An interference magnetic field separation method based on momentum wheel rotation speed information

Abstract

The application provides a momentum wheel rotating speed information-based interference magnetic field separation method, and relates to the technical field of high-precision magnetic field detection signal processing. The method comprises the following steps: installing two magnetometers on extension rods with different lengths from a spacecraft platform; establishing a satellite coordinate system, and positioning the position information of a plurality of momentum wheels and the two magnetometers based on the satellite coordinate system. An analytical method is used to model the magnetic dipole of each momentum wheel. Based on the real-time rotating speed information of the momentum wheel, the characteristic waveform of the interference magnetic field signal of the momentum wheel is reconstructed, the differential signal of the inner and outer magnetometers is calculated, and the independent components are analyzed from the mixed signal in combination with constraint optimization, so that the momentum wheel components and their coupling contributions in the magnetic field are reconstructed, and the accurate stripping of the momentum wheel interference signal in the on-board magnetic field data is realized. The method does not require additional sensors or frequency domain filtering processing, and is suitable for high-precision magnetic field measurement tasks on spacecraft platforms such as satellites and space stations.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of spacecraft platform magnetic field detection signal processing, and particularly relates to a disturbance magnetic field separation method based on multi-momentum rotation speed information. The method is suitable for high-precision space magnetic field measurement in different scenarios, including but not limited to space magnetic field measurement in satellite missions, geomagnetic mapping, navigation, and geological exploration in spacecraft platform missions. BACKGROUND

[0002] Space magnetic field detection is an indispensable key technology in most space science missions and deep space exploration missions, and plays a vital role in the field of space research. As a core element of space physics, magnetic field data is crucial for a deep understanding of the basic physical processes of space plasma motion, solar wind and Earth's magnetosphere interaction. For example, through accurate magnetic field detection data, researchers can predict space weather events such as solar storms, providing reliable protection for the safe operation of satellite communication and navigation systems, and avoiding huge economic losses caused by space weather disasters. As an in-situ detection technology, magnetic field detection requires precise deployment of magnetometers at the detection target location to directly obtain space magnetic field data. However, in the actual detection process, the signals collected by the magnetometer will inevitably superimpose the interference magnetic field generated by the spacecraft platform itself, which becomes a key factor affecting the detection accuracy.

[0003] Among the many interference sources of the spacecraft platform, the interference generated by the momentum wheel of the control system is particularly prominent. The momentum wheel maintains the attitude stability of the spacecraft by rotating at high speed, but the alternating magnetic field generated during its operation will affect the magnetic field signals collected by the magnetometer. With the development of spacecraft technology, the application of multi-momentum wheel systems is becoming increasingly common, which further exacerbates the complexity of magnetic field signal processing. Currently, common momentum wheel interference signal processing methods include filtering, independent component analysis, singular spectrum decomposition, modal decomposition (Imajo, Nosé et al. 2021, Ream, Weiss et al. 2022, Finley, Broadfoot et al. 2023) and the like. These methods can play a certain role in non-overlapping frequency bands, but when the frequency bands of the space magnetic field signal and the momentum wheel interference overlap, frequency domain suppression-based methods such as filtering and modal decomposition will mistakenly delete the target magnetic field signal while removing the interference. This technical bottleneck results in the fact that publicly available magnetic field data sets, such as the MAVEN, E-POP, and other mission data, generally do not exclude momentum wheel interference. The interference magnetic field signal will affect the scientific value of the data and make it difficult for scientific research to proceed.

[0004] Achieving high-precision space magnetic field signal extraction in the complex environment of multiple momentum wheel interference has become a pressing challenge in the field of space magnetic field detection. Overcoming this technical bottleneck will not only help improve the data quality of existing space science missions, but also lay a solid technical foundation for future, more complex deep space exploration missions, and has important practical significance for promoting the development of space science research. Summary of the Invention

[0005] The purpose of this application is to overcome the defect of the prior art that the interference cannot be removed when the frequency bands of the spatial magnetic field signal and the momentum wheel interference overlap.

