Online carrier magnetic compensation method based on hls tool

Abstract

The application discloses an online carrier magnetic compensation method based on an HLS tool, and aims at solving the problem of the ill-conditioned coefficient of traditional carrier magnetic compensation, wherein by researching the main source of magnetic interference, the influence of the eddy current field is ignored, the original model is simplified, a three-component magnetic compensation model is established, and thus the online carrier compensation method of the magnetic field component is proposed. Meanwhile, the method is embedded in the hardware simulation by using the high-level synthesis (HLS) tool of FPGA, the real-time performance and compensation accuracy of the method in the hardware system are verified, and the online carrier magnetic compensation has important practical significance.

Description

Technical Field

[0001] This invention belongs to the fields of airborne magnetic detection technology and magnetic compensation, specifically relating to an online carrier magnetic compensation method based on HLS tools. Background Technology

[0002] Magnetic anomaly detection involves using magnetic signal detection equipment to detect and locate magnetic targets that cause anomalous magnetic signals. Airborne magnetic detection, a widely used technique, uses magnetic detection devices mounted on aircraft to collect data on the surrounding magnetic field strength, thereby searching for magnetically related substances. During airborne magnetic measurements, improving the quality of the measured magnetic field data requires enhancing the level of magnetic sensors and interference compensation. In airborne magnetic measurements, mobile carriers are often composed of magnetic materials such as steel. Ferromagnetic materials, after being magnetized by the Earth's magnetic field, become magnetic, which affects magnetic detection. Magnetic interference from the carrier mainly includes the inherent magnetic field of hard magnetic materials, the induced magnetic field generated by soft magnetic materials, and the eddy current magnetic field generated by the rate of change of magnetic flux through the carrier. Furthermore, the magnitude and direction of the induced magnetic field change with the carrier's position and attitude. Carrier interference is an unavoidable error factor. The presence of interfering magnetic fields leads to significant errors in magnetic field measurements, resulting in inaccurate assessments of magnetic anomalies in the detected area and reduced measurement efficiency. Therefore, carrier magnetic compensation technology is one of the key technologies in the field of magnetic anomaly detection.

[0003] In 1950, Tolles and Lawson, through analysis, concluded that the magnetic interference of an aircraft carrier platform is related to the carrier's motion and expressed the interference as the sum of three field strengths: a constant field, an induced field, and an eddy current field. They established a related mathematical model—the TL model—to calculate carrier interference and achieve carrier compensation. The effectiveness of carrier compensation is not only affected by the interference from the carrier itself. In actual long-term, long-distance, high-precision flight exploration, due to the existence of the geomagnetic gradient, the geomagnetic field varies with location, and its variation lacks significant regularity. The geomagnetic gradient not only affects the calculation and compensation of the induced field and eddy current field but also directly acts as a dynamic signal superimposed on the total field signal, affecting the accuracy of carrier compensation. Furthermore, geomagnetic diurnal variation is influenced by solar activity, and the intensity and direction of the geomagnetic vector field exhibit irregular continuous changes over time. Williams proposed using a neural network with time series data as input nodes to fit the diurnal variation of the geomagnetic field (Williams P M. Aeromagnetic compensation using neural networks[J]. Neural Computing & Applications,1993 (1):207-214.). This method can only achieve high-precision compensation within a specific time period and does not have the ability to generalize parameters over long-term detection processes. Wu Dongling et al. used a genetic algorithm to solve for the compensation coefficients (Wu Dongling, Chen Zhengxiang, Wang Xiu. Magnetic interference compensation method based on genetic algorithm[J]. Journal of Detection and Control,2012,34(6):16-20.), but it suffers from relatively low computational efficiency.

[0004] Most carrier magnetic compensation based on the total field model is based on the premise that the geomagnetic field is constant, which leads to ill-conditioned problems in the solution coefficients and neglect of the influence of the geomagnetic gradient. Summary of the Invention

[0005] This invention addresses the ill-conditioned nature of traditional carrier magnetic compensation solution coefficients. By studying the main sources of magnetic interference and ignoring the influence of eddy current fields, the original model is simplified, a three-component magnetic compensation model is established, and an online carrier magnetic compensation method based on the HLS tool is proposed.

[0006] The technical solution to achieve the objective of this invention is as follows: Firstly, this invention provides an online carrier magnetic compensation method based on HLS tools, comprising the following steps:

[0007] Step 1: Select a spatial region with a stable magnetic field strength and a gradient less than a set threshold, and record the initial total magnetic field value.

