CN108873086B - A method for locating magnetic targets using the total geomagnetic field gradient array - Google Patents

Abstract

The invention belongs to earth magnetism detection technology fields, and in particular to a method of using geomagnetic total field gradient array to locating magnetic objects, comprising the following steps: the sensor array that building is made of seven scalar Magnetic Sensors；According to the sensor array, it is based on magnetic dipole far field model, establishes the relationship of geomagnetic total field gradient Yu magnetic target position coordinates and magnetic moment vector；The solution parameter member that disappears is reduced to three using matrixing by the relationship of the geomagnetic total field gradient and magnetic target position coordinates and magnetic moment vector；Fitness function is established, the minimum value of above-mentioned fitness function is calculated with particle swarm algorithm, solves target position.Resolution ratio is relatively high, can position to weak magnetic target, orientation distance is remote, range is big.It is designed by special algorithm, as long as without real-time measurement and each sensor attitude can be compensated, simplicity is quickly to avoid optical pumped magnetometer dead zone direction when the measurement of rapid solving positioning equation.

Description

A method for locating magnetic targets using the total geomagnetic field gradient array

technical field

The invention belongs to the technical field of geomagnetic detection, and in particular relates to a method for locating a magnetic target by using a geomagnetic total field gradient array.

Background technique

The geomagnetic field is a natural physical field of the earth. It has various origins and is formed by the superposition of magnetic field components with different changing laws. According to the location of the field source, the geomagnetic field can be divided into internal source field and external source field. If we consider the changing characteristics of the geomagnetic field with time, the geomagnetic field that changes rapidly with time becomes the changing magnetic field of the earth, and the geomagnetic field that changes slowly or basically remains unchanged with time becomes the stable magnetic field of the earth.

The magnetic field generated by a magnetic target will lead to changes in the distribution of the geomagnetic field in space, which can produce magnetic anomalies. When the distance between the observation point and the target is 2-3 times greater than the target scale, the magnetic field generated by the magnetic target at the observation point is generally regarded as the magnetic dipole far field.

When locating a magnetic target, a vector magnetic sensor capable of measuring the three components of geomagnetism or a scalar magnetic sensor capable of measuring the total geomagnetic field can be used. The installation and use of the vector magnetic sensor is relatively complicated, the initial attitude must be strictly calibrated, and the working attitude must be measured in real time. When the angle error of the vector magnetic sensor is 0.05°, the measured geomagnetic error is about 50nT. Therefore, it is necessary and difficult to compensate the influence of the attitude change of the vector magnetic sensor on the measurement in real time. In general, the resolution of common vector magnetic sensors (such as ordinary fluxgate magnetometers) on the market is relatively low (compared to scalar optical pump magnetometers), so the target positioning distance based on vector measurement cannot be too long, and the target magnetic moment Can't be too small.

To sum up, there are problems in the prior art such as low resolution, complex installation and use, short detection distance, high requirements for detected objects, and small application range.

Contents of the invention

The purpose of the present invention is to track and locate the magnetic target accurately, quickly and easily. First, a sensor array composed of seven scalar magnetic sensors (optical pump magnetometers) is constructed, and then based on the magnetic dipole far-field model, the relationship between the total geomagnetic field gradient and the magnetic target position coordinates and magnetic moment vector is established. Using matrix transformation, the solution parameter elimination is reduced to three. The fitness function is established, and the particle swarm algorithm is used to locate the magnetic target.

A method for locating a magnetic target using a geomagnetic total field gradient array, comprising the following steps:

(1) Build a sensor array consisting of seven scalar magnetic sensors;

(2) according to the sensor array, based on the magnetic dipole far-field model, establish the relationship between the total geomagnetic field gradient and the magnetic target position coordinates and the magnetic moment vector;

(3) according to the relationship between the total geomagnetic field gradient and the magnetic target position coordinates and the magnetic moment vector, matrix transformation is adopted to reduce the solution parameter elimination to three;

(4) Establish fitness function, use particle swarm algorithm to calculate the minimum value of the above fitness function, and solve the target position.

