A method, system and device for demodulating and anti-aliasing measurement of optical fiber FP sensor
Through adaptive sampling technology and adaptive model reference control, the sampling rate of the optical fiber F-P sensor is adjusted in real time, which solves the aliasing phenomenon in the vibration environment, realizes high-precision and high-resolution vibration measurement, and improves the robustness of the system and Demodulation bandwidth.
Patent Information
- Application Number
- CN202411381020.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-30
- Publication Date
- 2024-11-29
- Estimated Expiration
- 2044-09-30
AI Technical Summary
The aliasing phenomenon caused by insufficient sampling rate in the vibration and pressure measurement in extreme environments reduces measurement accuracy and stability. The existing technology has failed to effectively solve the problem of computing resource allocation and real-time, limiting understanding of bandwidth adjustment and applications of high frequency dynamic pressure measurement.
Adaptive sampling technology and adaptive model reference control are adopted to establish a sampling rate error model, adjust the sampling rate in real time, and equally spaced sampling is achieved through triangular fitting function interpolation method to improve the accuracy and resolution of signal reconstruction.
It realizes high-precision and high-resolution vibration measurement in extreme environments, dynamically adjusts computing resource allocation, enhances the system's robustness to unknown perturbations, and improves the accuracy of understanding bandwidth adjustment and signal reconstruction.
Smart Images

Figure CN118896636B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of optical fiber sensor demodulation, and in particular to an optical fiber FP sensor demodulation anti-aliasing measurement method, system and equipment. Background Art
[0002] Fiber FP sensor is a sensor commonly used for vibration measurement and pressure monitoring in extreme environments. However, random vibration poses a challenge to the performance of FP sensors because the optical path length in the FP air cavity will be affected by vibration and change linearly under changing vibration environment conditions. According to the Nyquist sampling theorem, if the spectral sampling rate is lower than twice the highest frequency of the vibration signal, the measured FP cavity length value will be periodically aliased, thereby reducing the accuracy and stability of the cavity length data under high-frequency signal measurement. This aliasing error will affect the credibility of the experimental results, resulting in severe distortion of the signal waveform and instability in frequency measurement. Therefore, there is an urgent need for a new solution that can effectively meet this challenge. In order to solve this problem, the sampling rate of the spectrometer is usually increased as a means of anti-aliasing, or a spectrometer with multiple channels is used to splice the spectral signal. By increasing the sampling rate, the bandwidth of the measurement system can be increased, and the upper limit of the measurable signal frequency can be increased.
[0003] The invention patent application document with application number 202011413621.4 discloses an adaptive infrared sensor signal sampling algorithm, which uses the mean square error as a judgment condition to improve the validity and stability of the sampling data, ensure the reliability and robustness of the temperature measurement results, and optimize the computing resources to a certain extent by dynamically adjusting the number of sampling data. The invention patent application document with application number 202110509184.4 discloses a high-frequency anti-aliasing bandpass optical analog-to-digital conversion device, which realizes digital signal anti-aliasing by bandpass filtering and alias-free digitization of the signal. However, both methods do not take into account the allocation and real-time performance of computing resources, and thus cannot guarantee demodulation accuracy and resolution.
[0004] In addition, the frame rate of the spectrometer limits the demodulation bandwidth of the sensor demodulation system, which in turn limits the application of the demodulation system in high-frequency dynamic pressure measurement. Therefore, it is of great significance to increase the frame rate of spectral acquisition and thus increase the demodulation bandwidth of the fiber FP sensor. By using time-sharing exposure and spectral signal fusion technology, multiple spectrometer modules are stacked to form a high-speed spectrometer module, thereby doubling the frame rate of the spectrometer and achieving high-frame-rate spectral acquisition. However, this method increases the size of the demodulation system, making it inconvenient to carry, and the non-equally spaced sampling caused by cross sampling will also cause errors in the demodulation data. Summary of the invention
[0005] In view of the above-mentioned problems, the present invention aims to provide a method, system and device for demodulating and anti-aliasing measurement of optical fiber FP sensor. Based on adaptive sampling technology and adaptive model reference control, the sampling rate is adjusted in real time to improve the accuracy of the overall sampling rate and signal reconstruction, and realize anti-aliasing measurement of optical fiber FP sensor demodulation.