[0008] Step 2: According to the TL carrier compensation model, the magnetic interference of the aircraft carrier platform is formed by the superposition of the magnetic field vectors of the constant field, the induced field and the eddy current field.

[0009] Step 3: Perform non-magnetic modification on the aircraft carrier;

[0010] Step 4: Record the measurements of the triaxial magnetometer and the total geomagnetic field during the uniform flight of the carrier;

[0011] Step 5: During the carrier magnetic compensation process, the three direction cosines of the geomagnetic field vector direction in the carrier coordinate system are obtained in real time.

[0012] Step 6: Use the HLS tool to encapsulate the online carrier magnetic compensation method into an IP core, and use the FPGA to call the IP core to solve for the carrier magnetic compensation coefficient, thereby realizing online carrier magnetic compensation.

[0013] In a second aspect, the present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the method described in the first aspect.

[0014] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in the first aspect.

[0015] Compared with existing technologies, this invention discusses the magnetic interference problem of the traditional TL model, analyzes the characteristics of carrier magnetic field interference, simplifies the traditional model, and proposes an online carrier magnetic compensation method based on vector measurement. This method can effectively improve the ill-conditioned problem existing in the traditional method when solving the compensation coefficients, and improve the compensation accuracy. Finally, the carrier compensation coefficients are updated in real time in hardware system simulation using the HLS tool. Compared with Verilog language, its hardware algorithm has higher accuracy and better compensation effect. Attached Figure Description

[0016] Figure 1 The simulation diagram shows the condition number comparison between the traditional method and the proposed method.

[0017] Figure 2 The simulation results show a comparison between the compensation results of the traditional method and the proposed method.

[0018] Figure 3 Simulation results of compensation using HLS tools and Verilog language.

[0019] Figure 4 The graph shows the results of resource consumption compensation for HLS tools and Verilog language. Detailed Implementation

[0020] This invention proposes an online carrier magnetic compensation method based on High-level Synthesis (HLS) tools, comprising the following steps:

[0021] Step 1: Select a spatial region with stable magnetic field strength and small gradient, and record the initial total magnetic field value. The component is .

[0022] Among them, the spatial region with stable magnetic field strength refers to the region where the magnitude of the total geomagnetic field fluctuates within 1 nT and the gradient of the total geomagnetic field is within 1 nT / m.

[0023] Step 2: According to the TL carrier compensation model, the constant field is a static magnetic field relative to the carrier; the induced field is a magnetic field formed by the magnetization of the carrier's own soft magnetic material; and the eddy current field is generated by the rate of change of magnetic flux through the carrier. The magnetic interference of the aircraft carrier platform is formed by the vector superposition of the constant field, induced field, and eddy current field magnetic fields.

[0024] (1)

[0025] in , , These are the projections of the constant field, the induced field, and the eddy current field onto the direction of the geomagnetic field, respectively.

[0026] Since the constant field is constant relative to the aircraft carrier and is independent of the magnitude of the Earth's magnetic field and aircraft maneuvers, it can be directly decomposed along the three axes of the carrier's coordinate system.

[0027] (2)

[0028] in , , These are the cosine values ​​of the angles between the three components of the magnetic field in the carrier coordinate system and the geomagnetic field. , , It is the projection of the constant field onto the three axes of the carrier coordinate system.

[0029] The induced field is formed by the magnetization of a portion of the carrier's soft magnetic material by the Earth's magnetic field, and is directly proportional to the strength of the magnetic field, but not necessarily aligned with its direction. Therefore, the induced field can be expressed as the sum of the induced magnetic fields generated by the magnetization of the carrier's soft magnetic material along the three axes of the carrier's coordinate system.

[0030] (3)

[0031] matrix Let A = (The induced field interference coefficient is given by the formula A = ...) .

[0032] The eddy current field is formed by the change in magnetic flux of the Earth's magnetic field passing through the carrier, and the magnitude of the eddy current field is directly proportional to the rate of change of the Earth's magnetic field strength. The eddy current field can be expressed as the sum of the eddy current magnetic fields generated by the changes in magnetic flux of the Earth's magnetic field along the three axes of the carrier's coordinate system.