The construction consists of a sensor array consisting of seven scalar magnetic sensors, including:

The center of the array is the coordinate origin o, seven sensors are located on the origin and three coordinate axes, and the coordinates of each sensor in the array are T ₀ (0,0,0), In order to prevent mutual interference between sensors, L≥1m;

Wherein, L is the distance between two remote sensors on each coordinate axis.

According to the sensor array, based on the magnetic dipole far-field model, the relationship between the total geomagnetic field gradient and the magnetic target position coordinates and magnetic moment vector is established, including:

The measured value T _i of each scalar magnetic sensor is the total field modulus of the superposition of the geomagnetic normal field B _E and the target abnormal field B _i , namely

T _i ＝|B _E +B _i |

U=[cos(I ₀ )cos(D ₀ )cos(I ₀ )sin(D ₀ )sin(I ₀ )]=[abc]

The far-field condition is expressed as:

3L _A ≤ r, L < r, B _i < B _E

Under far-field conditions:

T _i ＝|B _E +B _i |＝B _E +U·B _i

Among them, U is the direction vector of B _E , I ₀ is the geomagnetic inclination, D ₀ is the geomagnetic declination, B _E is the normal field of geomagnetic field, B _i is the abnormal field of the target, a=cos(I ₀ )cos(D ₀ ), b=cos(I ₀ )sin(D ₀ ), c=sin(I ₀ ), L _A is the maximum geometric scale of the target, r is the distance between the observation point and the target, L is the two far ends on each coordinate axis The distance of the sensor, B _E is the _modulus of B _E , and _Bi is the modulus of Bi.

According to the sensor array, based on the magnetic dipole far-field model, the relationship between the total geomagnetic field gradient and the magnetic target position coordinates and magnetic moment vector is established, including:

Let (x _A , y _A , z _A ) be the target position, and ( _xi , y _i , _zi ) be the position of sensor i, we get:

make have to:

T _i ＝B _E +ω·UP _i M

T _i ＝B _E +ω(Q _i M _x +S _i M _y +H _i M _z )

Among them, B _i is the target anomalous field, B _ix , B _iy , B _iz are the rectangular coordinate components of _Bi , M _x , M _y , M _z are the rectangular coordinate components of the target magnetic moment vector M,

Q _i =a·f _{i 11} +b·f _{i 21} +c·f _{i 31} , S _i =a·f _{i 12} +b·f _{i 22} +c·f _{i 32} ,H _i =a·f _{i 13} +b·f _{i 23} +c·f _{i 33} .

According to the relationship between the total geomagnetic field gradient and the magnetic target position coordinates and the magnetic moment vector, matrix transformation is adopted to reduce the solution parameter elimination to three, including:

Among the seven sensors, three groups of sensors are selected, each group consists of two magnetic sensors i and j, i, j=0, 1, 2...6. The difference ΔT _ij =T _i -T _j of the measured values of each group of sensors, and these three differences are linearly independent;

Express the magnetic moment component in terms of the position coordinates of the magnetic target:

The gradient G _i of the measured value T _i of the total geomagnetic field:

The total geomagnetic field gradient value G ₀ at the origin o is:

in, _U is the direction vector of BE, M _x , M _y , M _z are the rectangular coordinate components of the target magnetic moment vector M, and G _ix , G _iy , G _iz are the rectangular coordinate components of G _i .

Described establishment fitness function, calculates the minimum value of above-mentioned fitness function with particle swarm optimization algorithm, solves target position, comprises:

When L<<r in the far-field situation, each sensor measures synchronously to obtain ΔT ₁₂ , ΔT ₃₄ , ΔT ₅₆ , and the measured value of the total geomagnetic field gradient at the origin can be obtained in Yes Cartesian coordinate components of ,

Obviously:

The expression of the fitness function F necessary to establish the particle swarm algorithm solution is:

Among them, L is the distance between two remote sensors on each coordinate axis, r is the distance between the observation point and the target, Yes The Cartesian coordinate components of .

