CN113625204B - Atomic magnetometer magnetic field measurement method based on particle swarm algorithm - Google Patents
Atomic magnetometer magnetic field measurement method based on particle swarm algorithm Download PDFInfo
- Publication number
- CN113625204B CN113625204B CN202110908304.8A CN202110908304A CN113625204B CN 113625204 B CN113625204 B CN 113625204B CN 202110908304 A CN202110908304 A CN 202110908304A CN 113625204 B CN113625204 B CN 113625204B
- Authority
- CN
- China
- Prior art keywords
- magnetic field
- particle
- atomic magnetometer
- particle swarm
- particles
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R33/00—Arrangements or instruments for measuring magnetic variables
- G01R33/02—Measuring direction or magnitude of magnetic fields or magnetic flux
- G01R33/032—Measuring direction or magnitude of magnetic fields or magnetic flux using magneto-optic devices, e.g. Faraday or Cotton-Mouton effect
Abstract
The invention discloses a particle swarm algorithm-based atomic magnetometer magnetic field measuring method, which is characterized in that an applied single-frequency excitation magnetic field is used as virtual particles in a particle swarm algorithm, an interaction mechanism among alternating excitation magnetic fields with different frequencies is constructed through the particle swarm algorithm, an atomic spin resonance position is searched through cooperation and information sharing among individuals in a full frequency band, the atomic magnetometer magnetic field measuring efficiency is improved, and the improvement of the group cooperativity is realized.
Description
Technical Field
The invention relates to the field of magnetic detection, in particular to a magnetic field tracking measurement method of an atomic magnetometer.
Background
The atomic magnetometer has important application prospect in the fields of cardio-cerebral magnetic measurement, geomagnetic mapping, military field and the like.
Alkali metal atomic spin polarization, magnetic resonance, and spin detection are the three major physical processes in atomic magnetometers. The transfer of angular momentum of the circularly polarized laser light and the atomic spins causes a large number of atoms in the atomic ensemble to be in the same spin state, which macroscopically appears as a large number of atomic spins with uniform polarization orientation. The alkali metal atom spin can precess around the magnetic field direction in the external magnetic field, and the precession frequency omega is gamma B0Referred to as larmor frequency, where: γ is called gyromagnetic ratio and is only related to the kind of alkali metal atom; b is0Representing the magnitude of the external magnetic field. The atom spin precession can modulate the polarization direction of the linear polarization laser which passes through the atom medium, and the precession frequency of the atom spin can be obtained by measuring the polarization direction change of the linear polarization detection light. If an alternating excitation magnetic field is applied in the vertical direction of the external magnetic field, the alkali metal atoms can generate a magnetic resonance effect when the frequency of the alternating excitation magnetic field is equal to the larmor frequency of the spins of the alkali metal atoms. At this time, the modulation intensity of the linear polarization laser is the highest by the atomic mediumAnd the measured signal amplitude reaches the maximum value, and the external magnetic field intensity can be obtained by resolving by using the polarization modulation frequency of the detection light at the resonance point.
To create the magnetic resonance condition of the alkali metal atomic media, the frequency of the alternating excitation magnetic field needs to be scanned around the larmor frequency, and the scanning period determines the measuring speed of the atomic magnetometer. The general measurement method is to set the starting frequency and the ending frequency, change the frequency of the alternating excitation magnetic field with fixed step length, and obtain the peak frequency according to the resonance curve fitting. When the magnetic field to be measured is unknown or the amplitude of the magnetic field changes, the traditional frequency scanning method has low magnetic field measurement efficiency and the sampling rate is difficult to meet the actual requirement, and the application of the atomic magnetometer in a large-range and high-dynamic scene is limited.
