A signal processing method in a large plane magnetometry system

Abstract

The present application relates to a kind of signal processing methods in large plane magnetometry system, the magnetic signal of acquisition is preprocessed, and the magnetic signal after pre-processing is input into multistage error model and is processed by multistage error compensation, obtains the magnetic data required by system.The present application is targeted to realize the signal processing of magnetic data in magnetometry system, including signal acquisition, band filtering, linearity compensation, orthogonality compensation, vector axis alignment compensation, gradient error compensation, etc., eliminates band interference, linearity interference, orthogonality interference, vector axis alignment error, gradient error and other interference factors;Realize the whole process of acquisition, processing, compensation, display and storage required by system magnetometry, so that the output performance of magnetometry system can be realized;After completing the above signal processing, the output error of entire magnetometry system is not more than 20nT;In practical application, it is easy to implement.

Description

Technical Field

[0001] This invention relates to the field of measurement and testing, and particularly to a signal processing method in a large-plane magnetic measurement system for underwater magnetic measurement. Background Technology

[0002] The large-plane magnetic field measurement system is mainly used to test the magnetic field distribution characteristics of the target object and provide a compensation basis for demagnetization operations. The system is typically deployed along a defined plane, with probes installed at equal intervals. Before demagnetization, the system acquires the specific magnetic field distribution map of the target object and feeds it back to the operation center to guide the charging parameters of the demagnetization coil, forming a reverse canceling magnetic field that conforms to the magnetic field characteristics, thus eliminating the target object's magnetic field features. The system then acquires the magnetic field distribution map again after demagnetization to verify the demagnetization effect and generates a targeted reverse canceling magnetic field to continue eliminating areas that do not meet the requirements. This process is repeated until the requirements are met.

[0003] The core unit of the magnetic measurement system is the vector fluxgate magnetometer probe, which can acquire axial component magnetic fields in three orthogonal directions. A single vector magnetometer probe (referred to as a probe) can acquire vector magnetic field data at a single point. Multiple probes arranged at equal intervals on the same plane can form a probe matrix, which can acquire a vector magnetic field distribution map on the plane.

[0004] Normally, after the probe acquires magnetic data, it provides it to the host computer for processing and display, thus completing the magnetic measurement process. However, in actual engineering, a large number of probes are arranged on the same plane, each carrying a set of magnetic signals uploaded. In order to obtain accurate and reliable magnetic signals, it is necessary to perform data acquisition, interference processing, fusion, and correction on the uploaded magnetic signals, including:

[0005] 1) The original magnetic signal is an analog quantity and needs to be converted into a recognizable digital quantity;

[0006] 2) The magnetic signal contains unwanted frequency band interference, and invalid signals need to be removed;

[0007] 3) Correction of magnetic field module conversion error in a single axis;

[0008] 4) Correction of orthogonality error in the three axes of the probe;

[0009] 5) Correction of axial (vector axis) alignment error when installing multiple probes;

[0010] 6) Gradient error compensation at different locations in the plane.

[0011] The above-mentioned processing of magnetic signals is necessary for a magnetometry system to perform its normal function, and existing technologies are lacking in this regard. SUMMARY

[0012] The present application solves the problems in the prior art and provides a signal processing method in a large plane magnetic measurement system, which can realize complete processing of magnetic signals in the large plane magnetic measurement system, obtain accurate vector magnetic signals and provide accurate magnetic characteristic information of a magnetic target.

[0013] The technical scheme adopted by the present application is a signal processing method in a large plane magnetic measurement system, which pre-processes the obtained magnetic signals, inputs the pre-processed magnetic signals into a multi-stage error model for multi-stage error compensation processing and obtains the required magnetic data of the system.

[0014] Preferably, the pre-processing processes the obtained magnetic signals with a low-pass filter, extracts low-frequency useful signals and filters out useless signals in other frequency bands; and outputs the low-frequency useful signals after reducing the data frequency.

[0015] Preferably, the multi-stage error model comprises a linearity error model, an orthogonality error model, an alignment error model and a gradient error model connected in sequence.

[0016] The linearity error model completes compensation of linearity error in the analog-to-digital conversion process and eliminates linearity error caused by analog-to-digital conversion.

[0017] The orthogonality error model completes compensation of error caused by non-orthogonality of three axial directions in the probe.

[0018] The alignment error model compensates error caused by non-consistency of axial directions of each vector axis in different probes.

[0019] The gradient error model compensates gradient error caused by different positions of each probe in the geomagnetic field.