[0006] To achieve the above objectives, the present application proposes a method for separating interference magnetic fields based on momentum wheel speed information, comprising:

[0007] Two magnetometers are installed on extension poles at different lengths from the spacecraft platform. The one closer to the spacecraft platform is the inner magnetometer, and the one farther from the spacecraft platform is the outer magnetometer.

[0008] Obtain the magnetic field signals measured by the internal and external magnetometers, as well as the real-time rotational speed of each momentum wheel, and interpolate the rotational speed information sampling rate to the same as the magnetic field data sampling rate;

[0009] Calculate the differential signal based on the magnetic field signal data of the internal and external magnetometers Reconstruct the characteristic waveform of the magnetic field signal of each momentum wheel based on real-time speed information Constructing a differential signal for estimating magnetic interference characteristics;

[0010] Optimize parameters under preset constraints to minimize the mean square error between the measured differential signal and the estimated differential signal;

[0011] Calculate the ambient magnetic field based on the optimization results.

[0012] As an improvement to the above method, the magnetic field signals of the inner and outer magnetometers are respectively:

[0013] ;

[0014] ;

[0015] in, and It is the magnetic field signal data measured by the two inner and outer magnetometers; For the internal magnetometer i A momentum wheel interferes with the magnetic field modeling signal; n is the number of momentum wheels; For the i The ratio of the signals of the momentum wheel interference magnetic field at the outer magnetometer and the inner magnetometer; Represents the actual ambient magnetic field.

[0016] As an improvement to the above method, the characteristic waveform expression of the magnetic field signal of each momentum wheel is:

[0017] ;

[0018] in, Indicates the i Characteristic waveform of magnetic field signal of a momentum wheel; For the i Real-time speed information of each momentum wheel; For the i The initial phase of the momentum wheel; t Indicates time

[0019] As an improvement to the above method, the step of optimizing parameters under preset constraints to minimize the mean square error between the measured differential signal and the estimated differential signal includes:

[0020] The mean square error between the measured differential signal and the estimated differential signal is expressed as:

[0021] ;

[0022] in, represents the mean square error; represents the Euclidean norm; N represents the number of data points. Represents the measured differential signal:

[0023] ;

[0024] in, For the internal magnetometer i A momentum wheel interferes with the magnetic field modeling signal; n is the number of momentum wheels; For the i The ratio of the signals of the momentum wheel interference magnetic field at the outer magnetometer and the inner magnetometer;

[0025] Represents the estimated differential signal:

[0026] ;

[0027] in, Indicates the i The amplitude of the magnetic field disturbed by each momentum wheel; Indicates the i Characteristic waveform of magnetic field signal of a momentum wheel;

[0028] The optimal solution is obtained by using a nonlinear optimization algorithm based on initial value estimation of prior information. .

[0029] As an improvement of the above method, the preset constraint condition comprises:

[0030] ;

[0031] Wherein, The minimum and maximum value range of is determined by the installation position of the momentum wheel and the magnetometer, and is inversely proportional to the third power of the distance between the inner and outer magnetometers and the probe platform; The minimum and maximum value range of is determined by the amplitude of the target frequency band momentum wheel interference field; And The minimum and maximum value range of is 0 and .

[0032] As an improvement of the above method, the nonlinear optimization algorithm based on prior information initial value estimation is a quasi-Newton method with boundary constraint based on prior information initial value estimation.

[0033] As an improvement of the above method, the environmental magnetic field is calculated according to the optimization result, and the expression is:

[0034] ;

[0035] Wherein, Indicates the real environmental magnetic field; Indicates the magnetic field signal data measured by the outer magnetometer; Indicates the amplitude of the interference magnetic field of the i th momentum wheel at the outer magnetometer; Indicates the magnetic field signal characteristic waveform of the i th momentum wheel; The ratio of the signal of the i th momentum wheel interference magnetic field at the outer magnetometer and the inner magnetometer; n The number of momentum wheels.

