic eliminated fiber bundle pass through the branching positioning device to realize the parallel and uniform arrangement and spacing control of each fiber. Then, perform the second electrostatic elimination treatment to ensure the arrangement is flat and stable for subsequent bonding.

[0093] Bonding resin coating and molding control: Uniformly coat the UV cured bonding resin on the arranged fiber bundle, and use the coating system containing precise eye mold to control the fiber ribbon geometric size and resin layer distribution uniformity.

[0094] Cable curing and atmosphere control: Make the resin coated fiber bundle enter the UV curing system to realize complete resin curing under specific wavelength ultraviolet light and inert atmosphere, forming a predetermined core number ribbon structure; the pulling speed and curing conditions are matched, and the tension is controlled according to the core number.

[0095] Cured ribbon cable transportation and permanent marking: Transport the cured fiber ribbon through the transmission device to maintain the structural integrity, and apply permanent marking containing model, specification or batch information on its surface to realize product traceability.

[0096] Tension optimization and finished product winding: Optimize the winding process through dynamic tension adjusting device, and finally wind the formed fiber ribbon neatly into a reel. The whole process is coordinated and controlled by automatic system to ensure quality consistency.

[0097] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and / or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof. As used herein, the term "if' can be construed to mean "when" or "upon" or "in response to the happening of," unless explicitly indicated to the contrary. That is, if A, B and C are intended to convey if A, B, and C exist or occur, then "if A, B and C" is intended to mean A, B and C when A, B and C exist or occur, or upon the happening of A, B and C, or in response to the happening of A, B and C.

[0098] The preferred embodiments of the present application have been described above with the intent to be illustrative only and not limiting of the application as defined by the appended claims. Changes in form and detail can be made by those skilled in the art, as certain of the requirements may dictate, without departing from the spirit and scope of the application.

Claims

1. A self-decoupling method for multi-feature spectral inversion, characterized in that, The method includes: Acquire and measure the grating reflection spectrum and calibration reference value of the ribbon fiber Bragg grating sensing cable; The parallel-band fiber grating sensing cable is placed in a preset characteristic calibration state, the reflection spectrum corresponding to the parallel-band fiber grating sensing cable is collected, and the reflection spectrum is traversed by the main peak global extremum search algorithm to obtain the center wavelength of the parallel-band fiber grating sensing cable. Within the symmetrical wavelength windows on both sides of the center wavelength, the peak detection algorithm is used to extract the intensity of the first-order left sidelobe and the intensity of the first-order right sidelobe. Based on the calibration reference values, the wavelength reference change, left sidelobe reference change and right sidelobe reference change corresponding to the center wavelength, the intensity of the first-order left sidelobe and the intensity of the first-order right sidelobe are calculated respectively. Based on the wavelength reference change, the left sidelobe reference change, and the right sidelobe reference change, a response characteristic calibration model is constructed to solve for the temperature change and strain change of the parallel-band fiber optic grating sensing cable, respectively.

2. The self-decoupling method for multi-feature spectral inversion as described in claim 1, characterized in that, The step of placing the parallel fiber Bragg grating sensing cable in a preset characteristic calibration state specifically includes: The parallel-band fiber Bragg grating sensing cable is placed in a constant stress fixture, and the ambient temperature of the parallel-band fiber Bragg grating sensing cable is controlled in stages. While keeping the strain of the parallel-band fiber Bragg grating sensing cable at zero, the ambient temperature of the parallel-band fiber Bragg grating sensing cable is changed at a predetermined gradient, and the reflection spectrum of the parallel-band fiber Bragg grating sensing cable is collected.

3. The self-decoupling method for multi-feature spectral inversion as described in claim 1, characterized in that, The step of placing the parallel-band fiber Bragg grating sensing cable in a preset characteristic calibration state further includes: The parallel-band fiber Bragg grating sensing cable is fixed on a strain loading platform with constant temperature. An axial strain load is applied to the parallel-band fiber Bragg grating sensing cable, and the strain value borne by the parallel-band fiber Bragg grating sensing cable is changed in preset step size. The reflection spectrum of the parallel-band fiber Bragg grating sensing cable is collected.

4. The self-decoupling method for multi-feature spectral inversion as described in claim 1, characterized in that, The extraction of the first-order left sidelobe intensity and the first-order right sidelobe intensity specifically includes: Based on the initially calibrated wavelength offset reference, a peak detection algorithm is performed in the symmetrical regions on both sides of the center wavelength. The first-order left sidelobe intensity is extracted in the first predefined wavelength range based on the peak detection algorithm, and the first-order right sidelobe intensity is extracted in the second predefined wavelength range based on the peak detection algorithm. The range of the first predefined wavelength range is smaller than the range of the second predefined wavelength range.

5. The self-decoupling method for multi-feature spectral inversion as described in claim 1, characterized in that, The step of constructing a response characteristic calibration model based on the wavelength reference change, the left sidelobe reference change, and the right sidelobe reference change specifically includes: Based on the wavelength reference change, temperature sensitivity coefficient, and strain sensitivity coefficient, a linear characterization function for the center wavelength is constructed. Based on the left sidelobe reference change and the right sidelobe reference change, a sidelobe intensity response function is constructed. A response characteristic calibration model is constructed based on the center wavelength linear characterization function and the sidelobe intensity response function.

6. The self-decoupling method for multi-feature spectral inversion as described in claim 1, characterized in that, The calibration reference quantities include the initial center wavelength, the initial first-order left sidelobe intensity, and the initial first-order right sidelobe intensity of the parallel-band fiber grating sensing optical cable.

7. A parallel-band fiber Bragg grating sensing optical cable, characterized in that, For implementing the self-decoupling method of multi-feature spectral inversion according to any one of claims 1 to 6, the parallel-band fiber grating sensing optical cable includes multiple single-mode optical fibers arranged side by side, each single-mode optical fiber is coated with a coloring layer and an adhesive layer in sequence, the adhesive layer is used to form a strip structure on the surface of the single-mode optical fiber, at least one single-mode optical fiber in the parallel-band fiber grating sensing optical cable is inscribed with a uniform fiber grating that retains high-intensity sidelobes, and the adhesive layer is used to suppress fiber microbending and vibration.

8. A parallel-band fiber Bragg grating sensing optical cable as described in claim 7, characterized in that, The uniform fiber grating is a uniform refractive index modulation structure without apodization, and the reflection spectrum of the uniform fiber grating includes the main peak reflection spectrum, the first-order left sidelobe reflection spectrum, and the first-order right sidelobe reflection spectrum.

9. A parallel-band fiber Bragg grating sensing optical cable as described in claim 7, characterized in that, Both the adhesive layer and the coloring layer are UV-curable resin layers. The curing process of the single-mode optical fiber is carried out in an inert atmosphere, and the geometric dimensions and resin distribution uniformity of the parallel-band fiber grating sensing cable are controlled by a precision eye mold.

10. A parallel-band fiber Bragg grating sensing optical cable as described in claim 7, characterized in that, The parallel-band fiber Bragg grating sensing cable is matched with a multi-channel fiber Bragg grating demodulator, and multiple gratings are arranged along the length of the parallel-band fiber Bragg grating sensing cable to form a high-density sensing network.

Images