A high-precision magnetic flux measurement method and system based on three-dimensional geometric correction

Abstract

The application discloses a kind of high-precision magnetic flux measurement method and system based on three-dimensional geometry correction, belong to magnetism measurement and electromagnetic metrology technical field.The application is by constructing three-dimensional entity volume decomposition analytical model, and the space constant of coil is inverted, and standard volt-ampere method is combined to obtain coil constant;During system idle period, the integrated noise floor is captured as dynamic zero baseline;Utilize tightly coupled temperature sensor to realize hardware level dynamic thermal compensation;During measurement stage, the sampling signal is sequentially subjected to power frequency interference removal, dynamic zero baseline deduction, sliding average filtering and digital integration operation, and the net magnetic flux is output in combination with coil constant.The application solves the boundary field distortion problem of traditional one-dimensional model, suppresses drift interference through soft and hard collaborative mechanism, and improves the signal-to-noise ratio and stability of weak magnetic flux measurement.

Description

Technical Field

[0001] This invention relates to the fields of magnetic measurement and electromagnetic metrology, and in particular to a high-precision magnetic flux measurement method and system based on three-dimensional geometric correction. Background Technology

[0002] Magnetic measurement and electromagnetic metrology are fundamental technological fields in modern industry and scientific research. Accurate measurement of the magnetic moment of magnetic materials is a core aspect of evaluating their performance. Soft magnetic materials, as key components in aerospace, military defense, and power electronics systems, have magnetic moment parameters that directly affect the operational performance and reliability of these systems. Currently, vibrating sample magnetometers are widely used as the mainstream measurement equipment in the industry. These devices rely on standard samples for calibration and traceability, and detect the magnetization state of magnetic materials based on Faraday's law of electromagnetic induction. Induction coils, as the core sensing component for magnetic flux measurement, are widely used in various magnetometer systems. Working in conjunction with analog signal conditioning circuits and digital signal processing units, they complete the conversion and analysis from magnetic signals to electrical signals, providing data support for the performance evaluation of magnetic materials.

[0003] Current magnetic flux measurement techniques based on induction coils still suffer from several limitations. Traditional measurement methods typically treat the induction coil as an ideal single-turn or one-dimensional current wire structure, neglecting the actual cross-sectional dimensions and spatial geometry of the coil. This fails to accurately describe the local spatial leakage magnetic field and edge field distribution within the coil, leading to discrepancies between theoretically calculated coil constants and actual physical properties. In strong background magnetic fields, weak magnetic flux signals generated by changes in the sample's magnetization state are easily submerged by broadband common-mode noise in the environment, making effective extraction difficult. Furthermore, traditional analog integrators are susceptible to environmental temperature variations and inherent low-frequency noise from electronic components during long-term operation, resulting in accumulated zero-point drift and signal distortion. In addition, crosstalk between the front-end analog circuitry and the back-end digital circuitry also negatively impacts the measurement accuracy of weak signals. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the prior art and provide a high-precision magnetic flux measurement method and system based on three-dimensional geometric correction.

[0005] The objective of this invention is achieved through the following technical solution: A high-precision magnetic flux measurement method based on three-dimensional geometric correction is provided, which includes the following steps: S1. In the cylindrical coordinate system, the cross section of the induction coil is discretized into infinitesimal elements. Based on the two-dimensional solid integral model and the Biot-Savart law, considering the inverse square law of distance and the attenuation of spatial distance orthogonal projection, a global magnetic field multiple integral inversion operator at the central axis is established to map the uniform spatial sensitivity region inside the induction coil. The coil constant of the assembled induction coil is obtained by physical calibration using the standard voltammetry method. S2. In the idle state without sample excitation, start background monitoring, capture the overall background noise, and store it as a dynamic zero baseline; S3. The tightly coupled temperature sensor on the surface of the instrumentation amplifier picks up the physical temperature change of the microenvironment, and feeds back to the bias pin input of the instrumentation amplifier to generate a conditioned induction signal. S4. During the measurement phase, the conditioned induction signal is sampled, power frequency interference is filtered out from the sampled signal, the dynamic zero baseline is subtracted point by point from the real-time data stream, the debiased data is subjected to moving average filtering, the filtered data is injected into the digital integrator to complete the time-area discrete integration operation, and the net magnetic flux is output in combination with the coil constant.

[0006] Furthermore, step S1 includes the following sub-steps: S1.1. Determine the structural parameters of the induction coil, including the inner diameter, outer diameter, axial width, and center-to-center distance. S1.2. Based on the structural parameters of the induction coil, the cross-section of the induction coil is discretized into current micro-elements whose cross-sectional area is the product of radial micro-elements and axial micro-elements. The radial micro-elements are micro-elements cut off along the radial direction of the induction coil, and the axial micro-elements are micro-elements cut off along the axial direction of the induction coil. The current value of each current micro-element is calculated. S1.3. Based on the Biot-Savart law, the axial magnetic induction intensity generated by all current elements at the central axis is superimposed and summed to obtain the global magnetic field multiple integral inversion operator; S1.4. Based on the global magnetic field multiple integral inversion operator, the spatial sensitivity uniform region inside the induction coil is mapped, and the assembled induction coil is physically calibrated by the standard volt-ampere method to obtain the coil constant.