[0036] Compared with the prior art, the application has the advantages that:

[0037] The application obtains the real-time rotation speed information of each momentum wheel of the spacecraft platform, constructs the interference magnetic field characteristic waveform of each momentum wheel by using the rotation speed information, combines the prior parameters In a certain range, the accurate elimination parameters are obtained by solving the loss function ; and the environmental magnetic field is calculated The application utilizes real-time parameters of an aircraft platform, realizes accurate correction of variable frequency and variable amplitude data of a momentum wheel without increasing the number of devices, without using filtering technology, but directly modeling and separating a mixed signal for a single interference signal, and can still effectively separate an interference magnetic field in a scenario where a frequency domain of an environmental magnetic field and the interference magnetic field overlaps, and improve the accuracy of a space magnetic field. BRIEF DESCRIPTION OF DRAWINGS

[0038] Figure 1 Fig. 1 shows a structural schematic diagram of a multi-momentum wheel system.

[0039] Figure 2 Fig. 4 shows a schematic diagram of an interference magnetic field elimination process.

[0040] Figure 3 Fig. 5 shows a flowchart of an interference magnetic field elimination algorithm of a multi-momentum wheel system. DETAILED DESCRIPTION

[0041] The technical solutions of the application will be described in detail below with reference to the drawings.

[0042] The application provides an interference magnetic field separation method based on momentum wheel rotating speed information, and the core is that: the rotating speed information is used to reconstruct the interference magnetic field characteristic waveform of each momentum wheel, independent components are analyzed from a mixed signal and an environmental magnetic field is recovered by combining constraint optimization, and no filtering or additional hardware is needed.

[0043] The application embodiment provides an interference magnetic field elimination method based on momentum wheel rotating speed information, as shown in Fig. 1, and the technical solution comprises: Figure 2

[0044] 1. Signal modeling

[0045] Two magnetometers are installed on extension rods with different lengths from the spacecraft platform; a satellite coordinate system is established, and position information of a plurality of momentum wheels and the two magnetometers based on the satellite coordinate system is positioned.

[0046] The magnetic field signal of each momentum wheel is periodically changed by the rotating speed, and therefore is modeled as:

[0047]

[0048] In the above formula, is an amplitude, reflecting the intensity of the interference magnetic field of the momentum wheel, which is related to the physical characteristics of the momentum wheel, the current intensity, etc.; is a real-time rotating speed, which will change the frequency of the interference magnetic field signal, and the rotating speed signal sampling rate is interpolated to be the same as the magnetic field signal sampling rate; is an initial phase, which is determined by factors such as the starting time of the momentum wheel. The real-time rotating speed can be directly obtained, and the amplitude and the initial phase are solved in the constraint optimization link. ​​

[0049] For the inner and outer magnetometers installed in different positions of the spacecraft, their signal models are respectively:

[0050] Inner probe: ;

[0051] Outer probe: ;

[0052] where, n denotes the number of momentum wheels.

[0053] Here, and are the actual magnetic field data measured by the two magnetometers, denotes the real environmental magnetic field, according to the magnetic dipole far-field magnetic field model ( m is the magnetic moment vector of the magnetic dipole, r is the distance from the magnetic dipole to a point in space, μ 0 is the magnetic permeability in vacuum) it is known that the contribution of the momentum wheel interference magnetic field to the signals measured by the two magnetometers is inversely proportional to the third power of the distance, which is expressed here as . is the ratio of the signal of the i th momentum wheel interference magnetic field at the outer magnetometer and the inner magnetometer, which is related to the relative position of the installation of the momentum wheel and the magnetometer, and it reflects the difference in the influence of the interference magnetic field of different momentum wheels on the measurement values of different magnetometers. is solved in the constraint optimization link.

[0054] The difference signal is obtained by subtracting the measurement data of the two magnetometers:

[0055] ;

[0056] This difference signal can highlight the waveform characteristics of the momentum wheel interference magnetic field, providing key data for subsequent separation of the interference magnetic field.

