Method and system for short-term daily variation magnetic field interference suppression based on double-machine cooperation

Abstract

This invention proposes a method for suppressing short-duration diurnal magnetic field interference based on dual-aircraft collaboration. Addressing the challenges of establishing fixed diurnal variation monitoring stations and lacking effective diurnal variation reference information in offshore dual-aircraft collaborative airborne magnetic surveys, this method utilizes airborne magnetic survey data collected during near-shore airborne magnetic surveys and synchronous observation data from diurnal variation monitoring stations. It proposes a dual-aircraft collaborative diurnal magnetic interference suppression method based on a neural network-guided blind source separation algorithm. This method first uses a deep neural network to learn and extract prior features of diurnal variation interference from the dual-aircraft observation signals. These features are then used as guiding information in the blind source separation process to separate the independent source components corresponding to the diurnal magnetic interference, ultimately achieving high-precision identification and suppression of diurnal variation interference components in the original signal. This method can efficiently separate and suppress short-duration diurnal magnetic interference, significantly improve the signal-to-noise ratio of magnetic survey data, and effectively preserve the target magnetic anomaly characteristics, providing reliable technical support for high-precision airborne magnetic surveys.

Description

Technical Field

[0001] This invention relates to the field of target interference suppression signal processing technology, and in particular to a method and system for suppressing short-duration diurnal magnetic field interference based on dual-machine collaborative operation. Background Technology

[0002] Airborne magnetic anomaly detection is a passive target detection method that uses high-precision magnetic field measurements to detect magnetic anomaly signals, thereby determining the location and magnetic field strength of the object causing the anomaly. It plays a vital role in industries, mining, archaeology, and military fields, and can be applied to surface ship monitoring, moving target detection and identification, buried target surveying, mineral exploration, and archaeology. Therefore, airborne magnetic detection is of great significance from the perspectives of social life, national development, and defense construction.

[0003] Airborne magnetic anomaly detection technology mainly consists of two parts: interference suppression and target detection. Interference suppression directly affects the final detection results. In airborne magnetic surveys, the interfering magnetic fields measured by the magnetometer on the platform include diurnal variations, the Earth's main magnetic field, aircraft platform interference, equipment interference, geological interference during low-altitude flight, and marine environmental magnetic fields. Among these, short-term diurnal variations introduce significant errors and are highly random, making them difficult to model and suppress. Therefore, it is crucial to effectively eliminate interference anomalies caused by diurnal variations and to effectively correct the raw airborne magnetic survey data for diurnal variations.

[0004] Diurnal magnetic interference is a typical non-stationary time-varying magnetic field. Its variations are mainly influenced by the dynamic evolution of electron concentration in the upper ionosphere, exhibiting significant spatiotemporal nonuniformity and unpredictable modeling characteristics, making it difficult to effectively eliminate using traditional physical modeling or fixed-parameter filtering methods. In near-shore airborne magnetic surveying operations, high-stability diurnal variation monitoring stations can typically be deployed near the coastline to continuously observe regional geomagnetic field changes and obtain high-temporal-resolution diurnal variation reference signals, thereby achieving diurnal variation correction of the airborne magnetic data. However, in offshore or deep-sea operations, geographical conditions and deployment costs limit the ability to establish fixed monitoring stations, resulting in a lack of effective diurnal variation reference information, which severely restricts the accuracy and reliability of offshore magnetic survey data. Summary of the Invention

[0005] This invention provides a method and system for suppressing short-duration diurnal magnetic field interference based on dual-aircraft coordinated operation, addressing the challenges of setting up fixed diurnal variation monitoring stations and lacking effective diurnal variation reference information in offshore dual-aircraft coordinated airborne magnetic surveys. Furthermore, it considers the random and non-stationary characteristics of short-duration diurnal magnetic interference noise, which general filtering methods cannot suppress.

