Method and device for predicting parameters of fiber bragg grating based on SVR (Support Vector Register), and computer equipment

Through the prediction model based on SVR, the fiber grating sensitivity characteristic curve training model is used to perform nonlinear prediction of temperature and refractive index, which solves the problem of large measurement error of fiber grating under nonlinear sensitivity characteristics, and realizes accurate measurement of multiple parameters.

CN118898023APending Publication Date: 2024-11-05CHANGZHOU XINGYU AUTOMOTIVE LIGHTING SYST CO LTD

Patent Information

Application Number
CN202410972583.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-19
Publication Date
2024-11-05

AI Technical Summary

Technical Problem

When the sensitivity characteristic curve of fiber grating is nonlinear during multi-parameter measurement, it is difficult for traditional methods to achieve accurate measurement and easily introduce large measurement errors.

Method used

Using a prediction model based on support vector regressor (SVR), the temperature and refractive index SVR model are trained through data on the optical fiber grating sensitivity characteristic curve, and the nonlinear prediction of temperature and refractive index is used to use the resonant peak change.

Benefits of technology

The fiber grating is able to measure multiple environmental parameters stably and accurately, reduce measurement errors, and ensure the accuracy of the fiber grating sensor.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118898023A_ABST
    Figure CN118898023A_ABST
Patent Text Reader

Abstract

The embodiment of the invention discloses an SVR (Support Vector Register)-based fiber bragg grating parameter prediction method and device and computer equipment. The method comprises the following steps: acquiring harmonic peak variation of a fiber bragg grating spectrum; inputting the harmonic peak variation into a prediction model for temperature and refractive index prediction to obtain a prediction result; and outputting the prediction result. Wherein the prediction model is obtained by taking data on a fiber grating sensitivity characteristic curve as a sample set to train a support vector regression machine. By implementing the method provided by the embodiment of the invention, the problem that when the fiber bragg grating realizes multi-parameter measurement, the measurement cannot be realized or the error is large under the condition that the sensitivity characteristic curve is nonlinear can be solved, and the simultaneous measurement of a plurality of environment parameters based on the fiber bragg grating can be realized; therefore, the fiber bragg grating sensor can stably and accurately provide required environmental parameter measurement results.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of methods for predicting fiber Bragg grating parameters, and in particular to a method, device and computer equipment for predicting fiber Bragg grating parameters based on SVR. Background Art

[0002] Fiber Bragg grating is a sensor technology that uses changes in the frequency spectrum of light to detect environmental parameters. It uses the interaction between the periodic structure in the optical fiber and the physical parameters of the external environment, such as temperature, refractive index, pressure, stress, etc., and measures the size of these parameters in real time by observing changes in the spectrum. Fiber Bragg grating has multiple resonance peaks, each of which corresponds to a different environmental parameter. By sending out a light signal through the fiber Bragg grating, a spectrum analyzer is used to analyze the returned spectrum, which can accurately measure the position of the resonance peak or the frequency shift, thereby inferring the change in the environmental parameter.

[0003] However, fiber Bragg gratings often face challenges of nonlinearity or low linearity when measuring sensitivity characteristics, which means that changes in environmental parameters may not affect the resonance peak of the grating at a constant rate, but are reflected in the spectrum in a complex nonlinear manner. Specifically, the resonance peak position of the fiber Bragg grating may not be a simple linear function, but may show a complex nonlinear relationship with changes in environmental parameters. For example, when the temperature changes, the resonance peak frequency of the grating may show a nonlinear response, which makes it more complicated to accurately calculate changes in environmental parameters. The traditional sensitivity matrix method assumes that the sensitivity is linear and can simply infer the values ​​of multiple environmental parameters through changes in multiple resonance peaks. However, since the actual sensitivity curve may be nonlinear or have cross-influences, this method may introduce large measurement errors or fail to accurately infer multiple parameters.

