A method for realizing a nuclear magnetic resonance atomic gyroscope based on a self-calibrating magnetometer

Abstract

A method for implementing a nuclear magnetic resonance atomic gyroscope based on a self-calibrating magnetometer is conducive to achieving high-sensitivity and high-accuracy measurement of the magnetic field phase, thereby improving the long-term stability of the gyroscope. It is characterized in that the differential photodetector transmits the output electrical signal formed by the detection light on the light output side of the gas chamber to the first lock-in amplifier. The first lock-in amplifier performs the first demodulation on the output electrical signal, and respectively transmits the real part of the complex-valued signal after the first demodulation to the second lock-in amplifier and the third lock-in amplifier for the second demodulation. The second lock-in amplifier negatively feeds back the calibrated circularly polarized magnetic field phase after its second demodulation to the first lock-in amplifier through a proportional-integral controller to determine the self-calibration of the circularly polarized magnetic field. The third lock-in amplifier negatively feeds back the nuclear spin precession phase after its second demodulation to the excitation magnetic field frequency drive end through a phase-locked loop to maintain nuclear magnetic resonance and achieve closed-loop operation of the gyroscope.

Description

Technical Field

[0001] The present invention relates to the technical field of atomic devices based on nuclear magnetic resonance technology, and particularly to a method for realizing a nuclear magnetic resonance atomic gyroscope based on a self-calibrating magnetometer. Background Art

[0002] As a core component of inertial measurement, gyroscopes have always received extensive attention and have great demands in both military and civilian fields. Among them, atomic gyroscopes have developed rapidly in recent years, and there have been rapid improvements in performance and practicality. In basic physical research, they have good application prospects in fields closely related to inertial measurement, such as aviation, aerospace, navigation, and traffic navigation. Currently, in the field of atomic gyroscopes, they are mainly divided into gyroscopes using atomic spin and atomic interference. Gyroscopes using atomic spin can be divided into compensated gyroscopes and nuclear magnetic resonance gyroscopes, etc. The experimental setup of compensated gyroscopes is complex, and data processing is cumbersome, and there may be many difficulties in the future miniaturization and practical application process. Nuclear magnetic resonance gyroscopes may solve these problems. However, nuclear magnetic resonance gyroscopes have also encountered other challenges.

[0003] The nuclear magnetic resonance gyroscope works using an inert gas nuclear spin-alkali metal system. In our experiment, we use the Rb- 129 Xe / 131 Xe system. Rubidium atoms have two functions here. One function is to polarize xenon atomic nuclei through collisions, and the other function is to act as a built-in rubidium atom magnetometer to detect the precession signal of xenon nuclear spin. The xenon nuclear spin detects the rotation of the carrier through the change of the precession signal under different equivalent magnetic fields, which is the basis of the nuclear magnetic resonance gyroscope. Among inert gas atoms, xenon has the largest collision cross-section with alkali metal rubidium atoms. In this way, on the one hand, it can be quickly polarized by alkali metal rubidium atoms (the polarization time is about 1 minute), and on the other hand, the polarization magnetic field of xenon can be most sensitively detected by the alkali metal rubidium atom magnetometer, and the detected signal can be processed to obtain the system rotation information.

[0004] The long-term stability of an inert gas xenon nuclear spin co-gyroscope is ultimately directly proportional to the signal-to-noise ratio of the equivalent magnetic field signal generated by the precession of the xenon nuclear spin and inversely proportional to the resonance linewidth of xenon. However, in practice, due to various uncertain factors in the gyroscope system (current source, gas chamber temperature, light intensity and wavelength of the pumping light and the probing light), the error generated by the magnetometer system in measuring the phase of the magnetic field signal makes the long-term stability of the gyroscope unable to reach the theoretical limit determined by the signal-to-noise ratio and linewidth of the precession signal of the inert gas xenon nucleus. To address this, the present invention proposes adding a set of calibration circularly polarized magnetic fields with a determined frequency and amplitude, and through negative feedback control, keeping the measured phase of the calibration circularly polarized magnetic field signal unchanged, so as to fully suppress the error brought by the magnetometer detection system, thereby compensating for the drift of the measured magnetic field signal caused by experimental environmental variables and achieving the purpose of improving the long-term stability of the gyroscope system. Summary of the Invention

[0005] In view of the defects or deficiencies of the prior art, the present invention provides a method for implementing a nuclear magnetic resonance atomic gyroscope based on a self-calibrating magnetometer, which can actively apply a calibration circularly polarized magnetic field signal and negatively feedback and close the loop to compensate for the measurement errors caused by modulation and instruments, facilitating the realization of high-sensitivity and high-accuracy measurement of the magnetic field phase, and further improving the long-term stability of the gyroscope.

[0006] The technical solution of the present invention is as follows:

[0007] A method for implementing a nuclear magnetic resonance atomic gyroscope based on a self-calibrating magnetometer, characterized in that the differential optoelectronic detector transmits the differential optoelectronic detector output electrical signal formed by the probing light on the light output side of the gas chamber to the first lock-in amplifier. The first lock-in amplifier performs the first demodulation on the differential optoelectronic detector output electrical signal, and respectively transmits the real part of the complex value signal after the first demodulation to the second lock-in amplifier and the third lock-in amplifier for the second demodulation. The second lock-in amplifier negatively feeds back the calibration circularly polarized magnetic field phase after its second demodulation to the first lock-in amplifier through a proportional-integral controller to determine the self-calibration of the circularly polarized magnetic field. The third lock-in amplifier negatively feeds back the nuclear spin precession phase after its second demodulation to the excitation magnetic field frequency drive end through a phase-locked loop to maintain nuclear magnetic resonance and realize the closed-loop operation of the gyroscope.

[0008] The gas chamber is an atomic gas chamber containing alkali metal rubidium (Rb) atoms and inert gas xenon (Xe) atoms, where Rb includes 85 Rb and 87 Rb, and Xe includes 129 Xe and 131 Xe.

[0009] An induction furnace, an excitation coil, and a magnetic shielding cylinder are sequentially arranged outward around the gas chamber. The excitation coil includes an X-direction magnetic field coil, a Y-direction magnetic field coil, and a Z-direction magnetic field coil. The Z-direction magnetic field coil has two sets of coils to respectively apply a Z-direction bias magnetic field and a Z-direction modulation magnetic field. The Z-direction bias magnetic field is used to define the polarization axis direction of the system and is parallel to the pumping light direction. The pumping light continuously polarizes rubidium atoms, and the polarization is transferred to xenon nuclei through collisions. At this time, the polarizations of both rubidium atoms and xenon nuclei are along the direction of the pumping light. The frequency of the Z-direction modulation magnetic field is the Larmor precession frequency of alkali metal rubidium atoms. The X-direction magnetic field coil and the Y-direction magnetic field coil are used to generate a circularly polarized calibration magnetic field and a xenon nucleus precession excitation magnetic field.

[0010] The differential photodetector measures the optical powers of two linearly polarized components of the X-direction linearly polarized detection light, and then uses a differential circuit to obtain an output electrical signal of the optical polarization rotation. The self-calibration of the circularly polarized magnetic field is to transversely apply a circularly polarized calibration magnetic field with a determined frequency and amplitude, and feedback the signal phase of the calibrated magnetic field detected by the magnetometer to the phase shift of the first demodulation reference signal of the magnetometer through a proportional-integral controller so that the detected phase of the calibrated magnetic field remains unchanged. In this way, the self-calibration of the magnetometer for the determined circularly polarized magnetic field is completed.