[0006] According to one aspect of the present invention, a method for suppressing short-duration diurnal magnetic field interference based on dual-machine collaborative operation is provided, the method comprising:

[0007] Step 1: Based on the nearshore dual-aircraft collaborative aeromagnetic data and the synchronous observation results of the shore-based diurnal variation station, construct a labeled training database;

[0008] Step 2: Based on the dual-aircraft aeromagnetic input and the diurnal variation station label, a time-series encoder-decoder network and the MSE loss function are used for supervised training to achieve efficient extraction of diurnal variation interference features;

[0009] Step 3, Collaborative Data Acquisition for Long-Distance Exploration: During long-distance operations, two UAVs are arranged in a "||" shape to conduct exploration in the mission area. The raw data array X carried by the two UAVs is the effective payload. a and X b Data is transmitted down to the ground data station via a data link;

[0010] Step four: Preprocess the original data array to obtain the preprocessed first data array D. a Second data array D b ;

[0011] Step 5: Transfer the preprocessed dual-machine data array [D] a D b Input a pre-trained diurnal variation interference identification model, and output the estimated diurnal variation magnetic interference characteristic signal sig. rb ;

[0012] Step 6: Convert the diurnal variation characteristic interference signal sig rb First data array D a Second data array D b Combinatorial mixing matrix M = [sig rb D a D b The mixture matrix M = [sig] is solved using Fast Independent Component Analysis (FastICA). rb D a D b ], and extract independent daily variation components;

[0013] Step 7: Based on the independent diurnal variation components, process the preprocessed first data array D a Second data array D b Diurnal variation interference removal is performed to obtain clean components in the dual-machine data, thus completing the suppression of short-term diurnal variation magnetic field interference.

[0014] Furthermore, in step three, the sampling rates of the two sets of original data arrays are the same, and the two sets of original data arrays X a and Xb All of them contain the following information: total magnetic field, time, longitude, latitude, altitude, ground speed, heading angle, and geomagnetic components.

[0015] Furthermore, in step four, the two sets of original data arrays X are first separated using time information. a and X b Alignment, followed by extraction of the two sets of original data arrays X a and X b The data from the level flight phase of the UAV was used to design appropriate bandpass filters based on the upper bound estimation function of the characteristic frequency of the magnetic anomaly signal, and then applied to the two sets of original data arrays X. a and X b The data from the level flight phase of the UAV is filtered to obtain the first preprocessed data array D. a Second data array D b .

[0016] Furthermore, in step six, the mixture matrix M = [sig] is solved using Fast Independent Component Analysis (FastICA). rb D a D b The independent components S are obtained according to M = A·S, where A is the mixing matrix. The relationship between each independent component in S and sig is calculated. rb The correlation coefficients were used to extract the component with the largest absolute value of the correlation coefficient as an independent diurnal component.

[0017] Furthermore, in step seven, the preprocessed first data array D... a Second data array D b Subtracting the individual diurnal components separately yields the clean components in the dual-machine data.

[0018] According to another aspect of the present invention, a dual-machine collaborative short-duration diurnal variation magnetic field interference suppression system is provided, which uses the dual-machine collaborative short-duration diurnal variation magnetic field interference suppression method described above to suppress short-duration diurnal variation magnetic field interference.

[0019] Furthermore, the dual-drone collaborative short-duration diurnal variation magnetic field interference suppression system includes: a database construction unit, which collects aeromagnetic observation data during dual-drone collaborative aeromagnetic operations in nearshore areas and simultaneously acquires high-precision geomagnetic time-series signals recorded by shore-based diurnal variation monitoring stations to construct a training database containing dual-channel aeromagnetic signals and their corresponding diurnal variation reference labels; a diurnal variation interference feature identification network model training unit, which trains a time-series encoder-decoder network model based on the constructed labeled training database to achieve high-precision identification of diurnal variation interference features, providing prior model support for interference extraction under offshore operations; and a raw data acquisition unit, which acquires the raw data array X of the payload carried by the two UAVs. a and X b The preprocessing unit is used to preprocess the original data array to obtain the preprocessed first data array D. a Second data array D b The diurnal variation magnetic interference feature signal acquisition unit is used to input the preprocessed dual-channel data into the trained diurnal variation interference identification model and output the corresponding diurnal variation interference feature signal sig. rb As a guiding prior in the blind source separation process, it is used to enhance the identifiability and separation robustness of diurnal variation components; the diurnal variation component extraction unit is used to extract the diurnal variation characteristic interference signal sig rb First data array D a Second data array D b Combinatorial mixing matrix M = [sig rb D a D b The mixture matrix M = [sig] is solved using Fast Independent Component Analysis (FastICA). rb D a D b The system extracts independent diurnal variation components; a magnetic field interference suppression unit is used to process the preprocessed first data array D based on the independent diurnal variation components. a Second data array D b Diurnal variation interference removal is performed to obtain clean components in the dual-machine data, thus completing the suppression of short-term diurnal variation magnetic field interference.