[0004] Therefore, it is necessary to design a new method to solve the problem that the fiber Bragg grating cannot be measured or the error is large when the sensitivity characteristic curve is nonlinear when realizing multi-parameter measurement, and to realize the simultaneous measurement of multiple environmental parameters based on the fiber Bragg grating to ensure that the fiber Bragg grating sensor can stably and accurately provide the required environmental parameter measurement results. Summary of the invention

[0005] The purpose of the present invention is to overcome the defects of the prior art and provide a method, device and computer equipment for predicting fiber Bragg grating parameters based on SVR.

[0006] To achieve the above object, the present invention adopts the following technical solution: a method for predicting fiber Bragg grating parameters based on SVR, comprising:

[0007] Obtaining the resonance peak change of the fiber Bragg grating spectrum;

[0008] Inputting the resonance peak variation into a prediction model to predict temperature and refractive index to obtain a prediction result;

[0009] Outputting the prediction result;

[0010] The prediction model is obtained by training a support vector regression machine using data on a fiber Bragg grating sensitivity characteristic curve as a sample set.

[0011] A further technical solution is: the prediction model includes a temperature SVR model and a refractive index SVR model; wherein the temperature SVR model is used to predict temperature; and the refractive index SVR model is used to predict refractive index.

[0012] A further technical solution is that the prediction model is obtained by training a support vector regression machine using data on the fiber Bragg grating sensitivity characteristic curve as a sample set, and includes:

[0013] Obtain the resonance peak changes corresponding to different temperatures and different refractive indices in the sensitivity characteristic curve of the fiber Bragg grating to obtain a sample set;

[0014] Construct temperature SVR model and refractive index SVR model;

[0015] The cross-validation method is used to select the parameters of the temperature SVR model and the refractive index SVR model, and the parameters of the temperature SVR model and the refractive index SVR model are trained in combination with the sample set to obtain the prediction model.

[0016] A further technical solution is that the temperature SVR model and the refractive index SVR model are ε-SVR models respectively.

[0017] A further technical solution is that the change in the resonance peak of the fiber grating spectrum includes the change in the resonance peak and the reference position based on the normal temperature and the refractive index of air.

[0018] The present invention also provides a device for predicting fiber Bragg grating parameters based on SVR, comprising:

[0019] An acquisition unit, used for acquiring a resonance peak variation of a fiber grating spectrum;

[0020] A prediction unit, used for inputting the resonance peak variation into a prediction model to perform temperature and refractive index prediction to obtain a prediction result;

[0021] An output unit is used to output the prediction result.

[0022] Its further technical solution is: also includes:

[0023] The model generation unit is used to train a support vector regression machine using data on the fiber Bragg grating sensitivity characteristic curve as a sample set to obtain a prediction model.

[0024] Its further technical solution is: the model generation unit includes:

[0025] The sample set acquisition subunit is used to acquire the resonance peak change corresponding to different temperatures and different refractive indices in the fiber Bragg grating sensitivity characteristic curve to obtain a sample set;

[0026] A model building subunit, used to build a temperature SVR model and a refractive index SVR model;

[0027] The training subunit is used to select parameters of the temperature SVR model and the refractive index SVR model by using a cross-validation method, and train the parameters of the temperature SVR model and the refractive index SVR model in combination with the sample set to obtain a prediction model.

[0028] The present invention further provides a computer device, which includes a memory and a processor. The memory stores a computer program, and the processor implements the above method when executing the computer program.

[0029] Compared with the prior art, the present invention has the following beneficial effects: the present invention predicts the temperature and refractive index by using the change in the fiber Bragg grating resonance peak by using the SVR model, thereby solving the problem that the fiber Bragg grating cannot be measured or has a large error when the sensitivity characteristic curve is nonlinear when measuring multiple parameters, and realizes the simultaneous measurement of multiple environmental parameters based on the fiber Bragg grating, so as to ensure that the fiber Bragg grating sensor can stably and accurately provide the required environmental parameter measurement results.

[0030] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying any creative work.

