Magnetic field vector sensor based on large-angle tilted fiber Bragg grating

Abstract

The present application relates to a magnetic field vector sensor based on a large-angle tilted fiber grating. It includes: when a temperature drift compensation processor is used to correct the influence of temperature changes on the measurement results, in the temperature drift compensation processor, by obtaining the magnetic field detection spectrum generated by the spectrometer, and by real-time monitoring of the collected external environment temperature data through the temperature sensor, then introducing an image processing and data analysis algorithm based on artificial intelligence and deep learning at the back end to analyze the magnetic field detection spectrum and capture the time series characteristics of the external environment temperature data, thereby extracting and characterizing the cross-modal joint feature representation between the magnetic field detection spectrum characteristics and the external environment temperature time series characteristics, and generating a magnetic field intensity decoding value after temperature drift compensation through a magnetic field intensity detector, so as to achieve temperature drift compensation correction. In this way, it can help eliminate the measurement error caused by temperature changes, thereby improving the overall accuracy and reliability of the magnetic field vector sensor.

Description

Technical Field

[0001] The present application relates to the field of optical fiber sensing technology, and in particular to a magnetic field vector sensor based on a large-angle tilted optical fiber Bragg grating. Background Art

[0002] Traditional magnetic field sensors rely on principles such as the Hall effect, magneto-optical effect or hysteresis effect. Although these sensors perform well in many applications, they generally have problems such as low accuracy and limited sensitivity, especially when used in complex environments. There are many limitations, such as narrow bandwidth, easy magnetic saturation and even safety hazards. With the advancement of fiber optic sensing technology, these problems have been alleviated to a large extent. Fiber optic sensors are gradually replacing traditional sensors with their high precision, miniaturization, lightweight and strong anti-electromagnetic interference capabilities, and are widely used in many fields such as marine hydrology, geological disaster monitoring, building structural health monitoring and geophysical exploration.

[0003] However, even in the field of fiber optic sensing technology, there are certain limitations. For example, although the magnetic field sensor made by combining fiber optic Bragg gratings with magnetostrictive materials has been improved, it still faces the problems of complex manufacturing process and insufficient sensitivity, and it is difficult to meet the increasing accuracy requirements. In addition, although the simple combination of fiber optic Bragg gratings and magnetic fluids improves sensitivity, it loses the ability to detect the direction of the magnetic field.

[0004] In order to solve the above problems, Chinese patent CN109375124B proposes a magnetic field vector sensor based on large-angle tilted fiber Bragg grating, which utilizes the high sensitivity of large-angle tilted fiber Bragg grating to the refractive index change of the external environment, combined with the birefringence characteristics of the magnetic fluid, to achieve simultaneous measurement of the magnetic field intensity and direction. Specifically, when the magnetic field intensity changes, the refractive index of the magnetic fluid will also change accordingly, which will cause the spectrum of the large-angle tilted fiber Bragg grating to drift. By monitoring this drift, the intensity of the magnetic field can be determined. At the same time, since the refractive index sensitivity of the large-angle tilted fiber Bragg grating is directional, the refractive index changes in different directions can be obtained by rotating the sensor, thereby determining the direction of the magnetic field.

[0005] Although the above magnetic field vector sensor has high sensitivity and accuracy, in actual operation, temperature changes will cause changes in the refractive index of the optical fiber material itself, which in turn affects the spectral characteristics of the large-angle tilted fiber Bragg grating. Although the large-angle tilted fiber Bragg grating has a low temperature crosstalk characteristic, which makes it less affected by external temperature changes, it still cannot completely rule out the potential impact of temperature changes on the measurement of magnetic field strength and direction. Especially in environments where long-term continuous monitoring is required or where the temperature changes are large, this impact may be more significant.

[0006] Therefore, an optimized magnetic field vector sensor based on large-angle tilted fiber Bragg grating is desired. Summary of the invention

[0007] This summary is provided to introduce concepts in a brief form that will be described in detail in the detailed description below. This summary is not intended to identify key features or essential features of the claimed technical solution, nor is it intended to limit the scope of the claimed technical solution.

[0008] In a first aspect, the present application provides a magnetic field vector sensor based on a large-angle tilted fiber Bragg grating, the magnetic field vector sensor comprising: a broadband light source, a fiber circulator, a fiber rotary connector, a sensing probe and a spectrometer, the broadband light source is used to emit a broadband optical signal; the fiber circulator is used to guide the broadband optical signal to advance along a predetermined path, the fiber circulator is connected to the fiber rotary connector and the spectrometer by fiber fusion; the fiber rotary connector is used to control the rotation of the sensing probe; the sensing probe is used to receive and transmit reflected light signals, the sensing probe comprises a 45° tilted fiber Bragg grating, a large-angle tilted fiber Bragg grating, a chirped fiber Bragg grating and a glass capillary filled with a magnetic fluid, wherein the tilt angle of the large-angle tilted fiber Bragg grating is 69-89°; the spectrometer is used to perform spectral analysis processing on the reflected light signal to generate a magnetic field detection spectrum; the magnetic field vector sensor based on the large-angle tilted fiber Bragg grating also includes a temperature drift compensation processor, which is used to perform temperature drift compensation correction on the detected magnetic field intensity based on the magnetic field detection spectrum;

[0009] Wherein, the temperature drift compensation processor comprises:

[0010] A magnetic field detection spectrum acquisition module, used to obtain a magnetic field detection spectrum generated by the spectrometer;

[0011] The external environment temperature acquisition module is used to obtain the time queue of the external environment temperature collected by the temperature sensor;

[0012] An external environment temperature time series encoding module is used to perform sequence encoding on the time queue of the external environment temperature to obtain an external environment temperature time series associated implicit feature vector;

[0013] A magnetic field detection spectrum semantic feature extraction and optimization module is used to perform spectrum feature extraction and feature selection optimization processing on the magnetic field detection spectrum to obtain a magnetic field detection optimized spectrum feature map;

[0014] A temperature drift compensation magnetic field detection spectrum semantic expression module is used to input the magnetic field detection optimization spectrum feature map and the external environment temperature time series associated implicit feature vector into a gated cross-modal joint encoder based on an analog converter structure to obtain a temperature drift compensation magnetic field detection optimization spectrum feature map;

[0015] The magnetic field strength detection module is used to determine the detection result based on the temperature drift compensation magnetic field detection optimization spectrum characteristic diagram, and the detection result is used to represent the magnetic field strength decoding value.

[0016] Optionally, the external environment temperature time series encoding module includes: inputting the time queue of the external environment temperature into a sequence encoder based on a Bi-LSTM model to obtain an implicit feature vector associated with the external environment temperature time series.

[0017] Optionally, the magnetic field detection spectrum map semantic feature extraction optimization module includes: a magnetic field detection spectrum feature extraction unit, used to input the magnetic field detection spectrum map into a spectrum feature extractor based on a void convolutional neural network model to obtain a magnetic field detection spectrum feature map; a magnetic field detection spectrum feature selection optimization unit, used to input the magnetic field detection spectrum feature map into a feature selection optimization module based on a characteristic energy distribution spectrum to obtain the magnetic field detection optimized spectrum feature map.

[0018] Optionally, the magnetic field detection spectrum feature selection optimization unit includes: a vector expansion subunit, used to vectorize and expand the magnetic field detection spectrum feature map along the channel dimension to obtain a sequence of magnetic field detection spectrum channel feature vectors; a characteristic energy distribution spectrum vector calculation subunit, used to calculate the characteristic energy distribution spectrum vector of each magnetic field detection spectrum channel feature vector in the sequence of magnetic field detection spectrum channel feature vectors to obtain a sequence of magnetic field detection spectrum channel characteristic energy distribution spectrum vectors; a position mean vector calculation subunit, used to calculate the position mean vector of the sequence of magnetic field detection spectrum channel characteristic energy distribution spectrum vectors as the center vector of the magnetic field detection spectrum channel energy spectrum distribution; an energy distribution spectrum span factor calculation subunit, used to calculate the energy spectrum distribution of the magnetic field detection spectrum channel. The energy distribution spectrum span factor between the center vector and each magnetic field detection spectrum channel characteristic energy distribution spectrum vector in the sequence of the magnetic field detection spectrum channel characteristic energy distribution spectrum vector is used to obtain the sequence of magnetic field detection spectrum channel energy distribution spectrum span factors; a spectrum span factor and preset threshold comparison subunit is used to determine the sequence of selected optimized magnetic field detection spectrum channel characteristic vectors based on the comparison between each magnetic field detection spectrum channel energy distribution spectrum span factor in the sequence of the magnetic field detection spectrum channel energy distribution spectrum span factor and the preset threshold; a dimension reconstruction subunit is used to reconstruct the dimension of the selected optimized magnetic field detection spectrum channel characteristic vector sequence to obtain the magnetic field detection optimized spectrum feature map.