[0020] This invention provides a method for suppressing short-duration diurnal magnetic field interference based on dual-drone collaborative operation. The method first trains and constructs a deep neural network model for identifying diurnal interference based on near-shore dual-drone aeromagnetic observation data and reference signals from synchronously deployed diurnal variation stations. Then, during offshore operations, the method acquires raw data from the payloads of the two drones, including multi-source information such as total magnetic field strength, timestamp, longitude, latitude, altitude, ground speed, heading angle, and the three components of the geomagnetic field. The raw data is then preprocessed, including time alignment, level flight segment extraction, and bandpass filtering, and input into the trained model to achieve intelligent identification of diurnal interference characteristic signals. Furthermore, the method combines the FastICA (Fast Independent Component Analysis) algorithm, utilizing the prior features output by the neural network to guide the blind source separation process, accurately extracting the independent components of diurnal magnetic interference, and suppressing interference from the raw signals. In practical applications, this method can suppress diurnal interference and significantly improve the detection capability of underwater targets. Therefore, compared with the prior art, the dual-machine collaborative short-term diurnal variation magnetic field interference suppression method provided by this invention proposes a neural network-guided blind source separation diurnal variation interference suppression method. This method does not rely on fixed diurnal variation monitoring stations and still has strong adaptability under the condition of no reference signal. It significantly improves the anti-interference capability and data quality of the offshore aeromagnetic detection system and effectively enhances the identification and detection capability of underwater weak magnetic targets. Attached Figure Description

[0021] The accompanying drawings, which form part of this specification, are provided to further illustrate embodiments of the invention and, together with the textual description, explain the principles of the invention. It is obvious that the drawings described below are merely some embodiments of the invention, and those skilled in the art can obtain other drawings based on these drawings without any creative effort.

[0022] Figure 1 A flowchart of a method for suppressing short-duration diurnal magnetic field interference based on dual-machine collaborative operation according to a specific embodiment of the present invention is shown;

[0023] Figure 2 A schematic diagram of the structure of a time-series coding-decoding network diurnal interference identification model provided according to a specific embodiment of the present invention is shown;

[0024] Figure 3 A schematic diagram of a dual-aircraft collaborative exploration scenario in the open ocean is shown, according to a specific embodiment of the present invention.

[0025] Figure 4 A schematic diagram of the processing results of data 1 (with a target) before and after removing solar interference (the circled area represents the simulated target signal) provided according to a specific embodiment of the present invention is shown.

[0026] Figure 5 A schematic diagram of the processing results of data 2 (without target) before and after removal from diurnal variation interference provided according to a specific embodiment of the present invention is shown. Detailed Implementation

