Extreme learning machine assisted photoelectric oscillator magnetic field and temperature measuring device and method

Abstract

The invention relates to the technical field of measuring devices, in particular to an extreme learning machine assisted photoelectric oscillator magnetic field and temperature measuring device and method, and can solve the problems of poor stability and slow response speed of a magnetic field and temperature double-parameter measuring system in the existing measuring technology to a certain extent. According to the device, the OEO is used as a resonant cavity, the magnetostrictive material is pasted on the optical fiber, material stretching caused by an external magnetic field can be conducted to the optical fiber, FSR of the MZI is influenced, then the center frequency of the microwave photonic filter is influenced, magnetic field and temperature measurement is effectively converted into monitoring of oscillation signal frequency of the OEO, the measurement precision is improved, and the measurement accuracy is improved. By monitoring the center frequency and other mode frequencies of the OEO oscillation signal at the same time and demodulating through ELM, double-parameter measurement can be achieved under the simple and compact hardware condition, so that the problem of temperature cross sensitivity in magnetic field sensing is solved, and the measurement system has high system stability.

Description

Technical Field

[0001] The present application relates to the technical field of measuring devices, and in particular to an extreme learning machine-assisted optoelectronic oscillator magnetic field and temperature measurement device and method. Background Art

[0002] Due to the rapid development of applications in various fields such as magnetic resonance imaging, marine resource exploration, thermomagnetic material processing, navigation, geophysics, etc., the demand for high-precision synchronous detection of magnetic fields is increasing. Magnetic field and temperature are usually inseparable in measurement: since the generation of magnetic field is often accompanied by a large amount of heat, the magnetic field measurement process is easily affected by heat and radioactive interference in the measurement environment and wiring problems, which will limit the practical application of magnetometers under some special conditions such as small or high temperature and high humidity spaces.

[0003] In order to ensure high-precision magnetic field measurement, it is necessary to deploy temperature sensors to obtain temperature changes in real time, and combine decoupling compensation temperature to achieve accurate measurement of magnetic field. At the same time, in order to effectively ensure the operational safety of industrial applications and realize real-time monitoring and early warning, it is necessary to set up multi-parameter sensors to achieve panoramic perception of multiple physical parameters in industry. Among them, magnetic field and temperature are the two most common physical quantities in industry. Accurately obtaining magnetic field and temperature information can know the operating status of the Internet of Things, which is the primary task of establishing multi-parameter perception of the Internet of Things. Therefore, the deployment of high-precision, fast-response, and high-sensitivity magnetic field and temperature sensors plays a decisive role in the efficient production, transmission, and utilization of the power Internet of Things. The study of the dual-parameter measurement scheme of magnetic field and temperature is of great significance to the high-precision monitoring of magnetic fields.

[0004] In current industrial applications, there is an urgent need to set up sensors that can measure magnetic fields and temperatures to achieve panoramic perception. In addition, since magnetic fields and temperatures are generated simultaneously in the measurement environment, in order to achieve high-precision magnetic field measurement, corresponding temperature measurement solutions are needed to solve the cross-sensitivity problem.

[0005] However, traditional electrical sensors are susceptible to electromagnetic interference and have limited response time. Fiber optic measurement technology has problems such as low resolution and limited response speed. Microwave photon measurement solutions can improve the resolution, response speed and sensitivity of fiber optic sensors, but in order to achieve dual-parameter measurement of magnetic field and temperature, it is usually necessary to form a composite sensing system through multi-dimensional multiplexing, which increases the cost and structural complexity, and reduces the stability and response speed of the sensing system to a certain extent.

[0006] Therefore, in summary, in order to achieve high-precision magnetic field and temperature dual parameter measurement, it is urgent to propose a compact magnetic field and temperature dual parameter measurement scheme with high stability and high precision. Summary of the invention

[0007] In order to solve the problems of poor stability and slow response speed of the magnetic field and temperature dual parameter measurement system in the existing measurement technology, the present application provides an extreme learning machine-assisted optoelectronic oscillator magnetic field and temperature measurement device and method.

[0008] The embodiment of the present application is implemented as follows:

[0009] In a first aspect, the present application provides an extreme learning machine-assisted optoelectronic oscillator magnetic field and temperature measurement device, comprising:

[0010] An optoelectronic oscillator sensing system, in which a wide-spectrum light source, a modulator, a Mach-Zehnder interferometer, a dispersion element and a photodetector are constructed to form a finite impulse response microwave photon filter caused by the dispersion effect and an infinite impulse response filter based on an optical fiber ring, so as to realize the frequency selection of the oscillation signal generated by the optoelectronic oscillator;

[0011] The extreme learning machine post-processing system is decoupled by the extreme learning machine post-processing system. It only needs to collect the output signal of the optoelectronic oscillator and input it into the trained extreme learning machine post-processing system. The extreme learning machine post-processing system can directly and accurately output the magnetic field strength and temperature information.

[0012] In one possible implementation, the optoelectronic oscillator sensing system includes a light source, two 1×2 optical couplers, two polarization controllers, an adjustable optical delay line, an electro-optic modulator, a single-mode optical fiber, an optical amplifier, two 2×2 optical couplers, a photodetector and a microwave amplifier.

[0013] In a possible implementation, the finite impulse response microwave photonic filter is related to both magnetic field and temperature, and the change of its center frequency determines the change of the oscillation frequency of the optoelectronic oscillator, so as to achieve a mapping from the magnetic field and temperature to the main mode frequency of the oscillation signal of the optoelectronic oscillator.

[0014] In a possible implementation, the free spectrum range of the IIR filter determines the side mode frequency interval of the signal generated by the OEO oscillator, which is only related to temperature, so as to achieve mapping from temperature to the side mode frequency of the OEO oscillation signal.