[0057] 2. Feature waveform reconstruction

[0058] Based on the obtained real-time rotation speed information , the interference magnetic field feature waveform is generated:

[0059] ;

[0060] This interference magnetic field feature waveform reflects the change rule of the magnetic field of the momentum wheel in the ideal state, which is determined by the rotation speed. This interference magnetic field feature waveform is an amplitude-normalized signal.

[0061] The estimated difference signal is constructed:

[0062] ;

[0063] The estimated differential signal is constructed according to the characteristic waveform of the interference magnetic field , the amplitude and the installation position related coefficient , for subsequent comparison and optimization with the measured differential signal.

[0064] 3. Parameter optimization

[0065] The loss function is defined to measure the difference between the measured differential signal and the estimated signal, and the loss function is:

[0066] ;

[0067] Wherein, represents the Euclidean norm; N represents the number of data points.

[0068] At the same time, the parameter range is constrained as: .

[0069] Wherein, is a reasonable value range determined according to the installation position of the momentum wheel and the magnetometer, the amplitude of the specific frequency band momentum wheel interference field and other actual conditions. Can take 0, . By limiting the parameters in the above range, it can be ensured that the optimization result conforms to the actual physical meaning, and unreasonable parameter values are avoided.

[0070] The nonlinear optimization algorithm based on prior information initial value estimation is used to solve the optimal . The quasi-Newton method with boundary constraint (L-BFGS-B) based on prior information initial value estimation is used in this embodiment.

[0071] 4. Reconstruct the environmental magnetic field

[0072] After parameter optimization, the environmental magnetic field is recovered by the following formula:

[0073] ;

[0074] This formula uses the optimized parameters to subtract the influence of the momentum wheel interference magnetic field from the magnetometer measurement data , so as to obtain the accurate environmental magnetic field .

[0075] As Figure 3 shown, the following specific embodiments are used to introduce and explain in detail the interference magnetic field separation method based on momentum wheel speed information provided by the present application.

[0076] On the spacecraft platform, two magnetometers are installed as inner and outer probes according to the system structure diagram shown in FIG. 1. The inner probe is installed close to the spacecraft platform, and the outer probe is installed relatively far away from the spacecraft platform. At the same time, the real-time speed information of each momentum wheel is obtained through the spacecraft bus.

[0077] During the signal acquisition process, the inner and outer magnetometers continuously acquire magnetic field signals to obtain and data. At the same time, the real-time speed data of each momentum wheel is read in real time from the bus . The real-time speed signal sampling rate of each momentum wheel is interpolated to the same sampling rate as the magnetic field signal.

[0078] According to the formula , the magnetic field signal of each momentum wheel is modeled. Among them and are subsequently solved. Since the contribution of the momentum wheel interference magnetic field to the signals measured by the two magnetometers is inversely proportional to the cube of the distance, the ratio of the signals at the inner and outer probes can be expressed as . Combined with the and data collected by the inner and outer magnetometers, the signal model is constructed according to , , and the difference signal is calculated through to highlight the waveform features of the momentum wheel interference magnetic field.

[0079] The interference magnetic field characteristic waveform signal of each momentum wheel is established as , where is the real-time speed information of each momentum wheel obtained, and is the same as the in the aforementioned modeling of the magnetic field signal of each momentum wheel. The difference lies in that the characteristic waveform signal does not contain amplitude information here, and the signal here is an amplitude-normalized signal.

[0080] The difference signal estimation model is established.

[0081] The loss function is defined as , and the parameter range is set as . The nonlinear optimization algorithm is used to solve to obtain the optimal parameter value.

[0082] Using the optimized parameter value, the background space magnetic field is calculated through .

[0083] ​Finally, it should be noted that the above examples are only used to illustrate the technical solutions of the present application and are not limiting. Although the present application has been described in detail with reference to the examples, those skilled in the art should understand that modifications or equivalent replacements to the technical solutions of the present application can be made without departing from the spirit and scope of the present application, and all should be covered in the scope of the claims of the present application.