[0027] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the present invention or its application or use. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0028] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments according to this application. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0029] Unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps set forth in these embodiments do not limit the scope of the invention. It should also be understood that, for ease of description, the dimensions of the various parts shown in the drawings are not drawn to actual scale. Techniques, methods, and devices known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and devices should be considered part of the specification. In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values. It should be noted that similar reference numerals and letters in the following figures denote similar items; therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0030] like Figures 1 to 5As shown, a specific embodiment of the present invention provides a method for suppressing short-duration diurnal variation magnetic field interference based on dual-drone collaboration. This method includes: Step 1, constructing a labeled training database based on nearshore dual-drone collaborative aeromagnetic data and synchronous observations from shore-based diurnal variation stations; Step 2, using a time-series encoder-decoder network and the MSE loss function for supervised training based on dual-drone aeromagnetic input and diurnal variation station labels to achieve efficient extraction of diurnal variation interference features; Step 3, having two UAVs arranged in a "||" shape to conduct detection in the mission area, with the raw data array X of the payloads carried by the two UAVs... a and X b The data is transmitted down to the ground data station via a data link; step four involves preprocessing the original data array to obtain the preprocessed first data array D. a Second data array D b Step 5: The preprocessed data matrix [D] a D b The input is fed into the trained diurnal variation interference identification model to extract the diurnal variation magnetic interference feature signal sig. rb Step 6: Convert the diurnal variation characteristic interference signal sig rb First data array D a Second data array D b Combinatorial mixing matrix M = [sig rb D a D b The mixture matrix M = [sig] is solved using Fast Independent Component Analysis (FastICA). rb D a D b Step 7: Based on the independent diurnal variation components, process the preprocessed first data array D... a Second data array D b Diurnal variation interference removal is performed to obtain clean components in the dual-machine data, thus completing the suppression of short-term diurnal variation magnetic field interference.

[0031] This configuration provides a method for suppressing short-duration diurnal magnetic field interference based on dual-drone collaborative operation. The method first obtains the raw payload data of the two UAVs, then preprocesses the raw data, identifies diurnal interference characteristic signals, extracts independent components of diurnal interference using the FastICA algorithm, and finally removes the diurnal interference. Compared with existing technologies, the method provided by this invention can suppress diurnal interference in practical applications, significantly improving the detection capability of underwater targets. Therefore, compared with existing technologies, the dual-drone collaborative airborne magnetic detection diurnal magnetic interference suppression method provided by this invention proposes a neural network-guided blind source separation-based dual-drone collaborative airborne magnetic detection method. This method can suppress diurnal magnetic interference in magnetic measurement data with high precision and effectively improve the signal-to-noise ratio.

[0032] Specifically, this invention addresses a scenario where two UAVs simultaneously conduct magnetic surveys in the open ocean. Limited by geographical conditions and deployment costs, fixed monitoring stations cannot be established, resulting in a lack of effective diurnal variation reference information. Furthermore, short-term diurnal magnetic interference noise exhibits randomness and non-stationarity, making it difficult to suppress using conventional filtering methods. Neural networks possess powerful nonlinear fitting and feature learning capabilities, demonstrating significant advantages in identifying complex interference signals. Blind source separation refers to the process of recovering the independent components of a source signal from only the observed signal, based on the statistical characteristics of the input source signal, when the source signal and transmission channel parameters are unknown. In recent years, blind source separation has become a research hotspot in signal processing. Since Herault and Jutten proposed a neural-like blind source separation method in 1991, numerous blind source separation algorithms have emerged. Blind source separation has been widely applied in fields such as audio signal processing, biomedical engineering, communications and radar, image and video processing, and geophysics and environmental science. In dual-aircraft collaborative airborne magnetic surveying, each aircraft's magnetic detector acts as a signal receiving channel. Therefore, the entire dual-aircraft collaborative surveying system is equivalent to a multi-channel signal receiving system, meeting the basic requirements for blind source separation. Based on this, this invention proposes a method for suppressing diurnal magnetic interference in dual-aircraft collaborative airborne magnetic surveying based on neural network-guided blind source separation.

[0033] To achieve the above objectives, the technical solution provided by the present invention includes the following steps:

[0034] Step one involves conducting dual-aircraft coordinated aeromagnetic operations in the nearshore area while simultaneously deploying highly stable diurnal variation monitoring stations near the coastline of the survey area. Aeromagnetic observation data from both platforms, along with high-precision geomagnetic time-series signals recorded by the diurnal variation stations, are collected synchronously. A training database containing multiple sets of dual-channel aeromagnetic signals and corresponding diurnal variation reference signals is then constructed for subsequent model training.

[0035] Step 2: Construct a time-series coding-decoding network diurnal interference identification model, as shown in Table 1, using the dual-aircraft aeromagnetic signals [sig1(t), sig2(t)]. t＝1,...,N As a dual-channel input, the daily variable station observation signal y(t) is used. t＝1,...,N Mean squared error (MSE) was used as the supervisory label, and the network parameters were optimized through backpropagation. After training, the model was able to effectively identify diurnal variation interference features (sig) from complex mixed signals. rb (t) t＝1,...,N .