[0019] Optionally, the characteristic energy distribution spectrum vector calculation subunit includes: a characteristic distribution energy synergistic representation vector calculation secondary subunit, which is used to respectively calculate the characteristic distribution energy synergistic representation vectors between each magnetic field detection spectrum channel characteristic vector and each other magnetic field detection spectrum channel characteristic vector in the sequence of the magnetic field detection spectrum channel characteristic vector to obtain a sequence of multiple magnetic field detection spectrum channel characteristic distribution energy synergistic representation vectors; a characteristic distribution energy synergistic factor vector calculation secondary subunit, which is used to calculate the characteristic distribution energy synergistic factor vector of the sequence of each magnetic field detection spectrum channel characteristic distribution energy synergistic representation vector in the sequence of the multiple magnetic field detection spectrum channel characteristic distribution energy synergistic representation vectors to obtain the sequence of the magnetic field detection spectrum channel characteristic energy distribution spectrum vectors; wherein the characteristic distribution energy synergistic representation vector calculation secondary subunit is used to: extract a predetermined magnetic field detection spectrum channel characteristic vector from the sequence of characteristic energy distribution spectrum vectors of the magnetic field detection spectrum channel characteristic vector; and perform element-by-element maximum value extraction on the predetermined magnetic field detection spectrum channel characteristic vector and each other magnetic field detection spectrum channel characteristic vector to obtain a sequence of magnetic field detection spectrum channel characteristic distribution energy synergistic representation vectors corresponding to the predetermined magnetic field detection spectrum channel characteristic vector.

[0020] Optionally, the characteristic energy distribution spectrum vector calculation subunit is used to: calculate the characteristic distribution energy synergy factor vector of each magnetic field detection spectrum channel characteristic distribution energy synergy representation vector in the sequence of the multiple magnetic field detection spectrum channel characteristic distribution energy synergy representation vectors to obtain the sequence of the magnetic field detection spectrum channel characteristic energy distribution spectrum vectors, including: respectively extracting the maximum value of each magnetic field detection spectrum channel characteristic distribution energy synergy representation vector in the sequence of the magnetic field detection spectrum channel characteristic distribution energy synergy representation vector to obtain a sequence of magnetic field detection spectrum channel characteristic maximum values; respectively calculating the mean and variance of each magnetic field detection spectrum channel characteristic distribution energy synergy representation vector to obtain a sequence of magnetic field detection spectrum channel characteristic means and a sequence of magnetic field detection spectrum channel characteristic variances; adding the sequence of magnetic field detection spectrum channel characteristic variances and preset hyperparameters by position to obtain a sequence of magnetic field detection spectrum channel characteristic first energy distribution coefficients; calculating the maximum value of the magnetic field detection spectrum channel characteristic The sequence of magnetic field detection spectral channel difference values ​​is obtained by multiplying the sequence of magnetic field detection spectral channel characteristic variances by constant two according to position to obtain a sequence of modulated magnetic field detection spectral channel characteristic variances; after adding the sequence of magnetic field detection spectral channel difference values ​​and the sequence of modulated magnetic field detection spectral channel characteristic variances, each group of corresponding magnetic field detection spectral channel difference values ​​and modulated magnetic field detection spectral channel characteristic variances are added, the sum is added to the preset hyperparameter to obtain a sequence of magnetic field detection spectral channel characteristic second energy distribution coefficients; the sequence of magnetic field detection spectral channel characteristic first energy distribution coefficients is divided by the sequence of magnetic field detection spectral channel characteristic second energy distribution coefficients according to position to obtain the magnetic field detection spectral channel characteristic energy distribution spectrum vector composed of multiple characteristic distribution energy synergy factors.

[0021] Optionally, the energy distribution spectrum span factor calculation subunit is used to: calculate the positional difference between the characteristic energy distribution spectrum vector of the magnetic field detection spectrum channel and the center vector of the energy spectrum distribution of the magnetic field detection spectrum channel to obtain the differential characteristic vector of the magnetic field detection spectrum channel; calculate the covariance matrix between the center vector of the energy spectrum distribution of the magnetic field detection spectrum channel and the characteristic energy distribution spectrum vector of the magnetic field detection spectrum channel, and calculate the inverse of the covariance matrix to obtain the inverse covariance matrix of the magnetic field detection spectrum channel; calculate the transposed vector of the differential characteristic vector of the magnetic field detection spectrum channel and the product of the inverse covariance matrix of the magnetic field detection spectrum channel and the differential characteristic vector of the magnetic field detection spectrum channel to obtain the distribution value of the magnetic field detection spectrum channel; calculate the square root of the distribution value of the magnetic field detection spectrum channel to obtain the energy distribution spectrum span factor of the magnetic field detection spectrum channel.

[0022] Optionally, the temperature drift compensation magnetic field detection spectrum semantic expression module includes: a layer normalization and convolution coding processing unit, which is used to perform layer normalization and convolution coding processing on the magnetic field detection optimization spectrum feature map to obtain a magnetic field detection optimization spectrum space deepening feature map; a space deepening feature map copying unit, which is used to copy the magnetic field detection optimization spectrum space deepening feature map to obtain a backup magnetic field detection optimization spectrum space deepening feature map; a feature shape reshaping unit, which is used to reshape the backup magnetic field detection optimization spectrum space deepening feature map and the magnetic field detection optimization spectrum space deepening feature map respectively to obtain a backup magnetic field detection optimization spectrum space deepening feature matrix as an imitation query feature matrix and a magnetic field detection optimization spectrum space deepening feature matrix as an imitation key feature matrix; a transposed matrix calculation unit, which is used to calculate the product between the imitation query feature matrix and the transposed matrix of the imitation key feature matrix to obtain a magnetic field detection optimization spectrum local autocorrelation library matrix; an attention gating unit, which is used to input the magnetic field detection optimization spectrum local autocorrelation library matrix into a basis The attention gating unit of the GELU function is used to obtain the magnetic field detection optimization spectrum gating local autocorrelation library weight matrix; the vector product calculation unit is used to use the external environment temperature time series associated implicit feature vector as the value feature vector, calculate the product between the magnetic field detection optimization spectrum gating local autocorrelation library weight matrix and the external environment temperature time series associated implicit feature vector to obtain the external environment temperature cross-modal interaction fusion condition weight vector; the channel-by-channel weighted calculation unit is used to use the external environment temperature cross-modal interaction fusion condition weight vector as the weight vector, and perform channel-by-channel weighting on the magnetic field detection optimization spectrum feature map to obtain the temperature drift compensation magnetic field detection optimization spectrum feature map; wherein the layer normalization and convolution coding processing unit is used to: perform layer normalization processing on the magnetic field detection optimization spectrum feature map to obtain a layer normalized magnetic field detection optimization spectrum feature map; perform point convolution coding on the layer normalized magnetic field detection optimization spectrum feature map and then perform spatial convolution coding on the layer normalized magnetic field detection optimization spectrum feature map to obtain the magnetic field detection optimization spectrum space deepening feature map.

[0023] Optionally, the magnetic field strength detection module is used to: input the temperature drift compensation magnetic field detection optimization spectrum characteristic diagram into a decoder-based magnetic field strength detector to obtain the detection result.

[0024] By adopting the above technical solution, when using a temperature drift compensation processor to correct the influence of temperature change on the measurement result, in the temperature drift compensation processor, by obtaining the magnetic field detection spectrum generated by the spectrometer, and by real-time monitoring of the collected external environment temperature data through the temperature sensor, then introducing an image processing and data analysis algorithm based on artificial intelligence and deep learning at the back end to analyze the magnetic field detection spectrum and capture the time series characteristics of the external environment temperature data, thereby extracting and characterizing the cross-modal joint feature representation between the magnetic field detection spectrum characteristics and the external environment temperature time series characteristics, and generating a magnetic field intensity decoding value after temperature drift compensation through a magnetic field intensity detector, so as to realize temperature drift compensation correction. In this way, it can help eliminate the measurement error caused by temperature change, thereby improving the overall accuracy and reliability of the magnetic field vector sensor.