Disclosure of Invention
The invention provides a magnetic field tracking measurement method based on a particle swarm algorithm for improving the measurement speed of an atomic magnetometer, which comprises the following steps:
the method comprises the following steps: in an application scene of the atomic magnetometer, alternating magnetic field driving signals with the same strength and a plurality of frequencies are randomly set, each alternating magnetic field is regarded as a particle, the position of the particle is the frequency of the alternating magnetic field, the alternating magnetic field is applied to an excitation magnetic field coil of the atomic magnetometer, and the response of the amplitude of an output signal of the atomic magnetometer to the position of each particle is regarded as a fitness index of a particle swarm algorithm;
step two: initializing particle swarm algorithm parameters, setting a particle swarm scale n, a maximum iteration number G, a maximum particle speed vlim and position information as a whole search frequency domain;
step three: randomly initializing an initial position x ═ x _1, x _ 2., x _ n } and a velocity v ═ v _1, v _ 2.., v _ n } of the n particles;
step IV: the method comprises the steps that the optimal response of the atom magnetometer is searched for by each alternating magnetic field through a local search strategy, the current positions of all particles are superposed and then are applied to an excitation magnetic field coil of the atom magnetometer, the output signal of the atom magnetometer is subjected to frequency spectrum transformation, the amplitude value f (x _ i) of the response of the atom magnetometer to each particle is obtained and serves as the fitness value of the position, and the global optimal position of a group is obtained according to the fitness values of all the particles.
Step five: and (3) updating the speed and the position of the particles by a particle swarm algorithm, wherein an updating formula is shown as follows:
v_i＝w*V_i+C_1*Random(0,1)*(pbest_i-x_i)+C_2*Random(0,1)*(gbest-x_i)；
x_i＝x_i+v_i；
wherein v _ i represents the speed of the ith particle, x _ i represents the position of the ith particle, w is an inertia factor, C _1 and C _2 are acceleration constants, generally, C _1 is C _2 ∈ [0,4], Random (0,1) is a Random number in the [0,1] interval, pbesti represents the historical optimal position of the ith particle, and gbest represents the historical optimal position of the particle swarm;
step (c): if the current iteration times are larger than the maximum iteration times G or formants are successfully identified, executing the step (c), otherwise, returning to the step (c) for next iteration;
step (c): and all the particles move to the gbest, which is the resonance frequency of the atomic magnetometer system, and the magnetic field intensity to be measured of the atomic magnetometer is obtained by the gbest.
Optionally, the step (i) is specifically: randomly setting alternating magnetic field driving signals with multiple frequencies and the same current amplitude, connecting an alternating magnetic field driving current source with an atomic magnetometer excitation field coil, and measuring the amplitude of an output signal of the atomic magnetometer; each alternating magnetic field is regarded as a particle, the frequency of the alternating magnetic field is the position of the particle, and the response of the amplitude of the output signal of the atomic magnetometer to the position of each particle is used as the fitness index of the particle swarm algorithm.
Optionally, in the second step, the size n of the particle group is usually a value within a range of 10 to 80, the maximum iteration number G is usually a value within a range of 50 to 1000, and the maximum velocity vlim of the particle is usually a value within a range of 0.01 to 0.5.
The technical scheme of the invention has the following beneficial effects:
the method for measuring the magnetic field of the atomic magnetometer based on the particle swarm optimization disclosed by the invention has the advantages that the applied single-frequency excitation magnetic field is used as virtual particles in the particle swarm optimization, an interaction mechanism among alternating excitation magnetic fields with different frequencies is constructed through the particle swarm optimization, the atomic spin resonance position is searched through cooperation and information sharing among individuals in a full frequency band, the magnetic field measurement efficiency of the atomic magnetometer is improved, and the improvement of the group synergy is realized.
According to the method, the single-frequency excitation magnetic field is used as virtual particles in the particle swarm algorithm, the group particles are superposed and then applied to the atomic magnetometer excitation magnetic field coil in each iteration period, and the atomic magnetometer outputs a signal spectrum to respond to the particle fitness value, so that the optimization performance and robustness of the algorithm are effectively improved, the calculated amount is reduced, and the operation efficiency of the algorithm is improved.