[0007] Furthermore, step S2 includes the following sub-steps: S2.1. Detect the operating status and determine whether the operating status is an idle state without sample excitation; S2.2. When the operating state is an idle state without sample excitation, start the analog-to-digital converter to continuously sample and acquire the background signal within a set time period; S2.3. Calculate the arithmetic mean of the background signal and extract the comprehensive noise floor including quantization residual error and quasi-DC micro-bias; S2.4. Store the overall noise floor in the dynamic buffer as a dynamic zero baseline.

[0008] Furthermore, step S3 includes the following sub-steps: S3.1. The physical temperature data of the microenvironment in which the instrumentation amplifier is located is collected in real time by a temperature sensor tightly coupled to the surface of the instrumentation amplifier package; S3.2. Input the collected physical temperature data into the bias compensation circuit to generate the corresponding compensation voltage signal; S3.3. Input the compensation voltage signal to the reference pin of the instrumentation amplifier to adjust the bias of the instrumentation amplifier and generate the conditioned induction signal.

[0009] Furthermore, step S4 includes the following sub-steps: S4.1. Detect whether the operation status has entered the measurement stage. When a pulse excitation signal is detected, start the analog-to-digital converter to sample the conditioned induction signal at a set frequency in real time to obtain the sampled signal. S4.2. Perform power frequency interference filtering on the sampled signal to obtain the digital signal after power frequency interference is removed; S4.3. Read the dynamic zero baseline from the dynamic buffer, and perform digital subtraction operation point by point in the real-time data stream of the digital signal after filtering out power frequency interference to subtract the dynamic zero baseline and obtain the debiased data; S4.4. Perform a moving average filter on the debiased data to obtain the filtered data; S4.5. Input the filtered data into the digital integrator to perform time-area discrete integration. The time-area discrete integration is to accumulate the product of the signal value at discrete time points and the sampling time interval, and output the net magnetic flux in combination with the coil constant.

[0010] Furthermore, in step S1.1, the induction coil adopts a Helmholtz structure, and the induction coil includes two coaxial single coils, with the structural parameters of the two coaxial single coils being consistent.

[0011] Furthermore, in step S4.1, the conditioned inductive signal undergoes signal conditioning by an analog front-end conditioning unit before being input to the analog-to-digital converter. The signal conditioning includes signal amplification, signal filtering, and gain adjustment. The analog front-end conditioning unit includes a cascaded instrumentation amplifier, an LC filter, and a programmable gain amplifier.

[0012] Furthermore, in step S4.1, the analog front-end conditioning unit and the digital analysis and integration unit are physically isolated from each other. The physical isolation is achieved by using opto-isolators to block crosstalk generated by the clock signal at the set frequency.

[0013] Furthermore, in step S4.2, the power frequency interference filtering process adopts a power frequency suppression algorithm based on phase-locked loop to filter out power frequency harmonic signals of a set frequency. The power frequency harmonic signals are the fundamental frequency signal generated by the power grid and its integer multiples of harmonic signals.

[0014] A high-precision magnetic flux measurement system based on three-dimensional geometric correction is provided. The system includes a multi-dimensional physical induction coil unit, an analog front-end conditioning unit, and a digital analysis and integration unit. The signal output terminal of the multidimensional physical induction coil unit is connected to the signal input terminal of the analog front-end conditioning unit. The signal output terminal of the analog front-end conditioning unit is connected to the signal input terminal of the digital analysis and integration unit via an electrical isolation device. The multidimensional physical induction coil unit includes a coil frame made of non-magnetic material and an induction coil wound on the coil frame. The analog front-end conditioning unit includes a cascaded instrumentation amplifier, an LC filter, a programmable gain amplifier, a temperature sensor, and a bias compensation circuit. The temperature sensor is physically tightly coupled to the surface of the instrumentation amplifier package, and the signal output terminal of the temperature sensor is connected to the reference pin of the instrumentation amplifier via the bias compensation circuit. The digital analysis and integration unit includes a 24-bit set-frequency analog-to-digital converter and a microcontroller. The 24-bit set-frequency analog-to-digital converter is physically isolated from the analog front-end conditioning unit.

[0015] The beneficial effects of this invention are: (1) By constructing a three-dimensional solid integral analytical model to invert the spatial constant of the coil, the uniform spatial sensitivity region inside the coil is mapped, and the coil constant is obtained by physical calibration using the standard voltammetry method, which effectively reduces the spatial deformation error between the theoretical model and the actual measurement. (2) Dynamic thermal compensation is performed on the instrumentation amplifier at the physical source, and the dynamic zero baseline formed by the overall system noise is subtracted in real time in the digital domain. The dual mechanism works together to suppress drift interference during the measurement process. (3) The sampling signal is sequentially processed by power frequency interference filtering, moving average filtering and unbiased digital integration, and clock crosstalk is blocked by physical isolation between the analog domain and the digital domain, thereby improving the signal-to-noise ratio and stability of weak magnetic flux measurement. Attached Figure Description

[0016] Figure 1 A flowchart illustrating the steps of a high-precision magnetic flux measurement method based on three-dimensional geometric correction; Figure 2 A schematic diagram of the spatial discrete structure of an induction coil based on solid geometry modeling provided for an embodiment; Figure 3 The overall architecture block diagram of the signal extraction system with hardware and software collaborative anti-drift characteristics provided in the embodiment; Figure 4 The flowchart of the asynchronous dynamic zero-point tracking and integration algorithm provided for the embodiment. Detailed Implementation