[0011] The first lock-in amplifier includes a first demodulator having a first input terminal, a second input terminal, a third input terminal, and a fourth output terminal, a first oscillator having a fifth setting terminal and a sixth output terminal, and a proportional-integral controller having a nineteenth input terminal and a twentieth output terminal. The second lock-in amplifier includes a second demodulator having a seventh input terminal, an eighth input terminal, a ninth setting terminal, and a tenth output terminal, and a second oscillator having an eleventh setting terminal and a twelfth output terminal. The third lock-in amplifier includes a third demodulator having a thirteenth input terminal, a fourteenth input terminal, a fifteenth setting terminal, and a sixteenth output terminal, a third oscillator having a Y-direction magnetic field coil frequency driving terminal and a seventeenth output terminal, and a phase-locked loop having a twenty-first input terminal and a twenty-second output terminal. The differential photodetector is connected to the first input terminal. The sixth output terminal is respectively connected to the second input terminal and the voltage-controlled current source is connected to the Z-direction magnetic field coil. The fourth output terminal is respectively connected to the seventh input terminal and the thirteenth input terminal. The twelfth output terminal is respectively connected to the X-direction magnetic field coil, the Y-direction magnetic field coil, and the eighth input terminal. The tenth output terminal is connected to the nineteenth input terminal. The twentieth output terminal is connected to the third input terminal. The seventeenth output terminal is respectively connected to the Y-direction magnetic field coil and the fourteenth input terminal. The sixteenth output terminal is connected to the twenty-first input terminal. The twenty-second output terminal is connected to the Y-direction magnetic field coil frequency driving terminal.

[0012] The detection light of the gas cell comes from a detection light generator. The detection light emitted by the detection light generator sequentially passes through a second polarization beam splitter prism, a noise attenuator, a second Glan-Taylor prism, the gas cell, a half-wave plate, and a Wollaston prism and then enters the differential photodetector. The pumping light of the gas cell comes from a pumping light generator. The pumping light emitted by the pumping light generator sequentially passes through a first polarization beam splitter prism, a tapered amplifier, a first convex lens, an acousto-optic modulator, a diaphragm, a second fiber collimator, a third fiber collimator, a beam splitter, a mirror, a first Glan-Taylor prism, and a quarter-wave plate and then traverses the gas cell. The second polarization beam splitter prism is connected to a wavelength meter through a fourth fiber collimator. The first polarization beam splitter prism is connected to a wavelength meter through a first fiber collimator. The beam splitter is connected to a photodetector through a second convex lens.

[0013] including the following expression: B ac (t) = B1cos(ω0t + θ ac ), where B ac (t) is the modulation magnetic field in the Z direction, B1 is the amplitude of the modulation magnetic field in the Z direction, ω0 is the frequency of the modulation magnetic field in the Z direction, t is time, and θ ac is the phase of the modulation magnetic field in the Z direction.

[0014] The first demodulation reference signal where θ is the phase shift of the reference signal during the first demodulation, e is the natural constant, and i is the imaginary unit; the second demodulation reference signal where ω is the frequency of the transverse magnetic field.

[0015] The detection light on the light output side of the gas cell is divided into two polarized lights with different powers by the Wollaston prism and respectively incident on two diodes of the differential photodetector to obtain a Faraday rotation angle θ x proportional to the average value <S FR > of the rubidium atom spin ensemble in the X direction. The expression of θ FR is as follows:

[0016]

[0017] where; e is the classical electron radius, c is the speed of light in vacuum, is the oscillator strength of the rubidium atom D2 line, L is the length of the atomic gas cell, T is the temperature of the gas cell, and n Rb (T) is the rubidium atom number density dependent on the temperature T; v prFbe , are respectively the frequency of the detection laser and the resonance frequency of the rubidium atom D2 line, is the full width at half maximum of the rubidium atom absorption peak.

[0018] The phase of the calibrated circularly polarized magnetic field after the second demodulation is set as ∠C (2)(θ), then ∠C (2) The expression of (θ) is as follows:

[0019]

[0020] where k(η) is an intermediate quantity, k(η) = [J0(η) + J -2 (η)] / [J0(η) - J -2 (η)], η is the modulation depth, J0(η) is the Bessel function of the first kind of order zero with η as the argument, J -2 (η) is the Bessel function of the first kind of order -2 with η as the argument, θ is the phase shift of the reference signal during the first demodulation, Θ (y) is the working point phase at which the magnetometer most suppresses the magnetic field in the Y direction, ω is the frequency of the circularly polarized calibration magnetic field; Γ2 is the linewidth of rubidium.

[0021] The technical effects of the present invention are as follows: A method for implementing a nuclear magnetic resonance atomic gyroscope based on a self-calibrating magnetometer of the present invention mainly consists of two systems: 1. Measuring the magnetic field signal around the gas chamber using polarized alkali metal rubidium atoms; 2. Measuring the rotational signal of the equivalent carrier through the precession frequency signal of polarized noble gas xenon in a magnetic field. The core component of the invention is a glass gas chamber filled with two isotopic gases of alkali metal rubidium atoms and xenon gas. The polarization is transferred to the nuclear spin of the noble gas atom xenon through the spin-exchange collision of the polarized alkali metal rubidium atoms. The noble gas atoms are continuously driven by the excitation magnetic field, and then the noble gas starts to precess under the drive and generates an equivalent circularly polarized magnetic field, which is measured by the polarized alkali metal rubidium atom magnetometer. The frequency of the equivalent magnetic field generated by the precession is the precession frequency of the noble gas. The excitation magnetic field frequency is maintained in resonance through the negative feedback of the phase-locked loop, realizing the closed-loop operation of the gyroscope. In the magnetometer system, a calibration circularly polarized magnetic field with a fixed frequency and amplitude is actively applied, and the magnetometer monitors the phase of the calibration magnetic field and keeps the measured calibration magnetic field phase unchanged through the feedback of the phase-locked loop, thereby realizing the self-calibration of the magnetometer for the measurement of the determined circularly polarized magnetic field. At the same time, it can avoid the interference introduced by the instability of the external environment to the magnetometer measurement, thereby ultimately improving the long-term stability of the gyroscope system. At the same time, this method reduces the requirements for the stability of the electronic equipment of the gyroscope and does not require the introduction of more equipment, which is beneficial to the miniaturization of the gyroscope. Description of the Drawings

[0022] Figure 1 It is a schematic structural diagram of the device involved in implementing a method for implementing a nuclear magnetic resonance atomic gyroscope based on a self-calibrating magnetometer of the present invention.

[0023] Figure 2 It is a schematic structural diagram of the device for applying the calibration circularly polarized magnetic field and the excitation magnetic field.

[0024] Figure 3Schematic diagram of the optical path structures of the pumping light and the probing light.

[0025] Reference numerals are as follows: A1 - gas chamber (containing alkali metal rubidium Rb atoms + inert gas xenon Xe, Rb includes 85 Rb and 87 Rb, Xe includes 129 Xe and 131 Xe); A2 - heating furnace (such as boron nitride ceramic heating furnace); A3 - magnetic field coil in the X direction; A4 - magnetic field coil in the Y direction; A5 - magnetic field coil in the Z direction; A6 - pumping light generator or pumping light; A7 - probing light generator or probing light; A8 - magnetic shielding cylinder; A9 - differential photodetector; A10 - first lock-in amplifier; A11 - second lock-in amplifier; A12 - third lock-in amplifier; A13 - proportional-integral controller; A14 - phase-locked loop; A15 - frequency drive end of the magnetic field coil in the Y direction; A16 - output electrical signal of the differential photodetector; A17 - real part of the complex value signal after the first demodulation; A18 - calibrated circularly polarized magnetic field phase after the second demodulation; A19 - xenon nuclear spin precession phase after the second demodulation; Demod1 to Demod3 - the 1st demodulator to the 3rd demodulator; Osc1 to Osc3 - the 1st oscillator to the 3rd oscillator; B1 to B3 - the 1st to the 3rd input terminals; B4 - the 4th output terminal; B5 - the 5th setting terminal; B6 - the 6th output terminal; B7 to B8 - the 7th to the 8th input terminals; B9 - the 9th setting terminal; B10 - the 10th output terminal; B11 - the 11th setting terminal; B12 - the 12th output terminal; B13 to B14 - the 13th to the 14th input terminals; B15 - the 15th setting terminal; B16 - the 16th output terminal; B17 - the 17th output terminal; B18 - voltage-controlled current source; B19 - the 19th input terminal; B20 - the 20th output terminal; B21 - the 21st input terminal; B22 - the 22nd output terminal; C1 - tapered amplifier; C2 - wavemeter; C3 - first polarization beam splitter prism; C4 - first fiber optic collimator; C5 - first convex lens; C6 - acousto-optic modulator; C7 - aperture; C8 - second fiber optic collimator; C9 - third fiber optic collimator; C10 - beam splitter; C11 - second convex lens; C12 - photodetector; C13 - mirror (45°); C14 - first Glan-Taylor prism; C15 - quarter-wave plate; C16 - second polarization beam splitter prism; C17 - fourth fiber optic collimator; C18 - noise attenuator; C19 - second Glan-Taylor prism; C20 - half-wave plate; C21 - Wollaston prism; XYZ - three axes of the Cartesian coordinate system (x-axis, y-axis, z-axis). Detailed implementation manners

[0026] The present invention will be described below with reference to the accompanying drawings ( Figures 1-3 ) and embodiments.

[0027] Figure 1 Schematic diagram of the device structure involved in implementing a method for realizing a nuclear magnetic resonance atomic gyroscope based on a self-calibrating magnetometer according to the present invention. Figure 2 Schematic diagram of the device structure for applying a circularly polarized magnetic field and an excitation magnetic field for calibration. Figure 3 Schematic diagram of the optical path structure of the pumping light and the probing light. Refer to Figures 1 to 3 As shown, in a method for realizing a nuclear magnetic resonance atomic gyroscope based on a self-calibrating magnetometer, a differential photodetector A9 transmits a differential photodetector output electrical signal A16 formed by the probing light A7 on the light output side of the gas cell A1 to a first lock-in amplifier A10. The first lock-in amplifier A10 performs a first demodulation on the differential photodetector output electrical signal A16, and respectively transmits the real part A17 of the complex-valued signal after the first demodulation to a second lock-in amplifier A11 and a third lock-in amplifier A12 for respective second demodulations. The second lock-in amplifier A11 negatively feeds back the calibrated circularly polarized magnetic field phase A18 after its second demodulation to the first lock-in amplifier A10 through a proportional-integral controller A13 to determine the self-calibration of the circularly polarized magnetic field. The third lock-in amplifier A12 negatively feeds back the nuclear spin precession phase A19 after its second demodulation to the excitation magnetic field frequency driving end A15 through a phase-locked loop A14 to maintain nuclear magnetic resonance and realize the closed-loop operation of the gyroscope.

[0028] The gas cell A1 is an atomic gas cell containing alkali metal rubidium (Rb) atoms and inert gas xenon (Xe) atoms. Among them, Rb includes 85 Rb and 87 Rb, and Xe includes 129 Xe and 131 Xe. An electric heating furnace A2, an excitation coil, and a magnetic shielding cylinder A8 are sequentially arranged outward around the gas cell A1. The excitation coil includes an X-direction magnetic field coil A3, a Y-direction magnetic field coil A4, and a Z-direction magnetic field coil A5. The Z-direction magnetic field coil A5 has two sets of coils to respectively apply a Z-direction bias magnetic field and a Z-direction modulation magnetic field. The Z-direction bias magnetic field is used to define the polarization axis direction of the system and is parallel to the pumping light direction. The pumping light continuously polarizes the rubidium atoms, and transfers the polarization to the xenon nuclei through collisions. At this time, the polarizations of both the rubidium atoms and the xenon nuclei are along the direction of the pumping light. The frequency of the Z-direction modulation magnetic field is the Larmor precession frequency of the alkali metal rubidium atoms. The X-direction magnetic field coil and the Y-direction magnetic field coil are used to generate a circularly polarized calibration magnetic field and a xenon nucleus precession excitation magnetic field.

[0029] The differential photodetector A9 measures the optical powers of two linearly polarized components of the linearly polarized detection light in the X direction, and then uses a differential circuit to obtain an output electrical signal of optical polarization rotation. The self-calibration of the circularly polarized magnetic field is to transversely apply a circularly polarized calibration magnetic field with a determined frequency and amplitude, and the signal phase of the calibrated magnetic field detected by the magnetometer is negatively fed back to the phase shift of the first demodulation reference signal of the magnetometer through the proportional-integral controller A13 so that the detected phase of the calibrated magnetic field remains unchanged. In this way, the self-calibration of the magnetometer for the determined circularly polarized magnetic field is completed.

[0030] The first lock-in amplifier A10 includes a first demodulator Demod1 having a first input terminal B1, a second input terminal B2, a third input terminal B3, and a fourth output terminal B4, a first oscillator Osc1 having a fifth setting terminal B5 and a sixth output terminal B6, and a proportional-integral controller A13 having a nineteenth input terminal B19 and a twentieth output terminal B20. The second lock-in amplifier A11 includes a second demodulator Demod2 having a seventh input terminal B7, an eighth input terminal B8, a ninth setting terminal B9, and a tenth output terminal B10, and a second oscillator Osc2 having an eleventh setting terminal B11 and a twelfth output terminal B12. The third lock-in amplifier A12 includes a third demodulator Demod3 having a thirteenth input terminal B13, a fourteenth input terminal B14, a fifteenth setting terminal B15, and a sixteenth output terminal B16, a third oscillator Osc3 having a Y-direction magnetic field coil frequency drive terminal A15 and a seventeenth output terminal B17, and a phase-locked loop A14 having a twenty-first input terminal B21 and a twenty-second output terminal B22. The differential photodetector A9 is connected to the first input terminal B1. The sixth output terminal B6 is respectively connected to the second input terminal B2 and the voltage-controlled current source B18 which is connected to the Z-direction magnetic field coil A5. The fourth output terminal B4 is respectively connected to the seventh input terminal B7 and the thirteenth input terminal B13. The twelfth output terminal B12 is respectively connected to the X-direction magnetic field coil A3, the Y-direction magnetic field coil A4, and the eighth input terminal B8. The tenth output terminal B10 is connected to the nineteenth input terminal B19. The twentieth output terminal B20 is connected to the third input terminal B3. The seventeenth output terminal B17 is respectively connected to the Y-direction magnetic field coil A4 and the fourteenth input terminal B14. The sixteenth output terminal B16 is connected to the twenty-first input terminal B21. The twenty-second output terminal B22 is connected to the Y-direction magnetic field coil frequency drive terminal A15.

[0031] The detection light of the gas chamber A1 comes from the detection light generator A7. The detection light emitted by the detection light generator A7 passes through the second polarization beam splitter prism C16, the noise attenuator C18, the second Glan-Taylor prism C19, the gas chamber A1, the half-wave plate C20, and the Wollaston prism C21 in sequence and then enters the differential photodetector A9. The pumping light of the gas chamber A1 comes from the pumping light generator A6. The pumping light emitted by the pumping light generator A6 passes through the first polarization beam splitter prism C3, the tapered amplifier C1, the first convex lens C5, the acousto-optic modulator C6, the aperture C7, the second fiber collimator C8, the third fiber collimator C9, the beam splitter C10, the mirror C13, the first Glan-Taylor prism C14, and the quarter-wave plate C15 in sequence and then passes through the gas chamber A1. The second polarization beam splitter prism C16 is connected to the wavelength meter C2 through the fourth fiber collimator C17. The first polarization beam splitter prism C3 is connected to the wavelength meter C2 through the first fiber collimator C4. The beam splitter C10 is connected to the photodetector C12 through the second convex lens C11.

[0032] Includes the following expression: B ac (t) = B1cos(ω0t + θ ac ), where B ac (t) is the modulation magnetic field in the Z direction, B1 is the amplitude of the modulation magnetic field in the Z direction, ω0 is the frequency of the modulation magnetic field in the Z direction, t is time, and θ ac is the phase of the modulation magnetic field in the Z direction. The first demodulation reference signal where θ is the phase shift of the reference signal during the first demodulation, e is the natural constant, and i is the imaginary unit; The second demodulation reference signal where ω is the frequency of the transverse magnetic field.

[0033] The detection light on the light output side of the gas chamber is divided into two polarized lights with different powers by the Wollaston prism and are respectively incident on the two diodes of the differential photodetector to obtain the Faraday rotation angle θ x proportional to the average value <S FR > of the rubidium atom spin ensemble in the X direction. The expression of θ FR is as follows:

[0034]

[0035] where r e is the classical electron radius, c is the speed of light in vacuum, is the oscillator strength of the rubidium atom D2 line, L is the length of the atomic gas chamber, T is the temperature of the gas chamber, and n Rb (T) is the rubidium atom number density dependent on the temperature T; v prFbe , are respectively the frequency of the detection laser and the resonance frequency of the rubidium atom D2 line, is the full width at half maximum of the rubidium atom absorption peak.

[0036] The calibrated circularly polarized magnetic field phase after the second demodulation is set as ∠C (2) (θ), then the expression of ∠C (2) (θ) is as follows:

[0037]

[0038] where k(η) is an intermediate quantity, k(η) = [J0(η) + J -2 (η)] / [O0(η) - O -2 (η)], η is the modulation depth, O0(η) is the Bessel function of order 0 with η as the argument, O -2 (η) is the Bessel function of order -2 with η as the argument, θ is the phase shift of the reference signal during the first demodulation, Θ (y) is the working point phase at which the magnetometer most suppresses the magnetic field in the Y direction, ω is the frequency of the circularly polarized calibration magnetic field; Γ2 is the linewidth of rubidium.

[0039] The present invention relates to the field of atomic devices, and specifically to a method for implementing a nuclear magnetic resonance gyroscope that uses a self - calibrated magnetometer to improve the long - term stability of the gyroscope based on nuclear magnetic resonance technology.

[0040] A method for implementing a nuclear magnetic resonance atomic gyroscope with a self - calibrated magnetometer improves the accuracy of the magnetometer's detection of the magnetic field phase signal to within 3 mdeg (mdeg, millidegree), and has the advantages of simple operation, low cost, and high stability. In conventional experiments, the error in the magnetic field phase detection caused by the influence of experimental variables on the magnetometer itself is not considered, which limits the gyroscope from reaching the theoretical limit performance. The active method of using negative feedback to keep the calibrated circularly polarized magnetic field phase unchanged in the present invention can significantly improve the stability of the gyroscope after compensating for the error caused by experimental variables on the magnetic field phase detection. Its typical long - term stability parameter can reach: 0.2° / h.

[0041] A method for realizing a nuclear magnetic resonance atomic gyroscope based on a self-calibrating magnetometer, characterized in that: a gas chamber filled with alkali metal rubidium atoms and inert gas xenon is heated to reach the working state, and then circularly polarized pumping laser is used to polarize the alkali metal rubidium atoms, and linearly polarized detection laser is used to detect the Faraday rotation signal generated by the alkali metal rubidium atoms in the gas chamber; the pumping direction is set as the Z direction, the detection direction is the X direction, and the direction horizontally perpendicular to the detection direction is the Y direction, where the Z direction is the longitudinal direction and the X and Y directions are the transverse directions. We use the spin polarization of alkali metal rubidium atoms to polarize the nuclear spin of inert gas xenon, and the alkali metal rubidium atoms can be used as an internal magnetometer to detect the precession signal of xenon nuclear spin. To prepare the alkali metal rubidium atom magnetometer, we apply a fixed DC bias magnetic field in the Z direction and an AC magnetic field modulation with the same amplitude, where the frequency of the modulation magnetic field is the Larmor precession frequency of the alkali metal rubidium atoms. The Faraday rotation signal generated by the linearly polarized laser in the X direction passing through the gas chamber is converted into an electrical signal by a differential photodetector, and the electrical signal is referenced to the Larmor precession frequency and a suitable reference signal phase shift is selected for the first demodulation to obtain a magnetic field signal that is only proportional to the X direction. The signal after the first demodulation is then referenced to the frequency of the magnetic field to be measured for the second demodulation, and finally the phase and amplitude of the magnetic field signal of the transverse frequency to be measured are obtained. When we apply an excitation magnetic field equal to the Larmor frequency of the inert gas xenon in the Y direction to drive the nuclear spin of the inert gas xenon to generate a stable precession equivalent transverse magnetic field, and then the phase signal of the xenon nuclear precession in the X direction detected by the magnetometer is fed back to the frequency of the excitation magnetic field through a phase-locked loop circuit negative feedback, so as to realize the closed-loop operation. The function of the gyroscope is realized by using the internal alkali metal rubidium atom magnetometer to obtain the nuclear spin signal of the inert gas xenon to obtain the rotation information, and thus the preparation work of the system is completed. Finally, we apply a circularly polarized calibration magnetic field with a certain frequency and amplitude in the transverse direction, and the signal phase of the calibration magnetic field detected by the magnetometer is fed back to the first demodulation reference signal phase shift of the magnetometer through a proportional-integral controller to keep the detected calibration magnetic field phase unchanged, so that the self-calibration of the magnetometer for the determined circularly polarized magnetic field is completed.

[0042] A method for realizing a nuclear magnetic resonance atomic gyroscope based on a self-calibrating magnetometer according to the present invention includes: an atomic gas chamber containing alkali metal rubidium atoms and inert gas xenon, a temperature control unit, an excitation unit, a magnetic shielding barrel, a laser control unit, and a signal acquisition and processing unit;

[0043] The atomic gas chamber containing alkali metal rubidium atoms and inert gas xenon and the heating unit and the excitation coil are placed in a magnetic shielding barrel made of permalloy, and the magnetic shielding barrel shields the interference of the geomagnetic field and external stray magnetic fields;

[0044] The temperature control unit heats the atoms in the atomic cell containing alkali metal rubidium atoms and inert gas xenon, so that the rubidium atoms are in a stable high atomic number density; because the atomic number density of alkali metal rubidium atoms at room temperature is relatively low, the detection sensitivity of the magnetometer and the polarization rate of hyperpolarized xenon are low. Generally, the system needs to work at a temperature of about 100 °C to obtain a sufficiently large signal-to-noise ratio. The heating module is a polyimide heating film, which is driven by high-frequency alternating current to achieve heating. The heating sheet is closely attached to the four sides of the atomic cell, and a temperature sensor PT1000 is adhered to the bottom of the cell. The temperature of the cell is feedback-controlled within ±0.01 °C by a temperature control instrument;

[0045] The excitation unit includes excitation coils in three directions of X, Y, and Z, which are used to provide the DC bias magnetic field, modulation magnetic field, and excitation magnetic field required for the operation of the gyroscope respectively. There are two sets of coils in the Z direction, which apply the bias magnetic field and modulation magnetic field in the Z direction respectively. The bias magnetic field is used to define the polarization axis direction of the system and is parallel to the pumping light direction. The pumping light continuously polarizes the rubidium atoms, and transfers the polarization to the xenon atomic nucleus through collisions. At this time, the polarizations of both the rubidium atoms and the xenon atomic nucleus are along the direction of the pumping light. By applying the bias magnetic field and modulation magnetic field in the Z-direction magnetic field unit, the Larmor frequency and modulation frequency of the alkali metal rubidium atoms are determined by the bias magnetic field, and at the same time, the Larmor precession frequency of the xenon nucleus is also determined by the bias magnetic field; under the drive of the single-frequency alternating excitation magnetic field applied in the Y-direction magnetic field unit, part of the xenon nuclear polarization will precess transversely at the frequency of the excitation magnetic field. At the same time, a circularly polarized calibration magnetic field with a fixed frequency and amplitude needs to be actively applied transversely. The atomic cell and the heating sheet are both inside the excitation coils, and the atomic cell should be as close as possible to the center position of the magnetic field coils;

[0046] The signal acquisition and processing system consists of a differential photodetector and three lock-in amplifiers. The detector measures the optical powers of the two linearly polarized components of the linearly polarized detection light in the X direction, and then uses a differential circuit to obtain the signal of the optical polarization rotation. This signal is input to the input port of the lock-in amplifier for data processing; the lock-in amplifier includes a phase-locked loop or a proportional-integral controller and an oscillator. The lock-in amplifier performs phase-sensitive detection on the signal and separates and detects the part near a certain frequency in the signal; on the one hand, the oscillator signal of the lock-in amplifier is used as the output signal to drive the precession of the xenon isotope and modulate the alkali metal rubidium atoms. On the other hand, it is also used as a reference signal to demodulate the output signal of the photodetector. After demodulation, the precession signals of the respective xenon isotopes or the phase and amplitude signals of the transverse calibration circularly polarized magnetic field can be obtained; the xenon precession phase signal is fed back to the frequency of the excitation magnetic field through the phase-locked loop to achieve a closed loop, and the phase signal of the calibration circularly polarized magnetic field is fed back to the phase shift of the first demodulation reference signal through the proportional-integral controller to achieve a closed loop.

[0047] The implementation method of the present invention is as follows: An atomic cell containing alkali metal rubidium atoms and inert gas xenon, a heating sheet, and excitation coils in the X, Y, and Z directions are all placed in a three-layer magnetic shielding barrel, which plays a role in shielding external stray magnetic fields. Other component units are placed on a platform to collect experimental data, and then the data is processed; the designed heating sheet heats the atoms through alternating current of about 400 kHz, and the temperature needs to reach about 100 °C to keep the atoms in a stable high atomic number density to obtain a sufficiently high signal-to-noise ratio; during the operation of the atomic gyroscope, the pumping light continuously polarizes the rubidium atoms and transfers the polarization to the xenon atomic nucleus through collisions. A stable bias magnetic field B0 is generated by a highly stable current source on a set of magnetic field coils in the Z direction, where B0 = 180 mG. A modulation magnetic field B ac (t) is applied on another set of magnetic field coils in the Z direction, where t is time, and B ac (t) = B1cos(ω0t + θ ac ), where ω0 = γ Rb B0 = 2π × 85.5 kHz is the frequency of the modulation magnetic field, B1 = B0 is the amplitude of the modulation magnetic field, and θ ac is the phase of the modulation magnetic field. γ Rb is the gyromagnetic ratio of alkali metal rubidium atoms; under the working conditions of the bias magnetic field B0 and the modulation magnetic field B ac (t), the linearly polarized light in the X direction passes through the cell and uses the Faraday rotation effect to convert the transverse polarization signal of the alkali metal atoms into the rotation of the light polarization angle. Finally, the optical signal is extracted by a differential photodetector to detect the rotation of the light polarization and converted into an electrical signal. The electrical signal is input into a lock-in amplifier for the first demodulation by referring to the modulation magnetic field frequency. It is necessary to select an appropriate reference signal phase shift so that the signal after the first demodulation is only proportional to the magnetic field in the X direction. At this time, we actively apply a set of calibration circularly polarized calibration magnetic fields with a determined frequency and amplitude in the transverse direction, and the signal after the first demodulation is demodulated again by referring to the calibration circularly polarized magnetic field frequency. At this time, the amplitude and phase of the calibration circularly polarized magnetic field signal can be obtained. For the calibration magnetic field with a fixed frequency, amplitude, and phase, the phase of the calibration magnetic field measured by the magnetometer changes with the uncertain factors of the system, and the measurement error caused by this will directly affect the long-term stability of the gyroscope. Therefore, after the spin of the inert gas xenon nucleus is closed-loop, it is fed back to the phase shift of the first demodulation reference signal through a proportional-integral controller to keep the phase of the measured calibration magnetic field signal unchanged. In this way, the magnetometer calibrates the detection of the signal with a fixed frequency, phase, and amplitude in real time, fully compensating for the measurement error caused by the unstable factors of the system.

[0048] Before calibrating the phase of the closed-loop magnetometer, two sets related to 129 Xe and131 An alternating excitation magnetic field with the same Larmor frequency as Xe causes the polarization of some xenon nuclei to precess in the transverse direction at the frequency of the excitation magnetic field and generate an equivalent transverse magnetic field. Since our magnetometer is only sensitive to the magnetic field in the X direction, the magnetometer is not sensitive to the excitation magnetic field in the Y direction, and only the precession signals of two groups of xenon nuclei in the X direction are detected. Similarly, we can obtain the amplitudes and phases of the precession signals of the two groups of xenon nuclei, and use a phase-locked loop to close-loop feedback the precession phases of the two groups of xenon nuclei to two different excitation field frequencies respectively, so that the nuclear magnetic resonance atomic gyroscope is in a closed-loop working state. In this way, the external rotation is equivalent to the change in the polarized magnetic field, and its change will cause the change in the closed-loop feedback signal of the phase-locked loop, and the frequency of the excitation magnetic field output by the oscillator will also change accordingly. In this way, the feedback amount of the frequency is collected at the frequency measurement end to realize the detection of rotation, and the function of the nuclear magnetic resonance atomic gyroscope based on the self-calibrating magnetometer is realized.

[0049] Advantages and positive effects of the present invention: In the present invention, a modulation magnetic field is used to modulate the spin of alkali metal rubidium atoms, and then the low-frequency noise of the signal is suppressed by demodulation, improving the sensitivity of the magnetometer. And by actively applying a calibrated circularly polarized magnetic field signal and compensating for the measurement errors caused by modulation and the instrument through negative feedback closed-loop, the high-sensitivity and high-accuracy measurement of the magnetic field phase is realized, thereby improving the long-term stability of the gyroscope.

[0050] As Figure 1 shown, a method for realizing a nuclear magnetic resonance atomic gyroscope based on a self-calibrating magnetometer includes a gas chamber A1 of inert gas xenon of alkali metal rubidium atoms, a boron nitride ceramic heating furnace A2, an X-direction magnetic field coil A3, a Y-direction magnetic field coil A4, a Z-direction magnetic field coil A5, a pumping light A6, a detection light A7, a magnetic shielding cylinder A8, a differential photodetector A9, a first lock-in amplifier A10, a second lock-in amplifier A11, a third lock-in amplifier A12, a proportional-integral controller A13, a phase-locked loop A14, an oscillator frequency driving end A15 for driving the Y-direction magnetic field coil, an electrical signal A16 output by the differential photodetector, a real part A17 of the complex value signal after the first demodulation, a phase A18 of the calibrated circularly polarized magnetic field after the second demodulation, and a phase A19 of the xenon nuclear spin precession after the second demodulation.

[0051] As Figure 2As shown, a method for implementing a nuclear magnetic resonance atomic gyroscope based on a self-calibrating magnetometer includes a first demodulator Demod1 and a first oscillator Osc1 of a first phase-locked amplifier A10, an input B1 of the demodulated signal of the first demodulator Demod1, a reference signal input B2 of the first demodulator Demod1, a reference signal phase shift B3 of the first demodulator Demod1, an output B4 of the demodulated signal after the first demodulator Demod1, a frequency setting terminal B5 of the first oscillator Osc1, an oscillation signal output B6 of the first oscillator Osc1, a second demodulator Demod2 and a second oscillator Osc2 of a second phase-locked amplifier A11, a signal input B7 of the second demodulator Demod2, a reference signal input B8 of the second demodulator Demod2, a reference signal phase shift setting terminal B9 of the second demodulator Demod2, an output B10 of the demodulated signal after the second demodulator Demod2, a frequency B11 of the second oscillator Osc2, an oscillation signal output B12 of the second oscillator Osc2, a third demodulator Demod3 and a third oscillator Osc3 of a third phase-locked amplifier A12, a signal input B13 of the third demodulator Demod3, a reference signal input B14 of the third demodulator Demod3, a reference signal phase shift B15 of the third demodulator Demod3, an output B16 of the demodulated signal after the third demodulator Demod3, a frequency A15 of the third oscillator Osc3, an oscillation signal output B17 of the third oscillator Osc3, a voltage-controlled current source B18 for driving a Z-direction coil A5, a calibrated magnetic field phase signal input B19 of a proportional-integral controller A13, a phase feedback quantity output B20 of the proportional-integral controller A13, a xenon precession phase signal input B21 of a phase-locked loop A14, and a frequency feedback quantity output B22 of the phase-locked loop A14.

[0052] As Figure 3 shown, a method for implementing a nuclear magnetic resonance atomic gyroscope based on a self-calibrating magnetometer includes a conical amplifier C1, a wavemeter C2, a polarization beam splitter prism C3, an optical fiber collimator C4, a convex lens C5, an acousto-optic modulator C6, a diaphragm C7, an optical fiber collimator C8, an optical fiber collimator C9, a beam splitter C10, a convex lens C11, a photodetector C12, a 45° reflector C13, a Glan-Taylor prism C14, a quarter-wave plate C15, a polarization beam splitter prism C16, an optical fiber collimator C17, a noise attenuator C18, a Glan-Taylor prism C19, a half-wave plate C20, and a Wollaston prism C21.

[0053] As Figure 1As shown, a magnetic shield A8 and a gas chamber A1. The test of the present invention is carried out in a cylindrical magnetic shield barrel A8 made of three layers of permalloy. The shield barrel can shield the geomagnetic field and the interference of external stray magnetic fields to obtain a magnetic field with a controllable magnitude. Among them, the atomic gas chamber is the core part of the invention. The atomic gas chamber A1 is a glass wall filled with natural abundance rubidium atoms (where 85 Rb and 87 Rb have contents of 72.2% and 27.8% respectively), 4 Torr 12g Xe, 14 Torr 131 Xe, 450 Torr N2 as buffer gas.

[0054] Heating unit. We can control the number density of rubidium atoms in the gas chamber by controlling the temperature of the gas chamber A1. In the experiment, the signal intensity detected by the probe light is positively correlated with the number density of rubidium atoms. In order to optimize the signal-to-noise ratio of the signal, the gas chamber needs to be heated to about 100 °C. For this purpose, we designed a heating furnace A2 made of high thermal conductivity and non-magnetic material boron nitride ceramic to place the atomic gas chamber. The heating furnace is provided with a light passing hole to facilitate the use of laser pumping and probing atoms in the experiment. By heating the entire heating furnace, the temperature of the heating furnace and the gas chamber is controlled at the set temperature. The outer wall of the heating furnace is pasted with non-magnetic heating sheets and a temperature sensor PT1000. The temperature sensor is connected to a temperature controller. The internal PID algorithm of the temperature controller outputs a modulation voltage, which is fed back to the amplitude of the AC signal output by the signal generator. Finally, the amplified AC current is fed to the heating sheet through a power amplifier. The heating furnace and the heating sheet are wrapped by heat insulation materials, and the final temperature control can be accurate to ±0.01 °C. In the experiment, the working frequency of the magnetometer is usually about 100 kHz. In order to avoid the interference of the heating current on the signal of the magnetometer, we usually set the heating current frequency at a position far higher than the working frequency of the magnetometer, about 400 kHz. At the same time, a suitable low-pass filter is selected when processing the magnetometer or gyro signal to suppress the influence of electric heating to the lowest level.

[0055] Magnetic field coils A3, A4 and A5. When the magnetometer and the nuclear magnetic resonance gyroscope work, different magnetic fields need to be applied in all directions. Using coils and current sources can conveniently obtain the required magnetic field. There are two sets of Helmholtz coils A5 in the Z direction. One set of coils is driven by a low-noise current source to provide a bias magnetic field. The magnitude of the bias magnetic field B0 is 180 mG. The other set of Helmholtz coils is provided with an oscillation signal by the oscillator of the first lock-in amplifier and output to a voltage-controlled current source. Finally, the coil is driven by the voltage-controlled current source to generate a modulated magnetic field B ac (t), where B ac (t) = B1cos(ω0t + θ ac ), ω0 is the modulation magnetic field frequency and is the same as the Larmor frequency: ω0 = γ Rb B0 = 2π × 85.5 kHz, γRb is the gyromagnetic ratio of the alkali metal rubidium atom, B1 is the amplitude of the modulation magnetic field and B1 = B0, t is time, and θ ac is the phase of the modulation magnetic field. There is a set of saddle-shaped coils A3 and A4 in the X and Y transverse directions respectively. Both sets of coils are driven by the alternating voltage provided by the oscillator of the second lock-in amplifier A11 or the third lock-in amplifier A12 to generate the circularly polarized calibration magnetic field and the magnetic field for exciting xenon precession that we need.

[0056] As Figure 3 shown, the optical path system of the pumping light A6 and the probing light A7. The optical path is mainly divided into two beams. The pumping light A6 is a laser with the frequency of the D1 line of rubidium atoms, and the wavelength is about 795 nm. After exiting from the laser, a beam is split by a polarization beam splitter C3 to a wavemeter C2 for measuring and controlling the wavelength of the pumping light. The other beam is amplified in power by a tapered amplifier C1, and then passes through an acousto-optic modulator C6 and only the first-order diffracted light is allowed to pass through using a diaphragm C7. The power ratio of the first-order diffracted light can be adjusted by the modulation voltage of the acousto-optic modulation driver. As Figure 3 shown in the dashed box, we use a beam splitter C10 to split a fixed proportion of the sampling light near the shielding cylinder, use a photodetector C12 to measure the power of the pumping sampling beam, and then lock the sampling light power at a set value through PID control to make the power of the pumping light incident into the shielding cylinder stable. Subsequently, in order to make the atomic polarization magnetic field distribution in the gas chamber as uniform as possible, the pumping light needs to cover the cross-section of the gas chamber as much as possible. We choose a large-diameter output fiber collimator C9 to expand the beam. The beam expanded by the fiber collimator C9 is adjusted to circularly polarized light by a Glan-Taylor prism C14 and a quarter-wave plate C15 and is incident horizontally along the Z direction into the magnetic shielding cylinder; the probing light A7 is a laser with a slight detuning from the D2 line of rubidium atoms, and the wavelength is about 780 nm. After exiting from the laser, a beam of light is split by a polarization beam splitter prism C16 and connected to a wavemeter C2 through a fiber collimator C17, which is also used to measure and control the wavelength; the other transmitted beam is controlled by a liquid crystal noise attenuator C18 to control the optical power of the probing light. Subsequently, after being purified by a Glan-Taylor prism C19, the polarization is incident horizontally into the magnetic shielding cylinder in the X direction. After passing through the gas chamber, the probing light is split by a Wollaston prism C21 into two polarized lights with orthogonal polarizations and an included angle of 20°, which are respectively incident into the two diodes of a differential photodetector A9 to measure the Faraday rotation angle, and the rotation angle θ FR depends on the spin ensemble average <S x > of the alkali metal rubidium atoms in the X direction:

[0057]

[0058] where r e is the classical electron radius; c is the speed of light in vacuum; is the oscillator strength of the D2 line of rubidium atoms; L is the length of the atomic gas chamber; T is the temperature of the gas chamber, and n Rb(T) is the number density of rubidium atoms dependent on temperature; v prFbe 、 are the frequency of the probe laser and the resonance frequency of the rubidium atom D2 line respectively; is the full width at half maximum of the rubidium atom absorption peak.

[0059] The object of the present invention is achieved by the following technical solutions: The atomic gas cell is heated to about 100 °C, the bias magnetic field B0 and the pumping light direction are parallel, the rubidium atoms and the xenon nuclear spins are polarized in the Z direction, and at the same time there is a modulation magnetic field B controlled by the signal B6 output by the first oscillator Osc1 in the Z direction ac (t)=B1cos(ω0t + θ ac ), where ω0 is the frequency of the modulation field, B1 is the amplitude of the modulation field, t is time, and θ ac is the phase of the modulation field. When there is an alternating magnetic field b x0 / y0 cos(ωt + φ x / y ) in the transverse direction, where b x0 / y0 and φ x / y are the amplitude and phase of the magnetic field in the X / Y direction respectively, and ω is the frequency of the transverse magnetic field. The rubidium atom spins deviate from the Z direction and produce a projection in the direction of the probe light. Subsequently, a Faraday rotation angle is generated by the probe light and enters the differential photodetector A9 to obtain an electrical signal A16 proportional to the average value of the rubidium atom spin ensemble in the X direction. Because of the existence of the modulation magnetic field, the signal not only has a signal of frequency ω, but also the signal is modulated by the modulation magnetic field of frequency ω0. Therefore, we can obtain the amplitude and phase of the final transverse magnetic field to be measured through two demodulations. The process is as Figure 2 : The electrical signal A16 proportional to the average value of the rubidium atom spin ensemble in the X direction obtained from the differential photodetector A9 is directly input to the first demodulator signal input B1 of the first lock-in amplifier A10. First, the signal output B6 of the first oscillator Osc1 that drives the modulation magnetic field in the Z direction with a frequency of ω0 is used as the reference signal and input to the reference signal input port B2 of the first demodulator Demod1. Among them, the reference signal for the first demodulation where θ is the phase shift of the reference signal during the first demodulation, θ is controlled by the B3 port of the first demodulator; ω0 is the frequency of the modulation magnetic field; t is time; e is the natural constant; i is the imaginary unit. Then the reference signal is multiplied by the signal A16 of the average value of the rubidium atom spin system in the X direction and demodulated for the first time through a low-pass filter. The real part A17 of the demodulated complex value signal is output from the signal output port B4 and then simultaneously input to the signal input port B7 of the second demodulator Demod2 and the signal input port B13 of the third demodulator Demod3. Similarly, the output signal B12 of the second oscillator Osc2 that drives the transverse magnetic field in the second lock-in amplifier is used as the reference signal for the second demodulation Input to the reference signal input port B8 of the second demodulator, where ω is the transverse magnetic field frequency. After the same demodulation operation, the complex-valued demodulation signal C of the transverse magnetic field is obtained after the second demodulation (2) (θ):