[0036] Table 1. Layer structure of the temporal coding-decoding network diurnal interference identification model.

[0037]

[0038] Step 3: Two drones are positioned in a "||" shape to conduct reconnaissance in the mission area. The raw data array X carried by the two drones is the effective payload. a and X b The data is transmitted down to the ground data station via a data link. In step one, the sampling rates of the two sets of raw data arrays are the same, and the two sets of raw data arrays X... a and X b All of them contain the following information: total magnetic field, time, longitude, latitude, altitude, ground speed, heading angle, and geomagnetic components.

[0039] Step four: Preprocess the original data array to obtain the preprocessed first data array D. a Second data array D b Specifically, firstly, the two sets of original data arrays X are separated using time information. a and X b Alignment, followed by extraction of the two sets of original data arrays X a and X b The data from the level flight phase of the UAV was used to design appropriate bandpass filters based on the upper bound estimation function of the characteristic frequency of the magnetic anomaly signal, and then applied to the two sets of original data arrays X. a and X b The data from the level flight phase of the UAV is bandpass filtered to obtain the preprocessed first data array D. a Second data array D b .

[0040] Step 5: Process the preprocessed first data array D a Second data array D b The input is fed into the trained diurnal variation interference identification model, and the diurnal variation characteristic interference signal sig is obtained. rb .

[0041] Step 6: Convert the diurnal variation characteristic interference signal sigrb First data array D a Second data array D b Combinatorial mixing matrix M = [sig rb D a D b The mixture matrix M = [sig] is solved using Fast Independent Component Analysis (FastICA). rb D a D b ], and extract the independent diurnal variation components. Specifically, in step six, the mixture matrix M = [sig] is solved using Fast Independent Component Analysis (FastICA). rb D a D b The independent components S are obtained according to M = A·S, where A is the mixing matrix. The relationship between each independent component in S and sig is calculated. rb The correlation coefficients were used to extract the component with the largest absolute value of the correlation coefficient as an independent diurnal component.

[0042] Step 7: Based on the independent diurnal variation components, process the preprocessed first data array D a Second data array D b Diurnal variation interference removal is performed to obtain clean components from the dual-machine data, thus completing the suppression of short-term diurnal variation magnetic field interference. Specifically, in step seven, the preprocessed first data array D... a Second data array D b Subtracting the individual diurnal components separately yields the clean components in the dual-machine data.

[0043] According to another aspect of the present invention, a dual-machine collaborative short-duration diurnal variation magnetic field interference suppression system is provided, which uses the dual-machine collaborative short-duration diurnal variation magnetic field interference suppression method described above to suppress short-duration diurnal variation magnetic field interference.

[0044] This configuration provides a dual-aircraft collaborative short-duration diurnal variation magnetic field interference suppression system. The system first trains and constructs a deep neural network diurnal variation interference identification model based on nearshore dual-aircraft aeromagnetic observation data and reference signals from synchronously deployed diurnal variation stations. Then, during offshore operations, pre-processed dual-aircraft data is input into the trained model to achieve intelligent identification of diurnal variation interference characteristic signals. Furthermore, the FastICA (Fast Independent Component Analysis) algorithm is combined to accurately extract the independent components of diurnal variation magnetic interference, thereby suppressing interference from the original signal. Existing traditional filtering methods have limitations in suppressing non-stationary, offshore diurnal variation interference. Therefore, compared with existing technologies, the dual-aircraft collaborative short-duration diurnal variation magnetic field interference suppression system provided in this invention proposes a dual-aircraft collaborative aeromagnetic sounding diurnal variation magnetic interference suppression method based on a neural network-guided blind source separation algorithm. This method still exhibits strong adaptability under conditions without reference signals, significantly improving the anti-interference capability and data quality of offshore aeromagnetic sounding systems, and effectively enhancing the identification and detection capabilities of underwater weak magnetic targets.