[0025] Other features and advantages of the present application will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The above and other features, advantages and aspects of the embodiments of the present application will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the originals and elements are not necessarily drawn to scale. In the drawings:

[0027] Figure 1 It is a schematic structural diagram of a magnetic field vector sensor based on large-angle tilted fiber Bragg grating according to an exemplary embodiment.

[0028] Figure 2 It is a block diagram of the temperature drift compensation processor of a magnetic field vector sensor based on a large-angle tilted fiber Bragg grating according to an exemplary embodiment.

[0029] Figure 3 The present invention is a flow chart showing a magnetic field vector sensing method based on large-angle tilted fiber Bragg grating according to an exemplary embodiment.

[0030] Figure 4 It is a block diagram of an electronic device according to an exemplary embodiment.

[0031] Figure 5 This is a diagram of an application scenario of a magnetic field vector sensor based on a large-angle tilted fiber Bragg grating according to an exemplary embodiment. DETAILED DESCRIPTION

[0032] The embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be construed as being limited to the embodiments described herein. Instead, these embodiments are provided to provide a more thorough and complete understanding of the present application. It should be understood that the drawings and embodiments of the present application are only for exemplary purposes and are not intended to limit the scope of protection of the present application.

[0033] It should be understood that the various steps described in the method implementation of the present application can be performed in different orders and / or performed in parallel. In addition, the method implementation may include additional steps and / or omit the steps shown. The scope of the present application is not limited in this respect.

[0034] The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0035] It should be noted that the concepts such as "first" and "second" mentioned in this application are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0036] It should be noted that the modifications of "one" and "plurality" mentioned in the present application are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0037] The names of the messages or information exchanged between multiple devices in the embodiments of the present application are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0038] The specific implementation of the present application is described in detail below with reference to the accompanying drawings.

[0039] Figure 1 FIG. 1 is a schematic diagram of a magnetic field vector sensor based on a large-angle tilted fiber Bragg grating according to an exemplary embodiment. Figure 1 As shown, the magnetic field vector sensor includes:

[0040] A broadband light source 1, an optical fiber circulator 2, an optical fiber rotary connector 3, a sensing probe 4 and a spectrometer 5, wherein the broadband light source 1 is used to emit a broadband optical signal; the optical fiber circulator 2 is used to guide the broadband optical signal to advance along a predetermined path, and the optical fiber circulator 2 is connected to the optical fiber rotary connector 3 and the spectrometer 5 by optical fiber fusion; the optical fiber rotary connector 3 is used to control the sensing probe 4 to rotate; the sensing probe 4 is used to receive and transmit reflected light signals, and the sensing probe 4 includes a 45° tilted optical fiber grating, a large-angle tilted optical fiber grating, a chirped optical fiber grating and a glass capillary filled with magnetic fluid, wherein the tilt angle of the large-angle tilted optical fiber grating is 69-89°; the spectrometer 5 is used to perform spectral analysis processing on the reflected light signal to generate a magnetic field detection spectrum diagram; the magnetic field vector sensor based on the large-angle tilted optical fiber grating also includes a temperature drift compensation processor 6, which is used to perform temperature drift compensation correction on the detected magnetic field intensity based on the magnetic field detection spectrum diagram.

[0041] It should be understood that the wide-spectrum light signal emitted by the broadband light source generates linearly polarized light after passing through the 45° tilted fiber grating, and then the linearly polarized light enters the large-angle tilted fiber grating, and then is reflected to the spectrometer through the chirped fiber grating, so as to perform spectral analysis on the reflected light signal obtained by transmission and reflection of the wide-spectrum light signal through the spectrometer to generate a magnetic field detection spectrum diagram.

[0042] In particular, in the process of measuring magnetic field strength, temperature fluctuations will cause changes in the physical properties of the sensing material, thereby affecting the measurement accuracy. Therefore, how to eliminate or reduce the impact of temperature changes on the magnetic field measurement results has become an urgent problem to be solved. Based on this, in the above-mentioned magnetic field vector sensor based on large-angle tilted fiber grating, by introducing the temperature drift compensation processor, the detected magnetic field strength is compensated for temperature drift based on the magnetic field detection spectrum and the changes in the external ambient temperature, thereby correcting the impact of temperature changes on the measurement results, so as to achieve more accurate magnetic field strength detection.

[0043] Specifically, the technical concept of the present application is to use a temperature drift compensation processor to correct the influence of temperature changes on the measurement results. In the temperature drift compensation processor, the magnetic field detection spectrum generated by the spectrometer is obtained, and the external environment temperature data collected is monitored in real time by a temperature sensor, and then the image processing and data analysis algorithm based on artificial intelligence and deep learning is introduced at the back end to analyze the magnetic field detection spectrum and capture the timing characteristics of the external environment temperature data, so as to extract and characterize the cross-modal joint feature representation between the magnetic field detection spectrum characteristics and the external environment temperature timing characteristics, and generate the magnetic field intensity decoding value after temperature drift compensation by the magnetic field intensity detector, so as to realize temperature drift compensation correction. In this way, it can help eliminate the measurement error caused by temperature changes, thereby improving the overall accuracy and reliability of the magnetic field vector sensor. In addition, in other examples of the present application, it is also possible to consider taking further measures in material selection and structural design, such as using temperature compensation optical fiber materials or optimizing the formula of magnetic fluid to further reduce the influence of temperature on sensor performance.

[0044] In one embodiment of the present application, Figure 2 is a block diagram of a temperature drift compensation processor of a magnetic field vector sensor based on a large-angle tilted fiber Bragg grating according to an exemplary embodiment, such as Figure 2 As shown, the temperature drift compensation processor 6 includes:

[0045] A magnetic field detection spectrum acquisition module 101 is used to obtain a magnetic field detection spectrum generated by the spectrometer;

[0046] The external environment temperature acquisition module 102 is used to obtain the time queue of the external environment temperature collected by the temperature sensor;

[0047] An external environment temperature time series encoding module 103 is used to perform sequence encoding on the time queue of the external environment temperature to obtain an external environment temperature time series associated implicit feature vector;

[0048] A magnetic field detection spectrum semantic feature extraction and optimization module 104 is used to perform spectrum feature extraction and feature selection optimization processing on the magnetic field detection spectrum to obtain a magnetic field detection optimized spectrum feature map;

[0049] The temperature drift compensation magnetic field detection spectrum semantic expression module 105 is used to input the magnetic field detection optimization spectrum feature map and the external environment temperature time series associated implicit feature vector into a gated cross-modal joint encoder based on an analog converter structure to obtain the temperature drift compensation magnetic field detection optimization spectrum feature map;

[0050] The magnetic field strength detection module 106 is used to determine a detection result based on the temperature drift compensation magnetic field detection optimization spectrum characteristic diagram, and the detection result is used to represent the magnetic field strength decoding value.

[0051] More specifically, in the temperature drift compensation processor, first, a magnetic field detection spectrum generated by a spectrometer is obtained, and a time queue of the external environment temperature collected by a temperature sensor is obtained. Then, the time queue of the external environment temperature is input into a sequence encoder based on a Bi-LSTM model for encoding, so as to extract the temporal forward and backward implicit correlation features of the external environment temperature in the time dimension, thereby obtaining the temporal correlation implicit feature vector of the external environment temperature.

[0052] In one embodiment of the present application, the external environment temperature time series encoding module includes: inputting the time queue of the external environment temperature into a sequence encoder based on a Bi-LSTM model to obtain an implicit feature vector associated with the external environment temperature time series.

[0053] Then, the magnetic field detection spectrum graph is input into a spectrum feature extractor based on a void convolutional neural network model for feature mining to extract the magnetic field detection spectrum semantic feature information in the magnetic field detection spectrum graph, thereby obtaining a magnetic field detection spectrum feature graph.

[0054] In one embodiment of the present application, the magnetic field detection spectrum map semantic feature extraction optimization module includes: a magnetic field detection spectrum feature extraction unit, used to input the magnetic field detection spectrum map into a spectrum feature extractor based on a void convolutional neural network model to obtain a magnetic field detection spectrum feature map; a magnetic field detection spectrum feature selection optimization unit, used to input the magnetic field detection spectrum feature map into a feature selection optimization module based on a characteristic energy distribution spectrum to obtain the magnetic field detection optimized spectrum feature map.