[0017] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Example 1 See Figure 1 This embodiment provides a high-precision magnetic flux measurement method based on three-dimensional geometric correction, which includes the following steps: S1. In the cylindrical coordinate system, the cross section of the induction coil is discretized into infinitesimal elements. Based on the two-dimensional solid integral model and the Biot-Savart law, considering the inverse square law of distance and the attenuation of spatial distance orthogonal projection, a global magnetic field multiple integral inversion operator at the central axis is established to map the uniform spatial sensitivity region inside the induction coil. The coil constant of the assembled induction coil is obtained by physical calibration using the standard voltammetry method. S2. In the idle state without sample excitation, start background monitoring, capture the overall background noise, and store it as a dynamic zero baseline; S3. The tightly coupled temperature sensor on the surface of the instrumentation amplifier picks up the physical temperature change of the microenvironment, and feeds back to the bias pin input of the instrumentation amplifier to generate a conditioned induction signal. S4. During the measurement phase, the conditioned induction signal is sampled, power frequency interference is filtered out from the sampled signal, the dynamic zero baseline is subtracted point by point from the real-time data stream, the debiased data is subjected to moving average filtering, the filtered data is injected into the digital integrator to complete the time-area discrete integration operation, and the net magnetic flux is output in combination with the coil constant.

[0019] In some embodiments, step S1 includes the following sub-steps: S1.1. Determine the structural parameters of the induction coil, including the inner diameter, outer diameter, axial width, and center-to-center distance. S1.2. Based on the structural parameters of the induction coil, the cross-section of the induction coil is discretized into current micro-elements whose cross-sectional area is the product of radial micro-elements and axial micro-elements. The radial micro-elements are micro-elements cut off along the radial direction of the induction coil, and the axial micro-elements are micro-elements cut off along the axial direction of the induction coil. The current value of each current micro-element is calculated. S1.3. Based on the Biot-Savart law, the axial magnetic induction intensity generated by all current elements at the central axis is superimposed and summed to obtain the global magnetic field multiple integral inversion operator; S1.4. Based on the global magnetic field multiple integral inversion operator, the spatial sensitivity uniform region inside the induction coil is mapped, and the assembled induction coil is physically calibrated by the standard volt-ampere method to obtain the coil constant.

[0020] In some embodiments, step S2 includes the following sub-steps: S2.1. Detect the operating status and determine whether the operating status is an idle state without sample excitation; S2.2. When the operating state is an idle state without sample excitation, start the analog-to-digital converter to continuously sample and acquire the background signal within a set time period; S2.3. Calculate the arithmetic mean of the background signal and extract the comprehensive noise floor including quantization residual error and quasi-DC micro-bias; S2.4. Store the overall noise floor in the dynamic buffer as a dynamic zero baseline.

[0021] In some embodiments, step S3 includes the following sub-steps: S3.1. The physical temperature data of the microenvironment in which the instrumentation amplifier is located is collected in real time by a temperature sensor tightly coupled to the surface of the instrumentation amplifier package; S3.2. Input the collected physical temperature data into the bias compensation circuit to generate the corresponding compensation voltage signal; S3.3. Input the compensation voltage signal to the reference pin of the instrumentation amplifier to adjust the bias of the instrumentation amplifier and generate the conditioned induction signal.

[0022] In some embodiments, step S4 includes the following sub-steps: S4.1. Detect whether the operation status has entered the measurement stage. When a pulse excitation signal is detected, start the analog-to-digital converter to sample the conditioned induction signal at a set frequency in real time to obtain the sampled signal. S4.2. Perform power frequency interference filtering on the sampled signal to obtain the digital signal after power frequency interference is removed; S4.3. Read the dynamic zero baseline from the dynamic buffer, and perform digital subtraction operation point by point in the real-time data stream of the digital signal after filtering out power frequency interference to subtract the dynamic zero baseline and obtain the debiased data; S4.4. Perform a moving average filter on the debiased data to obtain the filtered data; S4.5. Input the filtered data into the digital integrator to perform time-area discrete integration. The time-area discrete integration is to accumulate the product of the signal value at discrete time points and the sampling time interval, and output the net magnetic flux in combination with the coil constant.

[0023] In some embodiments, in step S1.1, the induction coil adopts a Helmholtz structure, and the induction coil includes two coaxial single coils, the structural parameters of the two coaxial single coils being consistent.

[0024] In some embodiments, in step S4.1, the conditioned inductive signal is conditioned by an analog front-end conditioning unit before being input to the analog-to-digital converter. The signal conditioning includes signal amplification, signal filtering, and gain adjustment. The analog front-end conditioning unit includes a cascaded instrumentation amplifier, an LC filter, and a programmable gain amplifier.

[0025] In some embodiments, in step S4.1, the analog front-end conditioning unit and the digital analysis and integration unit are physically isolated from each other. The physical isolation is achieved by using opto-isolation devices to block crosstalk generated by clock signals of a set frequency.

[0026] In some embodiments, in step S4.2, the power frequency interference filtering process adopts a power frequency suppression algorithm based on phase-locked loop to filter out power frequency harmonic signals of a set frequency. The power frequency harmonic signals are the fundamental wave signal generated by the power grid and its integer multiples of the harmonic signals.

