Multi-channel sampling clutter processing method, system and equipment of optical fiber F-P sensor

Abstract

The invention relates to a multichannel sampling clutter processing method, system and device for an optical fiber F-P sensor, and the method comprises the steps: collecting reflection spectrum signals of the optical fiber F-P sensor in parallel, and obtaining multipath reflection spectrum data; taking data points corresponding to each path of reflection spectrum data at the same moment as a group of first data sets, and obtaining a plurality of groups of first data sets at continuous moments; obtaining a central point of each group of first data sets in sequence, and taking the central points as a second data set; performing Kalman filtering on the central point of the second data set to obtain wavelength signal data, and drawing the wavelength signal data into a waveform curve for output; according to the invention, the waveform burr of the multichannel sampling clutter output signal is effectively eliminated, so that the output curve obtained by demodulation is smoother, and the demodulation precision is improved.

Description

Technical Field

[0001] The invention belongs to the technical field of optical fiber sensor demodulation, and relates to a multi-channel sampling clutter processing method, system and equipment for an optical fiber FP sensor. Background Art

[0002] The fiber FP sensor is a sensor based on the Fabry-Perot interference principle, which is used to measure and monitor changes in physical quantities. This sensor uses the Fabry-Perot interference phenomenon outside the optical fiber to convert the optical signal into the corresponding physical quantity signal for measurement. The fiber FP sensor consists of a pair of reflectors at both ends of the optical fiber. The reflectors can be two reflective surfaces at the end of the optical fiber, or metal or dielectric reflective layers evaporated or welded on the optical fiber. When light enters from one end of the optical fiber, part of the light will be reflected back by the first reflective surface, and then reflected back by the second reflective surface to form interference. When the external physical quantity (such as temperature, pressure or deformation, etc.) changes, the length or refractive index of the optical fiber will change, and the position or intensity of its interference peak will also change accordingly. By measuring the movement or intensity change of the interference peak, the change of the physical quantity can be inferred.

[0003] When demodulating the fiber FP sensor, since the frame rate of the spectrometer can only reach about 25kHz, if a demodulation speed of 100kHz is to be achieved, four spectrometers are needed, and then four times the spectrum sampling frame rate is obtained by time-sharing exposure and parallel sampling of the four spectrometers. Therefore, the high-speed demodulator needs to drive the four spectrometers at the same time. Since the output signal of the spectrometer is an analog voltage signal, after the driving circuit of the four spectrometers, four high-speed AD sampling circuits are connected in parallel. Since there is time shift and offset between the four analog voltage signals, the cavity length output curve obtained after splicing the four analog voltage signals is locally jagged with burrs, and is not a smooth curve. Therefore, there will be certain errors in the information obtained after demodulation.

[0004] Publication No. CN 106706011 A discloses a method for filtering out pulse clutter in demodulation of a fiber Bragg grating sensor. The method comprises building a fiber Bragg grating demodulator optical path and circuit hardware platform, performing fiber Bragg grating demodulation, using a field programmable logic gate array for high-speed AD acquisition to read AD sampling values, and setting a "peak threshold"; when the current sampling value begins to be less than the threshold, calculating the distance from the initial value greater than the threshold to the initial value less than the threshold; filtering the pulse interference signal, and setting a "broadening threshold"; through the above steps, filtering of the pulse interference existing in the demodulation process is achieved, achieving the effect of improving the accuracy and stability of fiber Bragg grating demodulation. The scheme uses a threshold to determine whether the target information needs to be filtered out to achieve filtering, but for the multi-channel sampling clutter problem, since the sampling of the spectrometer is random, the threshold for filtering out clutter is not easy to set, so this method cannot solve the multi-channel sampling clutter problem well.

[0005] Publication No. CN114353844 A discloses a fiber Bragg grating signal demodulation method, which generates light waves of different wavelengths by setting the parameters of a laser generator to obtain an original signal containing noise; uses a wavelet denoising method to the original signal containing noise to obtain a signal with ordinary noise eliminated; performs optimal prediction variable threshold processing on the signal with ordinary noise eliminated to obtain a signal with special noise eliminated; performs 5-point smoothing processing on the signal with special noise eliminated to obtain a smoothed signal; derivates the smoothed signal to obtain a peak differential signal; when the peak differential signal is greater than the threshold, the signal is retained; when it is less than the threshold, it indicates that the signal has drifted and an early warning process is performed. This scheme uses a wavelet denoising method for noise reduction, but because the 100kHz high-speed demodulation requires signal processing at an extremely short speed, and the wavelet function for extracting the signal in each time period will occupy more resources of the demodulation chip, it cannot handle the multi-channel sampling noise problem well. Summary of the invention

[0006] In order to solve the above technical problems, the purpose of the present invention is to provide a multi-channel sampling clutter processing method, system and equipment for a fiber optic FP sensor. The present invention can solve the multi-channel sampling clutter problem during 100kHz high-speed demodulation by grouping the sampling data points to find the center point, and then performing Kalman filtering on the new data set composed of the center points of all groups, thereby eliminating the waveform burr problem caused by time shift and offset between the sampling data.