[0055] Furthermore, considering that different channels in the magnetic field detection spectrum feature map contain different semantic feature information about the magnetic field detection spectrum, these features contain not only information related to the magnetic field intensity detection, but also some irrelevant redundant information and noise interference. Therefore, in order to screen out the most representative and informative features from the original magnetic field detection spectrum feature map, thereby improving the accuracy and efficiency of subsequent processing steps, in the technical solution of the present application, the magnetic field detection spectrum feature map is further input into a feature selection optimization module based on a characteristic energy distribution spectrum to obtain a magnetic field detection optimized spectrum feature map. The feature selection optimization module based on a characteristic energy distribution spectrum is a feature selection technology based on energy distribution spectrum analysis, which aims to optimize the expression of the semantics of the magnetic field detection spectrum through the energy distribution of each quantized magnetic field detection spectrum channel feature.

[0056] Specifically, the magnetic field detection spectrum feature map is first vectorized and expanded along the channel dimension to obtain a sequence of magnetic field detection spectrum channel feature vectors, and the characteristic energy distribution spectrum vector is calculated for each magnetic field detection spectrum channel feature vector in the sequence. This step maps each magnetic field detection spectrum channel feature vector to an energy distribution spectrum space, in which the distribution of the magnetic field detection spectrum channel features at different frequencies or energy levels can be better captured. Then, the position mean vector of the sequence of magnetic field detection spectrum channel feature energy distribution spectrum vectors is calculated, that is, the so-called magnetic field detection spectrum channel energy spectrum distribution center vector, which represents the center position of the entire magnetic field detection spectrum channel feature in the energy distribution spectrum space. The center vector can be used as a reference point to evaluate the proximity of other magnetic field detection spectrum channel feature energy distribution spectrum vectors to the overall feature distribution of the magnetic field detection spectrum channel of the reference point, so as to help determine which magnetic field detection spectrum channel feature vectors are more representative. Further, the difference between the magnetic field detection spectrum channel energy spectrum distribution center vector and each magnetic field detection spectrum channel feature energy distribution spectrum vector is calculated to obtain the energy distribution spectrum span factor. The span factor quantifies the distance between each feature vector and the center vector, reflecting the degree of discreteness of each magnetic field detection spectral channel feature. By comparing the span factor with the preset threshold, it is possible to determine which magnetic field detection spectral channel feature vectors are within a certain range of distance from the center vector, so as to select those feature vectors that are closer to the center and more representative to form a sequence. In this way, feature vector screening is achieved by selecting feature vectors that are within a certain range of semantic distance from the center of the magnetic field detection spectral channel feature, which helps to remove noise and outliers and retain the most representative and informative features.

[0057] In one embodiment of the present application, the magnetic field detection spectrum feature selection optimization unit includes: a vector expansion subunit, which is used to vectorize and expand the magnetic field detection spectrum feature map along the channel dimension to obtain a sequence of magnetic field detection spectrum channel feature vectors; a characteristic energy distribution spectrum vector calculation subunit, which is used to calculate the characteristic energy distribution spectrum vector of each magnetic field detection spectrum channel feature vector in the sequence of magnetic field detection spectrum channel feature vectors to obtain a sequence of magnetic field detection spectrum channel characteristic energy distribution spectrum vectors; a position mean vector calculation subunit, which is used to calculate the position mean vector of the sequence of magnetic field detection spectrum channel characteristic energy distribution spectrum vectors as the center vector of the magnetic field detection spectrum channel energy spectrum distribution; an energy distribution spectrum span factor calculation subunit, which is used to calculate the energy spectrum distribution of the magnetic field detection spectrum channel. The energy distribution spectrum span factor between the center vector and each magnetic field detection spectrum channel characteristic energy distribution spectrum vector in the sequence of the magnetic field detection spectrum channel characteristic energy distribution spectrum vector is used to obtain the sequence of magnetic field detection spectrum channel energy distribution spectrum span factors; a spectrum span factor and preset threshold comparison subunit is used to determine the sequence of selected optimized magnetic field detection spectrum channel characteristic vectors based on the comparison between each magnetic field detection spectrum channel energy distribution spectrum span factor in the sequence of the magnetic field detection spectrum channel energy distribution spectrum span factor and the preset threshold; a dimension reconstruction subunit is used to reconstruct the dimension of the selected optimized magnetic field detection spectrum channel characteristic vector sequence to obtain the magnetic field detection optimized spectrum feature map.

[0058] Further, in one embodiment of the present application, the characteristic energy distribution spectrum vector calculation subunit includes: a characteristic distribution energy synergistic representation vector calculation secondary subunit, which is used to respectively calculate the characteristic distribution energy synergistic representation vector between each magnetic field detection spectrum channel characteristic vector and each other magnetic field detection spectrum channel characteristic vector in the sequence of the magnetic field detection spectrum channel characteristic vector to obtain a sequence of multiple magnetic field detection spectrum channel characteristic distribution energy synergistic representation vectors; a characteristic distribution energy synergistic factor vector calculation secondary subunit, which is used to calculate the characteristic distribution energy synergistic factor vector of the sequence of each magnetic field detection spectrum channel characteristic distribution energy synergistic representation vector in the sequence of the multiple magnetic field detection spectrum channel characteristic distribution energy synergistic representation vectors to obtain the sequence of the magnetic field detection spectrum channel characteristic energy distribution spectrum vectors; wherein the characteristic distribution energy synergistic representation vector calculation secondary subunit is used to: extract a predetermined magnetic field detection spectrum channel characteristic vector from the sequence of characteristic energy distribution spectrum vectors of the magnetic field detection spectrum channel characteristic vector; and perform element-by-element maximum value extraction on the predetermined magnetic field detection spectrum channel characteristic vector and each other magnetic field detection spectrum channel characteristic vector to obtain a sequence of magnetic field detection spectrum channel characteristic distribution energy synergistic representation vectors corresponding to the predetermined magnetic field detection spectrum channel characteristic vector.

[0059] Furthermore, in one embodiment of the present application, the characteristic energy distribution spectrum vector calculation subunit is used to: calculate the characteristic distribution energy synergy factor vector of each magnetic field detection spectrum channel characteristic distribution energy synergy representation vector in the sequence of the multiple magnetic field detection spectrum channel characteristic distribution energy synergy representation vectors to obtain the sequence of the magnetic field detection spectrum channel characteristic energy distribution spectrum vectors, including: respectively extracting the maximum value of each magnetic field detection spectrum channel characteristic distribution energy synergy representation vector in the sequence of the magnetic field detection spectrum channel characteristic distribution energy synergy representation vector to obtain a sequence of magnetic field detection spectrum channel characteristic maximum values; respectively calculating the mean and variance of each magnetic field detection spectrum channel characteristic distribution energy synergy representation vector to obtain a sequence of magnetic field detection spectrum channel characteristic means and a sequence of magnetic field detection spectrum channel characteristic variances; adding the sequence of magnetic field detection spectrum channel characteristic variances and preset hyperparameters by position to obtain a sequence of magnetic field detection spectrum channel characteristic first energy distribution coefficients; calculating the magnetic field detection The sequence of magnetic field detection spectral channel feature maximum values ​​and the sequence of magnetic field detection spectral channel feature mean values ​​corresponding to each group are squared to obtain a sequence of magnetic field detection spectral channel difference values; the sequence of magnetic field detection spectral channel feature variances is positionally multiplied by constant two to obtain a sequence of modulated magnetic field detection spectral channel feature variances; after adding the magnetic field detection spectral channel difference values ​​and the modulated magnetic field detection spectral channel feature variances corresponding to each group in the sequence of magnetic field detection spectral channel difference values ​​and the sequence of modulated magnetic field detection spectral channel feature variances, the sum is added to the preset hyperparameter to obtain a sequence of magnetic field detection spectral channel feature second energy distribution coefficients; the sequence of magnetic field detection spectral channel feature first energy distribution coefficients is positionally divided by the sequence of magnetic field detection spectral channel feature second energy distribution coefficients to obtain the magnetic field detection spectral channel feature energy distribution spectrum vector composed of multiple feature distribution energy synergy factors.