[0020] Preferably, the linearity error model is

[0021]

[0022] wherein X yi , Y yi and Z yi are original values of X component, Y component and Z component of the i-th probe;

[0023] X i , Y i and Z i are values of X component, Y component and Z component of the i-th probe after linearity error compensation;

[0024] a3, a2, a1, a0 are linear compensation factors of the X component of the i-th probe respectively, a3, a2, a1 ∈ [-1, 1], a0 ∈ [-500, 500];

[0025] b3, b2, b1, b0 are linear compensation factors of the Y component of the i-th probe respectively, b3, b2, b1 ∈ [-1, 1], b0 ∈ [-500, 500];

[0026] c3, c2, c1, c0 are linear compensation factors of the Z component of the i-th probe respectively, c3, c2, c1 ∈ [-1, 1], c0 ∈ [-500, 500].

[0027] Preferably, the orthogonality error model is,

[0028]

[0029] wherein X i , Y i and Z i are the X component, Y component and Z component values of the i-th probe after linear error compensation;

[0030] X ai , Y ai and Z ai are the X component, Y component and Z component values of the i-th probe after orthogonality error compensation;

[0031] α, β, γ, d x , d y , d z , e x , e y , e z are orthogonality compensation factors, α, β, γ ∈ [-1, 1], d x , d y , d z ∈ [0.9, 1.1], e x , e y , e z ∈ [-500, 500].

[0032] Preferably, the alignment error model is,

[0033]

[0034] wherein, and are the X component, Y component and Z component values of the i-th probe after orthogonality error compensation;

[0035] respectively are the X, Y and Z component values of the i-th probe after alignment error compensation;

[0036] respectively are the X, Y and Z component values of the i-th probe after alignment error compensation;

[0037] Preferably, the gradient error model is,

[0038]

[0039] wherein, respectively are the X, Y and Z component values of the i-th probe after alignment error compensation;

[0040] and respectively are the X, Y and Z component values of the i-th probe after alignment error compensation;

[0041] respectively are the X, Y and Z component values of the i-th probe before gradient error compensation at the initial time;

[0042] respectively are the X, Y and Z component values of the 1st probe before gradient error compensation at the initial time.

[0043] Preferably, the magnetic data required by the system includes magnetic field component data and magnetic field synthesis total field data.

[0044] Preferably, the pre-processed magnetic signal is subjected to a multi-stage error model, and the signal compensation is completed by the compensation factors of each stage. The compensated magnetic data is used for magnetic field component data output, storage, and magnetic field synthesis total field data calculation, respectively.

[0045] Preferably, the stored magnetic data is played back by signal extraction.

[0046] The theoretical basis of the present application is that when the magnetic field voltage signal is collected and converted into a digital signal at a high speed, the signal usually contains magnetic signals of different frequency bands, when the collection rate is set as 1KHz, the signal contains geomagnetic field signals (frequency <0.1Hz), magnetic target signals (frequency <1Hz), power frequency signals (frequency 50Hz) and high frequency interference signals (frequency >100Hz), through frequency band analysis, the frequency bands of the signals are independent of each other, and the useful magnetic field signals (geomagnetic field signals and magnetic target signals) are all in the low frequency band, therefore, the collected magnetic field signal can be substituted into a low-pass filter, the unwanted interference signals are removed, and the low frequency magnetic field is retained, so that the magnetic field signal filtering effect is achieved; the interference quantities contained in the magnetic field signal include linearity error caused by analog-to-digital conversion, orthogonality error caused by the non-orthogonality of the three axes of the probe, alignment error caused by the inconsistency of the axes corresponding to the plurality of probes, and gradient error caused by the different installation positions of the plurality of probes in the plane, the above errors can be obtained by establishing corresponding mathematical models to obtain model factors, and the models are independent of each other, therefore, the interference quantity compensation can be performed, so that the correction of each interference quantity is completed one by one.

[0047] Based on the theoretical basis, the present application provides a signal processing method in a large plane magnetic measurement system, the obtained magnetic signal is preprocessed, the preprocessed magnetic signal is input into a multi-stage error model for multi-stage error compensation processing, and the magnetic data required by the system is obtained.

[0048] The present application has the following beneficial effects:

[0049] (1) The signal processing of the magnetic data in the magnetic measurement system is realized, including signal collection, frequency band filtering, linearity compensation, orthogonality compensation, vector axis alignment compensation, gradient error compensation and the like, and the interference factors such as frequency band interference, linearity interference, orthogonality interference, vector axis alignment error and gradient error are removed;

[0050] (2) The entire process of collection, processing, compensation, display and storage required by the system for magnetic measurement is realized, so that the output performance of the magnetic measurement system is realized;

[0051] (3) After the above signal processing is completed, the output error of the entire magnetic measurement system is not more than 20nT;

[0052] (4) In practical application, it is easy to implement. BRIEF DESCRIPTION OF DRAWINGS

[0053] Figure 1 It is a flowchart for the magnetic data signal processing in the present application;

[0054] Figure 2 It is a schematic diagram of the signal compensation model in the present application. DETAILED DESCRIPTION

[0055] The application will be further described in detail in connection with the following examples, but the scope of protection of the application is not limited thereto.