[0045] Specifically, the dual-aircraft collaborative short-term diurnal variation magnetic field interference suppression system includes: a database construction unit, a diurnal variation interference feature identification network model training unit, a raw data acquisition unit, a preprocessing unit, a diurnal variation magnetic interference feature signal acquisition unit, a diurnal variation component extraction unit, and a magnetic field interference suppression unit. The database construction unit is used to construct a training database containing dual-channel aeromagnetic signals and their corresponding diurnal variation reference labels. The diurnal variation interference feature identification network model training unit is used to train a time-series encoder-decoder network model to achieve high-precision identification of diurnal variation interference features. The raw data acquisition unit is used to acquire the raw data array X of the payload carried by the two UAVs. a and X b The preprocessing unit is used to preprocess the original data array to obtain the preprocessed first data array D. a Second data array D b The diurnal variation magnetic interference characteristic signal acquisition unit is used to obtain signals from [D] a D b The diurnal variation magnetic interference characteristic signal sig in the trained model rb The diurnal variation component extraction unit is used to extract the diurnal variation characteristic interference signal sig. rb First data array D a Second data array D b Combinatorial mixing matrix M = [sig rb D a D b The mixture matrix M = [sig] is solved using Fast Independent Component Analysis (FastICA). rb D a D bThe independent diurnal variation components are extracted, and the magnetic field interference suppression unit is used to process the preprocessed first data array D based on the independent diurnal variation components. a Second data array D b Diurnal variation interference removal is performed to obtain clean components in the dual-machine data, thus completing the suppression of short-term diurnal variation magnetic field interference.

[0046] To gain a further understanding of the present invention, the following description is provided. Figures 1 to 5 The present invention provides a detailed description of the method for suppressing short-duration diurnal magnetic field interference based on dual-machine collaboration.

[0047] like Figures 1 to 5 As shown in the figure, a method for suppressing short-duration diurnal variation magnetic field interference based on dual-machine collaborative approach is provided according to a specific embodiment of the present invention. The basic idea is to integrate data-driven deep learning methods with an independent physical signal separation mechanism to suppress diurnal variation interference. The complete flowchart is shown below. Figure 1 As shown. First, the raw data arrays of the payloads carried by the two UAVs are transmitted to the ground data station via a data link. Second, the raw data arrays are preprocessed, including alignment, extraction of level flight segment data, and filtering. Then, the preprocessed data is input into a trained diurnal interference identification model to extract diurnal interference features. Subsequently, the diurnal interference features and the preprocessed data from both UAVs are combined into a matrix, and blind source separation is used to extract independent diurnal components, thereby suppressing diurnal interference for both UAVs. The specific implementation steps are as follows:

[0048] (1) While conducting dual-aircraft coordinated aeromagnetic operations in the nearshore area, high-stability diurnal variation monitoring stations were deployed near the coast of the survey area. Aeromagnetic observation data from both platforms and high-precision geomagnetic time series signals recorded by the diurnal variation stations were collected simultaneously to construct a training database containing multiple sets of dual-channel aeromagnetic signals and corresponding diurnal variation reference signals;

[0049] (2) Figure 2 As shown, a time-series encoder-decoder network model for identifying diurnal interference is constructed. The dual-aircraft aeromagnetic signals [sig1(t), sig2(t)] t=1,...,N are used as dual-channel inputs, and the diurnal station observation signals y(t) t=1,...,N are used as supervision labels. Mean squared error (MSE) is used as the loss function, and the network parameters are optimized through backpropagation. After training, the model can effectively identify the diurnal interference feature sig1(t), sig2(t)] from complex mixed signals. rb (t) t＝1,...,N ;

[0050] (3) Regarding such Figure 3 The scenario shown depicts a dual-drone collaborative detection system. Drone 1 and Drone 2 are positioned in a "||" shape within the mission area. The payloads carried by each drone transmit the raw data array X...a and X b The data is transmitted down to the ground data station. The two sets of raw data arrays have the same sampling rate. X a and X b All of them contain the following information: total magnetic field, time, longitude, latitude, altitude, ground speed, heading angle, and geomagnetic components;

[0051] (4) After receiving the raw data, the ground station first aligns the two sets of raw data arrays using time information, then extracts the data from the UAV's level flight segment, and finally designs a suitable bandpass filter based on the upper bound estimation function of the magnetic anomaly signal characteristic frequency to filter the data, thus obtaining the preprocessed data array D. a and D b ;