[0032] Figure 1 A schematic diagram of a method for predicting fiber Bragg grating parameters based on SVR provided in an embodiment of the present invention;

[0033] Figure 2 Schematic diagram of the effect of the refractive index and temperature change of the external environment on the fiber Bragg grating spectrum provided by the embodiment of the present invention Figure 1 ;

[0034] Figure 3 Schematic diagram of the effect of the refractive index and temperature change of the external environment on the fiber Bragg grating spectrum provided by the embodiment of the present invention Figure 2 ;

[0035] Figure 4 A schematic diagram of a sensitivity characteristic curve of the refractive index and temperature of the external environment to the resonance peak dip1 in the spectrum provided by an embodiment of the present invention;

[0036] Figure 5 A schematic diagram of a sensitivity characteristic curve of the refractive index and temperature of the external environment to the resonance peak dip2 in the spectrum provided by an embodiment of the present invention;

[0037] Figure 6 A schematic diagram of the error between the predicted result and the actual measured result of the temperature at different refractive indices provided by an embodiment of the present invention;

[0038] Figure 7 A schematic diagram of the error between the model prediction result and the actual measurement result of the refractive index at different temperatures provided by an embodiment of the present invention;

[0039] Figure 8 A schematic diagram of constructing a hyperplane for SVR provided in an embodiment of the present invention;

[0040] Fig. 9 A schematic diagram of a nonlinear SVR regression structure provided by an embodiment of the present invention;

[0041] Fig.10 A schematic block diagram of a device for predicting fiber Bragg grating parameters based on SVR provided by an embodiment of the present invention;

[0042] Fig.11 A schematic block diagram of a computer device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] It should be understood that when used in this specification and the appended claims, the terms "include" and "comprises" indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0045] It should also be understood that the terms used in this specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0046] It should be further understood that the term "and / or" used in the present description and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0047] See also Figure 1 , Figure 1 A schematic flow chart of a method for predicting fiber Bragg grating parameters based on SVR provided in an embodiment of the present invention. The method for predicting fiber Bragg grating parameters based on SVR is applied to a device with a fiber Bragg grating, and by introducing a support vector regression machine, it solves the problem that the fiber Bragg grating cannot be measured or has a large error when the sensitivity characteristic curve is nonlinear when realizing multi-parameter measurement.

[0048] Figure 1 FIG. 1 is a flow chart of a method for predicting fiber Bragg grating parameters based on SVR provided by an embodiment of the present invention. Figure 1 As shown, the method includes the following steps S110 to S130.

[0049] S110, obtaining a resonance peak change of the fiber Bragg grating spectrum.

[0050] In this embodiment, the resonance peak variation of the fiber Bragg grating spectrum includes the variation of the resonance peak and the reference position based on the normal temperature and the refractive index of air.

[0051] Fiber Bragg gratings usually show multiple resonance peaks in the spectrum, and the positions of these peaks can change with changes in environmental parameters such as temperature and refractive index. When the temperature and refractive index change at the same time, the position of each resonance peak on the spectrum will move accordingly.

[0052] In experimental observation, we can record the relationship between the position of these resonance peaks and the change of environmental parameters and describe this relationship with a mathematical model. The relationship between the influence of temperature and refractive index on the spectrum can be expressed as Δλ dip2 =K T2 ΔT+K RI2 ΔRI and Δλ dip2 =K T2 ΔT+K RI2ΔRI, where the position change of each resonance peak is related to the change of temperature and refractive index, which is specifically reflected in the corresponding sensitivity (K). If all sensitivities K are constants, that is, the change of environmental parameters is linearly related to the change of resonance peak position, then In this way, the simultaneous measurement of multiple parameters (such as temperature, refractive index, stress, etc.) can be achieved by measuring the sensitivity K.

[0053] However, in practical applications, the influence of environmental parameters is usually nonlinear, such as Figures 2 to 5 Therefore, it is impossible to directly obtain the constant sensitivity K, which makes it possible to use a simple linear relationship It becomes unfeasible to measure multiple parameters simultaneously. However, according to T = f1 (Δλ dip1 , Δλ dip2 ) and RI=f2(Δλ dip1 , Δλ dip2 ), we can try to use nonlinear functions to describe the relationship between temperature, refractive index, etc. and the change of resonance peak position to achieve multi-parameter estimation and measurement.