[0006] In order to achieve the above object, the technical solution adopted by the present invention is as follows:
[0007] A method for demodulating and anti-aliasing measurement of an optical fiber FP sensor comprises the following steps:
[0008] S1: Establish a sampling rate error model based on the cavity length signal input by the upper demodulation system;
[0009] S2: Based on the sampling rate error model, determine the adaptive control rate of the sampling rate so that the sampling rate can be adaptively adjusted when signal aliasing occurs;
[0010] S3: Based on the sampling rate obtained in step S2, further implementing equal-interval sampling through a triangular fitting function interpolation method;
[0011] S4: Sampling is performed according to the sampling points determined in step S3, and the signal frequency of the sampling points is calculated.
[0012] Furthermore, the specific operation of step S1 includes the following steps:
[0013] S101: According to the cavity length signal input by the upper demodulation system , determine the signal model after sampling ;
[0014] In the formula, n is the number of sampling points, It is the cavity length signal input to the upper demodulation system. t represents the sampling time, is the sampling rate in Hz, represents the unit impulse function;
[0015] S102: Calculate the signal phase at the current sampling rate according to the transfer equation;
[0016] S103: Calculate the phase in the previous sampling rate state according to the reference model;
[0017] S104: determining an error function according to a phase difference corresponding to two adjacent sampling rates;
[0018] S105: Determine the error e The calculation method of .
[0019] Further, in step S102, the signal phase at the current sampling rate is expressed as ;
[0020] In the formula, Represents the output of the signal, represents the phase of the signal, is the first derivative of the phase, which is the frequency of the signal; freq is the frequency calculation function, represents the feedback sampling signal, is the control coefficient of the forward channel, is controlled .
[0021] Further, in step S103, the phase in the previous sampling rate state is expressed as: ;
[0022] In the formula, is the signal output in the previous sampling rate state, Indicates the signal phase in the previous sampling rate state, Indicates the frequency in the previous sampling rate state, which is also the reference model.
[0023] Furthermore, the error function in step S104 is expressed as ;
[0024] In the formula, e Indicates the error, f Indicates the signal frequency at the current sampling rate, that is, ;Introduce parameters and right Control to change the sampling rate, Indicates the adjustment direction of the sampling rate, upward adjustment is positive, downward adjustment is negative, represents the adjustment amount, then ;like e If the sampling rate is greater than or equal to twice the highest frequency of the acquired signal, there is no need to increase the sampling rate. e If it is greater than the error threshold, it means that aliasing has occurred.
[0025] Furthermore, in step S105, the loss function calculation error is introduced e , the loss function is expressed as ;
[0026] In the formula, Indicates loss, The extreme point of The zero point.
[0027] Furthermore, the specific operation of step S2 includes the following steps:
[0028] S201: Adjust the sampling frequency and initial phase, then ;
[0029] in is a virtual interpolation point, and the real interpolation point is The lower integer of ;
[0030] S202: Determine an adaptive control rate;
[0031] The adaptive control rate is expressed as ;
[0032] S203: Determine using gradient descent method : ;
[0033] in, is the number of iterations, is the learning rate, is the loss function, Calculate the gradient.
[0034] Furthermore, the specific operation of step S3 includes the following steps:
[0035] S301: By sampling three points, three points are obtained , and ; Interpolation is obtained : ;
[0036] S302: Confirm location; .
[0037] Further, the present invention also includes a fiber optic FP sensor demodulation anti-aliasing measurement system, the anti-aliasing measurement system including a reference model module, a transfer equation module, an error identification module, an adaptive rate adjustment module, an interpolation module and a total control module;
[0038] The reference model module is used to calculate the phase under the previous sampling rate state, and the transfer equation module is used to calculate the signal phase under the current sampling rate; the error identification module determines whether the error exceeds the threshold after subtracting the signal phase calculated by the adaptive rate adjustment module from the reference model module and the transfer equation module; if the error does not exceed the threshold, the adaptive rate adjustment module adjusts the sampling rate through the interpolation module, and the overall control module calculates and outputs the current signal frequency according to the sampling rate output by the adaptive rate adjustment module;
[0039] The reference model module, the transfer equation module, the error identification module, the adaptive rate adjustment module, the interpolation module and the overall control module are implemented based on the anti-aliasing measurement method as described above.
[0040] Furthermore, the present invention also includes an electronic device, which includes at least one processor; and a memory communicatively connected to the processor; wherein the memory stores instructions executable by the processor, and the instructions are executed by the processor so that the processor can execute the optical fiber FP sensor demodulation anti-aliasing measurement method as described above.
[0041] The beneficial effects of the present invention are:
[0042] 1. The present invention adopts adaptive sampling technology and adaptive model reference control. According to the frequency information of the collected signal, the system sampling rate is adjusted in real time through the interpolation points of the adaptive rate to achieve higher accuracy and resolution in key areas. It not only solves the bandwidth limitation problem, but also can dynamically adjust the allocation of computing resources to ensure that the system can make a reasonable response to the changing needs and environment.