[0060] Furthermore, in one embodiment of the present application, the energy distribution spectrum span factor calculation subunit is used to: calculate the positional difference between the characteristic energy distribution spectrum vector of the magnetic field detection spectrum channel and the center vector of the energy spectrum distribution of the magnetic field detection spectrum channel to obtain the magnetic field detection spectrum channel differential characteristic vector; calculate the covariance matrix between the center vector of the energy spectrum distribution of the magnetic field detection spectrum channel and the characteristic energy distribution spectrum vector of the magnetic field detection spectrum channel, and calculate the inverse of the covariance matrix to obtain the magnetic field detection spectrum channel inverse covariance matrix; calculate the transposed vector of the magnetic field detection spectrum channel differential characteristic vector and the product of the inverse covariance matrix of the magnetic field detection spectrum channel and the differential characteristic vector of the magnetic field detection spectrum channel to obtain the magnetic field detection spectrum channel distribution value; calculate the square root of the magnetic field detection spectrum channel distribution value to obtain the magnetic field detection spectrum channel energy distribution spectrum span factor.

[0061] Specifically, the sequence of the magnetic field detection spectral channel feature vectors is processed by the feature selection optimization module based on the characteristic energy distribution spectrum according to the following feature selection optimization formula to obtain the sequence of the magnetic field detection spectral channel feature vectors after selection optimization;

[0062] Among them, the feature selection optimization formula is:

[0063]

[0064]

[0065]

[0066]

[0067]

[0068]

[0069]

[0070] in, and Respectively represent the first and The magnetic field detection spectral channel feature vector, It means taking the maximum value between two vectors element by element. For the and The magnetic field detection spectrum channel feature distribution energy synergistic representation vector between the magnetic field detection spectrum channel feature vectors, To extract the maximum value of a vector, and They are The mean and variance of To preset hyperparameters, is the first in the sequence of characteristic energy distribution spectrum vectors of the magnetic field detection spectral channel The characteristic energy distribution spectrum vector of the magnetic field detection spectral channel The characteristic distribution energy synergy factor, For the The characteristic energy distribution spectrum vector of the magnetic field detection spectral channel The characteristic distribution energy synergy factor, A sequence representing the characteristic energy distribution spectrum vector of the magnetic field detection spectral channel, is the first in the sequence of characteristic energy distribution spectrum vectors of the magnetic field detection spectral channel The characteristic energy distribution spectrum vector of the magnetic field detection spectral channel, is the number of vectors in the sequence of characteristic energy distribution spectrum vectors of the magnetic field detection spectral channel, is the center vector of the energy spectrum distribution of the magnetic field detection spectral channel, is the transpose of a vector, It is The covariance matrix between the characteristic energy distribution spectrum vector of the magnetic field detection spectrum channel and the center vector of the energy spectrum distribution of the magnetic field detection spectrum channel, For the The energy distribution spectrum span factor of the magnetic field detection spectrum channel corresponding to the characteristic energy distribution spectrum vector of the magnetic field detection spectrum channel, is the preset threshold, is the first in the sequence of the selected optimized magnetic field detection spectral channel feature vector The selected optimized magnetic field detection spectral channel feature vector, is the sequence of the selected and optimized magnetic field detection spectral channel feature vectors.

[0071] It should be understood that the magnetic field detection optimization spectrum feature map and the external environment temperature time series associated implicit feature vector respectively contain the semantic feature information of the magnetic field detection spectrum after feature optimization selection and the implicit feature information of the external environment temperature time series associated, and these two types of information belong to different modal features. In order to be able to use the semantics of the time series change of the ambient temperature to compensate for the temperature drift of the magnetic field spectrum feature, it is necessary to effectively fuse the feature information of these two different modalities. Based on this, in the technical solution of the present application, the magnetic field detection optimization spectrum feature map and the external environment temperature time series associated implicit feature vector are input into a gated cross-modal joint encoder based on an imitation converter structure to obtain a temperature drift compensated magnetic field detection optimization spectrum feature map. The gated cross-modal joint encoder based on an imitation converter structure is a multimodal data processing technology, which achieves the effective fusion of different modal features by imitating the self-attention mechanism of the Transformer structure and combining the gating unit.

[0072] Specifically, firstly, the magnetic field detection optimization spectrum feature map is subjected to point convolution encoding and spatial convolution encoding to extract deep spatial features and generate the magnetic field detection optimization spectrum spatial deepening feature map. This process enhances the spatial expression ability of the magnetic field detection optimization spectrum feature map, enabling it to capture the detailed information in the magnetic field detection more accurately. Then, the magnetic field detection optimization spectrum spatial deepening feature map is copied and reshaped to obtain the imitation query and imitation key feature matrices. By calculating the product of the imitation query feature matrix and the transpose of the imitation key feature matrix, the magnetic field detection optimization spectrum local autocorrelation library matrix is ​​constructed. This step realizes the correlation between the internal features of the modality. In addition, the autocorrelation weight of the magnetic field detection optimization spectrum local autocorrelation library matrix is ​​dynamically adjusted by using the attention gating unit based on the GELU function to realize the adaptive selection and enhancement of features. In this way, those feature parts that are more important for magnetic field detection can be highlighted, while irrelevant or noise features can be suppressed. Through the gating mechanism, combined with the implicit feature vector of the time series correlation of the external environment temperature, the cross-modal interactive fusion condition weight vector of the external environment temperature is calculated. This weight vector reflects the impact of temperature change on the magnetic field detection features, so that the magnetic field detection features can be conditionally adjusted during the fusion process. Then, the cross-modal interactive fusion condition weight vector of the external ambient temperature is used to weight the channel-by-channel optimized spectral feature map of magnetic field detection to generate a multimodal fusion feature map. This process enables the magnetic field detection feature map to take into account the impact of temperature changes and compensate for it accordingly. The whole process not only deepens the understanding of a single mode (magnetic field detection optimized spectral feature map), but also integrates the information of another mode (external ambient temperature time series associated implicit feature vector), enhances the model's ability to process multimodal data, and improves the accuracy and robustness of temperature drift compensation for magnetic field detection data. In addition, the consistency and accuracy of magnetic field detection results at different ambient temperatures are ensured through a cross-modal fusion mechanism.

[0073] In one embodiment of the present application, the temperature drift compensation magnetic field detection spectrum semantic expression module includes: a layer normalization and convolution coding processing unit, which is used to perform layer normalization and convolution coding processing on the magnetic field detection optimization spectrum feature map to obtain a magnetic field detection optimization spectrum space deepening feature map; a space deepening feature map copying unit, which is used to copy the magnetic field detection optimization spectrum space deepening feature map to obtain a backup magnetic field detection optimization spectrum space deepening feature map; a feature shape reshaping unit, which is used to reshape the backup magnetic field detection optimization spectrum space deepening feature map and the magnetic field detection optimization spectrum space deepening feature map respectively to obtain a backup magnetic field detection optimization spectrum space deepening feature matrix as an imitation query feature matrix and a magnetic field detection optimization spectrum space deepening feature matrix as an imitation key feature matrix; a transpose matrix calculation unit, which is used to calculate the product between the imitation query feature matrix and the transpose matrix of the imitation key feature matrix to obtain a magnetic field detection optimization spectrum local autocorrelation library matrix; an attention gating unit, which is used to convert the magnetic field detection optimization spectrum local autocorrelation library matrix into an attention gating unit. The matrix inputs an attention gating unit based on the GELU function to obtain a magnetic field detection optimization spectrum gated local autocorrelation library weight matrix; a vector product calculation unit is used to use the external environment temperature time series associated implicit feature vector as a value feature vector, and calculate the product between the magnetic field detection optimization spectrum gated local autocorrelation library weight matrix and the external environment temperature time series associated implicit feature vector to obtain the external environment temperature cross-modal interaction fusion condition weight vector; a channel-by-channel weighted calculation unit is used to use the external environment temperature cross-modal interaction fusion condition weight vector as a weight vector, and perform channel-by-channel weighting on the magnetic field detection optimization spectrum feature map to obtain the temperature drift compensation magnetic field detection optimization spectrum feature map; wherein the layer normalization and convolution coding processing unit is used to: perform layer normalization processing on the magnetic field detection optimization spectrum feature map to obtain a layer normalized magnetic field detection optimization spectrum feature map; perform point convolution coding on the layer normalized magnetic field detection optimization spectrum feature map and then perform spatial convolution coding on the layer normalized magnetic field detection optimization spectrum feature map to obtain the magnetic field detection optimization spectrum spatial deepening feature map.