[0054] In summary, the application of fiber Bragg grating in multi-parameter measurement needs to consider the nonlinear influence of environmental parameters and use appropriate mathematical models to describe the complex relationship between spectral characteristics and environmental parameters.

[0055] In this embodiment, a prediction model based on SVR (Support Vector Regression) is used to describe the complex relationship between spectral characteristics and environmental parameters.

[0056] S120, inputting the resonance peak change into a prediction model to perform temperature and refractive index prediction to obtain a prediction result.

[0057] In this embodiment, the prediction result refers to the predicted temperature and refractive index.

[0058] The prediction model is obtained by training a support vector regression machine using data on a fiber Bragg grating sensitivity characteristic curve as a sample set.

[0059] The prediction model includes a temperature SVR model and a refractive index SVR model; wherein the temperature SVR model is used to predict temperature; and the refractive index SVR model is used to predict refractive index.

[0060] As a regression algorithm, the essence of SVR is to fit the optimal function relationship through limited data samples to predict future data. If the fitted function relationship is linear, it is linear regression; if it is nonlinear, it is nonlinear regression. Therefore, SVR can solve both linear and nonlinear problems. Our research mainly focuses on the nonlinear relationship between temperature, refractive index and the change in the resonant wavelength of the fiber grating.

[0061] When SVR performs regression, it is necessary to construct a hyperplane to minimize the distance from the farthest sample point to this plane, such as Figure 8 As shown, such sample points are called support vectors. At the same time, in order to allow the error of the sample to exist, it is necessary to introduce an error term, which is called a penalty factor. The regression function also needs a loss function to ignore the error.

[0062] From the above, the SVR regression problem can be transformed into an optimization problem to find this hyperplane band:

[0063]

[0064] Using the Lagrange multiplier method, the above formula can be rewritten as:

[0065]

[0066] According to the above formula, we can get a i and Substituting it into the Lagrange multiplier method, we can obtain ω, and b can be obtained according to the KKT condition.

[0067] Therefore, the SVR function in the linear case can be expressed as:

[0068]

[0069] For nonlinear regression, it is necessary to introduce a kernel function K to replace the original inner product x i ·x j The kernel function needs to satisfy Mercer's theorem, which is defined as:

[0070] K(x i , x j )=[φ(x i )·φ(x j )];

[0071] The introduction of the kernel function K can map low-dimensional data samples to a high-dimensional feature space. That is, the original inner product space can be expanded to the kernel function space, and the original nonlinear regression problem is transformed into an approximate linear regression problem in the kernel function space.

[0072] The form and parameters of the kernel function determine the structure of the feature space, but when selecting a kernel function, it is not necessary to know how it is mapped or what the specific form is. The selection is mainly made by comparing the results. In this embodiment, the commonly used RBF kernel function is selected. Where σ is the kernel width.

[0073] The RBF kernel function has strong applicability in both high and low dimensions and large and small samples. The smaller σ is, the stronger its learning and generalization ability is. It can also be found from the above formula that the effective value range of the kernel function input is [0, 1], which greatly reduces the calculation time during SVR regression. Therefore, the RBF kernel function is widely used.

[0074] By introducing the kernel function K, the SVR of the nonlinear regression problem can be transformed into:

[0075] This summarizes the principle structure of SVR for solving nonlinear regression problems, such as Fig. 9 As shown,

[0076] Therefore, assuming that the external temperature T, the refractive index RI and the resonant wavelength change Δλ of the two resonant peaks dip1 and dip2 of the fine spiral grating dip1 , Δλ dip2 It has the following nonlinear functional relationship: RI = f(Δλ dip1 , Δλ dip2 ); T = f(Δλ dip1 , Δλ dip2 );

[0077] Through experimental measurements, Figure 4 and Figure 5 Select enough data samples from the sensitivity characteristic curve as the training set, that is, the changes in the corresponding resonance peaks dip1 and dip2 at different temperatures and different refractive indices. Use the cross-validation method to select the model parameters, and finally fit to obtain the nonlinear functions f1 and f2. Among them, the resonance peaks dip1 and dip2 refer to the first two resonance peaks in the spectrum. The position changes of these resonance peaks can be used to infer the environmental conditions to which the fiber Bragg grating is exposed, such as temperature or refractive index.