[0043] 2. In the present invention, through the interpolation points of the adaptive rate, the system can automatically adjust the sampling density according to the signal characteristics to ensure higher accuracy and resolution in key areas, thereby improving the overall sampling rate and the accuracy of signal reconstruction.
[0044] 3. The control system based on the adaptive model in the present invention uses an adaptive algorithm to iteratively adjust the controller parameters. This method can not only provide stability and fast convergence speed, but also enhance the robustness of the system to unknown disturbances, effectively cope with challenging vibration environment conditions, and provide an efficient and reliable solution for vibration measurement and pressure monitoring in extreme environments, which has broad application prospects in practical engineering. BRIEF DESCRIPTION OF THE DRAWINGS
[0045] Figure 1 It is a schematic diagram of the measurement of the optical fiber FP vibration sensor in the present invention.
[0046] Figure 2 It is a schematic diagram of the aliasing phenomenon in the present invention.
[0047] Figure 3 This is a block diagram of the demodulation and anti-aliasing measurement principle of the optical fiber FP sensor in the present invention.
[0048] Figure 4 Schematic diagram of multi-channel adaptive sampling anti-aliasing in the present invention. DETAILED DESCRIPTION
[0049] In order to enable those skilled in the art to better understand the technical solution of the present invention, the technical solution of the present invention is further described below in conjunction with the accompanying drawings and embodiments. Embodiment 1
[0050] Embodiment 1 provides a fiber FP sensor demodulation anti-aliasing measurement method. The light in the broadband light source is transmitted through the fiber channel. When it interferes with the vibration sensitive FP structure in the optical fiber, it is transmitted back through the fiber channel and transmitted to the demodulation system through the ring structure in the light source. A regularly changing sinusoidal cavity length data curve can be obtained through demodulation, as shown in the attached figure. Figure 1 If the sampling rate is not high enough, the spectral sidebands will be aliased and the frequency of the curve cannot be correctly obtained. Figure 2 As shown. In the present invention, the distance between each frequency band is widened by adaptively increasing the sampling rate, and finally the frequency of the signal is calculated by a frequency demodulation algorithm. The adaptive increase in sampling rate uses the phase as a judgment basis. If the phase difference between the previous sampling rate state and the current state does not exceed the threshold, there is no need to increase the sampling rate. If the phase difference between the previous sampling rate state and the current state exceeds the threshold, it proves that the sampling rate is not enough and needs to be increased. As shown in the attached Figure 3 This method eliminates the need for manual adjustment of the sampling rate, thereby increasing the system's adaptability to variable frequency signals.
[0051] Based on the above principle, the anti-aliasing measurement method in the present invention specifically includes the following steps:
[0052] S1: Establish a sampling rate error model based on the cavity length signal input by the upper demodulation system;
[0053] More specifically, S101: according to the cavity length signal input by the upper demodulation system , determine the signal model after sampling;
[0054] In the present invention, the sampling function adopts the Dirac comb function, and the signal model obtained after sampling according to the Dirac comb function is expressed as ;
[0055] In the formula, n is the number of sampling points, It is the cavity length signal input to the upper demodulation system. t represents the sampling time, is the sampling rate in Hz, represents the unit impulse function.
[0056] Through the derivation of Dirac comb function into Fourier series, we can see that the frequency domain of the input signal is also a discrete function. Therefore, by shifting the input signal spectrum, we can get ;
[0057] In the formula, is a constant; T s is the sampling time interval, is the sampling rate The reciprocal of , in s.
[0058] S102: Calculate the signal phase at the current sampling rate according to the transfer equation;
[0059] At the current sampling rate, the signal phase is calculated as ;
[0060] In the formula, Represents the output of the signal, represents the phase of the signal, is the first derivative of the phase, which is the frequency of the signal; freq is the frequency calculation function, represents the feedback sampling signal, is the control coefficient of the forward channel, is controlled .
[0061] S103: Calculate the phase in the previous sampling rate state according to the reference model;
[0062] The reference model is the non-controlled version of the transfer function, i.e. ;
[0063] In the formula, It is the signal output in the previous sampling rate state; Indicates the signal phase in the previous sampling rate state, Indicates the frequency in the previous sampling rate state, which is also the reference model.