[0015] In a possible implementation, the light source is a broadband light source, and the output optical signal is evenly divided into two paths through a first 1×2 optical coupler, transmitted along two arms of a Mach-Zehnder interferometer, and combined through a second 1×2 optical coupler.

[0016] In a possible implementation, the adjustable optical delay line is located on the upper arm of the Mach-Zehnder interferometer, and is used to match the arm lengths of the two arms and generate an initial length difference;

[0017] The lower arm of the Mach-Zehnder interferometer serves as the sensing arm, and the upper arm serves as the reference arm. The optical fiber in the sensing area of ​​the sensing arm is bonded with magnetostrictive material and is sensitive to both magnetic field and temperature.

[0018] In a possible implementation, the electro-optic modulator is used to modulate the microwave signal generated by the optoelectronic oscillator into an optical signal.

[0019] In one possible implementation, the modulated optical signal is delayed by a dispersion element and amplified by an optical amplifier before being converted into an electrical signal by a photodetector. After being amplified by a microwave amplifier, it is divided into two parts, one of which is fed back to the RF end of the electro-optic modulator to form a loop of the optoelectronic oscillator, and the other is used as the output of the microwave signal. The microwave signal is monitored and demodulated by the extreme learning machine post-processing system to obtain changes in the magnetic field and temperature.

[0020] In a possible implementation, a fiber ring is formed by a 2×2 optical coupler and a section of single-mode optical fiber, which is only sensitive to temperature in the sensing area.

[0021] In a second aspect, the present application provides an extreme learning machine-assisted method for measuring magnetic field and temperature of an optoelectronic oscillator, comprising:

[0022] The spectrum of the optical signal emitted by the wide-spectrum light source is cut by a Mach-Zehnder interferometer, and the free spectrum range of the cut optical signal is further changed. The cut optical signal is modulated by an electro-optic modulator, transmitted by a dispersion element, and detected by a photoelectric detector to obtain a finite impulse response microwave photon filter whose center frequency is affected by temperature and magnetic field.

[0023] The FSR of the infinite impulse response filter based on the fiber ring is affected by the temperature of the fiber ring, which changes the free spectral range of the infinite impulse response microwave photon filter;

[0024] Two types of microwave photon filters are embedded in the optoelectronic oscillator, and the mode of the oscillation signal of the optoelectronic oscillator is selected to realize the mapping of the dual parameters of magnetic field and temperature to the signal spectrum generated by the optoelectronic oscillator;

[0025] The center frequency of the main mode of the oscillation signal of the optoelectronic oscillator is determined by the center frequency of the finite impulse response microwave photon filter, and the frequency position varies with the magnetic field and temperature parameters;

[0026] The side mode frequency interval is determined by the free spectrum range of the infinite impulse response filter, which is only related to temperature. Therefore, by reading and decoupling the main mode frequency and side mode frequency of the output signal of the optoelectronic oscillator, the temperature and magnetic field information can be obtained.

[0027] In order to reduce the reading error of the main mode and side mode frequency and improve the magnetic-temperature dual-parameter sensing accuracy of the sensing system, an extreme learning machine is used for decoupling. It is only necessary to collect the output signal of the optoelectronic oscillator and input it into the trained extreme learning machine. The extreme learning machine can directly and accurately output the magnetic field strength and temperature information.

[0028] The technical solution provided by this application can at least achieve the following beneficial effects:

[0029] The extreme learning machine-assisted optoelectronic oscillator magnetic field and temperature measurement device provided in the present application can generate pure high-frequency microwave signals by utilizing an optoelectronic oscillator as a resonant cavity. The phase noise of the sensing signal is low and the stability is good. At the same time, no external microwave source is required, thereby reducing the complexity and cost of the system.

[0030] By sticking magnetostrictive material on the optical fiber, the material expansion and contraction caused by the external magnetic field can be transmitted to the optical fiber, affecting the free spectrum range of the Mach-Zehnder interferometer and further affecting the center frequency of the microwave photon filter, effectively converting the magnetic field measurement into monitoring the oscillation signal frequency of the optoelectronic oscillator, thereby increasing the measurement sensitivity.

[0031] At the same time, by monitoring the center frequency and other mode frequencies of the output signal of the optoelectronic oscillator and demodulating through ELM, dual-parameter measurement can be achieved under simple and compact hardware conditions. Since the temperature cross-sensitivity problem in magnetic field sensing is overcome and the measurement system has a high system stability, the invention has higher magnetic field and temperature dual-parameter measurement accuracy. The optical fiber sensor performs better than traditional electrical sensors in many extremely harsh conditions.

[0032] The measurement system of the present application has a simple structure, low cost and high stability. At the same time, by utilizing the excellent signal processing capability of ELM, the accuracy of the proposed magnetic field and temperature dual-parameter measurement scheme is greatly improved. The high-precision and high-stability magnetic field and temperature dual-parameter sensing technology can provide a solution for multi-parameter measurement needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0034] Figure 1 It is a structural schematic diagram of an extreme learning machine-assisted optoelectronic oscillator magnetic field and temperature measurement device shown in an exemplary embodiment of the present application;

[0035] Figure 2 It is a structural schematic diagram of a magnetic field and temperature dual parameter measurement scheme based on an optoelectronic oscillator according to an exemplary embodiment of the present application;

[0036] Figure 3 is a schematic diagram showing changes in an output signal of an optoelectronic oscillator when a cascaded microwave photon filter changes according to an exemplary embodiment of the present application;

[0037] Figure 4 It is a structural schematic diagram of an extreme learning machine for decoupling dual parameter measurement of magnetic field and temperature shown in an exemplary embodiment of the present application.