[0078] In one embodiment, the above prediction model is obtained by training a support vector regression machine using data on a fiber Bragg grating sensitivity characteristic curve as a sample set, and includes:

[0079] Obtain the resonance peak changes corresponding to different temperatures and different refractive indices in the sensitivity characteristic curve of the fiber Bragg grating to obtain a sample set;

[0080] Construct temperature SVR model and refractive index SVR model;

[0081] The cross-validation method is used to select the parameters of the temperature SVR model and the refractive index SVR model, and the parameters of the temperature SVR model and the refractive index SVR model are trained in combination with the sample set to obtain the prediction model.

[0082] In this embodiment, the temperature SVR model and the refractive index SVR model are ε-SVR models respectively.

[0083] The ε-SVR model is characterized by an insensitive loss function ε. When the input data error exceeds ε, the value of the ε function will be subtracted, which allows the error of the sample data to be between ±ε. The ε-SVR model has three important parameters to be selected, namely the value of the insensitive loss function ε, the parameter σ in the kernel function, and the penalty parameter c. The kernel function parameter σ mainly reflects the distribution of data in high-dimensional space. If σ is too large, the influence between support vectors will become stronger, thereby reducing the accuracy of the model; if σ is too small, the influence between support vectors will become too loose, which will reduce the generalization and evolution ability of the model. The penalty parameter c controls the degree of penalty for erroneous samples. Therefore, if c is larger, it will be more sensitive to errors. Although the accuracy of the algorithm will be higher, the overall algorithm will be more complicated, and the number of support vectors will be small, resulting in reduced generalization ability and over-learning. If c is too small, the training error will increase, which will also lead to reduced generalization ability and under-learning. The insensitive loss function ε controls the width of the insensitive region of the regression function to the sample data. Generally speaking, the smaller the value, the higher the accuracy of the model, but at the same time it will make the model more complex, require more running time, and reduce the generalization ability. Therefore, the selection of these three parameters needs to be adjusted according to the characteristics of the training set and the goals of the model.

[0084] Reasonable selection of model parameters is crucial to the overall performance of the model. Commonly used parameter selection methods include cross-validation, genetic algorithm, grid search, Bayesian optimization, differential evolution algorithm, etc.

[0085] In this embodiment, the cross-validation method is selected. Its advantages are mainly that it can obtain as much effective information as possible from limited data, and use different data division methods to increase the utilization and representativeness of data; it can select models and adjust parameters by comparing cross-validation errors under different models and parameters. The training set is divided into 10 equal subsets, each subset is used as a prediction set in turn, and the remaining 9 subsets are used as training sets for model parameter testing. This is repeated 10 times, and finally the most suitable σ and C are selected based on the average of the sum of squared errors.

[0086] In this embodiment, after the temperature (T) SVR model and the refractive index (RI) SVR model are trained, it is only necessary to observe the changes in the resonance peak and the reference position on the spectrometer based on room temperature and the refractive index of air, and input the data into the two models respectively, so as to obtain the temperature and refractive index of the space where the fiber grating is located at this time as output by the model.

[0087] According to Figure 6 and Figure 7 The comparison between the model prediction results and the actual results is shown. The maximum average error of temperature is 0.2℃, and the maximum average error of refractive index is 0.0007RIU.

[0088] Therefore, after obtaining the nonlinear functions f1 and f2, one only needs to observe the changes in the spectrum on the spectrometer, that is, the changes in the resonance peaks dip1 and dip2, and input them into the model to obtain the temperature T and refractive index RI at this time, thereby realizing simultaneous measurement of multiple parameters.

[0089] S130: Output the prediction result.

[0090] The predicted temperature and refractive index are output to the terminal for display.

