A two-dimensional weak vibration detector based on multi-core optical fiber and real-time analysis method

Abstract

The present invention relates to a two-dimensional weak vibration detector and real-time analysis method based on multi-core optical fiber, which belongs to the field of weak vibration measurement technology. After a beam of laser is emitted, it is reflected by a concave reflector connected to a vibration source, and the reflected light is split and coupled at the end faces of a single-core optical fiber and a four-core optical fiber respectively. When the vibration source drives the reflector to vibrate, the reflected light spot will be offset, thereby changing the coupling efficiency between light and optical fiber. The optical fiber output end is connected to a detector, and after detecting the size and distribution of the output light power, it is imported into a computer through a data acquisition card for automatic data collection and processing. In this experiment, the maximum coupling position of the light spot-single-core optical fiber is used as a reference to correct the measurement zero point, and the optical path is designed to cause the light spot to move through the reflector, and then static calibration is performed through the mapping relationship between the output light power and the displacement of the reflector, so as to obtain a quantitative relationship between the coupling efficiency and the position of the reflector, thereby realizing the real-time measurement of the displacement and frequency of dynamic vibration.

Description

Technical Field

[0001] The invention belongs to the technical field of weak vibration measurement, and specifically relates to a two-dimensional weak vibration detector based on multi-core optical fiber and a real-time analysis method. Background Art

[0002] Vibration is one of the most common phenomena in nature. From the Brownian motion of elementary particles to the eardrum and vocal cords of the human body, vibration is everywhere. In recent years, with the rapid development of micro-machining technology, the precise measurement of tiny deformations and tiny vibrations of objects has attracted more and more attention. The application of weak vibration measurement is very extensive. For example, in industrial production, weak vibration can be used to monitor the operating status of mechanical equipment in real time, and timely discover problems such as bearing failure, imbalance and looseness, effectively ensuring the reliability and stability of the equipment, which is of great significance for improving production efficiency and product quality. In the field of construction engineering, weak vibration measurement can be used to monitor and evaluate the vibration performance of engineering structures such as buildings and bridges, analyze vibration data to obtain the natural frequency, modal shape and vibration amplitude of the structure, prevent and discover hidden dangers, and ensure public safety. In the field of scientific research, especially in the fields of semiconductor manufacturing, optical instruments, aerospace, etc., precision equipment is very sensitive to vibration, and the results of some repetitive experiments also have very strict requirements on vibration perturbations. Therefore, the development and innovation of weak vibration detection methods have important application value.

[0003] At present, the main instruments for measuring vibration include piezoelectric vibration sensors, magnetoelectric vibration sensors, and Doppler radar measuring instruments. Among many methods, optical fiber vibration sensing technology has high sensitivity, wide frequency response range, fast response speed, anti-electromagnetic interference, and long transmission distance, and has broad application prospects in various fields. According to investigations, although the phase modulation and wavelength modulation sensors that are currently widely used have high accuracy, these methods generally have the disadvantages of complex structure, high demodulation difficulty, and susceptibility to other factors. In contrast, the principle of intensity modulation optical fiber vibration sensor is simple and intuitive, the demodulation difficulty is relatively low, and it is not easily affected by other factors. Therefore, the present invention will start with intensity modulation to design and manufacture a high-sensitivity, real-time imaging two-dimensional weak vibration optical fiber sensor.

[0004] Vibration generally refers to periodic motion, which has multiple physical characteristic parameters such as frequency (period), amplitude, displacement, velocity and acceleration, and these physical quantities are interrelated. This experiment chooses to use frequency and displacement to reflect vibration. Since weak vibrations are difficult to observe and measure, it is necessary to establish the relationship between vibration frequency and displacement and other easy-to-measure physical quantities to make the characterization of vibration simple and easy. Fiber optic sensing technology has a variety of intensity modulation methods and has high sensitivity. In view of this, this experiment starts from the properties of the fiber end face, in order to establish a quantitative relationship between coupling efficiency and displacement through the change in coupling efficiency caused by the offset of the light spot on the fiber end face, thereby realizing indirect measurement of vibration. Summary of the invention

[0005] In view of the shortcomings of the prior art, the technical problem to be solved by the present invention is to provide a two-dimensional weak vibration detector and a real-time analysis method based on multi-core optical fiber, taking the maximum coupling position of the light spot-single-core optical fiber as the reference to correct the measurement zero point, designing the optical path to cause the light spot to move through the reflector, and then performing static calibration through the mapping relationship between the output light power and the reflector displacement, so as to obtain the quantitative relationship between the coupling efficiency and the reflector position, thereby realizing the real-time measurement of the displacement and frequency of dynamic vibration.