According to the scheme of the invention, the dynamic inertia factor w is introduced into the particle speed and position updating formula, so that the global and local searching capabilities can be adjusted according to different atomic magnetometer magnetic detection application scenes, and various complex measurement scenes can be dealt with.
Drawings
FIG. 1 is a schematic flow chart of a magnetic field measurement method of an atomic magnetometer based on a particle swarm algorithm in an embodiment of the invention;
FIG. 2 is a graph of the initial position of particles and the amplitude of response in an atomic magnetometer system in an embodiment of the present invention;
FIG. 3 is a graph of the final state position of particles and the amplitude of response in an atomic magnetometer system in an embodiment of the present invention;
fig. 4 is a schematic diagram of the iteration number and convergence process of the algorithm in the embodiment of the present invention.
Detailed Description
The present invention will be further elucidated with reference to the accompanying drawings and specific embodiments, it being understood that the present embodiments are illustrative only and are not limiting to the scope of the invention, and that various equivalent modifications of the invention will occur to those skilled in the art upon reading the present invention and fall within the scope of the appended claims.
The embodiment is as follows:
referring to the attached figure 1, the atomic magnetometer magnetic field measurement method based on the particle swarm optimization specifically comprises the following steps:
the method comprises the following steps: in an application scene of the atomic magnetometer, alternating magnetic field driving signals with the same strength and a plurality of frequencies are randomly set, each alternating magnetic field is regarded as a particle, the position of the particle is the frequency of the alternating magnetic field, the alternating magnetic field is applied to an excitation magnetic field coil of the atomic magnetometer, the response of the amplitude of an output signal of the atomic magnetometer to the position of each particle is regarded as a fitness index of a particle swarm algorithm, and the stronger the output response of the magnetometer is, the better the fitness index is; the step is a defining method for linking the control quantity of the atomic magnetometer with the particle swarm algorithm variable, and the system connection mode (the application of a magnetic field and the acquisition of signal amplitude) can be directly realized by a person skilled in the art;
step two: initializing particle swarm algorithm parameters, setting the particle swarm size n to be 50, setting the maximum iteration number G to be 100, setting the maximum speed vlim of particles to be 0.1, and setting position information to be the whole search frequency domain;
step three: randomly initializing an initial position x ═ x _1, x _ 2., x _ n } and a velocity v ═ v _1, v _ 2.., v _ n } of the n particles;
step IV: the method comprises the steps that the optimal response of the atom magnetometer is searched for by each alternating magnetic field through a local search strategy, the current positions of all particles are superposed and then are applied to an excitation magnetic field coil of the atom magnetometer, the output signal of the atom magnetometer is subjected to frequency spectrum transformation, the amplitude value f (x _ i) of the response of the atom magnetometer to each particle is obtained and serves as the fitness value of the position, and the global optimal position of a group is obtained according to the fitness values of all the particles.
Step five: updating the speed and the position of the particles through a particle swarm algorithm (specifically comprising updating the historical optimal position of each particle, updating the global optimal position of the population and updating the speed and the position of each particle), wherein an updating formula is as follows:
v_i＝w*V_i+C_1*Random(0,1)*(pbest_i-x_i)+C_2*Random(0,1)*(gbest-x_i)；
x_i＝x_i+v_i；
wherein v _ i represents the velocity of the ith particle, x _ i represents the position of the ith particle, w is an inertia factor, w is 0.6, C _1 and C _2 are acceleration constants, C _1 is C _2 is 2, Random (0,1) is a Random number in the interval of [0,1], pbesti represents the historical optimal position of the ith particle, and gbest represents the historical optimal position of the particle swarm;
step (c): if the current iteration times are larger than or the maximum successful iteration times G identify the formants, executing the step (c), otherwise, returning to the step (c) for next iteration;
step (c): and all the particles move to the gbest which is the resonance frequency of the atomic magnetometer system, and the magnetic field intensity B to be measured of the atomic magnetometer is obtained from the gbest which is gamma B.