[0091] The method of this embodiment introduces a support vector regression machine to solve the problem that when using fiber Bragg gratings for multi-parameter measurement, the sensitivity characteristics are nonlinear and cannot be realized or the error is large. It reduces the spectral wavelength range of the measurement, makes the originally nonlinear curve approximately linear within a certain range, and has low cost.

[0092] The above-mentioned method for predicting fiber Bragg grating parameters based on SVR predicts temperature and refractive index through the change of fiber Bragg grating resonance peak by using the SVR model, which solves the problem that fiber Bragg grating cannot be measured or has large errors when the sensitivity characteristic curve is nonlinear when realizing multi-parameter measurement, and realizes the simultaneous measurement of multiple environmental parameters based on fiber Bragg grating to ensure that the fiber Bragg grating sensor can stably and accurately provide the required environmental parameter measurement results.

[0093] Fig.10 FIG. 3 is a schematic block diagram of a device 300 for predicting fiber Bragg grating parameters based on SVR provided by an embodiment of the present invention. Fig.10 As shown, corresponding to the above method for predicting fiber Bragg grating parameters based on SVR, the present invention also provides a device 300 for predicting fiber Bragg grating parameters based on SVR. The device 300 for predicting fiber Bragg grating parameters based on SVR includes a unit for executing the above method for predicting fiber Bragg grating parameters based on SVR, and the device can be configured in a device with a fiber Bragg grating. Specifically, please refer to Figure 5The device 300 for predicting fiber Bragg grating parameters based on SVR includes an acquisition unit 301 , a prediction unit 302 and an output unit 303 .

[0094] The acquisition unit 301 is used to acquire the resonance peak variation of the fiber Bragg grating spectrum; the prediction unit 302 is used to input the resonance peak variation into the prediction model to perform temperature and refractive index prediction to obtain the prediction result; the output unit 303 is used to output the prediction result.

[0095] In one embodiment, the above device further includes:

[0096] The model generation unit is used to train a support vector regression machine using data on the fiber Bragg grating sensitivity characteristic curve as a sample set to obtain a prediction model.

[0097] In one embodiment, the model generation unit includes:

[0098] The sample set acquisition subunit is used to obtain the corresponding resonance peak changes at different temperatures and different refractive indices in the fiber Bragg grating sensitivity characteristic curve to obtain a sample set; the model construction subunit is used to construct a temperature SVR model and a refractive index SVR model; the training subunit is used to select parameters of the temperature SVR model and the refractive index SVR model by using a cross-validation method, and train the parameters of the temperature SVR model and the refractive index SVR model in combination with the sample set to obtain a prediction model.

[0099] It should be noted that technicians in the relevant field can clearly understand that the specific implementation process of the above-mentioned SVR-based fiber Bragg grating prediction parameter device 300 and each unit can refer to the corresponding description in the aforementioned method embodiment, and for the convenience and brevity of description, it will not be repeated here.

[0100] The above-mentioned device 300 for predicting fiber Bragg grating parameters based on SVR can be implemented in the form of a computer program. The computer program can be used in Fig.11 Runs on the computer device shown.

[0101] See also Fig.11 , Fig.11 5 is a schematic block diagram of a computer device provided in an embodiment of the present application. The computer device 500 may be a server, wherein the server may be an independent server or a server cluster composed of multiple servers.

[0102] See also Fig.11 The computer device 500 includes a processor 502 , a memory and a network interface 505 connected via a system bus 501 , wherein the memory may include a non-volatile storage medium 503 and an internal memory 504 .

[0103] The non-volatile storage medium 503 can store an operating system 5031 and a computer program 5032. The computer program 5032 includes program instructions, and when the program instructions are executed, the processor 502 can execute a method for predicting fiber Bragg grating parameters based on SVR.

[0104] The processor 502 is used to provide computing and control capabilities to support the operation of the entire computer device 500 .

[0105] The internal memory 504 provides an environment for the operation of the computer program 5032 in the non-volatile storage medium 503. When the computer program 5032 is executed by the processor 502, the processor 502 can execute a method for predicting fiber Bragg grating parameters based on SVR.