The beneficial effects of the present invention are:

The resolution is relatively high, and it can locate weak magnetic targets with a long positioning distance and a large range. Through the special algorithm design, the positioning equation can be quickly solved, as long as the dead zone direction of the optical pump magnetometer is avoided during measurement, there is no need to measure and compensate the attitude of each sensor in real time, which is simple and fast.

Description of drawings

Figure 1 is a structure diagram of the geomagnetic total field gradient measurement array.

Fig. 2 is the relationship curve between the sensor measurement difference ΔT ₂₀ and the target position x _A.

FIG. 3 is _a graph showing the relationship between the sensor measurement difference ΔT ₄₀ and the target position xA.

FIG. 4 is a graph showing the relationship between the sensor measurement difference ΔT ₆₀ and the target position x _A .

Figure 5 is the result of magnetic positioning of moving targets.

Detailed ways

The present invention will be further described below in conjunction with the accompanying drawings.

The purpose of the present invention is to track and locate the magnetic target accurately, quickly and easily. First, a sensor array composed of seven scalar magnetic sensors (optical pump magnetometers) is constructed, and then based on the magnetic dipole far-field model, the relationship between the total geomagnetic field gradient and the magnetic target position coordinates and magnetic moment vector is established. Using matrix transformation, the solution parameter elimination is reduced to three. The fitness function is established, and the particle swarm algorithm is used to locate the magnetic target.

The invention provides a method for locating a magnetic target, which uses a plurality of scalar magnetic sensors to form a space array, measures the gradient of the total geomagnetic field, and locates the magnetic target through a specially designed algorithm. The algorithm transforms the total geomagnetic field gradient matrix, separates variables and eliminates the magnetic moment vector of the target, and uses the spatial position coordinates of the target to represent the three components of the magnetic moment vector of the target, reducing the six unknowns in the magnetic field gradient equation to three. The fitness function about the magnetic field gradient is established, and the particle swarm algorithm is used to solve the target position. The method for positioning a magnetic target proposed by the invention has high measurement accuracy, long detection distance, fast positioning, and simple and reliable operation. A new method is provided for magnetic target location. It provides a reference for the detection, positioning and identification of magnetic targets such as underground and underwater archaeology, pipeline detection, energy and mineral deposit survey, mine clearance and anti-submarine, and has certain application significance.

In this method, seven scalar magnetic sensors are used to construct the geomagnetic total field gradient measurement array, a target positioning algorithm is designed, and a new solution method is proposed, which can accurately and quickly locate the magnetic target.

(1) As shown in Figure 1, seven scalar magnetic sensors are used to form an array. The center of the array is the coordinate origin o, and the seven sensors are located on the origin and three coordinate axes, arranged symmetrically. The distance between the two remote sensors on each coordinate axis is L. The coordinates of each sensor in the array are T ₀ (0,0,0),

(2) Establish positioning algorithm equations. The process is: based on the above array, establish the relationship between the sensor measurement value and the magnetic target position vector and magnetic moment vector, as shown in formula (3). Using matrix transformation, the three components of the magnetic moment vector are represented by the coordinates of the target position, as shown in (6). The elimination of six unknowns about the target magnetic moment and position in equation (3) is reduced to three, and the total field gradient equations that can be solved are obtained, as shown in equation (12). Establish the fitness function about the total geomagnetic field gradient, as shown in formula (13). (3) The particle swarm optimization algorithm is used to solve the array equations of the geomagnetic total field gradient measurement. The key link is to use the particle swarm optimization algorithm to calculate the minimum value of the above fitness function, and solve the target position to realize the target positioning. There is a minimum limit on the distance L between the two sensors at the far end on each coordinate axis in the array. There is a minimum value of L, and the preset target magnetic moment, positioning range and positioning accuracy are substituted into the fitness function (13) to solve the minimum value of L. And there is another necessary condition, that is, there is a small amount of electromagnetic radiation when the sensors used are working. In order to prevent mutual interference, the distance between any two sensors is not less than 0.5m.

When the sensor selected by the present invention is used to construct an array and use the array, it is only necessary to avoid the direction of the dead zone of the optical pump magnetic sensor, and it is not necessary to measure and compensate the attitude of each magnetic sensor in real time. Therefore, the target location method is simple and fast.