[0038] Reference numerals:

[0039] 1. Light source; 2. 1×2 optical coupler; 3. Polarization controller; 4. Adjustable optical delay line; 5. Polarization controller; 6. Electro-optic modulator; 7. Single-mode optical fiber; 8. 1×2 optical coupler; 9. Dispersion element; 10. Optical amplifier; 11. 2×2 optical coupler; 12. Photodetector; 13. Microwave amplifier; 14. Extreme learning machine post-processing module. DETAILED DESCRIPTION

[0040] In order to make the purpose, implementation mode and advantages of the present application clearer, the exemplary implementation mode of the present application will be clearly and completely described below in conjunction with the drawings in the exemplary embodiments of the present application. Obviously, the described exemplary embodiments are only part of the embodiments of the present application, not all of the embodiments. It should be understood that the specific embodiments described here are only used to explain the present application and are not used to limit the present application.

[0041] It should be noted that the brief description of terms in this application is only for the convenience of understanding the embodiments described below, and is not intended to limit the embodiments of this application. Unless otherwise specified, these terms should be understood according to their common and usual meanings.

[0042] The terms "first", "second", "third", etc. in the specification and claims of this application and the above drawings are used to distinguish similar or similar objects or entities, and do not necessarily mean to limit a specific order or sequence, unless otherwise noted. It should be understood that the terms used in this way can be interchangeable under appropriate circumstances.

[0043] The terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover but not exclude inclusion, for example, a product or device comprising a list of components is not necessarily limited to all the components expressly listed but may include other components not expressly listed or inherent to such product or device.

[0044] To facilitate the technical solution of the application, some concepts involved in the application are first explained below.

[0045] Optoelectronic Oscillator (OEO): An optoelectronic hybrid resonant cavity that uses the energy storage characteristics of long optical fibers to form a low-loss optoelectronic positive feedback loop. It can generate high-frequency, low-phase noise, and high-bandwidth high-quality microwave signals. It is a photogenerated microwave technology.

[0046] Extreme Learning Machine (ELM): A single-hidden-layer feedforward neural network. Its improvement on the back-propagation algorithm can improve learning efficiency and simplify training costs. It has the advantages of fast learning speed and good generalization. In the present invention, it can effectively solve the measurement cross-sensitivity and improve measurement accuracy and speed.

[0047] Optical fiber magnetic field and temperature dual parameter measurement: By converting the changes in magnetic field and temperature into changes in parameters such as light intensity, wavelength, and phase of the transmitted light in the optical fiber, and monitoring the transmitted light signal at the receiving end, the measurement technology of simultaneous acquisition of magnetic field and temperature dual parameters is achieved.

[0048] Before explaining the extreme learning machine-assisted optoelectronic oscillator magnetic field and temperature measurement device provided in the embodiment of the present application, the application scenario and implementation environment of the embodiment of the present application are first introduced.

[0049] The demand for high-precision synchronous detection of magnetic field strength and temperature information in various fields such as the Internet of Things, magnetic resonance imaging, marine resource exploration, and thermomagnetic material processing is increasing. Real-time acquisition of magnetic field and temperature information can effectively reflect the status of the Internet of Things, exploration environment, and processing system. Therefore, the deployment of high-precision, fast-response dual-parameter measurement sensors is a necessary condition to ensure the safety of transmission, exploration, and processing.

[0050] Specifically, due to the rapid development of applications in various fields such as magnetic resonance imaging, marine resource exploration, thermomagnetic material processing, navigation, geophysics, etc., the demand for high-precision synchronous detection of magnetic fields is increasing. Magnetic field and temperature are usually inseparable in measurement:

[0051] Since the generation of magnetic field is often accompanied by a large amount of heat, the magnetic field measurement process is easily affected by heat, radioactive interference and wiring problems in the measurement environment, which will limit the practical application of magnetometers under some special conditions such as narrow or high temperature and high humidity spaces. In order to ensure high-precision magnetic field measurement, it is necessary to deploy temperature sensors at the same time to obtain temperature changes in real time, and combine decoupling compensation temperature to achieve accurate measurement of the magnetic field.

[0052] At the same time, in order to effectively ensure the operational safety of industrial applications and realize real-time monitoring and early warning, it is necessary to set up multi-parameter sensors to achieve panoramic perception of various physical parameters in industry. Among them, magnetic field and temperature are the two most common physical quantities in industry. Accurately obtaining magnetic field and temperature information can know the operating status of the Internet of Things, which is the primary task of establishing multi-parameter perception of the Internet of Things.

[0053] Therefore, the deployment of magnetic field and temperature sensors with high precision, fast response and high sensitivity plays a decisive role in the efficient production, transmission and utilization of the power Internet of Things. The research on the dual parameter measurement scheme of magnetic field and temperature is of great significance for high-precision monitoring of magnetic field.

[0054] Traditional electrical sensors are susceptible to electromagnetic interference and have limited response time, and cannot meet the requirements of simple structure and high precision at the same time. As a new application technology, fiber optic sensing technology can meet the high-sensitivity measurement requirements in many fields due to its advantages such as light weight, small size, low cost and high sensitivity. Compared with traditional electrical sensors, fiber optic sensing technology has the advantages of small size, low cost and strong resistance to chemical corrosion, and has attracted widespread attention from researchers. By setting a sensor head that is sensitive to the measured parameter, the measured parameter is mapped with the frequency, amplitude and phase information of the observed optical signal, and high-sensitivity perception can usually be effectively achieved by demodulating the changing parameters in the optical signal. However, fiber optic sensors still have problems such as low measurement resolution and slow demodulation speed. Limited by the slow response speed and complex demodulation system of fiber optic sensing technology, fiber optic sensors usually cannot perform well in high-speed real-time measurement scenarios. Microwave photon sensors combine the advantages of microwave and optical fields, and can effectively solve the problems of slow response speed and difficult demodulation in fiber optic sensors.

[0055] Microwave photonics, with its advantages of flexibility and speed, can further improve the response speed and resolution of sensors and achieve high-sensitivity, high-speed, and high-resolution sensing. OEO is an important device in microwave photonics, which usually includes light sources, electro-optic modulators, optical fibers, filters, microwave amplifiers or optical amplifiers, photodetectors and other optoelectronic components. These components are used to build a positive feedback loop to achieve self-oscillation. This classic structure was first proposed by Chinese Americans Yao Xiaotian (X.SteveYao) and L.Maleki at the Jet Propulsion Laboratory of the California Institute of Technology in 1995. OEO uses the high bandwidth of optical devices and the high energy storage and low loss characteristics of long optical fibers to obtain high-quality oscillation signals with ultra-low phase noise (up to -140dBc / Hz@10kHz) in high-frequency bands (microwave / millimeter wave bands). By establishing a connection between the physical quantity to be measured and the frequency, amplitude and other parameters of the oscillation signal, information about the physical quantity to be measured can be obtained by monitoring the high-quality microwave signal generated by OEO, which has the advantages of fast response and high resolution.

[0056] At present, fiber optic sensing technology for temperature and magnetic field is developing rapidly. Among them, temperature sensors utilize the temperature-sensitive properties of optical fibers themselves. By constructing optical sensor heads such as fiber interferometers or fiber gratings, they can effectively apply temperature changes to the transmission light signal in the fiber optic sensor head, and detect the changes in the light signal to achieve temperature measurement. OEO-based temperature sensors have the advantages of high resolution, high stability, and fast response. In 2022, Xie et al. proposed an enhanced vernier effect temperature sensor based on OEO that achieved a temperature measurement sensitivity of 141.13kHz / ℃.

[0057] In addition, the research on fiber optic magnetic field sensors mainly includes three technical routes: magnetostrictive effect based on Faraday rotation effect, magnetofluid optical effect and magnetostrictive effect based on magnetostrictive materials. By constructing optical structures such as optical resonant cavities, interferometers or fiber gratings, the magnetic field information is mapped to the amplitude, phase and frequency of the transmitted light signal for observation, thereby achieving magnetic field measurement.

[0058] Among them, in 2000, Mora et al. proposed a fiber magnetic field measurement scheme in which a fiber Bragg grating is pasted on a magnetostrictive material. The change in the length of the magnetostrictive material under a magnetic field is transmitted to the fiber Bragg grating through a photosensitive adhesive. The change in the magnetic field can be effectively identified by reading the wavelength drift of the dip peak in the fiber Bragg grating reflection spectrum. The magnetic field sensitivity finally achieved by this scheme is (2.31±0.05)×10-10nm / (A 2 m -2 ), and on this basis, combining OEO with optical fiber magnetic field sensors can significantly improve the resolution and response speed of the sensor.

[0059] In 2018, an OEO magnetic field sensing scheme based on a phase-shifted Bragg grating microwave photonic filter was proposed. The narrowband depression in the reflection spectrum of the phase-shifted grating was used to construct a microwave photonic filter to select the frequency of the OEO. When the magnetic field changes, the center frequency of the narrowband depression of the grating drifts, thereby changing the center frequency of the microwave photonic filter and the frequency of the OEO oscillation signal. Therefore, the magnetic field size can be obtained by monitoring the frequency changes of the microwave signal. The scheme ultimately achieved a high sensitivity of 38MHz / Oe.

[0060] However, in the process of magnetic field measurement, since the magnetic field generation process is often accompanied by heat generation, magnetic field temperature decoupling is a necessary step to ensure the accuracy of magnetic field measurement. In order to solve the cross-sensitivity characteristics in the magnetic field and temperature measurement process, a composite optical fiber sensor with integrated magnetic field and temperature dual parameter sensing function can significantly reduce the measurement error introduced by the cross-sensitivity characteristics in the magnetic field measurement process. By combining magnetic sensitive and thermal sensitive materials and designing an optical fiber sensing probe with a special structure, the magnetic field and temperature can be mapped to the interference spectrum and loss spectrum of the optical signal, and the magnetic field and temperature information can be demodulated. In 2014, Fu et al. proposed a high-sensitivity magnetic field and temperature dual parameter measurement scheme based on double-hole optical fiber to achieve high-sensitivity dual parameter measurement in a large measurement range, with a magnetic field sensitivity of up to 64000pm / mT. However, traditional optical fiber magnetic field and temperature dual parameter sensors have the problems of low measurement resolution and slow response speed. The high-resolution demodulation of optical signals through microwave photon technology can effectively improve the response speed of magnetic field and temperature dual parameter measurement.

[0061] At present, the dual-parameter measurement method based on microwave photonics can realize composite sensing by constructing multi-dimensional technologies such as dual wavelength and dual polarization. In 2021, Feng et al. proposed a dual-parameter sensor of magnetic field and temperature based on dual OEO. Two independent OEOs were constructed by cascading MA-FBG and FBG, and the magnetic field and temperature information were obtained by decoupling the generated signals of the dual OEO. However, this multi-dimensional composite dual-parameter sensing scheme requires the addition of lasers, modulators, photodetectors and other devices, which increases the structural complexity and cost of the sensor. In addition, the multiplexing technology will reduce the stability and response speed of the sensing system to a certain extent.

[0062] For example, in the above-mentioned magnetic field sensing scheme of OEO based on phase-shifted Bragg grating microwave photon filter, in order to realize the dual-parameter measurement of magnetic field and temperature, it is necessary to build a dual-frequency oscillation OEO, which has the problem of complex and unstable system structure, which will affect the measurement accuracy of the dual-parameter measurement scheme of magnetic field and temperature. Therefore, although the existing research on microwave photon magnetic field and temperature sensors has the advantages of high sensitivity, high resolution, high response speed, etc., when facing the dual-parameter measurement scenario, the existing microwave photon magnetic field and temperature dual-parameter measurement scheme usually adopts multiplexing technology, or adds multiple measurement systems to realize magnetic field and temperature measurement, which undoubtedly increases the layout cost and structural complexity of the sensor, and limits the practical application and application scenarios of dual-parameter sensors.

[0063] At the same time, the existing magnetic field and temperature dual parameter measurement system has problems such as poor stability and slow response speed.

[0064] The shortcomings of the existing technology are summarized as follows: in current industrial applications, it is urgent to set up sensors that can measure magnetic fields and temperatures to achieve panoramic perception. In addition, since the magnetic field and temperature are generated simultaneously in the measurement environment, in order to achieve high-precision magnetic field measurement, a corresponding temperature measurement solution is required to solve the cross-sensitivity problem.

[0065] Traditional electrical sensors are susceptible to electromagnetic interference and have limited response time. Fiber optic measurement technology has problems such as low resolution and limited response speed. Microwave photon measurement solutions can improve the resolution, response speed and sensitivity of fiber optic sensors. However, in order to achieve dual-parameter measurement of magnetic field and temperature, it is usually necessary to form a composite sensing system through multi-dimensional multiplexing. While increasing the cost and structural complexity, it also reduces the stability and response speed of the sensing system to a certain extent.

[0066] Therefore, in summary, in order to achieve high-precision magnetic field and temperature dual parameter measurement, it is urgent to propose a compact magnetic field and temperature dual parameter measurement scheme with high stability and high precision.

[0067] Based on this, the present application provides an extreme learning machine-assisted optoelectronic oscillator magnetic field and temperature measurement device and method, a magnetic field and temperature dual parameter sensing scheme integrating optoelectronic oscillator technology and extreme learning machine post-processing technology, and realizes the synchronous measurement of magnetic field and temperature dual parameters with high stability, high precision and fast response on compact hardware. At the same time, since the invention can realize the accurate perception of magnetic field and temperature dual parameters, it effectively overcomes the temperature cross-sensitivity problem in the current magnetic field measurement, improves the magnetic field measurement accuracy, and meets the needs of high-precision magnetic field perception in special scenarios with weak magnetic field and large temperature fluctuations.

[0068] Next, the technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems will be described in detail through embodiments and in combination with the accompanying drawings. The embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments.

[0069] Figure 1 It is a structural schematic diagram of an extreme learning machine-assisted optoelectronic oscillator magnetic field and temperature measurement device shown in an exemplary embodiment of the present application.

[0070] In an exemplary embodiment, Figure 1 As shown, an extreme learning machine-assisted optoelectronic oscillator magnetic field and temperature measurement device is provided, the device comprising:

[0071] An optoelectronic oscillator sensing system, in which a wide-spectrum light source, a modulator, a Mach-Zehnder interferometer, a dispersion element and a photodetector are constructed to form a finite impulse response microwave photon filter caused by the dispersion effect and an infinite impulse response filter based on an optical fiber ring, so as to realize the frequency selection of the oscillation signal generated by the optoelectronic oscillator;

[0072] The extreme learning machine post-processing system is decoupled by the extreme learning machine post-processing system. It only needs to collect the output signal of the optoelectronic oscillator and input it into the trained extreme learning machine post-processing system. The extreme learning machine post-processing system can directly and accurately output the magnetic field strength and temperature information.

[0073] Figure 2 It is a structural schematic diagram of a dual parameter measurement scheme of magnetic field and temperature based on an optoelectronic oscillator, shown in an exemplary embodiment of the present application.

[0074] In a possible implementation, the specific structure is as follows:

[0075] Firstly, a broadband light source, a modulator, a Mach–Zehnder Interferometer (MZI), a dispersion element and a photodetector constitute a finite impulse response microwave photonic filter caused by the dispersion effect. Among them, the MZI realizes the spectral cutting of the optical signal emitted by the broadband light source. Since the sensing optical fiber of the lower arm of the MZI is affected by the magnetic field and temperature at the same time, the changes in the magnetic field and temperature in the measured area will be converted into changes in the optical path difference between the two arms, further changing the free spectral range (FSR) of the optical signal after cutting.

[0076] Since the dispersion element introduces a time delay difference for the optical signals of different wavelengths after being cut, after detection by the photodetector, a finite impulse response microwave photon filter whose center frequency is affected by both temperature and magnetic field can be obtained, while the FSR of the infinite impulse response filter based on the fiber ring is affected by the temperature of the fiber ring: temperature changes can change the optical path in the fiber ring, thereby changing the FSR of the infinite impulse response microwave photon filter.

[0077] These two types of microwave photonic filters are embedded in the OEO, and the mode of the OEO's oscillation signal is selected to map the dual parameters of magnetic field and temperature to the signal spectrum generated by the OEO. The center frequency of the main mode of the OEO oscillation signal is determined by the center frequency of the finite impulse response microwave photonic filter, and the frequency position varies with the magnetic field and temperature parameters; while the side mode frequency interval is determined by the FSR of the infinite impulse response filter, which is only related to the temperature. Therefore, by reading and decoupling the main mode frequency and side mode frequency of the OEO's output signal, the temperature and magnetic field information can be obtained.

[0078] In order to reduce the reading errors of the main mode and side mode frequencies and improve the magnetic-temperature dual-parameter sensing accuracy of the sensing system, the present invention proposes to use ELM for decoupling. It only needs to collect the output signal of OEO and input it into the trained ELM. ELM can directly and accurately output the magnetic field strength and temperature information.

[0079] Specifically, the measurement device consists of two parts: the OEO sensing system and the ELM post-processing system. Figure 2 As shown, it includes a light source 1, two 1×2 optical couplers 2, 8, two polarization controllers 3, 5, an adjustable optical delay line 4, an electro-optic modulator 6, a single-mode optical fiber 7 with magnetostrictive material, a dispersion element 9, an optical amplifier 10, a 2×2 optical coupler 11, a photodetector 12, a microwave amplifier 13 and an extreme learning machine post-processing module 14.

[0080] The light source uses a wide-spectrum light source, and the output optical signal is evenly divided into two paths by the first 1×2 optical coupler and then transmitted along the two arms of the MZI. The adjustable optical delay line is located on the upper arm of the MZI, which is used to match the arm lengths of the two arms and generate an initial length difference. The lower arm of the MZI is used as the sensing arm, and the upper arm is the reference arm. The optical fiber in the sensing area of ​​the sensing arm is bonded with magnetostrictive material and is sensitive to magnetic fields and temperature. The electro-optic modulator modulates the microwave signal generated by the OEO onto the optical signal of the lower arm. The upper and lower arm signals are combined by the second 1×2 optical coupler to form an MZI structure that is sensitive to magnetic fields and temperature.

[0081] A 2×2 optical coupler and a section of single-mode optical fiber form a fiber ring that is only sensitive to temperature in the sensing area.

[0082] The output signal of the MZI is delayed by the dispersion element and amplified by the optical amplifier before being converted into an electrical signal by the photodetector 12. After being amplified by the microwave amplifier 13, it is divided into two parts, one of which is fed back to the RF end of the electro-optical modulator 6 to form an OEO loop, and the other is used as the output of the microwave signal. The microwave signal is monitored and demodulated by the extreme learning machine post-processing module 14 to obtain the changes in the magnetic field and temperature.

[0083] The output signal of the OEO is determined by two microwave photon filters. The finite pulse response microwave photon filter consists of a light source, an MZI, an electro-optic modulator, a dispersion element, and a photodetector. Its center frequency is related to the optical path difference between the upper and lower arms of the MZI, and its expression is:

[0084]

[0085] Where D is the total dispersion of the dispersion element, in ps / nm, Δl is the optical path difference between the upper and lower arms of the MZI, c is the speed of light, and n is the effective refractive index of the optical fiber. In the sensing area of ​​the sensing arm of the MZI, the optical fiber adheres to the magnetostrictive material in a racetrack shape after being de-coated, and is affected by the magnetic field and temperature, such as Figure 1 As shown in the illustration.

[0086] When an external magnetic field is applied to the sensing area, the magnetostrictive material will expand and contract as the magnitude of the magnetic field changes, further changing the refractive index and length of the optical fiber attached to the magnetostrictive material. At the same time, due to the thermo-optical effect and elastic-optical effect of the optical fiber, temperature changes also change the refractive index and length of the optical fiber in the sensing area.

[0087] Therefore, the change of the difference between the upper and lower arm lengths of the MZI leads to the change of the center frequency of the finite pulse response microwave photon filter, and the mapping relationship between the change of the magnetic field intensity ΔB and the change of the temperature ΔT is as follows:

[0088]

[0089] Where l is the length of the optical fiber affected by the magnetic field and temperature, ζ is the thermo-optical effect coefficient, and α F is the thermal expansion coefficient of the optical fiber, α M is the thermal expansion coefficient of the GMM, p e is the elasto-optical effect coefficient of the optical fiber, and k is a coefficient proportional to the magnetostriction constant.

[0090] It can be concluded that the center frequency of the finite impulse response microwave photonic filter is related to both the temperature and the magnetic field of the sensing area.

[0091] At the same time, the temperature-sensitive infinite impulse response filter is realized by the optical fiber ring. The FSR change of the infinite impulse response filter response is related to the temperature change of the optical fiber ring:

[0092]

[0093] Among them, L frr is the length of the optical fiber in the optical fiber ring. Therefore, as the temperature parameter changes, the FSR of the infinite impulse response filter changes.

[0094] By cascading a finite impulse response microwave photon filter and an infinite impulse response microwave photon filter, the spectrum of the OEO oscillation signal is affected by the response of the cascaded microwave photon filter: the center frequency position of the main mode of the OEO output signal is determined by the center frequency of the finite impulse response microwave photon filter, and the frequency position varies with the magnetic field and temperature parameters; at the same time, in addition to the main mode of the OEO output signal, there are other periodic side modes, and the side mode spacing is consistent with the FSR of the infinite impulse response filter response. Therefore, by observing the output signal spectrum of the OEO, the main mode frequency and the side mode frequency are obtained, and the information of the magnetic field and temperature changes can be obtained.

[0095] Figure 3 is a schematic diagram showing the change of the output signal of the optoelectronic oscillator under the change of the cascade microwave photon filter shown in an exemplary embodiment of the present application, Figure 4 It is a structural schematic diagram of an extreme learning machine for decoupling dual parameter measurement of magnetic field and temperature shown in an exemplary embodiment of the present application.

[0096] However, due to the mode competition effect of OEO, the suppression between the oscillating mode and the unoscillating side mode is relatively high, so the unoscillating side mode is easily submerged in the noise, and it is difficult to accurately distinguish the change in the side mode spacing caused by the change in the measured parameter, which brings a low signal-to-noise ratio problem to the sensing system and limits the accuracy of dual-parameter measurement. Figure 3 shown.

[0097] Therefore, the present invention intends to introduce an ELM post-processing module (such as Figure 4 ), effectively identifying the side mode spacing, can reduce the error introduced by human readings under the conditions of smaller data volume and worse frequency resolution, and achieve the purpose of high-precision magnetic field and temperature decoupling. The number of input layer nodes of ELM is equal to the length of the input single spectrum data, and the number of output layer nodes is two (output temperature and magnetic field respectively).

[0098] Through training with a certain amount of simulation data sets, ELM performs well in identifying experimental data. Therefore, this ELM-assisted OEO magnetic field and temperature measurement method and device constructs an optoelectronic measurement system with high sensitivity, high resolution, and high stability for magnetic field and temperature dual parameter measurement, and integrates an ELM post-processing module that improves the signal-to-noise ratio of the measurement signal, which can significantly improve the sensitivity and accuracy of the magnetic field and temperature dual parameter sensing scheme.

[0099] In a possible implementation, the more detailed structure is:

[0100] A finite impulse response microwave photonic filter is composed of a light source, an MZI, an electro-optic modulator, a dispersive element and a photodetector, and its center frequency is determined by the FSR of the MZI.