[0007] In order to achieve the above object, the technical solution adopted by the present invention is as follows: The present invention provides a multi-channel sampling clutter processing method for an optical fiber FP sensor, the method comprising the following steps: Step S1, collecting reflection spectrum signals of optical fiber FP sensors in parallel to obtain multi-channel reflection spectrum data; Step S2: Taking the data points corresponding to the reflection spectrum data of each path at the same moment as a set of first data sets, and obtaining multiple sets of the first data sets at consecutive moments; Step S3: Sequentially obtaining the center points of each set of the first data sets, and taking the center points as the second data set; Step S4: Performing Kalman filtering on the center points of the second data set to obtain wavelength signal data, and plotting the wavelength signal data as a waveform curve for output.

[0008] Furthermore, in the step S1, the reflection spectrum signals of the fiber optic F-P sensors are collected in parallel to obtain multiplexed reflection spectrum data, specifically: Inputting broadband light from the first port of the fiber optic circulator into the fiber optic F-P sensor, and outputting the reflected light of the fiber optic F-P sensor from the third port of the fiber optic circulator. The output reflected light is split and then enters multiple spectrometers respectively. Each spectrometer simultaneously obtains the electrical signals of the reflection spectrum data of the fiber optic F-P sensor, converts the electrical signals of each path into digital signals, and sends them to the FPGA controller for signal processing to obtain multiplexed reflection spectrum data.

[0009] Furthermore, in the step S3, when obtaining the center points of each set of the first data sets, specifically: Defining the center point The square of the distance to each data point in the first data set is: ; wherein, represents the distance from the th point to the center point , , are the coordinates of the center point , , are the coordinates of the th data point; Setting the function of the distance between the center point and each data point as: ; Using the iterative formula to solve the minimum value of the function , and the iterative formula is as follows: ; wherein, is the learning rate, and are the partial derivatives of the function with respect to the coordinates and coordinates The partial derivative of The partial derivative of is: ; Substitute the partial derivative into the iterative formula to obtain the center point Coordinates , .

[0010] Further, in step S4, the Kalman filter specifically includes: The center point The processing is performed by establishing the Kalman state prediction equation in turn. The Kalman state prediction equation is: ; in, yes The predicted value of the state at time is the state transition matrix, yes The optimal state estimate at time , is the control input matrix, yes Control input at all times; The forecast error covariance is: ; in, yes The forecast error covariance at time , yes The optimal estimation error covariance at time , yes The transposed matrix of is the process noise covariance.

[0011] Furthermore, the Kalman filter also includes: after establishing the Kalman state prediction equation for processing, the center point The update is performed by establishing a state update equation, which is: ; in, yes The Kalman gain at time t, is the observation matrix, is the observation noise covariance, yes The optimal state estimate at time , yes The observed value at time, yes The optimal estimation error covariance at time , is the identity matrix.

[0012] Furthermore, the Kalman state prediction equation is established through an FPGA controller.

[0013] Furthermore, the state update equation is established through an FPGA controller.

[0014] Furthermore, in step S4, the wavelength signal data is plotted as a waveform curve for output, specifically: the FPGA controller sends the wavelength signal data to the host computer software, and the wavelength signal data is plotted as a waveform curve and displayed by the host computer software.

[0015] The present invention also provides a multi-channel sampling clutter processing system for an optical fiber FP sensor, comprising: A reflection spectrum data acquisition module is used to collect reflection spectrum signals of optical fiber FP sensors in parallel to obtain multi-channel reflection spectrum data; A first data set module, used to take the data points corresponding to each channel of the reflection spectrum data at the same moment as a set of first data sets, and obtain a plurality of sets of the first data sets at consecutive moments; A second data set module, used for sequentially acquiring the center point of each group of the first data sets, and taking the center point as the second data set; The waveform curve output module is used to perform Kalman filtering on the center point of the second data set to obtain wavelength signal data, and draw the wavelength signal data into a waveform curve for output.