[0033] (4)

[0034] Step 3 involves modifying the aircraft carrier to be non-magnetic. This involves replacing large areas of bulk metallic conductive material in the fuselage with non-magnetic materials to weaken the eddy current magnetic field. For example, drones can undergo non-magnetic modifications to their wings or even the entire fuselage, effectively avoiding eddy current magnetic field interference. After the carrier ascends to cruising altitude, it is kept as stationary as possible in a straight line with minimal attitude changes to further reduce the impact of eddy current interference. The area threshold of the large-area bulk metallic conductive material is set according to the actual situation. Optionally, carbon fiber can be used as the non-magnetic material.

[0035] Step 4, due to neglecting the influence of the eddy current field, the magnetic interference of the aircraft carrier platform:

[0036] (5)

[0037] Step 5: During the uniform flight of the carrier, the measurements from the triaxial magnetometer are recorded as follows: The measured value of the total geomagnetic field is ,

[0038] (6)

[0039] Step 6: Based on the simplification of the carrier magnetic compensation model, a component model was established, which improved the ill-conditioned problem in solving the compensation coefficient and reduced the computational complexity.

[0040] Step 7: During the carrier magnetic compensation process, it is necessary to obtain the three direction cosines of the geomagnetic field vector direction in the carrier coordinate system in real time. , , Static compensation can be performed before the vehicle takes off to obtain accurate initial direction cosines. , , After performing rotation matrix operations with the attitude angle data measured in real time by the inertial navigation system, the real-time direction cosine value during the vehicle's flight process can be obtained, i.e.

[0041] = (7)

[0042] in, This indicates the rotation angle of the carrier about the X-axis of the carrier coordinate system; Indicates the rotation angle around the Y-axis; This represents the rotation angle around the Z-axis.

[0043] Step 8: Combining equations (3), (4), and (5), the magnetic interference of the aircraft carrier platform is decomposed into quantitative form:

[0044] (8)

[0045] The above formula , and use After making the substitution, we get:

[0046] (9)

[0047] As can be seen from equation (9), in this vector carrier magnetic compensation model, each measurement component has four independent compensation coefficients. Taking the X-axis as an example, the four independent compensation coefficients are as follows: , , , And the measured value The relationship between the compensation coefficient and the compensation coefficient is linear. Considering the presence of white noise, the compensation coefficient can be obtained simply by using the least squares method.

[0048] Step 9, use the compensation coefficients obtained from equation (8) to respectively apply , and Compensation is performed, and the post-compensation result is calculated. and with the actual Perform a difference comparison and compare it with the traditional total field method.

[0049] Traditional carrier compensation methods are all based on an 18-term compensation coefficient model of total field measurement.

[0050] (10)

[0051]

[0052]

[0053]

[0054]

[0055] in, , , , , , They are , , The derivative of , ; , .

[0056] Equation (10) contains 18 unknown variables, and there are serious correlations among the variables, which leads to ill-conditioning. Moreover, for the carrier compensation of the vector magnetometer, some coefficients cannot be combined, and there are actually more coefficients that need to be solved.

[0057] Step 10: The HLS tool encapsulates the online carrier magnetic compensation method into an IP core, which can be called on a Field Programmable Gate Array (FPGA) to solve for the carrier magnetic compensation coefficient in real time.

[0058] This invention transforms total magnetic field compensation into compensation for individual components. By altering the motion state of the aircraft carrier, the model is simplified, improving the ill-conditioned nature of solving the compensation coefficients. Finally, the online carrier magnetic compensation method is implemented in a hardware system using the HLS tool.

[0059] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0060] Example

[0061] Select a spatial region where the magnetic field gradient of the total geomagnetic field is required to vary approximately linearly. Record the initial value of the total geomagnetic field. , corresponding to its vector value .

[0062] After the carrier ascends to cruising altitude, it should maintain uniform linear motion and minimize attitude changes to ensure that the influence of the eddy field can be ignored.

[0063] Assuming the measurements of the triaxial magnetometer during the uniform flight of the carrier are recorded as follows: The attitude angle data of the inertial navigation system are , and .

[0064] The effectiveness of this method will be analyzed through simulation experiments in the following section.

[0065] The initial value of the total geomagnetic field is =50000nT, Initial direction cosine =0.44115、 , =0.65362, constant field disturbance coefficient The induced field interference coefficient matrix A= = The coefficients of the traditional total field method are... ,

[0066] = ,

[0067] = .