Magnetic detection has important application significance in underground and underwater mineral survey, pipeline monitoring, archaeology, shipwreck survey, mine sweeping and anti-submarine, etc. Because the above-mentioned targets cause anomalies in the geomagnetic field in this area, geomagnetic matching technology can be used to achieve target positioning and identification. In the field of oil and energy exploration, geomagnetic detection technology is often used as an auxiliary survey method. In navigation applications such as satellites, aircraft, ships, submarines, and vehicles, geomagnetic-assisted navigation provides a very important method and approach. Before and after the occurrence of natural disasters such as volcanoes, earthquakes, and tsunamis, magnetic storms and geomagnetic anomalies often occur together, which provides a possible technical approach for the location and prediction of disaster centers. Therefore, magnetic positioning technology has a wide range of applications. A method for locating a magnetic target by using the total geomagnetic field gradient proposed by the present invention can be applied to the above-mentioned technical fields.

The geomagnetic field is a natural physical field of the earth. It has various origins and is formed by the superposition of magnetic field components with different changing laws. According to the location of the field source, the geomagnetic field can be divided into internal source field and external source field. If we consider the changing characteristics of the geomagnetic field with time, the geomagnetic field that changes rapidly with time becomes the changing magnetic field of the earth, and the geomagnetic field that changes slowly or basically remains unchanged with time becomes the stable magnetic field of the earth.

The magnetic field generated by a magnetic target will lead to changes in the distribution of the geomagnetic field in space, which can produce magnetic anomalies. When the distance between the observation point and the target is 2-3 times greater than the target scale, the magnetic field generated by the magnetic target at the observation point is generally regarded as the magnetic dipole far field.

When locating a magnetic target, a vector magnetic sensor capable of measuring the three components of geomagnetism or a scalar magnetic sensor capable of measuring the total geomagnetic field can be used. The installation and use of the vector magnetic sensor is relatively complicated, the initial attitude must be strictly calibrated, and the working attitude must be measured in real time. When the angle error of the vector magnetic sensor is 0.05°, the measured geomagnetic error is about 50nT. Therefore, it is necessary and difficult to compensate the influence of the attitude change of the vector magnetic sensor on the measurement in real time. In general, the resolution of common vector magnetic sensors (such as ordinary fluxgate magnetometers) on the market is relatively low (compared to scalar optical pump magnetometers), so the target positioning distance based on vector measurement cannot be too long, and the target magnetic moment Can't be too small.

The invention relates to a method for locating a magnetic target by using a geomagnetic total field gradient array. The scalar magnetic sensor (optical pump magnetometer) adopted has a relatively high resolution and can locate weak magnetic targets with a long positioning distance and a large range. Through the special algorithm design, the positioning equation can be solved quickly. Just avoid the direction of the dead zone of the optical pump magnetometer during measurement, and there is no need to measure and compensate each sensor attitude in real time, which is simple and fast. This has certain application significance in energy and mineral exploration, underground and underwater pipeline monitoring and maintenance, archaeology, search and rescue of plane crashes and shipwrecks, mine clearance and anti-submarine, etc.

The purpose of the present invention is to track and locate the magnetic target accurately, quickly and easily. First, a sensor array composed of seven scalar magnetic sensors (optical pump magnetometers) is constructed, and then based on the magnetic dipole far-field model, the relationship between the total geomagnetic field gradient and the magnetic target position coordinates and magnetic moment vector is established. Using matrix transformation, the solution parameter elimination is reduced to three. The fitness function is established, and the particle swarm algorithm is used to locate the magnetic target.

Technical scheme of the present invention is realized through the following steps:

Step 1: Use seven scalar magnetic sensors T _i (i=0, 1, 2...6) to build a magnetic detector array as shown in Fig. 1 . The origin of the array coordinate system is o, the positive direction of the x-axis points to the geographic North Pole, the positive direction of the y-axis points to the east, and the positive direction of the z-axis is a right-handed spiral system. The Cartesian coordinates of each magnetic sensor in the array are T ₀ (0,0,0), The distance between the two remote sensors on each coordinate axis is L. In order to prevent mutual interference between the sensors, L≥1m.