[0027] A high-precision magnetic flux measurement system based on three-dimensional geometric correction is provided. The system includes a multi-dimensional physical induction coil unit, an analog front-end conditioning unit, and a digital analysis and integration unit. The signal output terminal of the multidimensional physical induction coil unit is connected to the signal input terminal of the analog front-end conditioning unit. The signal output terminal of the analog front-end conditioning unit is connected to the signal input terminal of the digital analysis and integration unit via an electrical isolation device. The multidimensional physical induction coil unit includes a coil frame made of non-magnetic material and an induction coil wound on the coil frame. The analog front-end conditioning unit includes a cascaded instrumentation amplifier, an LC filter, a programmable gain amplifier, a temperature sensor, and a bias compensation circuit. The temperature sensor is physically tightly coupled to the surface of the instrumentation amplifier package, and the signal output terminal of the temperature sensor is connected to the reference pin of the instrumentation amplifier via the bias compensation circuit. The digital analysis and integration unit includes a 24-bit set-frequency analog-to-digital converter and a microcontroller. The 24-bit set-frequency analog-to-digital converter is physically isolated from the analog front-end conditioning unit.

[0028] Example 2 This embodiment provides a specific implementation process for a high-precision magnetic flux measurement method based on three-dimensional geometric correction. This process achieves accurate inversion of the coil constant by constructing a three-dimensional spatial analytical model of a physical induction coil. Combined with a synergistic mechanism of hardware-level thermal compensation and digital domain dynamic baseline correction, it solves the problems of boundary field distortion and long-term integration zero-point drift in traditional magnetic flux measurement. It is applicable to magnetic measurement and electromagnetic metrology fields such as magnetic moment calibration of magnetic materials. The specific implementation process is as follows: Step 1. Inversion of coil space constant: Step 1.1. Determine the structural parameters of the induction coil: The induction coil is the sensing component for magnetic flux measurement, used to convert the magnetization state of the sample under test into an electrical signal. In this embodiment, the structural parameters of the induction coil are first determined, including the inner diameter, outer diameter, axial width, and center-to-center distance. In some embodiments, the induction coil adopts a Helmholtz structure, which consists of two coaxial and parallel single coils. The structural parameters of the two coaxial single coils are consistent. This structure can generate a magnetic field with good uniformity in the central region, and is suitable for magnetic flux measurement scenarios where magnetic field uniformity is required.

[0029] Step 1.2. Discretization of coil cross-section and calculation of current infinitesimal elements: Based on the structural parameters of the induction coil, the cross-section of the induction coil is discretized into current micro-elements whose cross-sectional area is the product of radial and axial micro-elements. The radial micro-elements are micro-elements intercepted along the radial direction of the induction coil, and the axial micro-elements are micro-elements intercepted along the axial direction of the induction coil. By discretizing the continuous coil cross-section into multiple tiny current units, the complex spatial magnetic field calculation can be transformed into a superposition calculation of the magnetic field generated by multiple micro-element currents, thus achieving a refined description of the spatial magnetic field distribution of the coil.

[0030] In some specific implementations, the induction coil is modeled in three-dimensional discrete form in cylindrical coordinate system (r, θ, z). The induction coil adopts a Helmholtz structure containing two coaxial single coils, such as... Figure 2 As shown, it includes an upper coil and a lower coil. The inner diameter of the upper and lower coils is Rin, the outer diameter is Rout, the cross-sectional width of a single coil is 2W, the center-to-center distance between the two coils on the Z-axis is 2D, and the number of turns on each side is N. The infinitesimal current within the cross-section is expressed as: ; Among them, by using the Biot-Savart law calculus inversion, and considering the nonlinear attenuation introduced by the inverse square law of distance and the orthogonal projection of spatial distance, the entity multiple integral analytical expression of the magnetic flux sensitivity at the target point P(z0) is expressed as: ; Where Bz(z0) is the axial magnetic induction intensity at the observation point z0 on the target axis; μ0 is the free permeability, a physical constant; N is the number of turns of a single-sided coil; I is the excitation current flowing through the induction coil; R in R is the inner diameter of the induction coil. out Let be the outer diameter of the induction coil; 2W be the axial width of a single-sided coil; D be half the center distance between the two coils; r be the radial integral variable; z be the axial integral variable; and z0 be the coordinates of the target observation point on the axis. By solving the above double integral, the axial magnetic induction intensity distribution at any position on the central axis of the induction coil can be obtained, thus mapping the spatial sensitivity uniform region inside the induction coil. The spatial sensitivity uniform region refers to the area inside the induction coil where the sensitivity variation is small; measurements within this region yield consistent results. Based on the distribution results of the spatial sensitivity uniform region, the cross-sectional dimensions of the induction coil can be optimized to expand the range of the uniform region and improve the applicability of the measurement.

[0031] Step 1.3. Construction of the global magnetic field multiple integral inversion operator: Based on the Biot-Savart law, the axial magnetic induction intensity produced by all current elements at the central axis is superimposed and summed to obtain the global magnetic field multiple integral inversion operator. The Biot-Savart law is a fundamental physical law describing the magnetic induction intensity produced by a current element at any point in space. This law states that the magnetic induction intensity produced by a current element at a point in space is directly proportional to the magnitude of the current element, inversely proportional to the square of the distance from the current element to that point, and directly proportional to the sine of the angle between the current element vector and the position vector.

[0032] By solving the double integral described above, the axial magnetic flux density distribution at any position on the central axis of the induction coil can be obtained, thus mapping the spatial sensitivity uniformity region inside the induction coil. The spatial sensitivity uniformity region refers to the area within the induction coil where the sensitivity variation is small; measurements taken within this region yield consistent results. Based on the distribution of the spatial sensitivity uniformity region, the cross-sectional dimensions of the induction coil can be optimized to expand the range of the uniformity region and improve the applicability of the measurement.

[0033] Step 1.4. Physical calibration of coil constant: Based on the global magnetic field multiple integral inversion operator mapping the uniform spatial sensitivity region inside the induction coil, the assembled induction coil is physically calibrated using the standard volt-ampere method to obtain the coil constant. The standard volt-ampere method is a general method for determining electrical parameters by measuring voltage and current. In this embodiment, by passing a known excitation current through the induction coil, the induced voltage across the coil is measured, and combined with known magnetic field parameters, the coil constant is calculated. The coil constant is an inherent parameter of the induction coil, used to convert the measured electrical signal into a corresponding magnetic flux value, and is the foundation for quantitative magnetic flux measurement.

[0034] In some embodiments, the calibration process of the coil constant can be carried out in a constant temperature environment to reduce the impact of temperature changes on the calibration results and improve the accuracy of the calibration.

[0035] Step 2. Dynamic zero baseline acquisition: Step 2.1. System operating status detection: The system's operating status is monitored to determine whether it is in an idle state without sample excitation. An idle state without sample excitation refers to a state where no AC or step magnetic field excitation is applied to the system, and no sample is placed. In this state, the system output signal is mainly the system's own noise floor signal and does not contain the magnetization information of the sample.

[0036] Step 2.2. Continuous sampling of background signal: When the operating state is an idle state without sample excitation, the analog-to-digital converter (ADC) is activated to continuously sample and acquire the background signal within a set time period. An ADC is an electronic device that converts analog signals into digital signals. It can convert continuously changing analog voltage signals into discrete digital signals, facilitating subsequent digital processing and analysis.

[0037] Step 2.3. Comprehensive background noise extraction: The arithmetic mean of the background signal is calculated to extract the combined noise floor, which includes quantization residual error and quasi-DC micro-bias. Quantization residual error is the error generated by the analog-to-digital converter (ADC) during the conversion of analog signals to digital signals due to limited quantization accuracy; quasi-DC micro-bias is a slowly varying DC offset signal present in the system, mainly caused by the characteristic drift of electronic components. Both of these signals interfere with the measurement of weak magnetic flux signals, leading to deviations in the measurement results.

[0038] Step 2.4. Dynamic Zero Baseline Storage: The overall noise floor is stored in a dynamic buffer as a dynamic zero baseline. The dynamic buffer is a memory area for temporary data storage, enabling fast data read and write operations, facilitating real-time reading of the dynamic zero baseline for subtraction during the measurement phase. In some embodiments, the dynamic zero baseline can be updated before each measurement or periodically at set time intervals to adapt to the slow changes in system noise floor over time.

[0039] Step 3. Hardware thermal compensation processing: Step 3.1. Microenvironment temperature data acquisition: A temperature sensor tightly coupled to the surface of the instrumentation amplifier package acquires real-time physical temperature data of the microenvironment in which the instrumentation amplifier resides. An instrumentation amplifier is an amplifier with high input impedance, high common-mode rejection ratio, and low noise, suitable for amplifying weak electrical signals. The temperature sensor, tightly coupled to the instrumentation amplifier's package surface, accurately measures the temperature changes within the instrumentation amplifier itself, as the temperature drift of the instrumentation amplifier is primarily caused by its own temperature variations.

[0040] In some embodiments, the temperature sensor may be a thermistor, which is a resistor whose resistance value changes significantly with temperature. The thermistor is small in size and has a fast response speed, enabling real-time temperature measurement.

[0041] Step 3.2. Generation of compensation voltage signal: The acquired physical temperature data is input to the bias compensation circuit, which generates a corresponding compensation voltage signal. The bias compensation circuit is a circuit that generates a corresponding compensation voltage based on the input temperature signal; its function is to counteract the bias voltage drift of the instrumentation amplifier caused by temperature changes. The bias compensation circuit can be implemented using analog circuitry or using digital circuitry combined with a digital-to-analog converter.

[0042] Step 3.3. Instrumentation amplifier bias adjustment: A compensation voltage signal is input to the reference pin of the instrumentation amplifier to adjust its bias and generate a conditioned inductive signal. By compensating for the temperature drift of the instrumentation amplifier at its physical source, the impact of temperature changes on the measurement signal can be effectively reduced, thus improving measurement stability.

[0043] In some embodiments, the bias compensation circuit may employ a proportional-integral control algorithm to automatically adjust the magnitude of the compensation voltage according to the rate and magnitude of temperature change, thereby achieving more accurate temperature drift compensation.

[0044] Step 4. Digital signal integration and output: Step 4.1. Real-time sampling of the inductive signal: To determine whether the system has entered the measurement phase, when a pulse excitation signal is detected, the analog-to-digital converter is activated to sample the conditioned inductive signal at a set frequency in real time, obtaining the sampled signal. The pulse excitation signal is used to excite the sample under test to produce a magnetization reversal. When this signal is detected, it indicates that the system has entered the measurement phase and needs to start acquiring the inductive signal.

[0045] In some specific implementations, the signal extraction system is strictly divided into an isolated analog front-end domain and a digital processing domain, such as... Figure 3 As shown, the digital and analog domains are physically isolated in their hardware interconnections to block high-speed clock crosstalk. In the analog front-end domain, the weak transient voltage output from the flux sensing coil is fed back to a cascaded instrumentation amplifier, LC filter, and programmable gain amplifier. A temperature sensor, attached to the surface of the instrumentation amplifier, monitors ambient thermal fluctuations in real time and injects a compensation voltage into the REF pin of the instrumentation amplifier through a bias circuit, suppressing the generation of millivolt-level thermal drift spurious signals at the physical source. In the digital processing domain, the conditioned signal is sampled at high frequency by a 24-bit analog-to-digital converter, and power frequency and harmonic noise are eliminated by a phase-locked loop-based power frequency suppression algorithm. Then, the dynamic zero-point baseline is subtracted point by point by a digital subtractor, followed by smoothing random white noise through a moving average filter. Finally, the signal is input to a digital integrator for unbiased time-area discrete integration to output the flux.

[0046] In some embodiments, the conditioned inductive signal undergoes signal conditioning by an analog front-end conditioning unit before being input to the analog-to-digital converter. Signal conditioning includes signal amplification, signal filtering, and gain adjustment. The analog front-end conditioning unit includes a cascaded instrumentation amplifier, an LC filter, and a programmable gain amplifier. An LC filter is a passive filter composed of inductors and capacitors, capable of filtering out interference signals of specific frequencies. A programmable gain amplifier is an amplifier whose gain can be adjusted digitally, automatically adjusting the gain according to the magnitude of the input signal to ensure that the output signal remains within the input range of the analog-to-digital converter.

[0047] In some embodiments, the analog front-end conditioning unit and the digital analysis and integration unit are physically isolated from each other. This physical isolation is achieved through electrical isolation using opto-isolators, blocking crosstalk generated by clock signals of a set frequency. An opto-isolator is a device that transmits electrical signals through optical signals. It enables electrical isolation between input and output, effectively blocking interference from high-frequency clock signals generated by digital circuits on the analog front-end circuit, thus improving the quality of the analog signal.

[0048] In some specific implementations, the execution flow of asynchronous dynamic zero-point tracking and integration algorithms is as follows: Figure 4As shown. The system first initializes its operating state, then determines the system status and detects the excitation trigger signal. When the system is in the measurement idle period without sample excitation, background noise monitoring is activated, automatically recording the quasi-DC micro-bias in the unexcited state, extracting the mean drift and quantization residual error, and updating the dynamic zero point written to the shared RAM dynamic buffer. When the system detects that pulse excitation has entered the measurement cycle, it starts high-frequency real-time sampling of the analog-to-digital converter to capture the weak induced electromotive force. It deeply suppresses power frequency interference through a nested phase-locked loop algorithm, then asynchronously reads the dynamic zero point reference, performs digital subtraction operation point by point in the real-time data stream to deduct the dynamic zero point, then balances random white noise through a moving average filter, and finally accumulates the discrete data stream point by point through an unbiased time-area integrator. Combined with the absolute coil constant extracted by the three-dimensional discrete volume integral stored in the ROM hard-core area, the absolute magnetic field inversion operator is reconstructed, and the magnetic flux is output.

[0049] In some specific implementations, during the physical hardware simulation domain conditioning phase, when receiving weak transient voltage feedback, a temperature sensor attached to the surface of the instrumentation amplifier monitors thermal fluctuations in real time. A compensation voltage is injected into the REF pin via a bias circuit to suppress temperature drift at its source. During the digital signal reconstruction phase, a silent scan is first performed to automatically record the quasi-DC micro-bias under no-excitation conditions to set a dynamic zero-point baseline. Secondly, after entering the pulse integration period, a digital phase-locked loop is used to eliminate power frequency harmonics, and the aforementioned dynamic baseline is subtracted point by point using an arithmetic formula. Finally, the de-divergent data stream enters the accumulator register for unbiased integration, preventing baseline divergence and directly calculating the magnetic flux.

[0050] In some specific implementations, during the three-dimensional inversion of the coil spatial constant, the cross-section of the induction coil is discretized into a cross-sectional area of ​​dr. The infinitesimal element dz is used to derive a two-dimensional solid integral model based on the coil cross-section. According to the Biot-Savart law, considering the inverse square law of distance and the nonlinear attenuation introduced by orthogonal projection of spatial distance, a double integral inversion operator Bz for the global magnetic field at z0 on the central axis is established. By solving the double solid integral of the above equation in a three-dimensional discrete coordinate system using a microcontroller, the corresponding sensitivity and uniform region distribution of samples of arbitrary shapes can be accurately mapped. This allows the extraction of the constant Kn, which excludes interference from the insulating layer and shape distortion, thereby establishing the accuracy of the collaborative correction between the underlying sensor and the algorithm.

[0051] Step 4.2. Power frequency interference filtering: The sampled signal is processed to remove power frequency interference, resulting in a digital signal with the interference removed. Power frequency interference is an AC signal and its harmonic signals generated by the power grid. It is one of the common sources of interference in magnetic measurements and can severely interfere with weak induced signals.

[0052] In some embodiments, the power frequency interference filtering process employs a power frequency suppression algorithm based on a phase-locked loop (PLL) to filter out power frequency harmonic signals at a set frequency. These harmonic signals are the fundamental frequency signal generated by the power grid and its integer multiples of harmonic signals. A PLL is a closed-loop control system capable of tracking the frequency and phase of an input signal. By comparing the phase of the input signal with a local reference signal, an error signal is generated, which is then used to adjust the frequency and phase of the local oscillator to synchronize it with the input signal, thereby achieving the tracking and filtering of signals at a specific frequency.

[0053] Step 4.3. Dynamic zero baseline point-by-point subtraction: The dynamic zero baseline is read from the dynamic buffer. A point-by-point digital subtraction operation is performed on the real-time data stream of the digital signal after filtering out power frequency interference to subtract the dynamic zero baseline, resulting in debiased data. By subtracting the dynamic zero baseline point by point, the influence of system noise floor on the measurement signal can be effectively eliminated, improving the signal-to-noise ratio of the measurement.

[0054] Step 4.4. Moving average filtering: The bias-free data is then subjected to a moving average filter to obtain the filtered data. Moving average filtering is a commonly used digital filtering method. Its principle is to perform an arithmetic average of multiple consecutive sample values ​​to obtain a smooth output value, effectively filtering out random white noise interference and improving signal smoothness.

[0055] Step 4.5. Discrete Integral over Time and Area and Magnetic Flux Output: The filtered data is input to a digital integrator for time-area discrete integration. This time-area discrete integration involves accumulating the product of the signal value at discrete time points and the sampling time interval, and then combining this product with the coil constant to output the net magnetic flux. A digital integrator is a program or circuit capable of integrating digital signals. By integrating the induced electromotive force signal, the corresponding magnetic flux value can be obtained.

[0056] In some embodiments, the digital integrator can be implemented using a software integrator. The software integrator, by running the integration algorithm within a microcontroller, offers high flexibility and ease of parameter adjustment. Furthermore, the software integrator can be equipped with an overflow protection mechanism to prevent saturation due to prolonged integration, ensuring the normal operation of the integration calculation.

[0057] This embodiment constructs a three-dimensional solid integral analytical model of the induction coil, achieving accurate mapping of the coil's spatial sensitivity uniformity region. This solves the boundary field distortion problem caused by neglecting the coil's cross-sectional dimensions in traditional one-dimensional approximation models, reducing the error between theoretical calculations and actual measurements. Simultaneously, this embodiment employs a hardware-software co-operated anti-drift mechanism combining hardware temperature compensation and software dynamic baseline subtraction. This suppresses temperature drift errors in the instrumentation amplifier at the physical source and subtracts system noise floor in real time in the digital domain, effectively improving the system's anti-drift capability and ensuring the signal-to-noise ratio in weak magnetic flux measurement environments. Furthermore, this embodiment blocks crosstalk from high-speed clock signals through physical isolation between the analog and digital domains, further improving measurement stability. The technical solution of this embodiment is applicable to magnetic moment calibration measurements of various magnetic materials, providing a high-precision, high-stability magnetic flux measurement method for the field of magnetic measurement, meeting the needs of aerospace, military defense, and power electronics for magnetic material performance evaluation.

[0058] The above description is merely a preferred embodiment of the present invention. It should be understood that the present invention is not limited to the forms disclosed herein and should not be construed as excluding other embodiments. It can be used in various other combinations, modifications, and environments, and can be altered within the scope of the concept described herein through the above teachings or related technologies or knowledge. Modifications and variations made by those skilled in the art that do not depart from the spirit and scope of the present invention should be within the protection scope of the appended claims.

Claims

1. A high-precision magnetic flux measurement method based on three-dimensional geometric correction, characterized in that, Includes the following steps: S1. In the cylindrical coordinate system, the cross section of the induction coil is discretized into infinitesimal elements. Based on the two-dimensional solid integral model and the Biot-Savart law, considering the inverse square law of distance and the attenuation of spatial distance orthogonal projection, a global magnetic field multiple integral inversion operator at the central axis is established to map the uniform spatial sensitivity region inside the induction coil. The coil constant of the assembled induction coil is obtained by physical calibration using the standard voltammetry method. S2. In the idle state without sample excitation, start background monitoring, capture the overall background noise, and store it as a dynamic zero baseline; S3. The tightly coupled temperature sensor on the surface of the instrumentation amplifier picks up the physical temperature change of the microenvironment, and feeds back to the bias pin input of the instrumentation amplifier to generate a conditioned induction signal. S4. During the measurement phase, the conditioned induction signal is sampled, power frequency interference is filtered out from the sampled signal, the dynamic zero baseline is subtracted point by point from the real-time data stream, the debiased data is subjected to moving average filtering, the filtered data is injected into the digital integrator to complete the time-area discrete integration operation, and the net magnetic flux is output in combination with the coil constant. Step S1 includes the following sub-steps: S1.

1. Determine the structural parameters of the induction coil, including the inner diameter, outer diameter, axial width, and center-to-center distance. S1.

2. Based on the structural parameters of the induction coil, the cross-section of the induction coil is discretized into current micro-elements whose cross-sectional area is the product of radial micro-elements and axial micro-elements. The radial micro-elements are micro-elements cut off along the radial direction of the induction coil, and the axial micro-elements are micro-elements cut off along the axial direction of the induction coil. The current value of each current micro-element is calculated. S1.

3. Based on the Biot-Savart law, the axial magnetic induction intensity generated by all current elements at the central axis is superimposed and summed to obtain the global magnetic field multiple integral inversion operator; S1.

4. Based on the global magnetic field multiple integral inversion operator, the spatial sensitivity uniform region inside the induction coil is mapped, and the assembled induction coil is physically calibrated by the standard volt-ampere method to obtain the coil constant.

2. The method according to claim 1, characterized in that, Step S2 includes the following sub-steps: S2.

1. Detect the operating status and determine whether the operating status is an idle state without sample excitation; S2.

2. When the running state is an idle state without sample excitation, start the analog-to-digital converter to continuously sample and acquire the background signal within a set time period; S2.

3. Calculate the arithmetic mean of the background signal and extract the comprehensive noise floor including quantization residual error and quasi-DC micro-bias; S2.

4. Store the overall noise floor in a dynamic buffer as a dynamic zero baseline.

3. The method according to claim 1, characterized in that, Step S3 includes the following sub-steps: S3.

1. The physical temperature data of the microenvironment in which the instrumentation amplifier is located is collected in real time by a temperature sensor tightly coupled to the surface of the instrumentation amplifier package; S3.

2. Input the collected physical temperature data into the bias compensation circuit to generate the corresponding compensation voltage signal; S3.

3. Input the compensation voltage signal to the reference pin of the instrumentation amplifier to adjust the bias of the instrumentation amplifier and generate the conditioned induction signal.

4. The method according to claim 1, characterized in that, Step S4 includes the following sub-steps: S4.

1. Detect whether the operation status has entered the measurement stage. When a pulse excitation signal is detected, start the analog-to-digital converter to sample the conditioned induction signal at a set frequency in real time to obtain the sampled signal. S4.

2. Perform power frequency interference filtering on the sampled signal to obtain the digital signal after power frequency interference is removed; S4.

3. Read the dynamic zero baseline from the dynamic buffer, and perform digital subtraction operation point by point in the real-time data stream of the digital signal after filtering out power frequency interference to subtract the dynamic zero baseline and obtain the debiased data; S4.

4. Perform a moving average filter on the debiased data to obtain the filtered data; S4.

5. Input the filtered data into the digital integrator to perform time-area discrete integration. The time-area discrete integration is to accumulate the product of the signal value at discrete time points and the sampling time interval, and output the net magnetic flux in combination with the coil constant.

5. The method according to claim 1, characterized in that, In step S1.1, the induction coil adopts a Helmholtz structure and includes two coaxial single coils, with the structural parameters of the two coaxial single coils being consistent.

6. The method according to claim 4, characterized in that, In step S4.1, the conditioned inductive signal undergoes signal conditioning by an analog front-end conditioning unit before being input to the analog-to-digital converter. The signal conditioning includes signal amplification, signal filtering, and gain adjustment. The analog front-end conditioning unit includes a cascaded instrumentation amplifier, an LC filter, and a programmable gain amplifier.

7. The method according to claim 6, characterized in that, In step S4.1, the analog front-end conditioning unit and the digital analysis and integration unit are physically isolated from each other. The physical isolation is achieved by using opto-isolation devices to block crosstalk generated by the clock signal at a set frequency.

8. The method according to claim 4, characterized in that, In step S4.2, the power frequency interference filtering process adopts a power frequency suppression algorithm based on phase-locked loop to filter out power frequency harmonic signals of a set frequency. The power frequency harmonic signals are the fundamental frequency signal generated by the power grid and its integer multiples of the harmonic signals.

9. A high-precision magnetic flux measurement system based on three-dimensional geometric correction, used to perform the method as described in any one of claims 1-8, characterized in that, The system includes a multi-dimensional physical induction coil unit, an analog front-end conditioning unit, and a digital analysis and integration unit; The signal output terminal of the multidimensional physical induction coil unit is connected to the signal input terminal of the analog front-end conditioning unit. The signal output terminal of the analog front-end conditioning unit is connected to the signal input terminal of the digital analysis and integration unit via an electrical isolation device. The multidimensional physical induction coil unit includes a coil frame made of non-magnetic material and an induction coil wound on the coil frame. The analog front-end conditioning unit includes a cascaded instrumentation amplifier, an LC filter, a programmable gain amplifier, a temperature sensor, and a bias compensation circuit. The temperature sensor is physically tightly coupled to the surface of the instrumentation amplifier package. The signal output terminal of the temperature sensor is connected to the reference pin of the instrumentation amplifier via the bias compensation circuit. The digital analysis and integration unit includes a 24-bit set-frequency analog-to-digital converter and a microcontroller. The 24-bit set-frequency analog-to-digital converter is physically isolated from the analog front-end conditioning unit.

Images