[0074] Specifically, the magnetic field detection optimization spectrum feature graph and the external environment temperature time series associated implicit feature vector are input into the gated cross-modal joint encoder based on the simulated converter structure and processed with the following multimodal fusion formula to obtain the temperature drift compensation magnetic field detection optimization spectrum feature graph;

[0075] The cross-domain joint coding formula is:

[0076]

[0077] in, optimizing a spectral signature for said magnetic field detection, is the layer normalization operation, is point convolutional coding, The convolution kernel is The convolution operation, optimizing the spectral spatial deepening feature map for said magnetic field detection, For copy operation, Optimizing the spectral spatial deepening feature map for the backup magnetic field detection, For the reshape operation, is the transpose of the feature matrix, is matrix multiplication, yes function, is the magnetic field detection optimized spectral gating local autocorrelation library weight matrix, is the implicit feature vector associated with the external environment temperature time series, is the weight vector of the cross-modal interaction fusion condition of the external environment temperature, This is the temperature drift compensation magnetic field detection optimization spectrum characteristic diagram.

[0078] Then, the temperature drift compensation magnetic field detection optimization spectrum characteristic diagram is input into the magnetic field strength detector based on the decoder to obtain the detection result, and the detection result is used to represent the magnetic field strength decoding value. In other words, the magnetic field detection optimization spectrum semantic expression characteristics after temperature drift supplementation are used to perform decoding regression, so as to generate the magnetic field strength decoding value after temperature drift compensation, and realize temperature drift compensation correction. In this way, it can help eliminate the measurement error caused by temperature change, thereby improving the overall accuracy and reliability of the magnetic field vector sensor. In addition, in other examples of the present application, it is also possible to consider taking further measures in material selection and structural design, such as using temperature compensation optical fiber materials or optimizing the formula of magnetic fluid, so as to further reduce the influence of temperature on sensor performance.

[0079] In one embodiment of the present application, the magnetic field strength detection module is used to: input the temperature drift compensation magnetic field detection optimization spectrum characteristic diagram into a decoder-based magnetic field strength detector to obtain the detection result.

[0080] Preferably, considering that the magnetic field detection optimized spectral feature graph and the external environment temperature time series associated implicit feature vector respectively represent the spectral image semantic enhancement features and the external environment temperature local time series associated features of the magnetic field detection spectrum graph obtained by void convolution coding and feature selection optimization, when the magnetic field detection optimized spectral feature graph and the external environment temperature time series associated implicit feature vector are gated cross-modal joint encoding based on the converter-like structure, the multimodal semantic fusion adaptive offset caused by the feature channel dimension alignment between the magnetic field detection optimized spectral feature graph and the external environment temperature time series associated implicit feature vector will be expected to further improve the multimodal semantic fusion adaptive regression comprehensibility of the temperature drift compensated magnetic field detection optimized spectral feature graph, thereby improving the accuracy of the detection results obtained by the decoder-based magnetic field strength detector.

[0081] Therefore, in a preferred embodiment of the present application, when the temperature drift compensation magnetic field detection optimization spectral characteristic diagram is passed through a decoder-based magnetic field strength detector to obtain a detection result, the temperature drift compensation magnetic field detection optimization spectral characteristic diagram is optimized, including the steps of:

[0082] Expanding the temperature drift compensation magnetic field detection optimization spectrum characteristic graph into a temperature drift compensation magnetic field detection optimization spectrum characteristic vector;

[0083] Subtract the zero norm of the temperature drift compensation magnetic field detection optimization spectrum feature vector from the length of the temperature drift compensation magnetic field detection optimization spectrum feature vector to obtain the temperature drift compensation magnetic field detection optimization spectrum isolated representation value;

[0084] Calculate the logarithm of the sum of the square of the temperature drift compensation magnetic field detection optimization spectrum isolated representation value and the temperature drift compensation magnetic field detection optimization spectrum isolated representation value with base 2 to obtain the temperature drift compensation magnetic field detection optimization spectrum information order value;

[0085] Calculate a power function with each eigenvalue of the temperature drift compensation magnetic field detection optimization spectrum feature vector as the base, the difference of the temperature drift compensation magnetic field detection optimization spectrum isolated representation value minus one as the exponent, and perform a point multiplication with the temperature drift compensation magnetic field detection optimization spectrum information order value to obtain the temperature drift compensation magnetic field detection optimization spectrum leading vector;

[0086] The temperature drift compensation magnetic field detection optimization spectrum feature vector is multiplied by the difference between the temperature drift compensation magnetic field detection optimization spectrum isolated representation value minus one, and then multiplied by the reciprocal of the temperature drift compensation magnetic field detection optimization spectrum isolated representation value to obtain the temperature drift compensation magnetic field detection optimization spectrum field constraint vector;

[0087] Calculating an exponential function with a natural constant as the base and each eigenvalue of the temperature drift compensation magnetic field detection optimization spectral field constraint vector as the exponent to obtain a temperature drift compensation magnetic field detection optimization spectral field bias vector; and

[0088] The temperature drift compensation magnetic field detection optimization spectrum pilot vector is dot-added with the temperature drift compensation magnetic field detection optimization spectrum field bias vector to obtain an optimized temperature drift compensation magnetic field detection optimization spectrum feature vector.

[0089] The temperature drift compensation magnetic field detection optimization spectral feature vector is, for example, recorded as The optimization process is expressed as:

[0090]

[0091] in, is the temperature drift compensation magnetic field detection optimization spectral feature vector, is the zero norm of the vector, is the length of the temperature drift compensation magnetic field detection optimization spectrum feature vector, is the value of the temperature drift compensation magnetic field detection optimization spectrum isolation representation, The power function is calculated with each eigenvalue of the temperature drift compensation magnetic field detection optimization spectrum feature vector as the base and the difference between the temperature drift compensation magnetic field detection optimization spectrum isolated representation value minus one as the exponent, is the logarithm with base 2, It is an exponential function that calculates the exponential function of each eigenvalue of a vector with a natural constant as the base. is the optimized temperature drift compensation magnetic field detection optimized spectral feature vector, It is the point multiplication by position. It is added by location point. It is the reciprocal of the temperature drift compensation magnetic field detection optimized spectral isolated representation value.

[0092] Therefore, the high-dimensional feature manifold of the temperature drift compensation magnetic field detection optimization spectrum feature vector is represented by the feature value of the feature set as the vector field of the aggregation dimension, and the superposition value of the vector field of the temperature drift compensation magnetic field detection optimization spectrum feature vector at the isolated zero position is used as the order information to fix the local position of the feature value of the feature set, and add the bias of the reversibility of the feature regression distribution field of the temperature drift compensation magnetic field detection optimization spectrum feature vector as a reward to achieve the mapping target tracking of the regression distribution of the temperature drift compensation magnetic field detection optimization spectrum feature vector for the feature value position, so that the feature set of the temperature drift compensation magnetic field detection optimization spectrum feature vector perceives the mapping migration to the aggregate distribution, thereby improving the accuracy of the detection result obtained by the temperature drift compensation magnetic field detection optimization spectrum feature vector through the decoder-based magnetic field strength detector by improving the multimodal semantic fusion adaptive regression understandability of the temperature drift compensation magnetic field detection optimization spectrum feature vector. In this way, the magnetic field intensity decoding value after temperature drift compensation can be generated more accurately, thereby realizing temperature drift compensation correction.

[0093] In summary, the above scheme is adopted, when using a temperature drift compensation processor to correct the influence of temperature change on the measurement result, in the temperature drift compensation processor, by obtaining the magnetic field detection spectrum generated by the spectrometer, and using the temperature sensor to monitor the collected external environment temperature data in real time, and then introducing the image processing and data analysis algorithm based on artificial intelligence and deep learning at the back end to analyze the magnetic field detection spectrum and capture the time series characteristics of the external environment temperature data, so as to extract and characterize the cross-modal joint feature representation between the magnetic field detection spectrum characteristics and the external environment temperature time series characteristics, and generate the magnetic field intensity decoding value after temperature drift compensation through the magnetic field strength detector, so as to realize temperature drift compensation correction. In this way, it can help eliminate the measurement error caused by temperature change, thereby improving the overall accuracy and reliability of the magnetic field vector sensor.