Step 2: Each magnetic sensor measures synchronously, and each measured value T _i is the total field modulus of the superposition of the normal geomagnetic field B _E and the target abnormal field B _i , that is, T _i =|B _E +B _i |. Let the modulus of B _E be B _E , B _E changes with time, and this change can be considered as synchronous and equal amplitude in the local space. Let U be the direction vector of B _E , I ₀ and D ₀ be the geomagnetic inclination and declination, then let a=cos(I ₀ )cos(D ₀ ), b=cos(I ₀ )sin(D ₀ ), c =sin(I ₀ ),

U=[cos(I ₀ )cos(D ₀ )cos(I ₀ )sin(D ₀ )sin(I ₀ )]=[abc]. B _ix , B _iy , B _iz are the _Cartesian coordinate components of Bi, and _Bi is the modulus of _Bi . The maximum geometric scale of the target is L _A , and the distance between the observation point and the target is r. The far-field conditions can be expressed as _3LA ≤r, L<<r, B _i <<B _E and so on. therefore

T _i ＝|B _E +B _i |≈B _E +U·B _i (1)

The magnetic field B _i generated by a magnetic target at sensor i is represented by a magnetic dipole far-field model:

In formula (2): μ ₀ =4π×10 ^-7 H/m is the magnetic permeability in vacuum. Let (x _A , y _A , z _A ) be the target position, ( _xi , y _i , z _i ) be the position of sensor i, M _x , M _y , M _z are the Cartesian coordinate components of the target magnetic moment vector M, make make another From (1) and (2) formula

T _i ＝B _E +ω·UP _i M (3)

Q _i =a·f _{i 11} +b·f _{i 21} +c·f _{i 31} , S _i =a·f _{i 12} +b·f _{i 22} +c·f _{i 32} ,H _i =a·f _{i 13} +b·f _{i 23} +c·f _{i 33} .

Formula (3) can be rewritten as: T _i ＝B _E +ω(Q _i M _x +S _i M _y +H _i M _z ) (4)

Step 3: Select three groups of sensors from the seven sensors, each group consists of two magnetic sensors i and j, i, j=0, 1, 2...6. The difference ΔT _ij =T _i -T _j of the measured values of each group of sensors. The selection principle is to ensure that the six differences of ΔT _ij , ΔT ₁₂ , ΔT ₃₄ , and ΔT ₅₆ are linearly independent. According to formula (3), select the following three sets of differences for example:

It can be seen that the geomagnetic normal field B _E is eliminated in ΔT _ij , and B _E is a function of time, so ΔT _ij is not affected by the change of the geomagnetic field with time. Transform (5), and express the magnetic moment component by the position coordinates of the magnetic target as:

At the sensor i, the gradient G _i of the total geomagnetic field measurement value T _i is defined according to formula (7), where G _ix , G _iy , and G _iz are the Cartesian coordinate components of G _i , i=0,1,2... ...6.

Substitute (4) into (7): Substitute formula (6) into formula (8):

According to formula (9), the total geomagnetic field gradient value G ₀ at the origin o is:

Step 4: In the far-field situation where L<<r, each sensor measures synchronously to obtain ΔT ₁₂ , ΔT ₃₄ , and ΔT ₅₆ . The measured value of the total geomagnetic field gradient at the origin can be obtained in Yes The Cartesian coordinate components of .

Simultaneously (10) and (11) two formulas, obviously

The vector equation (12) is a high-order ternary equation system whose target position (x _A , y _A , z _A ) is unknown, it contains three scalar equations, the constraints are complete, and there is a definite solution in the three-dimensional space. The number is limited, and some false solutions can be eliminated according to the actual situation. The expression of the fitness function F necessary to establish the particle swarm algorithm solution is:

The particle swarm algorithm is used to solve the minimum value of the fitness function F to realize the target positioning. Then substituting (6) to solve the target magnetic moment, the target can be initially identified.