[0101] Since the FSR of the MZI in the present invention is related to the magnitude of the magnetic field and temperature, the magnetic field and temperature to be measured can be effectively mapped to the center frequency of the microwave photon filter, and further mapped to the main mode frequency of the microwave signal generated by the OEO.

[0102] At the same time, due to the infinite pulse response based on the fiber ring in the OEO, there are side modes with fixed mode spacing in the output signal of the OEO. By monitoring the microwave signal generated by the OEO and monitoring the frequency changes of the extracted main mode and side mode, the temperature and magnetic field parameters can be demodulated:

[0103]

[0104] The present invention uses ELM for offline processing to further reduce the reading error in the measurement and improve the accuracy of magnetic field and temperature measurement. The ELM is trained by simulation data. When the measured spectrum is input into the trained ELM for post-processing, the accurate values ​​of the dual parameters can be output quickly and effectively, completing the magnetic field and temperature dual parameter demodulation work.

[0105] In addition, the present invention analyzes the change in FSR of OEO due to the change in OEO loop length under temperature and magnetic field, as shown in the following formula. Due to the use of dispersive elements, the total loop length of OEO is often several kilometers, while the change in OEO loop length caused by temperature and magnetic field changes is in the order of centimeters, as shown in the formula. Within a larger sensing range, the change in main mode frequency caused by the change in OEO length is very small, usually less than 1Hz, so the sensing error caused by the magnetic field and temperature on the OEO loop change can be ignored in the present invention.

[0106]

[0107] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the instructions, these steps are not necessarily executed in sequence according to the order of the instructions. Unless there is a clear explanation in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the above-mentioned embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of the steps or stages in other steps.

[0108] Corresponding to the aforementioned embodiment of an extreme learning machine-assisted optoelectronic oscillator magnetic field and temperature measurement device, the present application also provides an embodiment of an extreme learning machine-assisted optoelectronic oscillator magnetic field and temperature measurement method using the same technical concept.

[0109] In an exemplary embodiment, the extreme learning machine-assisted optoelectronic oscillator magnetic field and temperature measurement method includes:

[0110] The spectrum of the optical signal emitted by the wide-spectrum light source is cut by a Mach-Zehnder interferometer, and the free spectrum range of the cut optical signal is further changed. The cut optical signal is modulated by an electro-optic modulator, transmitted by a dispersion element, and detected by a photoelectric detector to obtain a finite impulse response microwave photon filter whose center frequency is affected by temperature and magnetic field.

[0111] The FSR of the infinite impulse response filter based on the fiber ring is affected by the temperature of the fiber ring, which changes the free spectral range of the infinite impulse response microwave photon filter;

[0112] Two types of microwave photon filters are embedded in the optoelectronic oscillator, and the mode of the oscillation signal of the optoelectronic oscillator is selected to realize the mapping of the dual parameters of magnetic field and temperature to the signal spectrum generated by the optoelectronic oscillator;

[0113] The center frequency of the main mode of the oscillation signal of the optoelectronic oscillator is determined by the center frequency of the finite impulse response microwave photon filter, and the frequency position varies with the magnetic field and temperature parameters;

[0114] The side mode frequency interval is determined by the free spectrum range of the infinite impulse response filter, which is only related to temperature. Therefore, by reading and decoupling the main mode frequency and side mode frequency of the output signal of the optoelectronic oscillator, the temperature and magnetic field information can be obtained.

[0115] In order to reduce the reading error of the main mode and side mode frequency and improve the magnetic-temperature dual-parameter sensing accuracy of the sensing system, an extreme learning machine is used for decoupling. It is only necessary to collect the output signal of the optoelectronic oscillator and input it into the trained extreme learning machine. The extreme learning machine can directly and accurately output the magnetic field strength and temperature information.

[0116] For the specific limitations of the extreme learning machine-assisted optoelectronic oscillator magnetic field and temperature measurement method, please refer to the above limitations of the extreme learning machine-assisted optoelectronic oscillator magnetic field and temperature measurement device, which will not be repeated here.

[0117] Working principle:

[0118] First of all, as a high-quality microwave signal generating device, the response of the microwave photon filter inside the OEO ring can determine the spectrum of the signal generated by the OEO.

[0119] Based on the above principles, we constructed a finite impulse response filter based on a wide-spectrum light source, MZI, dispersion elements and photodetectors in the OEO ring, as well as an infinite impulse response filter based on an optical fiber ring to achieve frequency selection of the oscillation signal generated by the OEO.

[0120] Among them, the response of the finite impulse response filter is related to both the magnetic field and temperature. The change of its center frequency determines the change of the oscillation frequency of the OEO, thereby achieving the mapping from the magnetic field and temperature to the main mode frequency of the OEO oscillation signal. The FSR of the infinite impulse response filter determines the side mode frequency interval of the OEO generated signal, which is only related to the temperature, thereby realizing the mapping from temperature to the side mode frequency of the OEO oscillation signal.

[0121] Since the main mode frequency and side mode frequency of the signal generated by OEO contain information about the magnetic field and temperature, by identifying the main mode and side mode frequency positions of the collected OEO output signal spectrum and decoupling them, independent magnetic field and temperature information can be obtained, thereby realizing the function of dual parameter sensing of magnetic field and temperature.

[0122] In order to further reduce the reading error and increase the measurement accuracy of the sensing system, an ELM is constructed to perform offline post-processing of the output signal, which can effectively demodulate the magnetic field intensity and temperature.

[0123] It can be seen that the measurement system of the present application has a simple structure, low cost and high stability. At the same time, by utilizing the excellent signal processing capability of ELM, the accuracy of the proposed magnetic field and temperature dual parameter measurement scheme is greatly improved. The high-precision and high-stability magnetic field and temperature dual parameter sensing technology proposed in the present invention can provide a solution for high-precision measurement needs.