[0068] The carrier flies at a constant speed along the X-axis and performs sinusoidal maneuvers with pitch, roll and yaw angles within ±10°, ±10° and ±10° respectively along the same route.

[0069] Commonly used methods for assessing ill-conditioned data include feature analysis, condition number, and variance inflation factor. The condition number is used as a criterion: if the condition number is less than 100, the ill-conditioned design array vector is relatively weak; if the condition number is greater than 100, the ill-conditioned design array vector is relatively strong; and if the condition number is greater than 1000, the ill-conditioned design array vector is very severe. This simulation experiment uses the condition number to evaluate the ill-conditioned nature of the data matrix. The simulation yielded 550 sets of data. The first 50 sets were used to solve for the compensation coefficients and the condition number, and the last 50 sets were used to verify the compensation effect. Figure 1 This shows the changes in the condition number between this method and the traditional total field method. Figure 2 The compensation results for both methods are presented. In summary, compared with the traditional total field method, this method greatly improves the ill-conditioned problem when solving coefficients using the traditional total field method, thereby improving the compensation accuracy.

[0070] This invention implements the solution of carrier compensation coefficients in the HLS tool, verifies online carrier compensation through simulation, and compares the simulation results with those of the HLS tool and Verilog language. This simulation experiment focuses on solving the four coefficients of the X component, using a Xilinx Kintex 7L series XC7K480TLFFV901-2L FPGA chip. The experiment involves 5000 sets of data, including attitude angle data and magnetic field component data. Four sets of data are selected each time as input signals for both the HLS tool and the Verilog language. The solved coefficients are then used to compensate the subsequent 50 sets of data. The standard deviations of the compensation results are shown below. Figure 3 As shown. Compare the resource consumption of the two, as follows: Figure 4 As shown in the figure. In summary, the HLS tool effectively improves the compensation accuracy compared to the Verilog language, but it also requires more resources.

Claims

1. An online carrier magnetic compensation method based on HLS tools, characterized in that, Includes the following steps: Step 1: Select a spatial region with a stable magnetic field strength and a gradient less than a set threshold, and record the initial total magnetic field value. Step 2: According to the TL carrier compensation model, the magnetic interference of the aircraft carrier platform is formed by the superposition of the magnetic field vectors of the constant field, the induced field and the eddy current field. Step 3: Perform non-magnetic modification on the aircraft carrier; Step 4: Record the measurements of the triaxial magnetometer and the total geomagnetic field during the uniform flight of the carrier; Step 5: During the carrier magnetic compensation process, the three direction cosines of the geomagnetic field vector direction in the carrier coordinate system are obtained in real time. Step 6: Use the HLS tool to encapsulate the online carrier magnetic compensation method into an IP core, and use the FPGA to call the IP core to solve for the carrier magnetic compensation coefficient, thereby realizing online carrier magnetic compensation.

2. The online carrier magnetic compensation method based on HLS tools according to claim 1, characterized in that, The stable magnetic field region mentioned in step 1 refers to the region where the magnitude of the total geomagnetic field fluctuates within 1 nT.

3. The online carrier magnetic compensation method based on HLS tools according to claim 1, characterized in that, The range of the total geomagnetic field gradient in step 1 is within 1 nT / m.

4. The online carrier magnetic compensation method based on HLS tools according to claim 1, characterized in that, Initial total field value of the magnetic field in step 1 This represents the average value of data measured over a period of time using the optical pump.

5. The online carrier magnetic compensation method based on HLS tools according to claim 1, characterized in that, In step 3, the aircraft carrier is modified to be non-magnetic by replacing the blocky conductive metal materials in the airframe with non-magnetic materials.

6. The online carrier magnetic compensation method based on HLS tools according to claim 1, characterized in that, In step 3, the carrier must maintain uniform linear motion during flight and perform sinusoidal maneuvers within the range of ±10°, ±10°, and ±10° respectively on the flight path.

7. The online carrier magnetic compensation method based on HLS tools according to claim 1, characterized in that, The data acquired during the flight of the carrier in step 4 are the magnetic field vector measurements of the corresponding triaxial magnetometer.

8. The online carrier magnetic compensation method based on HLS tools according to claim 1, characterized in that, In step 5, the direction cosine of the carrier coordinate system is obtained from the attitude angle using a rotation matrix.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the method as described in any one of claims 1-8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps of the method as described in any one of claims 1-8.

Images