Figure 1 Structure diagram of the geomagnetic total field gradient measurement array

Figure 2 The relationship curve between sensor measurement difference ΔT ₂₀ and target position x _A

Figure 3 The relationship curve between sensor measurement difference ΔT ₄₀ and target position x _A

Figure 4 The relationship curve between sensor measurement difference ΔT ₆₀ and target position x _A

Figure 5 Magnetic positioning results of moving targets

A target positioning experiment was carried out in a suburb along the Songhua River in Harbin City, and environmental magnetic field data was measured, and the embodiment of the present invention was illustrated with the experimental process.

The CS-L optical pump magnetic sensor is used to construct the geomagnetic total field gradient measurement array shown in Figure 1. The sensor resolution is 0.6pT, the highest sampling rate is 10Hz, and the measurement range is 15000nT-105000nT. The array parameter L=1m, the positive direction of the x-axis of the coordinate system where the array is located points to the geographic North Pole, the positive direction of the y-axis points to the east, and the positive direction of the z-axis points downward. The local geomagnetic inclination is 1.10rad, and the geomagnetic declination is -0.18rad. The magnetic moment of the magnetic target (permanent magnet) is 152Am ² , the inclination angle of the magnetic moment is 1.13rad, the declination angle of the magnetic moment is -0.18rad, and the geometric dimension of the target is 143mm×120mm×40mm. The direction of the magnetic moment remains unchanged during the target movement. The initial position of the target (3.4m, 4.21m, 0m), make a linear motion along the direction parallel to the x-axis, and the end position (-2.1m, 4.21m, 0m).

During the movement of the target, each sensor of the array measures the total geomagnetic field synchronously. The sampling rate of the sensor is selected as 5Hz, and the difference between the measured values of the sensor at the origin and the sensor on the negative half axis of the three coordinate axes is calculated ΔT ₂₀ , ΔT ₄₀ , and ΔT ₆₀ . The changes of the three differences with the target position x _A are shown in Fig. 2, Fig. 3 and Fig. 4 respectively. The blue dotted line represents the theoretical difference of the magnetic method, which is calculated by formula (5). The red circle point represents the difference value actually measured by the sensor. It can be seen that the theoretical curve basically coincides with the actual measurement point.

When the target moves to a certain position (x _A , y _A , z _A ), ΔT ₂₀ , ΔT ₄₀ , ΔT ₆₀ , ΔT ₁₂ , ΔT ₃₄ , ΔT ₅₆ are calculated according to the synchronous output data of the sensor. Substituting into formulas (10) and (11) respectively, the expressions of G _0x , G _0y , G _0z and value. Will Substituting G _0x , G _0y , and G _0z into formula (13), the fitness function F containing the target position (x _A , y _A , z _A ) is obtained. Using the Matlab particle swarm toolbox, the minimum value of the fitness function F is obtained, and the magnetic target position (x _A , y _A , z _A ) is solved to realize the target positioning.

Fig. 5 is the experimental result of this method for moving target localization. The blue circle point is the actual position of the target, and the red circle point is the magnetic positioning position. During the linear movement from the starting point (3.4m, 4.21m, 0m) to the end point (-2.1m, 4.21m, 0m), the target position x _A is from 2.4m to -2.1m, and magnetic positioning is performed every 0.5m. A total of 10 times. The positioning results show that the horizontal circular probability deviation CEP of the target positioning is equal to 0.287m at a vertical distance of 4.21m for a magnetic moving target with a size of 143mm×120mm×40mm and a magnetic moment of 152Am ² , and the positioning method is feasible.

Claims (5)

1. A method for positioning a magnetic target by adopting a geomagnetic total field gradient array is characterized by comprising the following steps:

(1) constructing a sensor array consisting of seven scalar magnetic sensors;

(2) establishing a relation between the geomagnetic total field gradient and the position coordinate and the magnetic moment vector of the magnetic target based on a magnetic dipole far field model according to the sensor array;