[0056] The application relates to a signal processing method in a large-plane magnetometer system.

[0057] The preprocessing is performed on the acquired magnetic signal by a low-pass filter to extract low-frequency useful signals and filter out useless signals in other frequency bands.

[0058] The multi-stage error model comprises a linearity error model, an orthogonality error model, an alignment error model and a gradient error model connected in sequence.

[0059] The linearity error model is used to compensate for the linearity error in the analog-to-digital conversion process and eliminate the linearity error caused by the analog-to-digital conversion.

[0060] The orthogonality error model is used to compensate for the error caused by the non-orthogonality of the three axial directions of the probe.

[0061] The alignment error model is used to compensate for the error caused by the non-consistency of the axial directions of the vector axes in different probes.

[0062] The gradient error model is used to compensate for the gradient error caused by the different positions of the probes in the geomagnetic field.

[0063] In the application, based on the theoretical basis, the acquired magnetic signal is filtered by a low-pass filter to extract low-frequency useful signals and filter out useless signals in other frequency bands, and then is substituted into the linearity error model, the orthogonality error model, the alignment error model and the gradient error model in sequence, each model eliminates the corresponding error amount in sequence, and finally the processed magnetic signal is obtained, which is used for the display, storage and calling of the subsequent magnetic field to obtain the required magnetic field information of the system.

[0064] In the application, specifically, the magnetic signal sensed by the probe is uploaded as an analog quantity, is first subjected to an analog-to-digital conversion module, the analog-to-digital conversion module realizes high-speed sampling of the analog magnetic signal at a sampling frequency of 1KHz and converts the analog magnetic signal into a digital signal, the digital signal is subjected to a Butterworth digital low-pass filter with a cutoff frequency of 5.0Hz, high-frequency magnetic signals with a frequency greater than 5.0Hz are filtered out, the geomagnetic field signal (frequency <0.1Hz) and the magnetic target signal (frequency <1Hz) are retained, and the data frequency is reduced from 1KHz to 10Hz through 100:1 interval sampling and is then output.

[0065] The linearity error model is,

[0066]

[0067] wherein, and are the X component, Y component and Z component original values of the i-th probe, respectively;

[0068] X i , Y i and Z i are the X component, Y component and Z component values of the i-th probe after linearity error compensation, respectively;

[0069] a3, a2, a1, a0 are the linearity compensation factors of the X component of the i-th probe, respectively, a3, a2, a1∈[-1, 1], a0∈[-500, 500];

[0070] b3, b2, b1, b0 are the linearity compensation factors of the Y component of the i-th probe, respectively, b3, b2, b1∈[-1, 1], b0∈[-500, 500];

[0071] c3, c2, c1, c0 are the linearity compensation factors of the Z component of the i-th probe, respectively, c3, c2, c1∈[-1, 1], c0∈[-500, 500].

[0072] The orthogonality error model is,

[0073]

[0074] wherein, X i , Y i and Z i are the X component, Y component and Z component values of the i-th probe after linearity error compensation, respectively;

[0075] and are the X component, Y component and Z component values of the i-th probe after orthogonality error compensation, respectively;

[0076] α, β, γ, d x , d y , d z , e x , e y , e z are the orthogonality compensation factors, α, β, γ∈[-1, 1], d x , d y , d z ∈[0.9, 1.1], e x , e y , e z ∈[-500, 500].

[0077] In the present application, the orthogonality error refers to the error caused by the non-orthogonality of the three axial directions of the probe, and α, β and γ respectively refer to the angle errors of the X component, the Y component and the Z component, which are easily understood by those skilled in the art.

[0078] The alignment error model is,

[0079]

[0080] Wherein, And Xi, Yi and Zi are the X component, the Y component and the Z component values of the i-th probe after the orthogonality error compensation;

[0081] Xi, Yi and Zi are the X component, the Y component and the Z component values of the i-th probe after the alignment error compensation;

[0082] δ, ε and θ are alignment error compensation factors, and δ, ε, θ ∈ [-1, 1].

[0083] In the present application, the alignment error refers to the error caused by the non-orthogonality of the three axial directions of the probe, and δ, ε and θ respectively correspond to the X axis rotation angle, the Y axis rotation angle and the Z axis rotation angle of the i-th probe before and after the alignment error compensation, which are easily understood by those skilled in the art.