[0094] Figure 3 is a flow chart of a magnetic field vector sensing method based on large-angle tilted fiber Bragg grating according to an exemplary embodiment. Figure 3 As shown, the method includes:

[0095] Step S201, obtaining a magnetic field detection spectrum generated by the spectrometer;

[0096] Step S202, obtaining a time queue of the external environment temperature collected by the temperature sensor;

[0097] Step S203, performing sequence encoding on the time queue of the external environment temperature to obtain an external environment temperature time series associated implicit feature vector;

[0098] Step S204, performing spectral feature extraction and feature selection optimization processing on the magnetic field detection spectrum to obtain a magnetic field detection optimized spectral feature graph;

[0099] Step S205, inputting the magnetic field detection optimization spectrum characteristic diagram and the external environment temperature time series associated implicit feature vector into a gated cross-modal joint encoder based on an analog converter structure to obtain a temperature drift compensated magnetic field detection optimization spectrum characteristic diagram;

[0100] Step S206: determining a detection result based on the temperature drift compensation magnetic field detection optimization spectrum characteristic diagram, wherein the detection result is used to represent a magnetic field intensity decoding value.

[0101] Reference below Figure 4 , which shows a schematic diagram of the structure of an electronic device 600 suitable for implementing the embodiment of the present application. The terminal device in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 4 The electronic device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0102] like Figure 4 As shown, the electronic device 600 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage device 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the electronic device 600 are also stored. The processing device 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0103] Typically, the following devices may be connected to the I / O interface 605: an input device 606 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 608 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 609. The communication device 609 may allow the electronic device 600 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 4 The electronic device 600 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0104] In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication device 609, or installed from the storage device 608, or installed from the ROM 602. When the computer program is executed by the processing device 601, the above-mentioned functions defined in the method of the embodiment of the present application are executed.

[0105] It should be noted that the computer-readable medium mentioned above in the present application may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0106] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (HyperText Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0107] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0108] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including, but not limited to, object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0109] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0110] The modules involved in the embodiments described in this application can be implemented by software or hardware. The name of the module does not limit the module itself in some cases. For example, the test parameter acquisition module can also be described as a "module for acquiring device test parameters corresponding to the target device".

[0111] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0112] In the context of the present application, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0113] Figure 5 FIG. 1 is an application scenario diagram of a magnetic field vector sensor based on a large-angle tilted fiber Bragg grating according to an exemplary embodiment. Figure 5 As shown, in this application scenario, first, a magnetic field detection spectrum generated by the spectrometer is obtained (for example, Figure 5 C1 shown in FIG); obtain the time queue of the external environment temperature collected by the temperature sensor (for example, Figure 5 Then, the obtained magnetic field detection spectrum and the time queue of the external environment temperature are input to a server (for example, Figure 5 In S) shown in the figure, the server can process the magnetic field detection spectrum and the time queue of the external environment temperature based on the magnetic field vector sensor algorithm of the large-angle tilted fiber Bragg grating to determine the detection result, and the detection result is used to represent the magnetic field intensity decoding value.

[0114] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in this application (but not limited to) by each other to form a technical solution.

[0115] In addition, although each operation is described in a specific order, this should not be construed as requiring these operations to be performed in the specific order shown or to be performed in a sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although some specific implementation details are included in the above discussion, these should not be construed as limiting the scope of the application. Some features described in the context of a separate embodiment can also be implemented in a single embodiment in combination. On the contrary, the various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination mode.

[0116] Although the subject matter has been described in language specific to structural features and / or method logic actions, it should be understood that the subject matter defined is not necessarily limited to the specific features or actions described above. On the contrary, the specific features and actions described above are merely exemplary forms of implementation. With respect to the apparatus in the above-mentioned embodiments, the specific manner in which each module performs the operation has been described in detail in the embodiments related to the method, and will not be described in detail here.

Claims

1. A magnetic field vector sensor based on large-angle tilted fiber Bragg grating, comprising: A broadband light source, an optical fiber circulator, an optical fiber rotary connector, a sensor probe and a spectrometer, wherein the broadband light source is used to transmit a wide-spectrum optical signal; The fiber optic circulator is used to guide the wide-spectrum optical signal to advance along a predetermined path, and the fiber optic circulator is connected to the fiber optic rotary connector and the spectrometer by means of fiber optic fusion; the fiber optic rotary connector is used to control the rotation of the sensing probe; the sensing probe is used to receive and transmit reflected light signals, and the sensing probe includes a 45° tilted fiber grating, a large-angle tilted fiber grating, a chirped fiber grating and a glass capillary filled with magnetic fluid, wherein the tilt angle of the large-angle tilted fiber grating is 69-89°; the spectrometer is used to perform spectral analysis on the reflected light signal to generate a magnetic field detection spectrum diagram; it is characterized in that the magnetic field vector sensor based on the large-angle tilted fiber grating also includes a temperature drift compensation processor, which is used to perform temperature drift compensation correction on the detected magnetic field intensity based on the magnetic field detection spectrum diagram; Wherein, the temperature drift compensation processor comprises: A magnetic field detection spectrum acquisition module, used to obtain a magnetic field detection spectrum generated by the spectrometer; The external environment temperature acquisition module is used to obtain the time queue of the external environment temperature collected by the temperature sensor; An external environment temperature time series encoding module is used to perform sequence encoding on the time queue of the external environment temperature to obtain an external environment temperature time series associated implicit feature vector; A magnetic field detection spectrum semantic feature extraction and optimization module is used to perform spectrum feature extraction and feature selection optimization processing on the magnetic field detection spectrum to obtain a magnetic field detection optimized spectrum feature map; A temperature drift compensation magnetic field detection spectrum semantic expression module is used to input the magnetic field detection optimization spectrum feature map and the external environment temperature time series associated implicit feature vector into a gated cross-modal joint encoder based on an analog converter structure to obtain a temperature drift compensation magnetic field detection optimization spectrum feature map; The magnetic field strength detection module is used to determine the detection result based on the temperature drift compensation magnetic field detection optimization spectrum characteristic diagram, and the detection result is used to represent the magnetic field strength decoding value.

2. The magnetic field vector sensor based on large-angle tilted fiber Bragg grating according to claim 1, characterized in that: The external environment temperature time series encoding module includes: inputting the time queue of the external environment temperature into a sequence encoder based on a Bi-LSTM model to obtain an implicit feature vector associated with the external environment temperature time series.

3. The magnetic field vector sensor based on large-angle tilted fiber Bragg grating according to claim 2, characterized in that: The magnetic field detection spectrum map semantic feature extraction optimization module includes: A magnetic field detection spectrum feature extraction unit, used for inputting the magnetic field detection spectrum graph into a spectrum feature extractor based on a hole convolutional neural network model to obtain a magnetic field detection spectrum feature graph; The magnetic field detection spectrum feature selection and optimization unit is used to input the magnetic field detection spectrum feature map into the feature selection optimization module based on the characteristic energy distribution spectrum to obtain the magnetic field detection optimized spectrum feature map.

4. The magnetic field vector sensor based on large-angle tilted fiber Bragg grating according to claim 3, characterized in that: The magnetic field detection spectrum feature selection and optimization unit comprises: A vectorized expansion subunit, used for vectorizing and expanding the magnetic field detection spectrum characteristic graph along the channel dimension to obtain a sequence of magnetic field detection spectrum channel characteristic vectors; A characteristic energy distribution spectrum vector calculation subunit is used to calculate the characteristic energy distribution spectrum vector of each magnetic field detection spectrum channel characteristic vector in the sequence of magnetic field detection spectrum channel characteristic vectors to obtain a sequence of magnetic field detection spectrum channel characteristic energy distribution spectrum vectors; A position mean vector calculation subunit is used to calculate the position mean vector of the sequence of characteristic energy distribution spectrum vectors of the magnetic field detection spectrum channel as the center vector of the energy spectrum distribution of the magnetic field detection spectrum channel; An energy distribution spectrum span factor calculation subunit is used to calculate the energy distribution spectrum span factor between the energy spectrum distribution center vector of the magnetic field detection spectrum channel and each magnetic field detection spectrum channel characteristic energy distribution spectrum vector in the sequence of the magnetic field detection spectrum channel characteristic energy distribution spectrum vector to obtain a sequence of magnetic field detection spectrum channel energy distribution spectrum span factors; A spectrum span factor and preset threshold comparison subunit, used to determine and select a sequence of optimized magnetic field detection spectrum channel feature vectors based on a comparison between each magnetic field detection spectrum channel energy distribution spectrum span factor in the sequence of magnetic field detection spectrum channel energy distribution spectrum span factors and a preset threshold; The dimension reconstruction subunit is used to reconstruct the dimension of the sequence of the selected optimized magnetic field detection spectrum channel feature vectors to obtain the magnetic field detection optimized spectrum feature map.