[0006] In order to solve the above technical problems, the technical solution adopted by the present invention is: a two-dimensional weak vibration detector based on multi-core optical fiber, including a laser, an aperture, a single-core optical fiber jumper, a four-core optical fiber jumper, a reflector, a first beam splitter, a second beam splitter, a detector, a vibration source and an information processing system, wherein the laser, the aperture, the first beam splitter, the second beam splitter and the reflector are arranged in sequence from left to right;

[0007] The laser is used to emit a laser beam, which passes through an aperture, a first beam splitter, a second beam splitter and a reflector in sequence, and a reflected light beam reflected from the reflector passes through the second beam splitter and the first beam splitter in sequence, and the reflected light beam is divided into a first light beam and a second light beam when passing through the second beam splitter, and the first light beam is coupled to the end face of a four-core optical fiber jumper, and the second light beam is coupled to the end face of a single-core optical fiber jumper after passing through the first beam splitter;

[0008] The aperture is used to filter stray light; the vibration source is used to provide vibration to the reflector; the output ends of the single-core fiber optic jumper and the four-core fiber optic jumper are both connected to the detector, and the information processing system is used to collect and process the voltage change after the photocurrent in the detector flows through the large resistor. When the information processing system moves through the reflector, the reflected light beam will be offset, causing the coupling efficiency of the single-core fiber optic jumper and the four-core fiber optic jumper to decrease, resulting in a change in the voltage signal received by the information processing system, thereby achieving real-time detection and analysis of vibration.

[0009] Further, the single-core optical fiber jumper is a quartz optical fiber with an operating band of 200-1100nm, a core diameter of 400μm, a cladding diameter of 440μm, a coating diameter of 700μm, a numerical aperture of 0.22, and a light receiving angle of 25.4°;

[0010] The material of the core in the single-core optical fiber jumper is pure quartz high hydroxyl, the material of the cladding in the single-core optical fiber jumper is fluorine-doped quartz, and the material of the coating in the single-core optical fiber jumper is double-layer ultraviolet curing acrylic resin.

[0011] Furthermore, the operating band of the four-core optical fiber patch cord is 650 nm, the total diameter of the four-core optical fiber patch cord is 1300 μm, and the diameter of a single fiber core in the four-core optical fiber patch cord is 500 μm.

[0012] Furthermore, the detector includes a photodetector, four PIN photodiodes and a data acquisition card. The photodetector and the four PIN photodiodes are respectively connected to the data acquisition card. The photodetector is connected to a single-core optical fiber jumper. The four PIN photodiodes are connected to a four-core optical fiber jumper. The four PIN photodiodes are all optical fiber-coupled visible light silicon PIN photodiodes with an operating band of 400-1100nm.

[0013] Furthermore, the vibration conducting rod comprises a rod body and a magnetic base, two ends of the rod body are respectively connected to the back side of the lens sleeve and the surface of the magnetic base, and the magnetic base is adsorbed on the surface of the vibration source.

[0014] Furthermore, the photodetector is a biased silicon photodiode with an operating band of 320-1100 nm.

[0015] A two-dimensional weak vibration real-time analysis method based on multi-core optical fiber comprises the following steps:

[0016] S1, emit a laser beam toward the reflector;

[0017] S2, the reflected laser reflected by the reflector is divided into a first light beam and a second light beam through a beam splitter, the first light beam is coupled with the end face of the four-core optical fiber, and the second light beam is coupled with the end face of the single-core optical fiber through another beam splitter;

[0018] S3. Collect the coupling efficiency of the single-core fiber end face and the four-core fiber end face. When vibration occurs, the coupling efficiency changes caused by the offset of the light spot on the single-core fiber end face and the four-core fiber end face, establish a quantitative relationship between the coupling efficiency and the displacement, and achieve indirect measurement of the vibration.

[0019] Furthermore, the fiber coupling efficiency is:

[0020] ; In the above formula: is the fiber mode field radius; It is the radial distance from any point on the end face of the optical fiber to the center; is the focal length of the coupling lens; is the wavelength; is the coupled aperture diaphragm; is the wave number; x is the offset distance; y is the coupling efficiency; J1 is the Legendre polynomial; i is the imaginary unit; x offset It is out of position.

[0021] Furthermore, in step S3, a Long Short-Term Memory (LSTM) model is used to perform time series modeling and prediction on the voltage signals at the end faces of the single-core optical fiber and the four-core optical fiber. The voltage signal decreases as the optical fiber coupling efficiency decreases.

[0022] Furthermore, in step S3, fast Fourier transform is used to process the collected voltage to obtain the vibration frequency.