(3) according to the relation between the geomagnetic total field gradient and the position coordinate and the magnetic moment vector of the magnetic target, adopting matrix transformation to reduce the solution parameter elimination to three;

of the seven sensors, three groups of sensors are selected, each group consisting of two magnetic sensors i and j, i, j being 0,1,2.. 6; difference value Delta T of each group of sensor measurement values_ij＝T_i-T_jAnd the three differences are linearly independent;

wherein: u is normal geomagnetic field B_EThe direction vector of (a);μ₀＝4π×10^-7h/m is magnetic permeability in vacuum; m_x、M_y、M_zIs the rectangular coordinate component of the target magnetic moment vector M; p_iThe calculation formula is as follows:

wherein: (x)_A，y_A，z_A) Is the target position, (x)_i，y_i，z_i) Is the sensor i position;

the magnetic moment component is expressed in the position coordinates of the magnetic target:

total geomagnetic field measurement T_iGradient G of_i：

Q_i＝a·f_{i 11}+b·f_{i 21}+c·f_{i 31}

S_i＝a·f_{i 12}+b·f_{i 22}+c·f_{i 32}

H_i＝a·f_{i 13}+b·f_{i 23}+c·f_{i 33}

a＝cos(I₀)cos(D₀)，b＝cos(I₀)sin(D₀)，c＝sin(I₀)；

Wherein: i is₀Is the geomagnetic inclination angle; d₀Is the geomagnetic declination; g_ix、G_iy、G_izIs G_iThe rectangular coordinate component of (a);

origin O geomagnetic total field gradient value G₀Comprises the following steps:

；

(4) and establishing a fitness function, calculating the minimum value of the fitness function by using a particle swarm algorithm, and solving the target position.

2. The method of claim 1, wherein the method further comprises the step of: in the step (1), a sensor array composed of seven scalar magnetic sensors is constructed, and the method comprises the following steps:

the center of the array is a coordinate origin o, seven sensors are positioned on the origin and three coordinate axes, and the coordinate of each sensor in the array is T₀(0，0，0)，In order to prevent mutual interference among the sensors, L is more than or equal to 1 m;

wherein L is the distance between the two remote sensors on each coordinate axis.

3. The method according to claim 1, wherein the step (2) of establishing the relationship between the geomagnetic total field gradient and the position coordinates and the magnetic moment vectors of the magnetic targets based on the magnetic dipole far-field model according to the sensor array comprises:

measurement value T of each scalar magnetic sensor_iAre all normal geomagnetic field B_EAnd a target abnormal field B_iSuperimposed total field modes, i.e.

T_i＝|B_E+B_i|

U＝[cos(I₀)cos(D₀) cos(I₀)sin(D₀) sin(I₀)]＝[a b c]

Far field conditions are expressed as:

3L_A≤r、L＜r、B_i＜B_E

in far field conditions:

T_i＝|B_E+B_i|＝B_E+U·B_i

wherein L is_AIs the maximum geometric dimension of the target, r is the distance between the observation point and the target, L is the distance between the two remote sensors on each coordinate axis, B_EIs B_EMold of (B)_iIs B_iThe die of (1).

4. A method according to claim 3, wherein the method comprises the steps of: in the step (2), a relationship between the geomagnetic total field gradient and the position coordinate and the magnetic moment vector of the magnetic target is established based on the magnetic dipole far-field model according to the sensor array, and the relationship comprises the following steps:

order toObtaining:

T_i＝B_E+ω·UP_iM

T_i＝B_E+ω(Q_iM_x+S_iM_y+H_iM_z)

wherein, B_ix、B_iy 、B_izIs B_iThe rectangular coordinate component of (a).

5. The method of claim 1, wherein the method further comprises the step of: the step (4) of establishing a fitness function, calculating the minimum value of the fitness function by using a particle swarm algorithm, and solving the target position comprises the following steps:

when L < r in far field, the sensors measure synchronously to obtain delta T₁₂、ΔT₃₄、ΔT₅₆Obtaining the gradient measurement value of the geomagnetic total field at the originWhereinIs thatThe rectangular-coordinate component of (a) is,

establishing an expression of a fitness function F necessary for solving by a particle swarm algorithm, wherein the expression is as follows:

wherein L is the distance between the two remote sensors on each coordinate axis, r is the distance between the observation point and the target, is thatThe rectangular coordinate component of (a).