5. The magnetic field vector sensor based on large-angle tilted fiber Bragg grating according to claim 4, characterized in that: The characteristic energy distribution spectrum vector calculation subunit includes: The characteristic distribution energy collaborative representation vector calculation secondary subunit is used to respectively calculate the characteristic distribution energy collaborative representation vector between each magnetic field detection spectrum channel characteristic vector and each other magnetic field detection spectrum channel characteristic vector in the sequence of the magnetic field detection spectrum channel characteristic vector to obtain a sequence of multiple magnetic field detection spectrum channel characteristic distribution energy collaborative representation vectors; The characteristic distribution energy synergy factor vector calculation secondary subunit is used to calculate the characteristic distribution energy synergy factor vector of the sequence of characteristic distribution energy synergy representation vectors of each magnetic field detection spectrum channel in the sequence of the plurality of magnetic field detection spectrum channel characteristic distribution energy synergy representation vectors to obtain the sequence of characteristic energy distribution spectrum vectors of the magnetic field detection spectrum channel; The feature distribution energy collaborative representation vector calculation secondary subunit is used to: Extracting a predetermined magnetic field detection spectrum channel feature vector from a sequence of feature energy distribution spectrum vectors of the magnetic field detection spectrum channel feature vector; The predetermined magnetic field detection spectrum channel feature vector and each other magnetic field detection spectrum channel feature vector are respectively subjected to element-by-element maximum value extraction to obtain a sequence of magnetic field detection spectrum channel feature distribution energy synergistic representation vectors corresponding to the predetermined magnetic field detection spectrum channel feature vector.

6. The magnetic field vector sensor based on large-angle tilted fiber Bragg grating according to claim 5, characterized in that: The characteristic energy distribution spectrum vector calculation subunit is used for: Calculating the characteristic distribution energy synergy factor vector of each sequence of the characteristic distribution energy synergy representation vector of the magnetic field detection spectrum channel in the sequence of the plurality of magnetic field detection spectrum channel characteristic distribution energy synergy representation vectors to obtain the sequence of the magnetic field detection spectrum channel characteristic energy distribution spectrum vectors, comprising: Respectively extracting the maximum value of each magnetic field detection spectrum channel characteristic distribution energy collaborative representation vector in the sequence of magnetic field detection spectrum channel characteristic distribution energy collaborative representation vectors to obtain a sequence of magnetic field detection spectrum channel characteristic maximum values; The mean and variance of the energy cooperative representation vector of each magnetic field detection spectrum channel characteristic distribution are calculated respectively to obtain a sequence of magnetic field detection spectrum channel characteristic means and a sequence of magnetic field detection spectrum channel characteristic variances; Adding the sequence of the magnetic field detection spectral channel characteristic variances and the preset hyperparameters by position to obtain a sequence of first energy distribution coefficients of the magnetic field detection spectral channel characteristics; Calculate the square of the difference between the maximum value of the magnetic field detection spectrum channel characteristic and the mean value of the magnetic field detection spectrum channel characteristic in each group corresponding to the sequence of the maximum value of the magnetic field detection spectrum channel characteristic and the sequence of the mean value of the magnetic field detection spectrum channel characteristic to obtain a sequence of magnetic field detection spectrum channel difference values; Multiplying the sequence of characteristic variances of the magnetic field detection spectrum channel by constant 2 by position to obtain a sequence of characteristic variances of the modulated magnetic field detection spectrum channel; After adding each group of corresponding magnetic field detection spectral channel difference values ​​and modulated magnetic field detection spectral channel characteristic variances in the sequence of magnetic field detection spectral channel difference values ​​and the sequence of modulated magnetic field detection spectral channel characteristic variances, the sum is added to the preset hyperparameter to obtain a sequence of second energy distribution coefficients of magnetic field detection spectral channel characteristics; The sequence of the first energy distribution coefficients of the magnetic field detection spectrum channel characteristics is divided by the sequence of the second energy distribution coefficients of the magnetic field detection spectrum channel characteristics by position to obtain the magnetic field detection spectrum channel characteristic energy distribution spectrum vector composed of multiple characteristic distribution energy synergy factors.

7. The magnetic field vector sensor based on large-angle tilted fiber Bragg grating according to claim 6, characterized in that: The energy distribution spectrum span factor calculation subunit is used to: Calculating the positional difference between the characteristic energy distribution spectrum vector of the magnetic field detection spectrum channel and the center vector of the energy spectrum distribution of the magnetic field detection spectrum channel to obtain a differential characteristic vector of the magnetic field detection spectrum channel; Calculating the covariance matrix between the energy spectrum distribution center vector of the magnetic field detection spectrum channel and the characteristic energy distribution spectrum vector of the magnetic field detection spectrum channel, and calculating the inverse of the covariance matrix to obtain the inverse covariance matrix of the magnetic field detection spectrum channel; Calculating the transposed vector of the magnetic field detection spectrum channel differential eigenvector and the product of the magnetic field detection spectrum channel inverse covariance matrix and the magnetic field detection spectrum channel differential eigenvector to obtain a magnetic field detection spectrum channel distribution value; The square root of the magnetic field detection spectrum channel distribution value is calculated to obtain the magnetic field detection spectrum channel energy distribution spectrum span factor.

8. The magnetic field vector sensor based on large-angle tilted fiber Bragg grating according to claim 7, characterized in that: The temperature drift compensation magnetic field detection spectrum semantic expression module includes: A layer normalization and convolution coding processing unit, used for performing layer normalization and convolution coding processing on the magnetic field detection optimization spectrum feature map to obtain a magnetic field detection optimization spectrum space deepening feature map; A space deepening feature map duplication unit, used for duplicating the magnetic field detection optimization spectrum space deepening feature map to obtain a backup magnetic field detection optimization spectrum space deepening feature map; A feature shape reshaping unit is used to reshape the backup magnetic field detection optimization spectrum space deepening feature map and the magnetic field detection optimization spectrum space deepening feature map to obtain a backup magnetic field detection optimization spectrum space deepening feature matrix as a simulated query feature matrix and a magnetic field detection optimization spectrum space deepening feature matrix as a simulated key feature matrix; A transposed matrix calculation unit, used for calculating the product between the transposed matrix of the simulated query feature matrix and the simulated key feature matrix to obtain a local autocorrelation library matrix of the magnetic field detection optimization spectrum; An attention gating unit, used for inputting the magnetic field detection optimization spectrum local autocorrelation library matrix into the attention gating unit based on the GELU function to obtain a magnetic field detection optimization spectrum gated local autocorrelation library weight matrix; A vector product calculation unit is used to calculate the product between the magnetic field detection optimization spectrum gating local autocorrelation library weight matrix and the external environment temperature time series associated implicit feature vector using the external environment temperature time series associated implicit feature vector as a value feature vector to obtain an external environment temperature cross-modal interactive fusion condition weight vector; A channel-by-channel weighted calculation unit, used to use the external environment temperature cross-modal interactive fusion condition weight vector as a weight vector, and perform channel-by-channel weighting on the magnetic field detection optimization spectrum feature map to obtain the temperature drift compensation magnetic field detection optimization spectrum feature map; Wherein, the layer normalization and convolutional coding processing unit is used to: Performing layer normalization processing on the magnetic field detection optimization spectrum characteristic graph to obtain a layer normalized magnetic field detection optimization spectrum characteristic graph; The layer normalized magnetic field detection optimization spectrum feature map is point-convolution encoded and then space-convolution encoded to obtain the magnetic field detection optimization spectrum space deepening feature map.

9. The magnetic field vector sensor based on large-angle tilted fiber Bragg grating according to claim 8, characterized in that: The magnetic field strength detection module is used to: input the temperature drift compensation magnetic field detection optimization spectrum characteristic diagram into a decoder-based magnetic field strength detector to obtain the detection result.

Images