[0064] S104: determining an error function according to a phase difference corresponding to two adjacent sampling rates;
[0065] The error function in the present invention uses the phase difference corresponding to two adjacent sampling rates, that is, ;
[0066] In the formula, e Indicates the error, f Indicates the signal frequency at the current sampling rate, that is, ;Introduce parameters and right To control, change the sampling rate, Indicates the adjustment direction of the sampling rate, upward adjustment is positive, downward adjustment is negative, represents the adjustment amount; hence we can get ;
[0067] According to the above formula, the initial phase of the next sampling rate can be obtained by The calculation is obtained because To introduce The binary function of frequency modulation, so only the control variable That's it.
[0068] Set the error threshold if e is less than the error threshold, which proves that the sampling rate is greater than twice or more of the highest frequency of the collected signal, and there is no need to increase the sampling rate; if e If the error is greater than the error threshold, it means that aliasing has occurred. In this case, interpolation is required to design the control rate. This design can make the system more stable. Is input The frequency modulation and initial phase are moved based on the frequency modulation and initial phase.
[0069] S105: Determine the error e Calculation method of
[0070] In the present invention, a loss function is introduced to calculate the error e , the loss function is expressed as ;
[0071] In the formula, Indicates loss, The extreme point of The zero point of The extreme point of The coefficient with the minimum error is obtained and , which is the optimal sampling rate.
[0072] Furthermore, step S2: based on the sampling rate error model, determine the adaptive control rate of the sampling rate, so that the sampling rate can be adaptively adjusted when signal aliasing occurs.
[0073] Specifically, S201: According to the loss function in step S1, it can be seen that the loss function includes the phase information of the signal and the sampling rate control factor If you want to achieve and maintain equal intervals, you need to adjust two parameters, namely frequency and initial phase, which can be expressed as ;
[0074] in is a virtual interpolation point, and the real interpolation point is Take the integer below, this formula can be obtained and relationship, we can get a Corresponding to one and an initial phase , that is, by changing value for phase control.
[0075] S202: Determine an adaptive control rate;
[0076] In the present invention, gradient descent is used for control rate design, and the rate of change of the sampling rate is expressed as ;
[0077] S203: Determine using gradient descent method ;
[0078] Specifically, we introduce the differential operator , ;
[0079] Then it can be deduced that ;
[0080] From the above formula, we can get that the control sampling rate needs to be found Calculated as ;
[0081] in, is the number of iterations, is the learning rate, is the loss function, Calculate the gradient.
[0082] Further, step S3: based on the sampling rate obtained in step S2, further implementing equal-interval sampling through a triangular fitting function interpolation method;
[0083] Specifically, the adjusted adaptive control rate can be obtained through step S2. and In order to achieve equal-interval sampling and further reduce the possibility of frequency mixing, the present invention uses a triangular fitting algorithm for interpolation. The specific formula is as follows:
[0084] By sampling three points, we can get the following three points , and , obtained by interpolation : ;
[0085] After trigonometric transformation, we can get ;
[0086] but .
[0087] After this step, the value of the interpolation point can be obtained, and the value is inserted into the calculated phase to complete the triangular fitting interpolation sampling rate improvement.
[0088] Further, step S4: sampling is performed according to the sampling points determined in step S3, and the signal frequency of the sampling points is calculated.
[0089] Through step S2 and step S3, the sampling rate can be adjusted in real time according to the collected signal frequency information to ensure that the frequency calculation result will not be affected by aliasing and cause errors within the working range. Figure 4 shown. Embodiment 2
[0090] Embodiment 2 provides a fiber FP sensor demodulation anti-aliasing measurement system, including a reference model module, a transfer equation module, an error identification module, an adaptive rate adjustment module, an interpolation module and a general control module;
[0091] The reference model module is used to calculate the phase under the previous sampling rate state, and the transfer equation module is used to calculate the signal phase under the current sampling rate; the error identification module determines whether the error exceeds the threshold after subtracting the signal phase calculated by the reference model module and the transfer equation module by the adaptive rate adjustment module; if the error does not exceed the threshold, the adaptive rate adjustment module adjusts the sampling rate through the interpolation module, and the overall control module calculates and outputs the current signal frequency according to the sampling rate output by the adaptive rate adjustment module.
[0092] It should be noted that the reference model module, the transfer equation module, the error identification module, the adaptive rate adjustment module, the interpolation module and the overall control module are implemented according to the method described in the first embodiment. Embodiment 3
[0093] Embodiment 3 provides an electronic device, comprising at least one processor; and a memory communicatively connected to the processor; wherein the memory stores instructions executable by the processor, and the instructions are executed by the processor so that the processor can execute the optical fiber FP sensor demodulation anti-aliasing measurement method described in Embodiment 1.