[0016] The present invention also 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 multi-channel sampling clutter processing method of the optical fiber FP sensor as described above.

[0017] The present invention adopts the above technical solution, which has the following advantages and effects: (1) The present invention provides a method, system and device for processing clutter in multi-channel sampling of an optical fiber FP sensor. The method, system and device are mainly used for processing multi-channel sampling data of a 100kHz high-speed demodulation. Since a spectrometer needs to perform parallel sampling, there is a time shift and offset between them. The present invention processes multi-channel sampling data by using a Kalman filtering method based on the center value of the data point, effectively eliminating the waveform burrs of the output signal, making the output curve obtained by 100kHz demodulation smoother, thereby improving the demodulation accuracy. The Kalman filtering based on the center value of the data point is an estimation algorithm that combines numerical calculation and recursive ideas. It can use numerical methods to solve the center point of a group of multi-channel collected data, and then perform Kalman filtering on the new data composed of the center points. Since the problem of solving the center point occupies less computing resources than solving the Kalman filtering equation, the waveform processing speed can be accelerated, thereby enhancing the stability and real-time performance of clutter processing.

[0018] (2) The multi-channel sampling clutter processing method, system and device of the optical fiber FP sensor of the present invention are not only suitable for demodulation of the optical fiber FP sensor, but can also be applied to other optical sensor systems that require high-speed and high-precision multi-channel sampling, and have broad application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 The present invention is a flowchart of a multi-channel sampling clutter processing method for an optical fiber FP sensor.

[0020] Figure 2 This is a four-channel sampling signal diagram of the present invention.

[0021] Figure 3 This is a waveform diagram after processing according to the present invention. DETAILED DESCRIPTION

[0022] The embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings so that the purpose, features and advantages of the present invention can be more clearly understood. It should be understood that the embodiments shown in the accompanying drawings are not intended to limit the scope of the present invention, but are only intended to illustrate the essential spirit of the technical solution of the present invention. Embodiment 1

[0023] like Figure 1 Embodiment 1 provides a multi-channel sampling clutter processing method for an optical fiber FP sensor, the method comprising the following steps: Step 1: Acquire multi-channel reflection spectrum data by collecting the reflected light signal of the optical fiber FP sensor in parallel through multiple channels.

[0024] Specifically, time-sharing exposure and parallel sampling are performed through a spectrometer, and the broadband light of the SLD light source is input from the first port of the optical fiber circulator and enters the optical fiber FP sensor through the second port of the optical fiber circulator. The reflected light of the optical fiber FP sensor enters from the second port of the optical fiber circulator, passes through the third port of the optical fiber circulator, and enters the input arm of the optical fiber beam splitter. The reflected light of the optical fiber FP sensor is divided into n light signals by the optical fiber beam splitter and enters n spectrometers respectively. The n spectrometers simultaneously obtain electrical signals of n-channel reflection spectrum data of the optical fiber FP sensor, and each electrical signal is converted into a digital signal by an AD sampling circuit. Finally, the converted digital signal is sent to the FPGA controller for signal processing to obtain n-channel reflection spectrum data.

[0025] As a preferred embodiment, the present invention uses four spectrometers, each with a sampling frequency of 25 kHz. The number of sampling channels is 4 and the sampling rate is 100k.

[0026] Step 2: taking the data points corresponding to each channel of reflection spectrum data at the same moment as a first data set, and obtaining a plurality of first data sets at consecutive moments.

[0027] Specifically, since the sampling frequency of the spectrometer is 25kHz, that is, 25k data are collected in 1s, n spectrometers will generate n data points in 1 / 25k second. The n data points corresponding to n channels of reflected spectral data at the same time are taken as the first data set, and multiple groups of first data sets at consecutive moments are obtained.

[0028] Step 3, sequentially obtain the center point of each group of the first data set, and use the center point as the second data set.

[0029] Specifically, the center point is calculated using a numerical solution method and processed by the FPGA controller. , according to the distance formula between two points, write out the distance between n data points and the center point The distance and summation can obtain a binary function , find the binary function The minimum point of The coordinates of .

[0030] The numerical solution method is specifically as follows: assuming that the first data set includes n data points , followed by P 1 (x 1 ,y 1 ), P 2 (x 2 ,y 2 ), P 3 (x 3,y 3 )... P n-1 (x n-1 ,y n-1 ), P n (x n ,y n ), so that the center point To these n data points The sum of squares of the distances is minimal.

[0031] Define the center point To each data point The square of the distance is: ; (1) (1) In the formula, Indicates Point to center The distance , Is the center point The coordinates of , It is The coordinates of the data points; Set the center point With each data point Function of the distance for: ; (2) (2) In the formula, ∑ represents the sum, that is, n data points To center point The distances are summed; Use gradient descent to solve the function The minimum value of , the iterative formula of the gradient descent method is: ; (3) (3) In the formula, is the learning rate (step size), and They are functions Coordinates and coordinates The partial derivative of The partial derivative of is: ; (4) Substitute the partial derivative of 4 (formula) into the iterative formula (3) to obtain the center point Coordinates , .

[0032] Step 4: After performing Kalman filtering on the center point of the second data set, wavelength signal data is obtained, and the wavelength signal data is plotted into a waveform curve for output.

[0033] Specifically, the center point The FPGA controller processes the Kalman state prediction equation. When establishing the Kalman state prediction equation, the parameters of the Kalman filter can be configured, such as the state transfer matrix, observation matrix, noise covariance, etc. The parameters are adjusted according to the shape of the actual generated curve. The wavelength signal data is obtained after being processed by the FPGA controller, and is sent to the host computer software through the high-speed communication interface. The wavelength signal data is drawn into a waveform curve through the drawing library in the host computer software, and the waveform curve is displayed by the host computer.

[0034] Among them, the Kalman state prediction equation is: ; (5) (5) In the formula, yes The predicted value of the state at time is the state transition matrix, yes The optimal state estimate at time , is the control input matrix, yes Control input at any time.

[0035] Among them, the prediction error covariance is: ; (6) (6) In the formula, yes The forecast error covariance at time , yes The optimal estimation error covariance at time , yes The transposed matrix of is the process noise covariance.

[0036] Furthermore, in order to enable the collected spectral data to be used effectively for state estimation and prediction, the center point After being processed by the Kalman state prediction equation, the FPGA controller updates it through the state update equation to obtain accurate wavelength signal data.

[0037] Among them, the state update equation is: ; (7) (7) In the formula, yes The Kalman gain at time t, is the observation matrix, is the observation noise covariance, yes The optimal state estimate at time , yes The observed value at time, yes The optimal estimation error covariance at time , is the identity matrix.

[0038] The steps of simulation verification using 4 sampling channels in Example 1 are as follows: First, the mathematical model of the fiber FP sensor was constructed using MATLAB software to simulate its interference phenomenon and the generation of spectral signals. The simulation parameters were set as follows: the number of sampling channels was 4, the sampling rate was 100k, the frame rate of the spectrometer was 25kHz, the simulation time was 1s, and a 50Hz simplified sine wave signal was used to generate the simulated spectral signal.

[0039] Second, add time shift and random offset to the sine wave signal to automatically generate three other signals with time shift and offset, and obtain 4 sampling channel signals, such as Figure 2 shown.

[0040] Third, four sampling channel signals are introduced into the mathematical model, and the time shift and offset between the four sampling channels are simulated to generate a reflection spectrum data signal with noise.

[0041] Fourth, the reflectance spectrum data signal is discretized to convert it into data points, and the center point is calculated by taking a group of 4 data points collected by 4 sampling channels as a set. After Kalman filtering, each center point is processed and the processed waveform is plotted, as shown in the figure. Figure 3 As shown. Figure 2 , Figure 3 It can be seen that after Kalman filtering, the previous four waveform curves with time shift and offset have been processed into smooth waveform curves. Embodiment 2

[0042] Embodiment 2 provides a multi-channel sampling clutter processing system for an optical fiber FP sensor, based on a multi-channel sampling clutter processing method for an optical fiber FP sensor in Embodiment 1, the system comprises: A reflection spectrum data acquisition module is used to collect reflection spectrum signals of optical fiber FP sensors in parallel to obtain multi-channel reflection spectrum data; A first data set module, used to take the data points corresponding to each channel of reflection spectrum data at the same moment as a set of first data sets, and obtain multiple sets of first data sets at consecutive moments; A second data set module, used to sequentially obtain the center point of each group of first data sets, and use the center point as the second data set; The waveform curve output module is used to perform Kalman filtering on the center point of the second data set to obtain wavelength signal data, and draw the wavelength signal data into a waveform curve for output. Embodiment 3

[0043] Embodiment 3 provides 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 a multi-channel sampling clutter processing method for an optical fiber FP sensor as in Embodiment 1.