[0052] (5) The preprocessed data [D a D b The diurnal variation interference signal sig is extracted from the pre-trained diurnal variation interference identification model. rb The diurnal variation characteristic interference signal sig rb With the preprocessed magnetic field data D a D b Combinatorial mixing matrix M = [sig rb D a D b Using FastICA, independent component analysis was performed on M to obtain independent components S, where M = A·S, and A is the mixture matrix. The relationship between each independent component in S and sig was calculated. rb The correlation coefficients were analyzed, and the component with the largest absolute value of the correlation coefficient was extracted as an independent diurnal variation component, thereby improving the preprocessed magnetic field data D. a D b Diurnal interference removal was performed to obtain clean components from the dual-machine data. A method for suppressing short-term diurnal magnetic interference based on dual-machine collaborative suppression was completed.

[0053] In summary, this invention provides a method for dual-drone collaborative suppression of short-term diurnal magnetic interference. This method first obtains the raw payload data of two UAVs, including information such as total magnetic field, time, longitude, latitude, altitude, ground speed, heading angle, and the three components of the geomagnetic field. Then, the raw data undergoes preprocessing, diurnal interference characteristic signal discrimination, extraction of independent diurnal interference components using the FastICA algorithm, and finally, removal of the diurnal interference. Compared with existing technologies, the method provided by this invention integrates a data-driven deep learning method with a physical signal separation mechanism based on statistical independence. In practical applications, it can suppress diurnal interference and significantly improve the detection capability of underwater targets.

[0054] For ease of description, spatial relative terms such as "above," "on top of," "on the upper surface of," "above," etc., are used herein to describe the spatial positional relationship of a device or feature as shown in the figures to other devices or features. It should be understood that spatial relative terms are intended to encompass different orientations in use or operation beyond the orientation of the device as described in the figures. For example, if the device in the figures were inverted, a device described as "above" or "on top of" other devices or structures would subsequently be positioned as "below" or "under" other devices or structures. Thus, the exemplary term "above" can include both "above" and "below." The device may also be positioned in other different ways (rotated 90 degrees or in other orientations), and the spatial relative descriptions used herein will be interpreted accordingly.

[0055] Furthermore, it should be noted that the use of terms such as "first" and "second" to define components is merely for the purpose of distinguishing the corresponding components. Unless otherwise stated, the above terms have no special meaning and therefore should not be construed as limiting the scope of protection of this invention.

[0056] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for suppressing short-duration diurnal magnetic field interference based on dual-machine collaborative operation, characterized in that, The method for suppressing short-duration diurnal magnetic field interference based on dual-machine collaborative operation includes: Step 1: Based on the nearshore dual-aircraft collaborative aeromagnetic data and the synchronous observation results of the shore-based diurnal variation station, construct a labeled training database; Step 2: Based on the training database, a temporal encoder-decoder network and the MSE loss function are used for supervised training to achieve efficient extraction of diurnal variation interference features; Step 3, Collaborative Data Acquisition for Long-Distance Exploration: During long-distance operations, two UAVs are arranged in a "||" shape to conduct exploration in the mission area. The raw data array X carried by the two UAVs is the effective payload. a and X b Data is transmitted down to the ground data station via a data link; Step four: Preprocess the original data array to obtain the preprocessed first data array D. a Second data array D b ; Step 5: Transfer the preprocessed dual-machine data array [D] a D b Input a pre-trained diurnal variation interference identification model, and output the estimated diurnal variation magnetic interference characteristic signal sig. rb ; Step 6: Convert the diurnal variation characteristic interference signal sig rb First data array D a Second data array D b Combinatorial mixing matrix M = [sig rb D a D b The mixture matrix M = [sig] is solved using Fast Independent Component Analysis (FastICA). rb D a D b ], and extract independent daily variation components; Step 7: Based on the independent diurnal variation components, process the preprocessed first data array D a Second data array D b Diurnal variation interference removal is performed to obtain clean components in the dual-machine data, thus completing the suppression of short-term diurnal variation magnetic field interference.