[0094] The above shows and describes the basic principles, main features and advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.
Claims
1. A demodulation and anti-aliasing measurement method for optical fiber FP sensor, It is characterized in that The following steps are involved: S1: Establish a sampling rate error model based on the cavity length signal input by the upper demodulation system; S2: Based on the sampling rate error model, determine the adaptive control rate of the sampling rate so that the sampling rate can be adaptively adjusted when signal aliasing occurs; S3: Based on the sampling rate obtained in step S2, further implementing equal-interval sampling through a triangular fitting function interpolation method; S4: Sampling is performed according to the sampling points determined in step S3, and the signal frequency of the sampling points is calculated; The specific operation of step S1 includes the following steps: S101: According to the cavity length signal input by the upper demodulation system , determine the signal model after sampling ; In the formula, n is the number of sampling points, It is the cavity length signal input by the upper demodulation system. t represents the sampling time, is the sampling rate in Hz, represents the unit impulse function; S102: Calculate the signal phase at the current sampling rate according to the transfer equation; S103: Calculate the phase in the previous sampling rate state according to the reference model; S104: determining an error function according to a phase difference corresponding to two adjacent sampling rates; S105: Determine the error e Calculation method of In step S102, the signal phase at the current sampling rate is expressed as ; In the formula, Represents the output of the signal, represents the phase of the signal, is the first derivative of the phase, which is the frequency of the signal; freq is the frequency calculation function, represents the feedback sampling signal, is the control coefficient of the forward channel, is controlled ; In step S103, the phase in the previous sampling rate state is expressed as: ; In the formula, is the signal output in the previous sampling rate state, Indicates the signal phase in the previous sampling rate state, Indicates the frequency at the previous sampling rate state, which is also the reference model; The error function in step S104 is expressed as ; In the formula, e Indicates the error, f Indicates the signal frequency at the current sampling rate, that is, ;Introduce parameters and right Control to change the sampling rate, Indicates the adjustment direction of the sampling rate, upward adjustment is positive, downward adjustment is negative, represents the adjustment amount, then ; like e If the sampling rate is greater than or equal to twice the highest frequency of the acquired signal, there is no need to increase the sampling rate. e If it is greater than the error threshold, it means that aliasing has occurred; The specific operation of step S2 includes the following steps: S201: Adjust the sampling frequency and initial phase, then ; in is a virtual interpolation point, and the real interpolation point is The lower integer of ; S202: Determine an adaptive control rate; The adaptive control rate is expressed as ; S203: Determine using gradient descent method : ; Among them, The number of iterations, is the learning rate, is the loss function, is the gradient calculation; The specific operation of step S3 includes the following steps: S301: By sampling three points, three points are obtained , and ; Interpolation is obtained : ; S302: Confirm location; 。 2. A fiber optic FP sensor demodulation anti-aliasing measurement method according to claim 1, It is characterized in that In step S105, the loss function calculation error is introduced e , the loss function is expressed as ; In the formula, Indicates loss, The extreme point of is the zero point of .
3. A fiber optic FP sensor demodulation anti-aliasing measurement system, Features: It includes a reference model module, a transfer equation module, an error identification module, an adaptive rate adjustment module, an interpolation module and a total control module; The reference model module is used to calculate the phase at the previous sampling rate state, and the transfer equation module is used to calculate the signal phase at the current sampling rate; The error identification module determines whether the error exceeds a threshold value after subtracting the signal phase calculated by the adaptive rate adjustment module for the reference model module and the transfer equation module; If the error does not exceed the threshold, the adaptive rate adjustment module adjusts the sampling rate through the interpolation module, and the overall control module calculates and outputs the current signal frequency according to the sampling rate output by the adaptive rate adjustment module; The reference model module, the transfer equation module, the error identification module, the adaptive rate adjustment module, the interpolation module and the overall control module are implemented based on the anti-aliasing measurement method described in claim 1 or 2.
4. An electronic device, Features: The electronic device includes at least one processor; and a memory communicatively connected to the processor; wherein the memory stores instructions executable by the processor, and the instructions are executed by the processor so that the processor can execute the optical fiber FP sensor demodulation anti-aliasing measurement method described in claim 1 or 2.
Citation Information
Patent Citations
Self-adaptive infrared sensor signal sampling algorithm
CN112461376A
High-frequency anti-aliasing band-pass dimmable analog-to-digital conversion device
CN113359369A
Anti-aliasing for frequency output sensors
US5095263A
Devices, systems, and methods for composite signal decoding
US5357544A