The particle group size is limited by the computing power of hardware (computer or server), the more the particle group size, the greater the computing pressure, but if the computing power of hardware can satisfy the convergence speed certainly faster, the preferred range of 10-80 is in this embodiment. The maximum iteration number has a large relationship with the signal quality of the atomic magnetometer system and is influenced by the particle swarm size. If the signal-to-noise ratio of the atomic magnetometer system is high (above 1000), the algorithm can be converged quickly, and the maximum iteration number can be selected to be small, and the range of 50-1000 is preferred in the embodiment.
Specifically, in the present embodiment, as shown in fig. 2, the particle group size n is determined to be 50, 50 driving signals of an alternating magnetic field having the same intensity are applied to the excitation magnetic field coil of the atomic magnetometer, the frequency of the alternating magnetic field is random, and the 50 frequencies are the initial positions of 50 particles. The response of the output signal amplitude of the atomic magnetometer to the alternating magnetic field with each frequency is the fitness index of the particle swarm algorithm. And searching the optimal response of the atomic magnetometer through a local search strategy by each alternating magnetic field, and obtaining the global optimal position of the population according to the fitness values of all the particles. As shown in fig. 3, in this embodiment, the resonant frequency of the atomic magnetometer system is 34.98kHz, and the magnetic field strength to be measured of the atomic magnetometer obtained according to gbest ═ γ B is 10 μ T. The schematic diagram of the iteration number and the convergence process of the algorithm in this embodiment is shown in fig. 4.
The technical scheme of the invention has the following effects:
the invention provides a particle swarm algorithm-based atomic magnetometer magnetic field measuring method, which is characterized in that an applied single-frequency excitation magnetic field is used as virtual particles in a particle swarm algorithm, an interaction mechanism among alternating excitation magnetic fields with different frequencies is constructed through the particle swarm algorithm, an atomic spin resonance position is searched through cooperation and information sharing among individuals in a full frequency band, the atomic magnetometer magnetic field measuring efficiency is improved, and the group synergy performance is improved. According to the method, the single-frequency excitation magnetic field is used as virtual particles in the particle swarm optimization, the group particles are superposed and then applied to the excitation magnetic field coil of the atomic magnetometer in each iteration period, the output signal frequency spectrum of the atomic magnetometer is used for responding to the particle fitness value, the optimization performance and robustness of the algorithm can be effectively improved, the calculation amount of the algorithm is reduced, and the algorithm operation efficiency is improved. According to the invention, the dynamic inertia factor w is introduced into the particle speed and position updating formula, so that the global and local searching capabilities can be adjusted according to different atom magnetometer magnetic detection application scenes, and various complex measurement scenes can be dealt with. The method is simple, convenient to operate, capable of being used in different atomic magnetometer application scenes, strong in transportability and suitable for practicality.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (3)
1. A particle swarm algorithm-based atomic magnetometer magnetic field measurement method is characterized by comprising the following steps:
in an application scene of an atomic magnetometer, randomly setting alternating magnetic field driving signals with multiple frequencies and the same intensity, regarding each alternating magnetic field as a particle, regarding the position of the particle as the frequency of the alternating magnetic field, applying the alternating magnetic field to an excitation magnetic field coil of the atomic magnetometer, and using the response of the amplitude of an output signal of the atomic magnetometer to the position of each particle as a fitness index of a particle swarm algorithm;