2. The method for suppressing short-duration diurnal magnetic field interference based on dual-machine collaborative operation according to claim 1, characterized in that, In step three, the sampling rates of the two sets of original data arrays are the same, and the two sets of original data arrays X a and X b All of them contain the following information: total magnetic field, time, longitude, latitude, altitude, ground speed, heading angle, and geomagnetic components.

3. The method for suppressing short-duration diurnal magnetic field interference based on dual-machine collaborative operation according to claim 2, characterized in that, In step three, the two sets of original data arrays X are first separated using time information. a and X b Alignment, followed by extraction of the two sets of original data arrays X a and X b The data from the level flight phase of the UAV was used to design appropriate bandpass filters based on the upper bound estimation function of the characteristic frequency of the magnetic anomaly signal, and then applied to the two sets of original data arrays X. a and X b The data from the level flight phase of the UAV is filtered to obtain the first preprocessed data array D. a Second data array D b .

4. The method for suppressing short-duration diurnal magnetic field interference based on dual-machine collaborative operation according to claim 3, characterized in that, In step five, the preprocessed dual-machine data [D] a D b The input is fed into the trained diurnal variation feature identification model, and the diurnal variation feature signal sig has been obtained. rb .

5. The method for suppressing short-duration diurnal magnetic field interference based on dual-machine collaborative operation according to claim 4, characterized in that, In step six, the mixture matrix M = [sig] is solved using Fast Independent Component Analysis (FastICA). rb D a D b The independent components S are obtained according to M = A·S, where A is the mixing matrix. The relationship between each independent component in S and sig is calculated. rb The correlation coefficient was used to extract the component with the largest absolute value of the correlation coefficient as an independent diurnal component.

6. The method for suppressing short-duration diurnal magnetic field interference based on dual-machine collaborative operation according to claim 5, characterized in that, In step seven, the preprocessed first data array D a Second data array D b Subtracting the individual diurnal components separately yields the clean components in the dual-machine data.

7. A dual-machine collaborative short-duration diurnal magnetic field interference suppression system, characterized in that, The dual-machine collaborative short-duration diurnal variation magnetic field interference suppression system uses the dual-machine collaborative short-duration diurnal variation magnetic field interference suppression method as described in any one of claims 1 to 7 to suppress short-duration diurnal variation magnetic field interference.

8. The dual-machine collaborative short-duration diurnal variation magnetic field interference suppression system according to claim 7, characterized in that, The dual-machine collaborative short-duration diurnal variation magnetic field interference suppression system includes: The database construction unit is used to collect aeromagnetic observation data during dual-aircraft coordinated aeromagnetic operations in the nearshore area, and simultaneously acquire high-precision geomagnetic time series signals recorded by shore-based diurnal variation monitoring stations to construct a training database containing dual-channel aeromagnetic signals and their corresponding diurnal variation reference labels. The training unit for the diurnal variation interference feature identification network model is used to train the time-series encoder-decoder network model to achieve high-precision identification of diurnal variation interference features. The raw data acquisition unit is used to acquire the raw data array X of the payloads carried by the two UAVs. a and X b ; The preprocessing unit is used to preprocess the original data array to obtain a preprocessed first data array D. a Second data array D b ; The diurnal variation magnetic interference characteristic signal acquisition unit, wherein the diurnal variation magnetic interference characteristic signal is used to acquire the preprocessed dual-machine data [D a D b Input the trained diurnal variation interference identification model and output the corresponding diurnal variation interference feature signal sig. rb ; The diurnal variation component extraction unit is used to extract the diurnal variation characteristic interference signal sig. rb First data array D a Second data array D b Combinatorial mixing matrix M = [sig rb D a D b The mixture matrix M = [sig] is solved using Fast Independent Component Analysis (FastICA). rb D a D b ], and extract independent daily variation components; Magnetic field interference suppression unit, which is used to suppress the interference of the first data array D based on independent diurnal variation components. a Second data array D b Diurnal variation interference removal is performed to obtain clean components in the dual-machine data, thus completing the suppression of short-term diurnal variation magnetic field interference.

Images