[0124] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0125] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. An extreme learning machine-assisted optoelectronic oscillator magnetic field and temperature measurement device, characterized in that: include: An optoelectronic oscillator sensing system, in which a wide-spectrum light source, a modulator, a Mach-Zehnder interferometer, a dispersion element and a photodetector are constructed to form a finite impulse response microwave photon filter caused by the dispersion effect and an infinite impulse response filter based on an optical fiber ring, so as to realize the frequency selection of the oscillation signal generated by the optoelectronic oscillator; The extreme learning machine post-processing system is decoupled by the extreme learning machine post-processing system. It only needs to collect the output signal of the optoelectronic oscillator and input it into the trained extreme learning machine post-processing system. The extreme learning machine post-processing system can directly and accurately output the magnetic field strength and temperature information.

2. The device for measuring magnetic field and temperature of an optoelectronic oscillator assisted by an extreme learning machine as claimed in claim 1, characterized in that: The optoelectronic oscillator sensing system includes a light source, two 1×2 optical couplers, two polarization controllers, an adjustable optical delay line, an electro-optic modulator, a single-mode optical fiber, an optical amplifier, two 2×2 optical couplers, a photodetector and a microwave amplifier.

3. The device for measuring magnetic field and temperature of an optoelectronic oscillator assisted by an extreme learning machine as claimed in claim 2, characterized in that: The finite impulse response microwave photon filter is related to both magnetic field and temperature. The change of its center frequency determines the change of the oscillation frequency of the optoelectronic oscillator, so as to achieve the mapping from magnetic field and temperature to the main mode frequency of the oscillation signal of the optoelectronic oscillator.

4. The device for measuring magnetic field and temperature of an optoelectronic oscillator assisted by an extreme learning machine as claimed in claim 3, characterized in that: The free spectrum range of the infinite impulse response filter determines the side mode frequency interval of the signal generated by the optoelectronic oscillator, which is only related to the temperature, so as to realize the mapping from the temperature to the side mode frequency of the oscillation signal of the optoelectronic oscillator.

5. The device for measuring magnetic field and temperature of an optoelectronic oscillator assisted by an extreme learning machine as claimed in claim 4, characterized in that: The light source adopts a wide-spectrum light source, and the output optical signal is evenly divided into two paths through the first 1×2 optical coupler, transmitted along the two arms of the Mach-Zehnder interferometer, and combined through the second 1×2 optical coupler.

6. The extreme learning machine-assisted optoelectronic oscillator magnetic field and temperature measurement device according to claim 5, characterized in that: The adjustable optical delay line is located on the upper arm of the Mach-Zehnder interferometer and is used to match the arm lengths of the two arms and generate an initial length difference; The lower arm of the Mach-Zehnder interferometer serves as the sensing arm, and the upper arm serves as the reference arm. The optical fiber in the sensing area of ​​the sensing arm is bonded with magnetostrictive material and is sensitive to both magnetic field and temperature.

7. The device for measuring magnetic field and temperature of an optoelectronic oscillator assisted by an extreme learning machine as claimed in claim 6, characterized in that: The electro-optic modulator is used to modulate the microwave signal generated by the optoelectronic oscillator into an optical signal.

8. The extreme learning machine-assisted optoelectronic oscillator magnetic field and temperature measurement device according to claim 7, characterized in that: The modulated optical signal generates a time delay through the dispersion element and is amplified by the optical amplifier before being converted into an electrical signal by the photodetector. After being amplified by the microwave amplifier, it is divided into two parts. One part is fed back to the RF end of the electro-optical modulator to form a loop of the optoelectronic oscillator, and the other part is used as the output of the microwave signal. The microwave signal is monitored and demodulated by the extreme learning machine post-processing system to obtain changes in the magnetic field and temperature.

9. The extreme learning machine-assisted optoelectronic oscillator magnetic field and temperature measurement device according to claim 8, characterized in that: A fiber ring is formed by a 2×2 optical coupler and a section of single-mode optical fiber, which is only sensitive to temperature in the sensing area.

10. An extreme learning machine-assisted method for measuring magnetic field and temperature of an optoelectronic oscillator, applied to the measuring device according to any one of claims 1 to 9, characterized in that: include: The spectrum of the optical signal emitted by the wide-spectrum light source is cut by a Mach-Zehnder interferometer, and the free spectrum range of the cut optical signal is further changed. The cut optical signal is modulated by an electro-optic modulator and transmitted through a dispersion element, and then detected by a photodetector, thereby obtaining a finite impulse response microwave photon filter whose center frequency is affected by temperature and magnetic field. The FSR of the infinite impulse response filter based on the fiber ring is affected by the temperature of the fiber ring, which changes the free spectral range of the infinite impulse response microwave photon filter; Two types of microwave photon filters are embedded in the optoelectronic oscillator, and the mode of the oscillation signal of the optoelectronic oscillator is selected to realize the mapping of the dual parameters of magnetic field and temperature to the signal spectrum generated by the optoelectronic oscillator; The center frequency of the main mode of the oscillation signal of the optoelectronic oscillator is determined by the center frequency of the finite impulse response microwave photon filter, and the frequency position varies with the magnetic field and temperature parameters; The side mode frequency interval is determined by the free spectrum range of the infinite impulse response filter, which is only related to temperature. Therefore, by reading and decoupling the main mode frequency and side mode frequency of the output signal of the optoelectronic oscillator, the temperature and magnetic field information can be obtained. In order to reduce the reading error of the main mode and side mode frequency and improve the magnetic-temperature dual-parameter sensing accuracy of the sensing system, an extreme learning machine is used for decoupling. It is only necessary to collect the output signal of the optoelectronic oscillator and input it into the trained extreme learning machine. The extreme learning machine can directly and accurately output the magnetic field strength and temperature information.

Images