[0060]

[0061] where θ is the phase shift of the reference signal during the first demodulation, and θ is controlled by the B3 port of the first demodulator; b x0 / y0 and φ x / y are the amplitudes and phases of the magnetic fields in the X / Y directions respectively, driven by the second oscillator Osc2; <S z > is the average value of the rubidium atom spin ensemble in the Z direction; γ Rb is the gyromagnetic ratio of the rubidium atom; Γ2 is the line width of the rubidium atom; e is the natural constant; i is the imaginary unit; G (2 x / 2y) (θ) is the complex-valued gain function of the magnetometer in the X / Y directions determined by θ:

[0062]

[0063]

[0064]

[0065]

[0066] θ ac is the phase of the modulation magnetic field; η is defined by us as the modulation depth, and the modulation depth is the ratio of the Rabi frequency related to the amplitude B1 of the modulation field to the modulation field frequency: η = γ Rb B1 / ω0, J n is the Bessel function of the nth order with the modulation depth η as the argument; n is an integer; z n = nω0 / Γ2+(γ Rb B0 - ω0) / Γ2; x = ω / Γ2.

[0067] Therefore, we know that the signal finally obtained after the second demodulation, that is, the transverse magnetic field signal detected by the magnetometer, is determined by the amplitudes and phases of the magnetic fields in the X and Y directions and the corresponding complex-valued gain functions in the corresponding directions. The modulus and phase of this complex-valued signal are the amplitude and phase of the magnetic field to be measured detected by the magnetometer respectively.

[0068] Therefore, when finding the working point of the magnetometer, first actively apply a linearly polarized magnetic field (b x0 = 0) through the signal output B2 of the second oscillator Osc2 in the Y-direction coil A4, and by selecting an appropriate reference signal phase shift θ, make |C (2)(θ) | ≈ 0, the magnetometer is in the operating state where the magnetic field in the Y direction is most suppressed, and at this time θ = Θ (y) is the operating point of our magnetometer:

[0069]

[0070] where θ ac is the phase of the modulation magnetic field; q(η) = [J0(η) - J -2 (η)] / [J0(η) + J -2 (η)]; δ = (γ Rb B0 - ω0) / Γ2; ζ = ω0 / Γ2.

[0071] When the magnetometer is at an operating point where the phase shift θ of the first demodulation reference signal is near Θ (y) , we drive the X and Y direction coils with the signal output B12 of the second oscillator Osc2 to apply a left-handed circularly polarized calibration magnetic field with a frequency of ω. At this time, the phase ∠C (2) (θ) of the calibration magnetic field detected by the magnetometer after secondary demodulation is:

[0072]

[0073] ∠C (2) (θ) is also Figure 2 A18 in, where k(η) = [J0(η) + J -2 (η)] / [J0(η) - J -2 (η)], η is the modulation depth, γ Rb is the gyromagnetic ratio of rubidium; J0(η) / J -2 (η) is the 0th order / -2nd order Bessel function with the modulation depth η as the argument; ω is the frequency of the circularly polarized calibration magnetic field; Γ2 is the linewidth of rubidium; θ is the phase shift of the reference signal during the first demodulation. From the above, we can see that even if we apply a fixed circularly polarized magnetic field and a fixed phase shift of the first reference signal, the phase of the calibration circularly polarized magnetic field detected by the magnetometer still changes with the operating point Θ (y) and the change of the linewidth Γ2 of rubidium. The change of the operating point Θ (y) is related to the drift of the bias magnetic field B0 and the phase θ of the modulation magnetic field acIt is related to drift. These two physical drifts are related to the instability of the current source of the driving magnetic field; the change in the linewidth Γ2 of rubidium is related to the change in the gas cell temperature and the light intensity of the pumping light. Therefore, the magnetic field phase detected by our magnetometer will drift due to instrument, gas cell temperature, and laser instability factors. At this time, we input the phase A18 of the calibrated circularly polarized magnetic field detected into the input port B19 of the proportional-integral controller A13. The closed-loop set value of the proportional-integral controller is 0°. The phase feedback quantity output B20 is negatively fed back to the phase shift θ of the reference signal during the first demodulation controlled by the B3 port of the first demodulator, so that the measured phase of the calibrated circularly polarized magnetic field ∠C (2) (θ) = 0. In this way, we have completed the self-calibration of the magnetometer. By applying the calibrated circularly polarized calibration magnetic field and negatively feeding back the phase shift of the first reference signal, we can compensate for the measurement errors caused by various instability factors of the system.

[0074] Due to the characteristic that the phase of the calibrated circularly polarized magnetic field is proportional to the drift of the magnetic field in the Z direction within the working point range, we can realize a magnetometer for simultaneously measuring magnetic fields in three directions. Among them, the change in the magnitude of the magnetic field in the Z direction is proportional to the phase of the calibrated circularly polarized magnetic field, and the magnetic field in the X direction is proportional to the amplitude after the first demodulation signal. We can also find the working point where the amplitude of the signal after the first demodulation most suppresses the magnetic field in the X direction and is only sensitive to the magnetic field in the Y direction according to the same operation as above. At this time, we can obtain magnetic field signals in three directions simultaneously and realize the function of a three-axis magnetometer.

[0075] Before completing the self-calibration by negatively feeding back the phase shift of the first reference signal, we first need to make the xenon nuclear spin precess and close the loop to measure the precession frequency, as Figure 2 shown. We use the signal output B17 of the third oscillator Osc3 to drive the Y-direction coil A4 to generate a linearly polarized magnetic field that resonates with the Larmor frequencies of 129 Xe and 131 Xe, so that the xenon nuclei generate stable spin precession in the transverse direction and produce a projection in the direction of the probe light. Similarly, the oscillator signal driving the Y coil is input to the B14 port of the third demodulator Demod3 as the reference signal for the second demodulation signal A17. The measured phase signal A19 of the xenon nuclear precession is input to the input port B21 of the phase-locked loop A14 through the signal output port B16 of the third demodulator Demod3. The closed-loop set value of the phase-locked loop is 0°. The frequency feedback quantity B22 of the phase-locked loop is input to the frequency A15 of the third oscillator Ocs3, so that the measured phase of the xenon nuclear precession is always 0°. In this way, the frequency offset output by the phase-locked loop is the equivalent rotation information of the system through calculation, and the function of the nuclear magnetic resonance atomic gyroscope based on the self-calibrated magnetometer is realized.

[0076] The content not described in detail in the specification of the present invention belongs to the prior art well-known to those skilled in the art. It is hereby specified that the above description helps those skilled in the art to understand the present invention, but does not limit the protection scope of the present invention. Any implementation that makes equivalent substitutions, modifications and improvements and / or simplifies the above description without departing from the essential content of the present invention falls within the protection scope of the present invention.

Claims

1. A method for realizing a nuclear magnetic resonance atomic gyroscope based on a self-calibrating magnetometer, characterized in that, The differential optoelectronic detector transmits the differential optoelectronic detector output electrical signal formed by the detection light on the light-emitting side of the gas chamber to the first lock-in amplifier. The first lock-in amplifier performs the first demodulation on the differential optoelectronic detector output electrical signal, and respectively transmits the real parts of the complex-valued signals after the first demodulation to the second lock-in amplifier and the third lock-in amplifier for the second demodulation. The second lock-in amplifier negatively feeds back the calibrated circularly polarized magnetic field phase after its second demodulation to the first lock-in amplifier through a proportional-integral controller to determine the self-calibration of the circularly polarized magnetic field. The third lock-in amplifier negatively feeds back the nuclear spin precession phase after its second demodulation to the excitation magnetic field frequency driving end through a phase-locked loop to maintain nuclear magnetic resonance and realize the closed-loop operation of the gyroscope.

2. The implementation method of the nuclear magnetic resonance atomic gyroscope based on a self-calibrating magnetometer according to claim 1, wherein The gas cell is an atomic gas cell containing alkali metal rubidium (Rb) atoms and inert gas xenon (Xe), where Rb includes 85 Rb and 87 Rb, and Xe includes 129 Xe and 131 Xe.

3. The implementation method of the nuclear magnetic resonance atomic gyroscope based on a self-calibrating magnetometer according to claim 1, wherein A heating furnace, an excitation coil, and a magnetic shielding cylinder are sequentially arranged outward around the gas chamber. The excitation coil includes an X-direction magnetic field coil, a Y-direction magnetic field coil, and a Z-direction magnetic field coil. The Z-direction magnetic field coil has two sets of coils to respectively apply a Z-direction bias magnetic field and a Z-direction modulation magnetic field. The Z-direction bias magnetic field is used to define the polarization axis direction of the system and is parallel to the pumping light direction. The pumping light continuously polarizes rubidium atoms and transfers the polarization to xenon nuclei through collisions. At this time, the polarizations of both rubidium atoms and xenon nuclei are along the direction of the pumping light. The frequency of the Z-direction modulation magnetic field is the Larmor precession frequency of alkali metal rubidium atoms. The X-direction magnetic field coil and the Y-direction magnetic field coil are used to generate a circularly polarized calibration magnetic field and a xenon nucleus precession excitation magnetic field.

4. The implementation method of a nuclear magnetic resonance atomic gyroscope based on a self-calibrating magnetometer according to claim 1, characterized in that, The differential optoelectronic detector measures the optical powers of two linearly polarized components of the X-direction linearly polarized detection light, and then uses a differential circuit to obtain the output electrical signal of the optical polarization rotation. The self-calibration of the circularly polarized magnetic field is to transversely apply a circularly polarized calibration magnetic field with a determined frequency and amplitude, and negatively feed back the signal phase of the calibrated magnetic field detected by the magnetometer to the phase shift of the first demodulation reference signal of the magnetometer through a proportional-integral controller so that the detected calibrated magnetic field phase remains unchanged. In this way, the self-calibration of the circularly polarized magnetic field by the magnetometer is completed.

5. The implementation method of a nuclear magnetic resonance atomic gyroscope based on a self-calibrating magnetometer according to claim 1, wherein The first lock-in amplifier includes a first demodulator having a first input terminal, a second input terminal, a third input terminal, and a fourth output terminal, a first oscillator having a fifth setting terminal and a sixth output terminal, and a proportional-integral controller having a nineteenth input terminal and a twentieth output terminal. The second lock-in amplifier includes a second demodulator having a seventh input terminal, an eighth input terminal, a ninth setting terminal, and a tenth output terminal, and a second oscillator having an eleventh setting terminal and a twelfth output terminal. The third lock-in amplifier includes a third demodulator having a thirteenth input terminal, a fourteenth input terminal, a fifteenth setting terminal, and a sixteenth output terminal, a third oscillator having a Y-direction magnetic field coil frequency drive terminal and a seventeenth output terminal, and a phase-locked loop having a twenty-first input terminal and a twenty-second output terminal. A differential photodetector is connected to the first input terminal. The sixth output terminal is respectively connected to the second input terminal and a voltage-controlled current source connected to the Z-direction magnetic field coil. The fourth output terminal is respectively connected to the seventh input terminal and the thirteenth input terminal. The twelfth output terminal is respectively connected to the X-direction magnetic field coil, the Y-direction magnetic field coil, and the eighth input terminal. The tenth output terminal is connected to the nineteenth input terminal. The twentieth output terminal is connected to the third input terminal. The seventeenth output terminal is respectively connected to the Y-direction magnetic field coil and the fourteenth input terminal. The sixteenth output terminal is connected to the twenty-first input terminal. The twenty-second output terminal is connected to the Y-direction magnetic field coil frequency drive terminal.

6. The implementation method of a nuclear magnetic resonance atomic gyroscope based on a self-calibrating magnetometer according to claim 1, characterized in that The detection light of the gas chamber comes from a detection light generator. The detection light emitted by the detection light generator sequentially passes through a second polarization beam splitter prism, a noise attenuator, a second Glan-Taylor prism, the gas chamber, a half-wave plate, and a Wollaston prism and then enters the differential photodetector. The pump light of the gas chamber comes from a pump light generator. The pump light emitted by the pump light generator sequentially passes through a first polarization beam splitter prism, a tapered amplifier, a first convex lens, an acousto-optic modulator, a diaphragm, a second fiber collimator, a third fiber collimator, a beam splitter, a mirror, a first Glan-Taylor prism, and a quarter-wave plate and then traverses the gas chamber. The second polarization beam splitter prism is connected to a wavelength meter through a fourth fiber collimator. The first polarization beam splitter prism is connected to a wavelength meter through a first fiber collimator. The beam splitter is connected to a photodetector through a second convex lens.

7. The implementation method of the nuclear magnetic resonance atomic gyroscope based on a self-calibrating magnetometer according to claim 1, characterized in that, including the following expression: B ac (t) = B1 cos(ω0t + θ ac ), where B ac (t) is the modulation magnetic field in the Z direction, B1 is the amplitude of the modulation magnetic field in the Z direction, ω0 is the frequency of the modulation magnetic field in the Z direction, t is time, and θ ac is the phase of the modulation magnetic field in the Z direction.

8. The implementation method of a nuclear magnetic resonance atomic gyroscope based on a self-calibrating magnetometer according to claim 1, characterized in that, First demodulation reference signal where θ is the phase shift of the reference signal during the first demodulation, E is the natural constant, and i is the imaginary unit; Second demodulation reference signal where ω is the transverse magnetic field frequency.

9. The implementation method of the nuclear magnetic resonance atomic gyroscope based on a self-calibrating magnetometer according to claim 1, characterized in that, The probe light on the light-emitting side of the gas chamber is divided into two polarized lights with different powers by a Wollaston prism, and are respectively incident on the two diodes of the differential photodetector to obtain a Faraday rotation angle θ proportional to the average value <S x > of the rubidium atom spin ensemble in the X direction FR , θ FR The expression is as follows: where r e is the classical electron radius, c is the speed of light in vacuum, is the oscillator strength of the D2 line of rubidium atoms, L is the length of the atomic cell, T is the temperature of the cell, n Rb (T) is the rubidium atomic number density dependent on the temperature T; v probe 、 are the frequency of the probe laser and the resonance frequency of the D2 line of rubidium atoms, respectively, is the full width at half maximum of the rubidium atomic absorption peak.

10. The method for implementing a nuclear magnetic resonance atomic gyroscope based on a self-calibrating magnetometer according to claim 1, wherein The calibrated circularly polarized magnetic field phase after the second demodulation is set as ∠C (2) (θ), then the expression of ∠C (2) (θ) is as follows: where k(η) is an intermediate quantity, k(η) = [J0(η) + J -2 (η)] / [J0(η) - J -2 (η)], η is the modulation depth, J0(η) is the Bessel function of the zero-th order with η as the argument, J -2 (η) is the Bessel function of the -2nd order with η as the argument, θ is the phase shift of the reference signal during the first demodulation, Θ (y) is the working point phase at which the magnetometer most suppresses the magnetic field in the Y direction, ω is the frequency of the circularly polarized calibration magnetic field; Γ2 is the linewidth of rubidium.

Images