Claims

1. A method for separating interference magnetic fields based on momentum wheel speed information, comprising: Two magnetometers are installed on extension poles at different lengths from the spacecraft platform. The one closer to the spacecraft platform is the inner magnetometer, and the one farther from the spacecraft platform is the outer magnetometer. Obtain the magnetic field signals measured by the internal and external magnetometers, as well as the real-time rotational speed of each momentum wheel, and interpolate the rotational speed information sampling rate to the same as the magnetic field data sampling rate; Calculating a differential signal based on magnetic field signal data from internal and external magnetometers; Reconstruct the characteristic waveform of the magnetic field signal of each momentum wheel based on the real-time speed information; construct the differential signal of magnetic interference characteristic estimation; Optimize parameters under preset constraints to minimize the mean square error between the measured differential signal and the estimated differential signal; Calculate the ambient magnetic field based on the optimization results.

2. The interference magnetic field separation method based on momentum wheel speed information according to claim 1 is characterized in that: The magnetic field signals of the inner and outer magnetometers are: ； ； in, and It is the magnetic field signal data measured by the two inner and outer magnetometers; For the internal magnetometer i A momentum wheel interferes with the magnetic field modeling signal; n is the number of momentum wheels; For the i The ratio of the signals of the momentum wheel interference magnetic field at the outer magnetometer and the inner magnetometer; Represents the actual ambient magnetic field.

3. The interference magnetic field separation method based on momentum wheel speed information according to claim 1 is characterized in that: The characteristic waveform expression of the magnetic field signal of each momentum wheel is: ； in, Indicates the i Characteristic waveform of magnetic field signal of a momentum wheel; For the i Real-time speed information of each momentum wheel; For the i The initial phase of the momentum wheel; t Indicates time.

4. The interference magnetic field separation method based on momentum wheel speed information according to claim 1 is characterized in that: Optimizing parameters under preset constraints to minimize the mean square error between the measured differential signal and the estimated differential signal includes: The mean square error between the measured differential signal and the estimated differential signal is expressed as: ； in, represents the mean square error; represents the Euclidean norm; N represents the number of data points; Represents the measured differential signal: ； in, For the internal magnetometer i A momentum wheel interferes with the magnetic field modeling signal; n is the number of momentum wheels; For the i The ratio of the signals of the momentum wheel interference magnetic field at the outer magnetometer and the inner magnetometer; Represents the estimated differential signal: ； in, Indicates the i The amplitude of the magnetic field disturbed by each momentum wheel; Indicates the i Characteristic waveform of magnetic field signal of a momentum wheel; The optimal solution is obtained by using a nonlinear optimization algorithm based on initial value estimation of prior information. .

5. The interference magnetic field separation method based on momentum wheel speed information according to claim 4 is characterized in that: The preset constraints include: ； in, express The minimum and maximum value ranges are determined by the installation positions of the momentum wheel and the magnetometer, and are inversely proportional to the cube of the distances between the inner and outer magnetometers and the detector platform; express The minimum and maximum value range is determined by the amplitude of the interference field of the target frequency driven momentum wheel; and express The minimum and maximum values ​​range, which are 0 and .

6. The interference magnetic field separation method based on momentum wheel speed information according to claim 4 is characterized in that: The nonlinear optimization algorithm based on initial value estimation of prior information is a quasi-Newton method with boundary constraints based on initial value estimation of prior information.

7. The interference magnetic field separation method based on momentum wheel speed information according to claim 1 is characterized in that: The ambient magnetic field is calculated based on the optimization results, and the expression is: ； in, Represents the real environmental magnetic field; Represents the magnetic field signal data measured by the external magnetometer; Indicates the i The amplitude of the magnetic field disturbed by the momentum wheel at the external magnetometer; Indicates the i Characteristic waveform of magnetic field signal of a momentum wheel; For the i The ratio of the signals of the momentum wheel interference magnetic field at the outer magnetometer and the inner magnetometer; n is the number of momentum wheels.

Images