[0084] The gradient error model is,

[0085]

[0086] Wherein, Xi, Yi and Zi are the X component, the Y component and the Z component values of the i-th probe after the alignment error compensation;

[0087] And Xi, Yi and Zi are the X component, the Y component and the Z component values of the i-th probe after the gradient error compensation;

[0088] Xi, Yi and Zi are the X component, the Y component and the Z component values of the i-th probe at the initial time before the gradient error compensation;

[0089] Xi, Yi and Zi are the X component, the Y component and the Z component values of the first probe at the initial time before the gradient error compensation.

[0090] The magnetic data required by the system includes magnetic field component data and magnetic field synthesis total field data.

[0091] The preprocessed magnetic signal is compensated by compensation factors of each stage of error model, and the compensated magnetic data is used for magnetic component data output, storage, and magnetic field synthesis total field data calculation.

[0092] The stored magnetic data is played back by signal extraction.

[0093] In the present application, the compensated magnetic data, including magnetic component data and magnetic field synthesis total field data, is post-processed to have man-machine interaction functions such as display, storage, and signal extraction.

[0094] In the present application, the magnetic signal sensed by the probe is collected at a sampling frequency of 1KHz to complete analog-digital conversion; the converted digital magnetic signal is filtered by a Butterworth digital low-pass filter with a cutoff frequency of 5.0Hz provided by a filter coefficient to filter out high-frequency magnetic signals with a frequency greater than 5.0Hz, retain geomagnetic field signals (frequency <0.1Hz) and magnetic target signals (frequency <1Hz), and output the data frequency reduced from 1KHz to 10Hz by 100:1 interval sampling; the filtered magnetic signal is compensated by an internal mathematical model to complete signal compensation by a coefficient provided by a compensation coefficient; the compensated magnetic data is used for magnetic component output and storage, and total field calculation to obtain a fitted total field value and realize magnetic total field output; the stored magnetic signal can be used for magnetic component output and magnetic total field output by signal extraction to realize magnetic signal playback.

Claims

1. A method of signal processing in a large plane magnetometric system, characterized by: The method pre-processes the acquired magnetic signal, inputs the pre-processed magnetic signal into a multi-stage error model to perform multi-stage error compensation processing, and obtains magnetic data required by the system; The multi-stage error model comprises a linearity error model, an orthogonality error model, an alignment error model and a gradient error model connected in sequence; The linearity error model is used to complete compensation of linearity error in the analog-digital conversion process and eliminate linearity error caused by the analog-digital conversion; the linearity error model is, ， wherein, , and are the X, Y and Z component raw values of the first probe, respectively. , and The first The X, Y, and Z component values ​​of each probe after linearity error compensation; , , , The first Linearity compensation factor for the X component of each probe. , , , ; , , , The first Linearity compensation factor for the Y component of each probe. , , , ; , , , The first Linearity compensation factor for the Z component of each probe. , , , ; The orthogonality error model is used to complete compensation of error caused by non-orthogonality of three axial directions in the probe; the orthogonality error model is, ， wherein, , and are the X, Y and Z component values after linearity error compensation for the first th probe, respectively. , and The first The X, Y, and Z component values ​​after probe orthogonality error compensation; 、 、 、 、 、 、 、 、 is an orthogonality compensation factor, , , , , , , , , ; The alignment error model is used to compensate error caused by non-identity of axial directions of each vector axis in different probes; the alignment error model is, ， wherein, , and are the compensated X, Y and Z component values of the first probe orthogonality error, respectively. , , are the X, Y and Z component values of the first probe alignment error compensation, respectively; and are the X, Y and Z component values of the second probe alignment error compensation, respectively. , and are alignment error compensation factors, , , ; The gradient error model is used to compensate gradient error caused by different positions of each probe in the geomagnetic field; the gradient error model is, ， wherein, , , are the compensated X, Y and Z component values of the first th probe alignment error, respectively. , and The first The X, Y, and Z component values ​​after gradient error compensation for each probe; , , are the X component, Y component, Z component values of the 1st probe gradient error compensation before the initial moment, respectively; are the X component, Y component, Z component values of the 1st probe gradient error compensation before the initial moment, respectively; , , are respectively the initial moment X component, Y component, Z component values before the first probe gradient error compensation The pre-processed magnetic signal passes through the multi-stage error model, and signal compensation is completed through compensation factors of each stage; the compensated magnetic data is used for magnetic field component data output, storage, magnetic field synthesis total field data calculation.

2. The signal processing method in a large plane magnetometric system according to claim 1, characterized in that: The pre-processing processes the acquired magnetic signal by using a low-pass filter, extracts low-frequency useful signals, and filters out useless signals in other frequency bands; the data frequency of the low-frequency useful signals is reduced and then output.

3. The method of claim 1, wherein: The magnetic data required by the system comprises magnetic field component data and magnetic field synthesis total field data.

4. The method of claim 1, wherein: The stored magnetic data is played back by signal extraction.

Images