initializing particle swarm algorithm parameters, setting a particle swarm scale n, the maximum iteration times G, the maximum speed vlim of particles and the position information of the whole search frequency domain;
randomly initializing initial positions x ═ x _1, x _2,. and x _ n } and velocities v ═ v _1, v _2,. and v _ n } of the n particles;
fourthly, searching the optimal response of the atomic magnetometer through a local search strategy by each alternating magnetic field; the current positions of all the particles are superposed and then applied to an atomic magnetometer excitation magnetic field coil, and the output signal of the atomic magnetometer is subjected to frequency spectrum transformation to obtain the amplitude value f (x _ i) of the response of the atomic magnetometer to each particle as the fitness value of the position; obtaining a global optimal position of a group according to the fitness values of all the particles;
fifthly, updating the speed and the position of the particles through a particle swarm algorithm, wherein an updating formula is as follows:
v_i=w*v_i+C_1*Random(0,1)*(pbest_i-x_i)+C_2*Random(0,1)*(gbest-x_i)；
x_i= x_i+v_i；
wherein: v _ i represents the velocity of the ith particle; x _ i represents the position of the ith particle; w is an inertia factor; c _1 and C _2 are acceleration constants; random (0,1) is a Random number over the [0,1] interval; pbest _ i represents the historical optimal location of the ith particle; the gbest represents the historical optimal position of the particle swarm;
step (c): if the current iteration times are larger than the maximum iteration times G or formants are successfully identified, executing the step (c), otherwise, returning to the step (c) for next iteration;
step (c): and all the particles move to the gbest, and the magnetic field intensity to be measured of the atomic magnetometer is obtained from the gbest.
2. The atomic magnetometer magnetic field measuring method based on the particle swarm optimization according to claim 1, wherein the step (r) is specifically: randomly setting alternating magnetic field driving signals with multiple frequencies and the same current amplitude, connecting an alternating magnetic field driving current source with an atomic magnetometer excitation field coil, and measuring the amplitude of an output signal of the atomic magnetometer; each alternating magnetic field is regarded as a particle, the frequency of the alternating magnetic field is the position of the particle, and the response of the amplitude of the output signal of the atomic magnetometer to the position of each particle is used as the fitness index of the particle swarm algorithm.
3. The atomic magnetometer magnetic field measuring method based on the particle swarm optimization according to claim 1, wherein in the step (ii), the particle group size n is set to a value within a range of 10 to 80, the maximum iteration number G is set to a value within a range of 50 to 1000, and the maximum velocity vlim of the particles is set to a value within a range of 0.01 to 0.5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110908304.8A CN113625204B (en) | 2021-08-09 | 2021-08-09 | Atomic magnetometer magnetic field measurement method based on particle swarm algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110908304.8A CN113625204B (en) | 2021-08-09 | 2021-08-09 | Atomic magnetometer magnetic field measurement method based on particle swarm algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113625204A CN113625204A (en) | 2021-11-09 |
| CN113625204B true CN113625204B (en) | 2022-05-06 |
Family
ID=78383627
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110908304.8A Active CN113625204B (en) | 2021-08-09 | 2021-08-09 | Atomic magnetometer magnetic field measurement method based on particle swarm algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113625204B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2001051035A (en) * | 1999-08-10 | 2001-02-23 | Kansai Research Institute | Magnetic-field detecting apparatus |
| CN103558566A (en) * | 2013-11-11 | 2014-02-05 | 东北石油大学 | High-sensitivity all-optical cesium atom magnetometer |
| CN106872911A (en) * | 2017-03-03 | 2017-06-20 | 中国人民解放军国防科学技术大学 | Atom magnetometer and application method under a kind of excitation field high |