[0023] Compared with the prior art, the present invention has the following beneficial effects:

[0024] 1. First, the application scenarios are wide and the expansibility is strong: This experiment jumps out of the conventional application scenario where multi-core optical fiber is only used for data transmission. It uses four-core optical fiber combined with LSTM deep learning algorithm to design and build a two-dimensional weak vibration detection system that can be imaged in real time. This system can be used to monitor the operating status of mechanical equipment in real time, monitor and evaluate the vibration performance of engineering structures such as buildings and bridges, and detect vibration perturbations of scientific research-level precision equipment. The experimental method based on this experiment can realize one- to three-dimensional weak vibration detection and measurement of other physical quantities related to optical fiber coupling efficiency (such as concentration, temperature, etc.).

[0025] 2. The second aspect is that it is easy to operate, intelligent and efficient, and easy to promote and apply: This experiment uses a concave reflector to replace the traditional plane reflector to realize the two-dimensional movement of the laser spot. At the same time, the addition of four-core optical fiber combined with the LSTM deep learning algorithm solves the problem of weak two-dimensional random vibration measurement; the reflector-vibration source clamping and transmission device is independently designed to realize the in-situ detection of integrated calibration and measurement, and enhance the practicality and innovation of the device; the maximum coupling position of the single-core optical fiber is used as the reference for optical path calibration and participates in the synchronous calibration of the four-core optical fiber, and the disturbance in the acquisition process is corrected. With the independently developed static calibration system and real-time acquisition imaging system, it has a high degree of automation, fast response speed and high precision. The frequency response range of the device is 0~2500Hz, the effective response range of the amplitude is 0.400mm, and the percentage accuracy is ±2.0%FS. Compared with phase modulation and wavelength modulation sensors, the present invention has the advantages of simple structure, easy demodulation, and not easily affected by other factors.

[0026] Thirdly, the experiment has rich content and good teaching effect: it involves knowledge of optics, acoustics, electricity, sensing, computer programming, neural networks, artificial intelligence, etc.; based on the fiber coupling efficiency theory, it can realize accurate measurement of one-dimensional to three-dimensional weak vibrations. The measurement principle is novel and the phenomenon is intuitive and visible, breaking the "black box" limitation. It can deepen students' understanding of physical processes and provide a novel and interesting experimental scheme for the theoretical teaching of fiber optic sensors. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a three-dimensional structural schematic diagram of a two-dimensional weak vibration detector based on multi-core optical fiber in the present invention;

[0028] Figure 2 It is a schematic diagram of the principle structure of a two-dimensional weak vibration detector based on multi-core optical fiber in the present invention;

[0029] Figure 3 It is a schematic diagram of the three-dimensional structure of the self-centering lens holder in the present invention;

[0030] Figure 4 It is a partial cross-sectional structural schematic diagram of the vibration source, the vibration conducting rod and the lens sleeve after being matched in the present invention;

[0031] Figure 5 For the present invention Figure 4 A schematic diagram of the local enlarged structure at point A in the middle;

[0032] Figure 6 is a circuit diagram of the detector bias and data acquisition circuit in the present invention;

[0033] Figure 7 It is a configuration diagram of the data acquisition part in the present invention;

[0034] Figure 8 It is a configuration diagram of the data visualization part in the present invention;

[0035] Fig. 9 is a curve of the relationship between optical fiber coupling efficiency and offset distance in the present invention;

[0036] Fig.10 is a diagram of main parameters of the absolute maximum ratings of the PIN photodiode in the present invention;

[0037] Fig.11 Graph showing the main parameters of the photoelectric performance of the PIN photodiode in the present invention.

[0038] Markings in the figure: 1. Laser; 2. Aperture; 3. Single-core fiber optic patch cord; 4. Four-core fiber optic patch cord; 5. Reflector; 6. First beam splitter; 7. Second beam splitter; 8. Detector; 81. Photodetector; 82. PIN photodiode; 83. Data acquisition card; 9. Vibration source; 91. Vibration transmission rod; 911. Rod body; 912. Magnetic base; 92. Three-dimensional adjustment frame; 93. Self-centering lens bracket; 931. Claw; 932. Connector; 94. Lens sleeve; 941. Retaining ring; 10. Information processing system. DETAILED DESCRIPTION

[0039] In order to make the above features and advantages of the present invention more obvious and easy to understand, embodiments are given below with reference to the accompanying drawings for detailed description as follows.

[0040] like Figure 1-Figure 6 As shown, this embodiment provides a two-dimensional weak vibration detector based on multi-core optical fiber, including a laser 1, an aperture 2, a single-core optical fiber jumper 3, a four-core optical fiber jumper 4, a reflector 5, a first beam splitter 6, a second beam splitter 7, a detector 8, a vibration source 9, a vibration conduction rod 91, a three-dimensional adjustment frame 92, a self-centering lens bracket 93, a lens sleeve 94 and an information processing system 10.

[0041] The laser 1, the aperture 2, the first beam splitter 6, the second beam splitter 7 and the reflector 5 are arranged in sequence from left to right; specifically, the laser 1 adopts a red light laser 1 with a wavelength of 650nm and is powered by its matching first power supply; the aperture 2 is used to filter stray light; the first beam splitter 6 and the second beam splitter 7 adopt a 50:50 beam splitting prism, the first beam splitter 6 and the second beam splitter 7 are installed on an optical breadboard and supported by an optical bracket, the optical breadboard and the optical bracket adopt the structure commonly used in the prior art, which will not be repeated here.

[0042] The reflector 5 is a silver-plated concave reflector 5 with a diameter of ∅25.4 and f101.6. The reflector 5 is connected to the vibration source 9 through a vibration conducting rod 91. Specifically, the bottom of the self-centering lens bracket 93 is connected to the three-dimensional adjustment bracket 92. The reflector 5 is mounted on a lens sleeve 94. The lens sleeve 94 is located at the positioning center of the self-centering lens bracket 93. The self-centering lens bracket 93 is a self-centering lens bracket 93 of the three-claw 931 type commonly used in the prior art. The self-centering lens bracket 93 used in this embodiment has three movable claws 931 arranged at annular intervals on its inner circumference, and a connecting piece 932 for limiting the movement of the claws 931. As long as the self-centering lens bracket 93 can realize the clamping, centering and locking functions, whether or not any other structure is used, it is the same as the self-centering lens bracket 93 described in this embodiment. The vibration conduction rod 91 includes a rod body 911 and a magnetic base 912. The two ends of the rod body 911 are respectively connected to the back side of the lens sleeve 94 and the surface of the magnetic base 912. The magnetic base 912 is adsorbed on the surface of the vibration source 9. The vibration source 9 adopts a magnetic metal box with a built-in Bluetooth speaker, such as an iron box, and is also equipped with a control software for controlling the Bluetooth speaker. The Bluetooth speaker control software is downloaded to a mobile phone, and the Bluetooth speaker is controlled by the mobile phone.

[0043] The steps for installing and measuring the reflector 5 are as follows: install the reflector 5 into the lens sleeve 94, fix the two ends with the SM1 retaining ring 941, and clamp the assembled lens sleeve 94 on the self-centering lens holder, as shown in FIG. Figure 1 and Figure 3 As shown, by adjusting the rotating shaft of the three-dimensional adjustment frame 92, the movement of the reflector 5 is simulated, and the voltage collected after the movement is calibrated; on the basis of static calibration, the adjustment frame is calibrated to the center position of the calibration, wherein the magnetic base 912 (M6 screw NdFeB magnetic base 912) is used to adsorb the vibration source 9, and the adapter end of the rod body 911 is connected to the lens sleeve 94 through a thread to achieve connection with the reflector 5. The rod body 911 is made of 7075 aluminum alloy that is not easy to deform and is rigidly connected through threads. The magnetic base 912 connected to the vibration source 9 has a vertical suction force of 70 pounds, so the vibration process can be approximately regarded as a rigid connection. After the rod is installed in the clamped state and the vibration source 9 is adsorbed, the self-centering lens bracket 93 is loosened and locked with a hexagonal wrench to perform measurement.

[0044] The single-core optical fiber jumper 3 is located on one side of the emission surface of the first beam splitter 6. Specifically, the single-core optical fiber jumper 3 is a quartz optical fiber with an operating band of 200-1100nm, a core diameter of 400μm, a cladding diameter of 440μm, a coating diameter of 700μm, a numerical aperture of 0.22, and a light receiving angle of 25.4°; the material of the core in the single-core optical fiber jumper 3 is pure quartz high hydroxyl, the material of the cladding in the single-core optical fiber jumper 3 is fluorine-doped quartz, and the material of the coating in the single-core optical fiber jumper 3 is a double-layer UV-cured acrylic resin, and the core optical fiber jumper has SMA905 connectors at both ends, good coupling, high light transmittance, good fatigue resistance, and is suitable for long-term operation of the sensor.

[0045] At present, the application scenarios of multi-core optical fibers are mainly concentrated in data transmission. By integrating multiple independent optical channels into one optical fiber, the transmission efficiency is significantly improved to meet the needs of high-density, large-capacity and long-distance transmission in data centers. Therefore, multi-core optical fibers currently work in communication bands such as C and L. This embodiment jumps out of the conventional application scenarios of multi-core optical fibers, and instead uses the coupling properties of its end face that are different from conventional optical fibers for vibration detection. The four-core optical fiber jumper 4 is located on one side of the emission surface of the second beam splitter 7. Specifically, the operating band of the four-core optical fiber jumper 4 is 650nm, the total diameter of the four-core optical fiber jumper 4 is 1300μm, the single core diameter of the four-core optical fiber jumper 4 is 500μm, and the four-core optical fiber jumper 4 has an FC / PC connector.

[0046] The output ends of the single-core optical fiber jumper 3 and the four-core optical fiber jumper 4 are both connected to the detector 8. Specifically, the detector 8 includes a photodetector 81, four PIN photodiodes 82 and a data acquisition card 83. The photodetector 81 and the four PIN photodiodes 82 are connected to the data acquisition card 83. The photodetector 81 is connected to the single-core optical fiber jumper 3, and the four PIN photodiodes 82 are connected to the four-core optical fiber jumper 4. The four PIN photodiodes 82 are optical fiber-coupled visible light silicon PIN diodes with an operating band of 400-1100nm.

[0047] The working principle of the four PIN photodiodes 82 in this embodiment is as follows: When light shines on the PN junction on the photodiode, the photons are absorbed by the atoms or molecules in the depletion layer to generate electron-hole pairs. Under the action of the built-in electric field, the electrons move to the N region and the holes move to the P region to form a photocurrent. During the operation of the photodiode, it is usually necessary to apply a reverse bias voltage to it to further enhance the effect of the built-in electric field so that the photoelectric potential is more obvious. At the same time, the reverse bias voltage can also suppress the dark current and improve the sensitivity and signal-to-noise ratio of the photodiode. In this embodiment, a 4.5V reverse bias voltage is used.

[0048] The main parameters of the four PIN photodiodes 82 are as follows:

[0049] Absolute Maximum Ratings (Tc = 25 ° C): Fig.10 shown.

[0050] Optical-Electro Characteristics(Tc=25℃): Fig.11 shown.

[0051] It should be noted that the rise / fall time of the photocurrent is only 3.0ns, which is much higher than the acquisition interval of the data acquisition card 83. Therefore, the upper limit of the frequency response range of the detector 8 in this embodiment is mainly limited by the acquisition rate of the data acquisition card 83 and the data processing speed of the host computer.

[0052] like Figure 6 As shown, this embodiment also provides a detector 8 bias and data acquisition circuit, firstly, four PIN photodiodes 82 are connected to four 1M The resistors are connected in series, and the resistor voltage is collected using the data acquisition card 83. At the same time, in order to eliminate the common mode deviation, the bias voltage applied by the data acquisition card 83 and the photodiode is treated as a common ground.

[0053] The information processing system 10 is connected to the data acquisition card 83. In this embodiment, the information processing system 10 is a computer containing a LabVIEW program and a Python program.

[0054] like Figure 1-Figure 2 As shown, Figure 2 The middle arrows represent each light beam. In this embodiment, the emission direction of the laser 1 is used as the main optical axis. The laser emitted by the laser 1 is converged by its built-in lens, and after passing through the aperture 2 to filter out stray light, it is first divided into two beams by the first beam splitter 6. One beam is emitted to the right side of the optical axis (not used), and the other beam (laser beam) is firstly divided by the second beam splitter 7 in the same way as the first beam splitter 6. The laser beam is reflected (reflected beam) after reaching the concave mirror, and the reflected beam passes through the second beam splitter 7. When passing through the second beam splitter 7, the reflected beam is divided into the first beam and the second beam. The first beam is coupled to the end face of the four-core fiber jumper 4, and the second beam passes through the first beam splitter 6 again and is coupled to the end face of the single-core fiber jumper 3. The photodetector 81 is connected to the single-core fiber jumper 3, and the four PIN photodiodes 82 are connected to the four-core fiber jumper 4. After the photocurrent flows through the large resistor, the voltage is collected by the data acquisition card 83 and transmitted to the computer for processing. When the reflector 5 moves (that is, vibrates), the reflected light spot will shift and the fiber coupling efficiency will decrease, which is reflected as a decrease in the collected voltage. Then, the real-time detection of the vibration is achieved through the processing of the LabVIEW program and the Python program.

[0055] This embodiment also provides a specific installation and operation steps of a two-dimensional weak vibration detector based on a multi-core optical fiber, as shown below:

[0056] Step 1: Adjust the light path alignment. Turn on the first power supply and adjust the output voltage to 5V. After the laser 1 is started for about 30 minutes, the laser 1 enters a relatively stable working state. Adjust the laser 1 to align the output light, place the concave reflector 5 in the clamping position and adjust the three-dimensional adjustment frame 92 so that the reflected light spot and the light outlet of the aperture 2 coincide. At this time, the laser has irradiated the center position of the concave reflector 5, and the light path is aligned. After that, adjust the two-dimensional adjustment frames of the single-core fiber jumper 3 and the four-core fiber jumper 4 respectively, so that the coupling efficiency of the single-core fiber jumper 3 is maximized, and the light spot is at the end face center of the four-core fiber jumper 4. At this time, the coupling efficiency of the four fiber cores is similar.

[0057] Step 2 uses the LabVIEW program to perform static calibration and construct a database. (1) Record the voltage of each channel at the center position and the X and Y axis coordinates of the concave reflector 5 and the three-dimensional adjustment frame 92. (2) Rotate the three-dimensional adjustment frame 92 along the negative direction of the Y axis until the two channel collection voltages of the four-core fiber jumper 4 are less than 10mV, eliminate the idle error, and record the current position and the voltage of each channel; rotate 0.010mm at a time in the positive direction, and record the current position and the voltage of each channel until the two channel voltages of the four-core fiber jumper 4 are less than 10mV again. (3) Rotate the three-dimensional adjustment frame 92 along the negative direction of the X axis by 0.010mm, and repeat the above step (2) until the voltage values ​​of the four channels of the four-core fiber jumper 4 are close to 10mV. (4) Re-adjust the three-dimensional adjustment frame 92 to the initial position. After eliminating the idle error, rotate the three-dimensional adjustment frame 92 in the positive direction of the X axis by 0.010mm, and the remaining steps are the same as step (3). (5) Repeat the second calibration in the opposite direction of step (2), step (3), and step (4), and average the two calibration data to generate a database, which is imported into LSTM deep learning to construct a voltage-position relationship model.

[0058] Step 3: Install the vibration transmission rod 91 and the vibration source 9. Install the rod body 911 on the back of the concave reflector 5, and adsorb the magnetic base 912 at the other end on the surface of the vibration source 9. After ensuring that the reflected light spot still coincides with the light outlet of the aperture 2, loosen the three claws 931 of the self-centering lens bracket 93 to disengage the three claws 931 from the concave reflector 5, and use a hexagonal wrench to lock the connector 932 to limit the movement of the claws 931.

[0059] Step 4: Dynamic weak vibration measurement. Turn on and connect the Bluetooth speaker in the vibration source 9, use the mobile phone software Phyphox to play audio with adjustable frequency, amplitude (volume) and waveform, wait for the vibration to stabilize, start the Python real-time imaging program, observe the vibration frequency, and the components of the vibration on the X and Y axes, and select a path to save the data. After the measurement is completed, re-lock the reflector 5 on the self-centering lens bracket 93 and repeat the above operation in step 4. Change the waveform, frequency, amplitude, etc. of the played audio, re-measure, and save the data.

[0060] Step 5: Use LabVIEW to write a calibration system. First, perform data acquisition and calibration. Since the voltage value collected instantaneously by the data acquisition card 83 will fluctuate within a small range, in order to obtain a steady-state value, this embodiment uses a method of continuously measuring a certain number of times and taking the average value to obtain calibration data. Figure 7-Figure 8 The code in can realize the setting of initialization parameters such as the number of sampling points, physical channels and sampling frequency, while meeting the voltage output requirements. Figure 7 Configure the acquisition part and select the appropriate number and mode of physical channels according to the needs, and then configure the input terminals, sampling mode, sampling frequency and sampling clock to realize data acquisition. The main parameters that need to be configured are: Physical channel: This acquisition card contains 8 reference single-ended ground channels and 4 differential channels; Input terminal configuration: Here we configure it as single-ended ground (RSE); Sampling mode: Here we choose the continuous sampling mode - "Continuous"; Sampling clock: We choose the clock of the host computer - "OnboardClock"; Sampling frequency: In order to meet the requirements, we choose the highest frequency allowed by this acquisition card as 1000. Figure 8 In the configuration data visualization part, the collected multi-channel data is stored and processed by channel, and the original waveform is displayed first using the waveform chart control; considering the high accuracy of the data acquisition card, the collected data will have a certain fluctuation. After collecting the voltage of each channel for a certain period of time and the collected data points, the average value is calculated, and then the waveform chart control is also used for display. Secondly, the vibration spectrum diagram is obtained by applying Fourier transform. In order to simplify the code and improve portability, this part of the code in this embodiment is mainly written using express VI and DAQ assistant in LabVIEW. With the help of this program, the dynamic measurement results can be observed on the front panel, that is, the original waveform collected when the vibration is applied and its Fourier transform graph. After the Fourier transform, the relationship between the peak value and the frequency of the signal can be observed.

[0061] Step 6: Use Python to write a real-time acquisition imaging system. Use Python to write a control panel to set initialization parameters such as voltage observation of each channel, sampling frequency control, filter cutoff frequency, and data storage path, and embed the LSTM deep learning model to realize real-time acquisition and imaging of vibration displacement and frequency.

[0062] This embodiment also provides a two-dimensional weak vibration real-time analysis method based on multi-core optical fiber, which is implemented by using the above-mentioned two-dimensional weak vibration detector based on multi-core optical fiber, and includes the following steps:

[0063] S1, emitting a laser beam toward the reflector 5;

[0064] S2, the reflected laser reflected by the reflector 5 is divided into a first light beam and a second light beam through a beam splitter, the first light beam is coupled with the end face of the four-core optical fiber, and the second light beam is coupled with the end face of the single-core optical fiber through another beam splitter;

[0065] S3. Collect the coupling efficiency of the single-core fiber end face and the four-core fiber end face. When vibration occurs, the coupling efficiency changes caused by the offset of the light spot on the single-core fiber end face and the four-core fiber end face, establish a quantitative relationship between the coupling efficiency and the displacement, and achieve indirect measurement of the vibration.

[0066] The specific deduction steps of the quantitative relationship between the coupling efficiency and the displacement in step S3 are as follows:

[0067] The electromagnetic field on the end face of an optical fiber can generally be approximated by a Gaussian distribution:

[0068]

[0069] In the formula, is the fiber mode field radius, It is the radial distance from any point on the end face of the optical fiber to the center.

[0070] Assuming that the light wave reaches the receiving aperture from infinity, it can be regarded as a plane wave (parallel light), and the electromagnetic field distribution at the focus of the receiving lens is:

[0071]

[0072] In the formula, is the focal length of the coupling lens; is the wavelength, is the coupled aperture diaphragm; is the wave number.

[0073] To maximize the coupling efficiency, two different light fields need to match. The simplest form is that the two light fields are exactly the same. If this requirement cannot be met, the two light fields need to be as similar as possible. In short, it is to compare the focus Airy disk and the fiber end face mode field to minimize the difference between the Airy spot diffraction light field and the fiber end face.

[0074] The fiber coupling efficiency is defined as the ratio of the optical power coupled into the fiber to the optical power on the receiving plane of the coupling system, which can be written as:

[0075]

[0076] therefore,

[0077]

[0078] When the center of the light spot is offset, the electromagnetic field distribution on the fiber end face is:

[0079]

[0080] At this time, the fiber coupling efficiency can be written as:

[0081]

[0082] is the fiber mode field radius, is the radial distance from any point on the fiber end face to the center, is the focal length of the coupling lens; is the wavelength, is the coupled aperture diaphragm; is the wave number, x is the offset distance, y is the coupling efficiency, J1 is the Legendre polynomial, i is the imaginary unit, x offset It is out of position.

[0083] The attenuation curve is similar to a Gaussian curve and can be approximated by a Gaussian curve, such as Fig. 9 shown.

[0084] In step S3, the Long Short-Term Memory model is used to model and predict the voltage signals of the single-core fiber end face and the four-core fiber end face in time series. The voltage signal decreases as the fiber coupling efficiency decreases. By introducing the LSTM model, the historical data of the vibration signal can be combined with the current input to make a high-precision prediction of the displacement of the vibration source. The specific principle of LSTM is as follows: LSTM is a special recurrent neural network that can process and predict long dependencies based on time series, solving the problem that traditional RNN is difficult to remember long time interval information. Its core lies in the ability to control the inflow, retention and output process of information, and is suitable for processing data with time series characteristics, such as the voltage signals received at equal time intervals and the real-time position of the movement in this embodiment.

[0085] In step S3, fast Fourier transform is used to process the collected voltage to obtain the vibration frequency. In this embodiment, fast Fourier transform is used to process the collected voltage to obtain the vibration frequency. The principle of fast Fourier transform is as follows: Fast Fourier transform (FFT) is an efficient algorithm for calculating discrete Fourier transform (DFT) and its inverse transform. The most famous FFT algorithm was proposed by Cooley-Tukey in 1965. The key idea of ​​FFT is to decompose the original DFT into smaller DFTs to utilize repeated calculations and reduce the overall computational complexity. FFT is a basic tool in fields such as signal processing, image processing, and audio analysis because it can quickly convert time domain signals to frequency domain representations, thereby providing the frequency composition and phase information of the signal.

[0086] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for illustrating 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 shall 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 two-dimensional weak vibration detector based on multi-core optical fiber, characterized in that: It includes a laser, an aperture, a single-core optical fiber jumper, a four-core optical fiber jumper, a reflector, a first beam splitter, a second beam splitter, a detector, a vibration source and an information processing system, wherein the laser, the aperture, the first beam splitter, the second beam splitter and the reflector are arranged in sequence from left to right; The laser is used to emit a laser beam, which passes through an aperture, a first beam splitter, a second beam splitter and a reflector in sequence, and a reflected light beam reflected from the reflector passes through the second beam splitter and the first beam splitter in sequence, and the reflected light beam is divided into a first light beam and a second light beam when passing through the second beam splitter, and the first light beam is coupled to the end face of a four-core optical fiber jumper, and the second light beam is coupled to the end face of a single-core optical fiber jumper after passing through the first beam splitter; The aperture is used to filter stray light; the vibration source is used to provide vibration to the reflector; The output ends of the single-core fiber optic patch cord and the four-core fiber optic patch cord are both connected to the detector. The information processing system is used to collect and process the voltage change after the photocurrent in the detector flows through the large resistor. When the information processing system moves through the reflector, the reflected light beam will be offset, causing the coupling efficiency of the single-core fiber optic patch cord and the four-core fiber optic patch cord to decrease, resulting in a change in the voltage signal received by the information processing system, thereby achieving real-time detection and analysis of vibration.

2. A two-dimensional weak vibration detector based on multi-core optical fiber according to claim 1, characterized in that: The single-core optical fiber jumper is a quartz optical fiber with an operating band of 200-1100nm, a core diameter of 400μm, a cladding diameter of 440μm, a coating diameter of 700μm, a numerical aperture of 0.22, and a light receiving angle of 25.4°; The material of the core in the single-core optical fiber jumper is pure quartz high hydroxyl, the material of the cladding in the single-core optical fiber jumper is fluorine-doped quartz, and the material of the coating in the single-core optical fiber jumper is double-layer ultraviolet curing acrylic resin.

3. A two-dimensional weak vibration detector based on multi-core optical fiber according to claim 1, characterized in that: The operating wavelength band of the four-core optical fiber jumper is 650 nm, the total diameter of the four-core optical fiber jumper is 1300 μm, and the diameter of a single fiber core in the four-core optical fiber jumper is 500 μm.

4. The two-dimensional weak vibration detector based on multi-core optical fiber according to claim 1, characterized in that: The detector also includes a vibration transmission rod, a three-dimensional adjustment frame, a self-centering lens holder and a lens sleeve. The vibration transmission rod is used to transmit the vibration emitted by the vibration source to the reflector. The self-centering lens holder is connected to the three-dimensional adjustment frame at the bottom, and the reflector is installed on the lens sleeve. The lens sleeve is located at the positioning center of the self-centering lens holder.

5. A two-dimensional weak vibration detector based on multi-core optical fiber according to claim 4, characterized in that: The vibration conducting rod comprises a rod body and a magnetic base, two ends of the rod body are respectively connected to the back side of the lens sleeve and the surface of the magnetic base, and the magnetic base is adsorbed on the surface of the vibration source.

6. A two-dimensional weak vibration detector based on multi-core optical fiber according to claim 1, characterized in that: The detector includes a photodetector, four PIN photodiodes and a data acquisition card. The photodetector and the four PIN photodiodes are respectively connected to the data acquisition card. The photodetector is connected to a single-core optical fiber jumper. The four PIN photodiodes are connected to a four-core optical fiber jumper. The four PIN photodiodes are all optical fiber-coupled visible light silicon PIN photodiodes with an operating band of 400-1100nm.

7. A two-dimensional weak vibration real-time analysis method based on multi-core optical fiber, characterized by: The following steps are involved: S1, emit a laser beam toward the reflector; S2, the reflected laser reflected by the reflector is divided into a first light beam and a second light beam through a beam splitter, the first light beam is coupled with the end face of the four-core optical fiber, and the second light beam is coupled with the end face of the single-core optical fiber through another beam splitter; S3. Collect the coupling efficiency of the single-core fiber end face and the four-core fiber end face. When vibration occurs, the coupling efficiency changes caused by the offset of the light spot on the single-core fiber end face and the four-core fiber end face, establish a quantitative relationship between the coupling efficiency and the displacement, and achieve indirect measurement of the vibration.

8. The method for real-time analysis of two-dimensional weak vibrations based on multi-core optical fiber according to claim 7, characterized in that: In step S3, the Long Short-Term Memory model is used to perform time series modeling and prediction on the voltage signals at the end faces of the single-core optical fiber and the four-core optical fiber. The voltage signal decreases as the optical fiber coupling efficiency decreases.

9. The method for real-time analysis of two-dimensional weak vibration based on multi-core optical fiber according to claim 7, characterized in that: In step S3, fast Fourier transform is used to process the collected voltage, thereby obtaining the vibration frequency.

Images