[0106] The network interface 505 is used to communicate with other devices over the network. Fig.11 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present application, and does not constitute a limitation on the computer device 500 to which the solution of the present application is applied. The specific computer device 500 may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0107] The processor 502 is used to run the computer program 5032 stored in the memory to implement the following steps:

[0108] Obtaining the resonance peak variation of the fiber Bragg grating spectrum; inputting the resonance peak variation into a prediction model to perform temperature and refractive index prediction to obtain a prediction result; and outputting the prediction result;

[0109] The prediction model is obtained by training a support vector regression machine using data on a fiber Bragg grating sensitivity characteristic curve as a sample set.

[0110] The prediction model includes a temperature SVR model and a refractive index SVR model; wherein the temperature SVR model is used to predict temperature; and the refractive index SVR model is used to predict refractive index.

[0111] The temperature SVR model and the refractive index SVR model are respectively ε-SVR models.

[0112] The resonance peak variation of the fiber Bragg grating spectrum includes the variation of the resonance peak and the reference position based on the normal temperature and the refractive index of air.

[0113] In one embodiment, when the processor 502 implements the step of training a support vector regression machine using data on the fiber Bragg grating sensitivity characteristic curve as a sample set to implement the prediction model, the processor 502 specifically implements the following steps:

[0114] The changes in the resonance peaks corresponding to different temperatures and refractive indices in the sensitivity characteristic curve of the fiber Bragg grating are obtained to obtain a sample set; a temperature SVR model and a refractive index SVR model are constructed; the parameters of the temperature SVR model and the refractive index SVR model are selected by cross-validation method, and the parameters of the temperature SVR model and the refractive index SVR model are trained in combination with the sample set to obtain a prediction model.

[0115] It should be understood that in the embodiment of the present application, the processor 502 may be a central processing unit (CPU), and the processor 502 may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0116] It can be understood by those skilled in the art that all or part of the processes in the method for implementing the above embodiment can be completed by instructing the relevant hardware through a computer program. The computer program includes program instructions, and the computer program can be stored in a storage medium, which is a computer-readable storage medium. The program instructions are executed by at least one processor in the computer system to implement the process steps of the embodiment of the above method.

[0117] Therefore, the present invention also provides a storage medium. The storage medium may be a computer-readable storage medium. The storage medium stores a computer program, wherein when the computer program is executed by a processor, the processor executes the following steps:

[0118] Obtaining the resonance peak variation of the fiber Bragg grating spectrum; inputting the resonance peak variation into a prediction model to perform temperature and refractive index prediction to obtain a prediction result; and outputting the prediction result;

[0119] The prediction model is obtained by training a support vector regression machine using data on a fiber Bragg grating sensitivity characteristic curve as a sample set.

[0120] The prediction model includes a temperature SVR model and a refractive index SVR model; wherein the temperature SVR model is used to predict temperature; and the refractive index SVR model is used to predict refractive index.

[0121] The temperature SVR model and the refractive index SVR model are respectively ε-SVR models.

[0122] The resonance peak variation of the fiber Bragg grating spectrum includes the variation of the resonance peak and the reference position based on the normal temperature and the refractive index of air.

[0123] In one embodiment, when the processor executes the computer program to implement the step of training a support vector regression machine using data on a fiber Bragg grating sensitivity characteristic curve as a sample set, the prediction model specifically implements the following steps:

[0124] The changes in the resonance peaks corresponding to different temperatures and refractive indices in the sensitivity characteristic curve of the fiber Bragg grating are obtained to obtain a sample set; a temperature SVR model and a refractive index SVR model are constructed; the parameters of the temperature SVR model and the refractive index SVR model are selected by cross-validation method, and the parameters of the temperature SVR model and the refractive index SVR model are trained in combination with the sample set to obtain a prediction model.

[0125] The storage medium may be a USB flash drive, a mobile hard disk, a read-only memory (ROM), a magnetic disk, or an optical disk, etc., which are computer-readable storage media that can store program codes.