| CN110261797A (en) * | 2019-07-23 | 2019-09-20 | 中国人民解放军军事科学院国防科技创新研究院 | A kind of optical pumping atom magnetometer based on multi-way annular optical cavity |
| CN110554333A (en) * | 2019-02-21 | 2019-12-10 | 天津职业技术师范大学(中国职业培训指导教师进修中心) | digital detection system for optical magnetic resonance signals of potassium optical pump magnetometer |
| CN112945414A (en) * | 2021-01-19 | 2021-06-11 | 北京航空航天大学 | Atomic gas chamber temperature measurement system and method based on cross absorption peak speed selection |
-
2021
- 2021-08-09 CN CN202110908304.8A patent/CN113625204B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2001051035A (en) * | 1999-08-10 | 2001-02-23 | Kansai Research Institute | Magnetic-field detecting apparatus |
| CN103558566A (en) * | 2013-11-11 | 2014-02-05 | 东北石油大学 | High-sensitivity all-optical cesium atom magnetometer |
| CN106872911A (en) * | 2017-03-03 | 2017-06-20 | 中国人民解放军国防科学技术大学 | Atom magnetometer and application method under a kind of excitation field high |
| CN110554333A (en) * | 2019-02-21 | 2019-12-10 | 天津职业技术师范大学(中国职业培训指导教师进修中心) | digital detection system for optical magnetic resonance signals of potassium optical pump magnetometer |
| CN110261797A (en) * | 2019-07-23 | 2019-09-20 | 中国人民解放军军事科学院国防科技创新研究院 | A kind of optical pumping atom magnetometer based on multi-way annular optical cavity |
| CN112945414A (en) * | 2021-01-19 | 2021-06-11 | 北京航空航天大学 | Atomic gas chamber temperature measurement system and method based on cross absorption peak speed selection |
Non-Patent Citations (1)
| Title |
|---|
| 用于原子干涉测量的多段圆柱形磁屏蔽装置性能分析及实验研究;王先华 等;《仪器仪表学报》;20200531;第41卷(第05期);第1-9页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113625204A (en) | 2021-11-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Dergachev et al. | Sensitivity improvements in the search for periodic gravitational waves using O1 LIGO data | |
| EP2656093B1 (en) | Parallel mri method using calibration scan, coil sensitivity maps and navigators for rigid motion compensation | |
| CN102866371B (en) | The Chemical Exchange saturation transfer imaging that magnetic field is insensitive | |
| US4682107A (en) | Method of operating a nuclear magnetic resonance spectrometer | |
| CN1190572A (en) | Method for reducing Maxwell term false image in fast spin echo magnetic rsonance image | |
| CN111721795A (en) | Material measurement method and system based on nuclear magnetic resonance system | |
| JP2003506174A (en) | Method and apparatus for mapping internal and global motion of a target by phase labeling in magnetic resonance imaging | |
| Goncharov et al. | All-sky radiometer for narrowband gravitational waves using folded data | |
| JPH07323018A (en) | Mr imaging apparatus | |
| CN114487954B (en) | Multichannel transmitting-receiving NMR method for accurately measuring field intensity and distribution of electromagnet | |
| CN113625204B (en) | Atomic magnetometer magnetic field measurement method based on particle swarm algorithm | |
| CN1867835A (en) | System and method for magnetic resonance imaging | |
| Banagiri et al. | Search strategies for long gravitational-wave transients: Hidden Markov model tracking and seedless clustering | |
| KR101878570B1 (en) | Modified truefisp sequence for parallel mr data capture | |
| CN1636144A (en) | Magnetic resonance method | |
| Shao et al. | DePos: Accurate orientation-free indoor positioning with deep convolutional neural networks | |
| CN109143389B (en) | A kind of three-dimensional industrial frequency interference source for nuclear-magnetism quantifies orienting device and measurement method | |
| US6777935B2 (en) | MR method for generating navigator pulses | |
| Tan et al. | A frequency measurement method using rising-falling edge of square wave for increasing proton magnetometer precision | |
| CN114518553A (en) | Broadband frequency conversion pseudo two-dimensional spectrum NMR method for accurately measuring field intensity of electromagnet | |
| US20200049638A1 (en) | Method for measuring change of cell in real time and device therefor | |
| WO2021130101A1 (en) | 3d mr imaging with intrinsic motion detection | |
| US6769612B1 (en) | Determining properties of magnetic elements | |
| JP2003135418A (en) | Zero-order phase detection method and mri device | |
| GB2264174A (en) | Method of and apparatus for obtaining NMR information |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |