A method for excitation-demodulation of a high-stability magnetic modulator without correction of the winding

Abstract

The application provides a high-stability magnetic modulator excitation demodulation method without correction winding, and relates to the technical fields of precise electromagnetic measurement and signal processing. The application adjusts the turn number difference of the excitation windings of two magnetic cores in a double-magnetic-core differential magnetic modulator, and uses a microcontroller to control the positive and negative duty ratio superposition of a low-frequency component of a square wave excitation source to equivalently inject a low-frequency current signal in the magnetic modulator; in combination with digital acquisition and Fourier analysis, a low-frequency voltage sequence and a second harmonic sequence are constructed, and a low-frequency proportional coefficient and an actual direct current proportional coefficient are calculated, so that the online calibration of a direct current measurement value is realized. The application can eliminate the measurement error caused by time and environmental changes without adding a correction winding or a correction power supply, and significantly improves the long-term measurement precision and stability of the magnetic modulator.

Description

Technical Field

[0001] This invention relates to the field of precision electromagnetic measurement and signal processing technology, and in particular to a high-stability magnetic modulator excitation and demodulation method that does not require correction windings. Background Technology

[0002] The problems caused by insulation defects in high-voltage direct current (HVDC) equipment are becoming increasingly prominent, seriously hindering the safe and stable operation of power systems. However, the progress of online monitoring technology for the insulation condition of HVDC equipment has been relatively slow, mainly because the amplitude of the most critical DC leakage current is extremely small, only around 10 μA under normal insulation conditions, which cannot be met by existing non-contact current sensing technology.

[0003] Among existing non-contact DC sensors, magnetic modulators offer advantages such as high resolution, high sensitivity, and high accuracy, making them well-suited for measuring DC leakage currents in the 10μA range. However, in practical applications, existing magnetic modulators only achieve non-contact measurement of DC currents in the hundreds of μA range. This is because, after prolonged operation, the DC proportional coefficient of the magnetic modulator is affected by factors such as ambient temperature and humidity, often exhibiting non-periodic and unexpected fluctuations, severely impacting the measurement accuracy of 10μA-level DC signals. Common methods to address this issue include adding additional correction modules, such as correction windings and correction power supplies. However, these modules may affect the performance level of the magnetic modulator and suffer from poor reliability and practicality. Therefore, a high-stability demodulation method for magnetic modulators that eliminates the need for correction windings is urgently needed to reduce the impact of DC proportional coefficient fluctuations, laying the foundation for further improving measurement accuracy. Summary of the Invention

[0004] To overcome the shortcomings of the prior art, the purpose of this invention is to provide a high-stability magnetic modulator excitation and demodulation method that does not require a calibration winding. Without increasing the calibration winding and calibration power supply, the method achieves online calibration of the DC proportional coefficient of the magnetic modulator by coordinating the control of the number of turns of the excitation winding and the duty cycle of the square wave excitation source. This effectively eliminates measurement errors caused by changes in working time and ambient temperature and humidity, and significantly improves the measurement stability and long-term accuracy of the magnetic modulator.

[0005] To achieve the above objectives, the present invention provides the following solution:

[0006] A high-stability magnetic modulator excitation and demodulation method that does not require correction windings includes:

[0007] Adjust the number of turns of the excitation windings of the two magnetic cores in the dual-core differential magnetic modulator so that the number of turns of the excitation windings of the two magnetic cores differs by a preset number of turns;

[0008] The positive and negative duty cycles of the square wave excitation source of the magnetic modulator are adjusted using a microcontroller; the positive and negative duty cycles are determined by the superposition of a constant duty cycle and a low-frequency duty cycle, so as to effectively inject a low-frequency current signal into the magnetic modulator.

[0009] The low-frequency timing data of the duty cycle of the square wave excitation source is obtained by using digital communication, and the analog signal of the output voltage of the magnetic modulator is converted into a digital signal by using digital acquisition equipment to obtain the timing data of the output voltage signal.

[0010] Based on the timing data of the multiple low-frequency timing data and the output voltage signal, a low-frequency voltage sequence and an output voltage sequence are constructed respectively.

[0011] Fourier analysis was performed on the low-frequency voltage sequence to obtain the low-frequency signal amplitude;

[0012] The output voltage sequence is divided into multiple subsequences, and Fourier analysis is performed on each subsequence to obtain the corresponding second harmonic signal amplitude.

[0013] A second harmonic sequence is constructed using the amplitudes of each of the second harmonic signals, and the actual low-frequency scaling factor of the magnetic modulator is determined based on the second harmonic sequence.

[0014] The actual DC proportional coefficient is obtained based on the actual low-frequency proportional coefficient, the factory low-frequency proportional coefficient of the magnetic modulator, and the factory DC proportional coefficient.

[0015] The calibrated high-stability DC measurement value is determined based on the second harmonic sequence and the actual DC proportionality coefficient.

[0016] Preferably, the preset number of turns ranges from 1 to 5, and the preset number of turns is less than the number of turns of the magnetic modulator excitation winding.

[0017] Preferably, the formulas for calculating the positive duty cycle and the negative duty cycle are as follows:

[0018] ;

[0019] in, For a constant duty cycle, For low frequency duty cycle, The frequency of the low-frequency duty cycle. The range is set to 0.01Hz~0.1Hz. The positive duty cycle is... For the negative duty cycle, For time variables, The amplitude coefficient of the low-frequency duty cycle is given.

[0020] Preferably, the formula for the length of the low-frequency voltage sequence or the output voltage sequence is:

[0021] ;

[0022] in, The length of the low-frequency voltage sequence or the output voltage sequence. The sampling rate of the digital acquisition device. The frequency of the low-frequency duty cycle is denoted as .

[0023] Preferably, a second harmonic sequence is constructed using the amplitudes of each of the second harmonic signals, and the actual low-frequency scaling factor of the magnetic modulator is determined based on the second harmonic sequence, including:

[0024] A second harmonic sequence is constructed using the amplitudes of each of the second harmonic signals, and Fourier analysis is performed on the second harmonic sequence to obtain the low-frequency signal output value.

[0025] The actual low-frequency scaling factor of the magnetic modulator is determined based on the amplitude of the low-frequency signal and the output value of the low-frequency signal.

[0026] Preferably, the formula for the actual DC proportional coefficient is:

[0027] ;

[0028] in, This refers to the actual DC proportional coefficient. This refers to the factory-issued DC proportional coefficient. This refers to the actual low-frequency proportionality coefficient. This refers to the factory-issued low-frequency ratio coefficient.

[0029] Preferably, the formula for the actual low-frequency proportionality coefficient is:

[0030] ;

[0031] in, This refers to the actual low-frequency proportionality coefficient. The low-frequency signal output value, The amplitude of the low-frequency signal is denoted as .

[0032] Preferably, the expression for the high-stability DC measurement value is:

[0033] ;

[0034] in, The high-stability DC measurement value is... The number of subsequences. For the first The second harmonic amplitude of each subsequence This is the actual DC proportional coefficient.

[0035] Preferably, a second harmonic sequence is constructed using the amplitudes of each of the second harmonic signals, and the actual low-frequency scaling factor of the magnetic modulator is determined based on the second harmonic sequence, including:

[0036] A zero-mean pseudo-random coding sequence is applied to the low-frequency duty cycle to form a coded low-frequency duty cycle;

[0037] The second harmonic sequence is constructed using the amplitudes of each of the second harmonic signals. The second harmonic sequence is then subjected to matched correlation processing with the pseudo-random coding sequence as a reference to obtain the output-side correlation peak amplitude.

[0038] Using the time sequence of the encoded low-frequency duty cycle as input and the pseudo-random encoded sequence as a reference, a matching correlation process is performed to obtain a low-frequency side encoding reference; the low-frequency side encoding reference is the corresponding correlation peak amplitude;

[0039] The ratio of the output-side correlated peak amplitude to the low-frequency side encoded reference is determined as the actual low-frequency proportional coefficient.

[0040] Preferably, a matching correlation process is performed using the timing of the encoded low-frequency duty cycle as input and the pseudo-random encoded sequence as a reference to obtain a low-frequency side encoded reference, including:

[0041] Within a relevant window, the timing of the encoded low-frequency duty cycle is sampled at equal intervals with a preset chip period to obtain a discrete sequence;

[0042] Generate a zero-mean pseudo-random encoded sequence locally in the same manner as at the injection end;

[0043] For the discrete sequence With the pseudo-random coding sequence Perform a matched correlation and take the normalized correlation peak value to obtain the normalized correlation peak amplitude, which serves as the coding reference for the low-frequency side. The calculation formula is: ;in, To encode the low-frequency duty cycle in the first... The discrete sequence sampled from the center of each chip; Zero mean generated locally The pseudo-random encoded sequence; This refers to the number of chips within the relevant window, i.e., the length of the discrete sequence; The chip period; For the code phase search offset and in 0 to It takes values ​​within the range; Indicates to to Summation; max This indicates that the largest amplitude value is taken within the specified offset range; The normalized correlation peak amplitude is the low-frequency side coding reference.

[0044] The present invention discloses the following technical effects:

[0045] This invention provides a high-stability magnetic modulator excitation and demodulation method that does not require a correction winding. It eliminates the need for additional correction windings, correction power supplies, or other modules. By slightly adjusting the number of turns in the excitation winding and the control method of the square wave excitation source in a traditional magnetic modulator, a low-frequency current signal with a constant amplitude can be effectively injected into the magnetic modulator, laying the foundation for improving the high-stability demodulation of the magnetic modulator.

[0046] This invention provides a high-stability magnetic modulator excitation and demodulation method that does not require correction windings. By adjusting the square wave excitation source to inject an equivalent low-frequency signal current, the low-frequency proportional coefficient can be obtained in real time, thereby realizing the online calibration of the DC proportional coefficient of the magnetic modulator. This solves the measurement error problem introduced by the DC proportional coefficient due to factors such as working time and ambient temperature and humidity, and further improves the measurement accuracy and stability of the magnetic modulator. Attached Figure Description

[0047] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1 This is a flowchart of a method provided in an embodiment of the present invention. Detailed Implementation

[0049] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only 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.

[0050] The purpose of this invention is to provide a high-stability magnetic modulator excitation and demodulation method that does not require correction windings. By introducing a controllable turns difference and low-frequency duty cycle modulation mechanism in a dual-core differential structure, the low-frequency and DC proportional coefficients of the magnetic modulator are self-calibrated. The coefficient offset caused by temperature drift and time drift can be compensated without additional hardware, thereby significantly improving the long-term stability and environmental adaptability of the measurement system while simplifying the structure.

[0051] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0052] Figure 1 The method flowchart provided in the embodiments of the present invention is as follows: Figure 1 As shown, the present invention provides a high-stability magnetic modulator excitation and demodulation method that does not require correction windings, comprising:

[0053] Step 100: Adjust the number of turns of the excitation windings of the two magnetic cores in the dual-core differential magnetic modulator so that the number of turns of the excitation windings of the two magnetic cores differs by a preset number of turns;

[0054] Step 200: Use a microcontroller to adjust the positive and negative duty cycles of the square wave excitation source of the magnetic modulator; the positive and negative duty cycles are determined by the superposition of a constant duty cycle and a low-frequency duty cycle, so as to effectively inject a low-frequency current signal into the magnetic modulator.

[0055] Step 300: Use digital communication to acquire low-frequency timing data of the duty cycle of the square wave excitation source, and use digital acquisition equipment to convert the analog signal of the output voltage of the magnetic modulator into a digital signal to obtain the timing data of the output voltage signal.

[0056] Step 400: Based on the timing data of multiple low-frequency timing data and the output voltage signal, construct the low-frequency voltage sequence and the output voltage sequence respectively;

[0057] Step 500: Perform Fourier analysis on the low-frequency voltage sequence to obtain the low-frequency signal amplitude;

[0058] Step 600: Divide the output voltage sequence into multiple subsequences on an average basis, and perform Fourier analysis on each subsequence in the output voltage sequence to obtain the corresponding second harmonic signal amplitude;

[0059] Step 700: Construct a second harmonic sequence using the amplitudes of each second harmonic signal, and determine the actual low-frequency scaling factor of the magnetic modulator based on the second harmonic sequence;

[0060] Step 800: Obtain the actual DC proportional coefficient based on the actual low-frequency proportional coefficient, the factory low-frequency proportional coefficient of the magnetic modulator, and the factory DC proportional coefficient;

[0061] Step 900: Determine the calibrated high-stability DC measurement value based on the second harmonic sequence and the actual DC proportionality coefficient.

[0062] As another optional embodiment of the present invention, the technical route of this embodiment includes the following steps:

[0063] S1. Based on a conventional dual-core differential magnetic modulator, the number of turns in the excitation windings of the two cores is manually adjusted so that the difference between their number of turns is significant. N 1 turn;

[0064] S2. Use a microcontroller to adjust the positive and negative duty cycles of the square wave excitation source of the magnetic modulator. D p and D n They are powered by a constant duty cycle. D dc Low frequency duty cycle D ac These are superimposed, thus effectively injecting low-frequency current signals into the magnetic modulator;

[0065] S3. Use digital communication to acquire low-frequency timing data of the duty cycle of the square wave excitation source; use digital acquisition equipment to convert the analog output voltage signal of the magnetic modulator into a digital signal to obtain the timing data of the output voltage signal;

[0066] S4. Utilize N a A low-frequency voltage sequence is constructed from the timing data of the low-frequency timing data and the output voltage signal. U ac Output voltage sequence U d ;

[0067] S5. For low-frequency voltage sequences U ac Perform Fourier analysis to obtain the amplitude of the low-frequency signal. U ac_f ;

[0068] S6. Output voltage sequence U d Average score N 2 A sequence, and for each voltage sequence U d.i (i=1,2,…, N 2 Fourier analysis was performed to obtain the amplitude of the second harmonic signal for each sequence. U d.i ;

[0069] S7. Utilize N2 The amplitude of the second harmonic signal U d.i Constructing second harmonic sequences U d2 The low-frequency signal output value was obtained by performing Fourier analysis on it. U d2_f ;

[0070] S8. Utilizing the amplitude of low-frequency signals U ac_f With low-frequency signal output value U d2_f Obtain the actual low-frequency scaling factor of the magnetic modulator CF ac_real ;

[0071] S9. Based on the factory low-frequency ratio coefficient of the magnetic modulator. CF ac_set DC proportional coefficient CF dc_set Obtain the actual DC proportional coefficient CF dc_real ;

[0072] S10. Utilizing second harmonic sequences U d2 Compared with the actual DC proportional coefficient CF dc_real Obtain high-stability DC measurement values ​​after calibration I m .

[0073] Optionally, the number of turns in step S1 N 1 can usually be set to 1~5, which is much smaller than the number of turns of the excitation winding of the magnetic modulator.

[0074] Optionally, the positive and negative duty cycles of the square wave excitation source in step S2... D p and D n The calculation expression is as follows:

[0075]

[0076] Among them, constant duty cycle D dc It is usually set to 0.5, low-frequency duty cycle. D ac The coefficient is usually set to 0.01~0.1, low-frequency duty cycle D ac frequency It is usually set to 0.01~0.1 Hz.

[0077] Optionally, the sampling rate of the digital acquisition device in step S3 is defined as follows: .

[0078] Optionally, the low-frequency voltage sequence in step S4 U ac Output voltage sequence U d length N a It is determined by the following formula:

[0079]

[0080] Optionally, the amplitude of the low-frequency signal in step S5 U ac_f This refers to the low-frequency range in the Fourier spectrum. The signal amplitude at that location.

[0081] Optionally, the number of sequences in step S6 N 2 Usually more than 100;

[0082] Optionally, the low-frequency signal output value in step S7 U d2_f This refers to the low-frequency range in the Fourier spectrum. The signal amplitude at that location.

[0083] Optionally, the actual low-frequency ratio coefficient in step S8 CF ac_real The calculation expression is: .

[0084] Optionally, the actual DC proportional coefficient in step S9 dc_real The expression is:

[0085]

[0086] Optionally, in step S10, the high-stability DC measurement value is... I m The expression is:

[0087]

[0088] As an optional implementation, the specific steps for demodulating the current value to be measured in this embodiment are as follows:

[0089] Step S1: Based on the conventional dual-core differential magnetic modulator, adjust the number of turns of the excitation windings of the two magnetic cores of the magnetic modulator so that the difference between their number of turns is significant. N 1 = 1 turn;

[0090] Step S2: Adjust the positive and negative duty cycles of the square wave excitation source of the magnetic modulator using a microcontroller. D p and D n This effectively injects a low-frequency current signal into the magnetic modulator:

[0091]

[0092] Step S3: Acquire low-frequency timing data of the duty cycle of the square wave excitation source using digital communication; use digital acquisition equipment to sample at a rate of... The analog signal of the output voltage of the magnetic modulator is converted into a digital signal to obtain the timing data of the output voltage signal;

[0093] Step S4, using A low-frequency voltage sequence is constructed from the low-frequency timing data and the timing data of the output voltage signal. U ac Output voltage sequence U d ;

[0094] Step S5: Process the low-frequency voltage sequence U ac Fourier analysis was performed to obtain the signal amplitude at a low frequency of 0.1 Hz. U ac_f ;

[0095] Step S6: Convert the output voltage sequence U d Average score N 2 =100 sequences, and for each voltage sequence U d.i Fourier analysis was performed on (i=1,2,…,100) to obtain the amplitude of the second harmonic signal for each sequence. U d.i ;

[0096] Step S7: Utilize the amplitude values ​​of the 100 second harmonic signals U d.i Constructing second harmonic sequences U d2 The signal output value at a low frequency of 0.1 Hz was obtained by performing Fourier analysis on the signal. U d2_f ;

[0097] Step S8: Utilize the amplitude of the low-frequency signal U ac_f With the low-frequency signal output value U d2_f Obtain the actual low-frequency scaling factor of the magnetic modulatorCF ac_real :

[0098]

[0099] Step S9: Based on the factory low-frequency ratio coefficient of the magnetic modulator CF ac_set DC proportional coefficient CF dc_set Obtain the actual DC proportional coefficient CF dc_real ;

[0100]

[0101] Step S10: Utilize the second harmonic sequence U d2 Compared with the actual DC proportional coefficient CF dc_real Obtain high-stability DC measurement values ​​after calibration I m :

[0102]

[0103] Specifically, this embodiment also discloses another calculation process for the actual low-frequency proportional coefficient, namely, after step S2, it further includes:

[0104] A pseudo-random coding sequence with zero mean is applied to the low-frequency duty cycle to form a coded low-frequency duty cycle;

[0105] After obtaining the second harmonic sequence, a matching correlation process is performed on the second harmonic sequence with a pseudo-random coding sequence as a reference to obtain the correlation peak amplitude on the output side.

[0106] Using the time sequence of the encoded low-frequency duty cycle as input and a pseudo-random encoded sequence as a reference, a matching correlation process is performed to obtain the low-frequency side encoded reference; the low-frequency side encoded reference is the corresponding correlation peak amplitude.

[0107] The ratio of the output-side correlation peak amplitude to the low-frequency side coding reference is determined as the actual low-frequency proportional coefficient.

[0108] Specifically, in this embodiment, after step S2, the controller superimposes a zero-mean pseudo-random encoded sequence onto the existing low-frequency duty cycle to obtain the encoded low-frequency duty cycle used to drive the square wave excitation source. The pseudo-random encoded sequence refers to a binary sequence with symmetrical values ​​and a mean of zero. The sequence length is preferably 127 or 255. The chip duration refers to the time interval during which a single chip remains unchanged, preferably 50 milliseconds or 100 milliseconds. The encoding amplitude is preferably 2% to 3% of the amplitude of the basic low-frequency duty cycle. To ensure that the injection does not introduce additional magnetic bias, the encoding amplitude does not exceed the upper limit of the above range. The controller aligns the chip boundary with the duty cycle update timing. If necessary, it performs a slight correction on the duty cycle refresh clock to make the chip switching point coincide with the duty cycle refresh point. If there is a small timing error in the field equipment, it is compensated by code phase search in subsequent processing. Code phase search refers to gradually shifting the alignment position of the reference sequence and the measured sequence within a complete sequence period to find the offset point with the highest correlation.

[0109] In this embodiment, after obtaining the second harmonic sequence according to the existing process, a correlation window is set and matching correlation processing is performed. The correlation window refers to the data interval participating in one matching correlation calculation. Its duration is preferably exactly covering one complete sequence period. For example, when the sequence length is 127 and the chip duration is 50 milliseconds, the correlation window duration is approximately 6.35 seconds. Matching correlation processing involves calculating the correlation degree for all possible code phase offsets within a correlation window, using a known pseudo-random coded sequence as a reference, and taking the correlation result with the largest amplitude within the correlation window as the correlation peak amplitude. To enhance robustness, a moving average is performed on the correlation peak amplitudes of multiple consecutive correlation windows. The moving average length is preferably 3 to 5 correlation windows, and an outlier removal strategy is adopted, with the outlier discrimination threshold preferably set at 60% of the baseline correlation amplitude. The output-side correlation peak amplitude obtained after the above processing serves as the unique response fingerprint of the measured output to the known coded excitation, used for subsequent coefficient calculation.

[0110] This embodiment synchronously records the timing of the encoded low-frequency duty cycle and uses the same correlation window, code phase search granularity, moving average length, and outlier discrimination threshold as the output side for matching correlation processing. This yields a low-frequency side encoding reference, which refers to the maximum correlation amplitude obtained after aligning with the reference sequence within each correlation window. This reference serves as a normalization benchmark for the effective intensity of the excitation end, offsetting uncertainties caused by amplitude setting errors, quantization errors, or nonlinearities of the execution unit. To ensure consistency, this embodiment uses the same sequence length, the same chip duration, and the same correlation window start and end times on both the output and low-frequency sides, and performs window alignment under the same time reference.

[0111] This embodiment calculates the ratio of the output-side correlation peak amplitude to the low-frequency side coding reference to obtain the actual low-frequency proportional coefficient. To suppress the influence of occasional noise, a finite-length moving average and outlier removal are applied to the obtained coefficient sequence. The moving average length is preferably 3 to 5 correlation windows, and the outlier discrimination threshold is preferably 60% of the baseline correlation amplitude. The actual low-frequency proportional coefficient is written into the parameter area as an intermediate parameter and passed to subsequent steps, participating in the calculation of the DC proportional coefficient and the high-stability DC measurement value. For ease of implementation and review, this embodiment also records the implementation parameters such as sequence length, chip duration, correlation window duration, outlier discrimination threshold, moving average length, and coding amplitude range. Example values ​​are given: sequence length 127 or 255, chip duration 50 ms or 100 ms, correlation window covering a complete sequence period, outlier discrimination threshold 60%, moving average length 3 to 5 correlation windows, and coding amplitude 2% to 3% of the baseline low-frequency duty cycle amplitude. The above parameters can be set within the range as needed on site without changing the overall technical solution logic.

[0112] Specifically, using the timing sequence of the encoded low-frequency duty cycle as input and the pseudo-random encoded sequence as a reference, a matching correlation process is performed to obtain a low-frequency side encoded reference, including:

[0113] Within a relevant window, the timing of the encoded low-frequency duty cycle is sampled at equal intervals with a preset chip period to obtain a discrete sequence;

[0114] Generate a zero-mean pseudo-random encoded sequence locally in the same manner as at the injection end;

[0115] For the discrete sequence With the pseudo-random coding sequence Perform a matched correlation and take the normalized correlation peak value to obtain the normalized correlation peak amplitude, which serves as the coding reference for the low-frequency side. The calculation formula is: ;in, To encode the low-frequency duty cycle in the first... The discrete sequence sampled from the center of each chip; Zero mean generated locally The pseudo-random encoded sequence; This refers to the number of chips within the relevant window, i.e., the length of the discrete sequence; The chip period; For the code phase search offset and in 0 to It takes values ​​within the range; Indicates to to Summation; max This indicates that the largest amplitude value is taken within the specified offset range; The normalized correlation peak amplitude is the low-frequency side coding reference.

[0116] Specifically, this embodiment sets a correlation window within a time interval covering the complete encoded sequence period. The "correlation window" refers to the data interval participating in a single matching correlation calculation, used to ensure alignment and peak determination are completed within a single processing step. Within this correlation window, the timing of the "coded low-frequency duty cycle" is sampled at equal intervals with a preset "chip period," and the sampling time is taken at the center of each chip to reduce the impact of sampling edge jitter. The "coded low-frequency duty cycle" refers to the duty cycle timing formed by superimposing pseudo-random encoding on an existing low-frequency duty cycle, used to inject discernible weak stimuli without changing the hardware structure. The "chip period" refers to the time interval during which a single chip remains unchanged. Example parameters: If the sequence length is 127 or 255, and the chip period is 50 milliseconds or 100 milliseconds, then when the sequence length is 127 and the chip period is 50 milliseconds, the duration of a single correlation window is approximately 6.35 seconds.

[0117] This embodiment generates a "pseudo-random encoded sequence" locally that is identical to the one injected at the receiving end. This refers to a binary sequence with symmetrical values ​​and a mean of zero, exhibiting good autocorrelation properties, and is used to achieve high-resolution identification at the receiving end. To ensure consistency, the local reference sequence and the injected sequence use the same sequence length and chip period, typically the same as described above (e.g., length 127 or 255, chip period 50 milliseconds or 100 milliseconds). To handle potential startup deviations or link delays, this embodiment performs a "code phase search" within a correlation window. This refers to progressively shifting the relative positions of the reference sequence and the discrete sequence under test within the correlation window and calculating a consistency score. The preferred step size is one chip, until the alignment point with the highest consistency is found, achieving precise synchronization of their time bases.

[0118] This embodiment performs "matching correlation processing" on the sampled discrete sequence and the local pseudo-random encoded sequence. This refers to traversing all possible code-phase offsets within a correlation window, calculating the similarity metric between the two sequences at each offset point, and selecting the result with the largest amplitude as the candidate peak of that window. To suppress amplitude setting errors and quantization errors, the correlation results are represented using a "normalized" method, ensuring that the results only reflect shape similarity and are as unaffected as possible by amplitude scale. The "normalized correlation peak amplitude" refers to the maximum similarity metric obtained within the current correlation window after normalization, typically ranging from 0 to 1, with larger values ​​indicating better alignment. To improve stability, an "outlier detection threshold" can be set and smoothed, typically set to a threshold of 60%, with a moving average length of 3 to 5 correlation windows.

[0119] This embodiment uses the timing of the encoded low-frequency duty cycle as input and employs the same correlation window length, sequence length, chip period, code phase search granularity, and outlier handling strategy as the output side, performing the same matching correlation and normalization process as described above. The "low-frequency side encoded reference" refers to the maximum normalized correlation amplitude obtained after aligning with the reference sequence within the current correlation window. This amplitude serves as the normalization benchmark for the effective intensity of the excitation end, offsetting uncertainties caused by injection amplitude setting errors, quantization errors, or nonlinearity of the execution unit. To ensure data consistency at both ends, this embodiment aligns the start and end times of the correlation windows on the low-frequency side and the output side to the same time base. The parameter examples are the same as described above (e.g., sequence length 127, chip period 50 milliseconds, correlation window approximately 6.35 seconds, outlier threshold 60%, moving average length 3 to 5 correlation windows).

[0120] This embodiment compares the normalized correlation peak amplitude obtained from the output side with the low-frequency side coding reference and takes the ratio to obtain the "actual low-frequency proportional coefficient". This name refers to the online estimation result reflecting the low-frequency proportional relationship between the injection end and the output end, and is used to compensate for coefficient offsets caused by environmental changes and device drift. To suppress the influence of occasional noise, this embodiment further applies finite-length smoothing and outlier removal to the coefficient sequence. The typical smoothing length is 3 to 5 correlation windows, and the outlier threshold is 60%. The final actual low-frequency proportional coefficient is written into the parameter area as an intermediate parameter and is passed to subsequent steps to participate in the calculation of DC proportional coefficient and high-stability DC measurement value according to the existing process. Without changing the overall technical solution logic, the above parameters can be set within a given range (such as coding amplitude of 2% to 3% of the basic low-frequency duty cycle amplitude, sequence length of 127 or 255, and chip period of 50 milliseconds or 100 milliseconds) to meet the stability and real-time requirements of different application scenarios.

[0121] The high-stability magnetic modulator excitation and demodulation method of the present invention aims to solve the measurement error problem caused by the influence of factors such as working time, ambient temperature and humidity on the DC proportional coefficient of existing magnetic modulators without adding additional correction windings, correction power supplies and other modules, and further improve the accuracy and stability of the measurement.

[0122] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0123] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A high-stability magnetic modulator excitation and demodulation method that requires no correction winding, characterized in that, include: Adjust the number of turns of the excitation windings of the two magnetic cores in the dual-core differential magnetic modulator so that the number of turns of the excitation windings of the two magnetic cores differs by a preset number of turns; The positive and negative duty cycles of the square wave excitation source of the magnetic modulator are adjusted using a microcontroller; the positive and negative duty cycles are determined by the superposition of a constant duty cycle and a low-frequency duty cycle, so as to effectively inject a low-frequency current signal into the magnetic modulator. The low-frequency timing data of the duty cycle of the square wave excitation source is obtained by using digital communication, and the analog signal of the output voltage of the magnetic modulator is converted into a digital signal by using digital acquisition equipment to obtain the timing data of the output voltage signal. Based on the timing data of the multiple low-frequency timing data and the output voltage signal, a low-frequency voltage sequence and an output voltage sequence are constructed respectively. Fourier analysis was performed on the low-frequency voltage sequence to obtain the low-frequency signal amplitude; The output voltage sequence is divided into multiple subsequences, and Fourier analysis is performed on each subsequence to obtain the corresponding second harmonic signal amplitude. A second harmonic sequence is constructed using the amplitudes of each of the second harmonic signals, and the actual low-frequency scaling factor of the magnetic modulator is determined based on the second harmonic sequence. The actual DC proportional coefficient is obtained based on the actual low-frequency proportional coefficient, the factory low-frequency proportional coefficient of the magnetic modulator, and the factory DC proportional coefficient. The calibrated high-stability DC measurement value is determined based on the second harmonic sequence and the actual DC proportionality coefficient.

2. The high-stability magnetic modulator excitation and demodulation method without correction windings according to claim 1, characterized in that, The preset number of turns ranges from 1 to 5, and the preset number of turns is less than the number of turns of the magnetic modulator excitation winding.

3. The high-stability magnetic modulator excitation and demodulation method without correction windings according to claim 1, characterized in that, The formulas for calculating the positive duty cycle and the negative duty cycle are as follows: ； in, For a constant duty cycle, For low frequency duty cycle, The frequency of the low-frequency duty cycle. The range is set to 0.01Hz~0.1Hz. The positive duty cycle is... For the negative duty cycle, For time variables, The amplitude coefficient of the low-frequency duty cycle is given.

4. The high-stability magnetic modulator excitation and demodulation method without correction windings according to claim 1, characterized in that, The formula for the length of the low-frequency voltage sequence or the output voltage sequence is: ； in, The length of the low-frequency voltage sequence or the output voltage sequence. The sampling rate of the digital acquisition device. The frequency of the low-frequency duty cycle is denoted as .

5. The high-stability magnetic modulator excitation and demodulation method without correction windings according to claim 1, characterized in that, Constructing a second harmonic sequence using the amplitudes of each of the aforementioned second harmonic signals, and determining the actual low-frequency scaling factor of the magnetic modulator based on the second harmonic sequence, including: A second harmonic sequence is constructed using the amplitudes of each of the second harmonic signals, and Fourier analysis is performed on the second harmonic sequence to obtain the low-frequency signal output value. The actual low-frequency scaling factor of the magnetic modulator is determined based on the amplitude of the low-frequency signal and the output value of the low-frequency signal.

6. The high-stability magnetic modulator excitation and demodulation method without correction windings according to claim 5, characterized in that, The formula for the actual DC proportional coefficient is: ； in, This refers to the actual DC proportional coefficient. This refers to the factory-issued DC proportional coefficient. This refers to the actual low-frequency proportionality coefficient. The low-frequency ratio coefficient is the one used at the factory.

7. The high-stability magnetic modulator excitation and demodulation method without correction windings according to claim 1, characterized in that, The formula for the actual low-frequency ratio coefficient is: ； in, This refers to the actual low-frequency proportionality coefficient. The low-frequency signal output value, The amplitude of the low-frequency signal is denoted as .

8. The high-stability magnetic modulator excitation and demodulation method without correction windings according to claim 1, characterized in that, The expression for the high-stability DC measurement value is: ； in, The high-stability DC measurement value is... The number of subsequences. For the first The second harmonic amplitude of each subsequence This is the actual DC proportional coefficient.

9. The high-stability magnetic modulator excitation and demodulation method without correction windings according to claim 1, characterized in that, Constructing a second harmonic sequence using the amplitudes of each of the aforementioned second harmonic signals, and determining the actual low-frequency scaling factor of the magnetic modulator based on the second harmonic sequence, including: A zero-mean pseudo-random coding sequence is applied to the low-frequency duty cycle to form a coded low-frequency duty cycle; The second harmonic sequence is constructed using the amplitudes of each of the second harmonic signals. The second harmonic sequence is then subjected to matched correlation processing with the pseudo-random coding sequence as a reference to obtain the output-side correlation peak amplitude. Using the time sequence of the encoded low-frequency duty cycle as input and the pseudo-random encoded sequence as a reference, a matching correlation process is performed to obtain a low-frequency side encoding reference; the low-frequency side encoding reference is the corresponding correlation peak amplitude; The ratio of the output-side correlated peak amplitude to the low-frequency side encoded reference is determined as the actual low-frequency proportional coefficient.

10. The high-stability magnetic modulator excitation and demodulation method without correction windings according to claim 9, characterized in that, Using the time sequence of the encoded low-frequency duty cycle as input and the pseudo-random encoded sequence as a reference, a matching correlation process is performed to obtain a low-frequency side encoded reference, including: Within a relevant window, the timing of the encoded low-frequency duty cycle is sampled at equal intervals with a preset chip period to obtain a discrete sequence; Generate a zero-mean pseudo-random encoded sequence locally in the same manner as at the injection end; For the discrete sequence With the pseudo-random coding sequence Perform a matched correlation and take the normalized correlation peak value to obtain the normalized correlation peak value amplitude, which serves as the coding reference for the low-frequency side. The calculation formula is: ;in, To encode the low-frequency duty cycle in the first... The discrete sequence sampled from the center of each chip; Zero mean generated locally The pseudo-random encoded sequence; This refers to the number of chips within the relevant window, i.e., the length of the discrete sequence; The chip period; For the code phase search offset and in 0 to The range of values; indicating that for to Summation; max This indicates that the largest amplitude value is taken within the specified offset range; The normalized correlation peak amplitude is the low-frequency side coding reference.

Images