[0126] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0127] In the several embodiments provided by the present invention, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of each unit is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed.

[0128] The steps in the method of the embodiment of the present invention can be adjusted in order, combined and deleted according to actual needs. The units in the device of the embodiment of the present invention can be combined, divided and deleted according to actual needs. In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0129] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, terminal, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention.

[0130] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A method for predicting fiber Bragg grating parameters based on SVR, characterized in that: include: Obtaining the resonance peak change of the fiber Bragg grating spectrum; Inputting the resonance peak variation into a prediction model to predict temperature and refractive index to obtain a prediction result; Outputting the prediction result; The prediction model is obtained by training a support vector regression machine using data on a fiber Bragg grating sensitivity characteristic curve as a sample set.

2. The method for predicting fiber Bragg grating parameters based on SVR according to claim 1, characterized in that: The prediction model includes a temperature SVR model and a refractive index SVR model; wherein the temperature SVR model is used to predict temperature; and the refractive index SVR model is used to predict refractive index.

3. The method for predicting fiber Bragg grating parameters based on SVR according to claim 2, characterized in that: The prediction model is obtained by training a support vector regression machine using data on the fiber Bragg grating sensitivity characteristic curve as a sample set, and includes: Obtain the resonance peak changes corresponding to different temperatures and different refractive indices in the sensitivity characteristic curve of the fiber Bragg grating to obtain a sample set; Construct temperature SVR model and refractive index SVR model; The cross-validation method is used to select the parameters of the temperature SVR model and the refractive index SVR model, and the parameters of the temperature SVR model and the refractive index SVR model are trained in combination with the sample set to obtain the prediction model.

4. The method for predicting fiber Bragg grating parameters based on SVR according to claim 2, characterized in that: The temperature SVR model and the refractive index SVR model are respectively ε-SVR models.

5. The method for predicting fiber Bragg grating parameters based on SVR according to claim 1, characterized in that: The resonance peak variation of the fiber Bragg grating spectrum includes the variation of the resonance peak and the reference position based on the normal temperature and the refractive index of air.

6. A device for predicting fiber Bragg grating parameters based on SVR, characterized in that: include: An acquisition unit, used for acquiring a resonance peak variation of a fiber grating spectrum; A prediction unit, used for inputting the resonance peak variation into a prediction model to perform temperature and refractive index prediction to obtain a prediction result; An output unit is used to output the prediction result.

7. The device for predicting fiber Bragg grating parameters based on SVR according to claim 6, characterized in that: Also includes: The model generation unit is used to train a support vector regression machine using data on the fiber Bragg grating sensitivity characteristic curve as a sample set to obtain a prediction model.

8. The device for predicting fiber Bragg grating parameters based on SVR according to claim 7, characterized in that: The model generation unit comprises: The sample set acquisition subunit is used to acquire the resonance peak change corresponding to different temperatures and different refractive indices in the fiber Bragg grating sensitivity characteristic curve to obtain a sample set; A model building subunit, used to build a temperature SVR model and a refractive index SVR model; The training subunit is used to select parameters of the temperature SVR model and the refractive index SVR model by using a cross-validation method, and train the parameters of the temperature SVR model and the refractive index SVR model in combination with the sample set to obtain a prediction model.

9. A computer device, characterized in that: The computer device comprises a memory and a processor, the memory stores a computer program, and the processor implements the method according to any one of claims 1 to 5 when executing the computer program.

Citation Information

Patent Citations

  • Multi-parameter identification system and method of tapered fiber

    CN106482766A

  • Method for realizing simultaneous sensing of temperature, strain and refractive index through single fiber bragg grating

    CN106706030A

  • Roadway nonlinear deformation prediction method based on fiber bragg grating sensing network data

    CN107368463A

  • Composite fiber grating sensor and dual-parameter measurement method of refractive index and temperature

    CN109709070A

Cited By

  • Prediction model construction method, temperature measurement method and module thereof, and measurement device

    CN121275182A

  • Prediction model construction method, temperature measurement method and module, and measurement device

    CN121275182B