Claims

1. A multi-channel sampling clutter processing method for an optical fiber FP sensor, characterized in that: The method comprises the following steps: Step S1, collecting reflection spectrum signals of optical fiber FP sensors in parallel to obtain multi-channel reflection spectrum data; Step S2, taking the data points corresponding to each channel of the reflection spectrum data at the same moment as a set of first data sets, and obtaining a plurality of sets of the first data sets at consecutive moments; Step S3, sequentially obtaining the center point of each group of the first data set, and taking the center point as the second data set; Step S4, performing Kalman filtering on the center point of the second data set to obtain wavelength signal data, and drawing the wavelength signal data into a waveform curve for output.

2. The multi-channel sampling clutter processing method of a fiber FP sensor according to claim 1, characterized in that: The step S1, collecting the reflection spectrum signals of the optical fiber FP sensors in parallel to obtain multi-channel reflection spectrum data, is specifically: Broadband light is input into the optical fiber FP sensor from the first port of the optical fiber circulator, and the reflected light of the optical fiber FP sensor is output from the third port of the optical fiber circulator. The output reflected light enters a multi-channel spectrometer after beam splitting, and each spectrometer simultaneously obtains an electrical signal of the reflected spectrum data of the optical fiber FP sensor. The electrical signal of each channel is converted into a digital signal and sent to the FPGA controller for signal processing to obtain multi-channel reflected spectrum data.

3. The multi-channel sampling clutter processing method of a fiber FP sensor according to claim 1, characterized in that: In step S3, the center point of each group of the first data set is obtained, specifically: Define the center point To each data point of the first data set The square of the distance is: in, Indicates Point to center The distance , Is the center point The coordinates of , It is The coordinates of the data points; Set the center point With each data point Function of the distance for: ； Solve the function using iterative formula The minimum value of , the iterative formula is as follows: ； in, is the learning rate, and They are functions Coordinates and coordinates The partial derivative of The partial derivative of is: ； Substitute the partial derivative into the iterative formula to obtain the center point Coordinates , .

4. The multi-channel sampling clutter processing method of a fiber FP sensor according to claim 1, characterized in that: In step S4, the Kalman filter specifically includes: The center point The processing is performed by establishing the Kalman state prediction equation in turn. The Kalman state prediction equation is: ； in, yes The predicted value of the state at time is the state transition matrix, yes The optimal state estimate at time , is the control input matrix, yes Control input at all times; The forecast error covariance is: ； in, yes The forecast error covariance at time , yes The optimal estimation error covariance at time , yes The transposed matrix of is the process noise covariance.

5. The multi-channel sampling clutter processing method of a fiber FP sensor according to claim 4, characterized in that: The Kalman filter also includes: after establishing the Kalman state prediction equation for processing, the center point The update is performed by establishing a state update equation, which is: ； in, yes The Kalman gain at time t, is the observation matrix, is the observation noise covariance; yes The optimal state estimate at time , yes The observed value at time, yes The optimal estimation error covariance at time , is the identity matrix.

6. The multi-channel sampling clutter processing method of a fiber FP sensor according to claim 4, characterized in that: The Kalman state prediction equation is established through an FPGA controller.

7. The multi-channel sampling clutter processing method of a fiber FP sensor according to claim 5, characterized in that: The state update equation is established by the FPGA controller.

8. The multi-channel sampling clutter processing method of a fiber FP sensor according to claim 1, characterized in that: In the step S4, the wavelength signal data is plotted as a waveform curve for output, specifically: the FPGA controller sends the wavelength signal data to the host computer software, and the wavelength signal data is plotted as a waveform curve and displayed by the host computer software.

9. A system based on the multi-channel sampling clutter processing method of a fiber FP sensor according to any one of claims 1 to 8, characterized in that: include: A reflection spectrum data acquisition module is used to collect reflection spectrum signals of optical fiber FP sensors in parallel to obtain multi-channel reflection spectrum data; A first data set module, used to take the data points corresponding to each channel of the reflection spectrum data at the same moment as a set of first data sets, and obtain a plurality of sets of the first data sets at consecutive moments; A second data set module, used for sequentially acquiring the center point of each group of the first data sets, and taking the center point as the second data set; The waveform curve output module is used to perform Kalman filtering on the center point of the second data set to obtain wavelength signal data, and draw the wavelength signal data into a waveform curve for output.

10. An electronic device, characterized in that: It comprises at least one processor; and a memory in communication connection with 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 a multi-channel sampling clutter processing method for an optical fiber FP sensor as described in any one of claims 1 to 8.

Images