High-precision voltage measurement method and system based on multi-parameter adaptive compensation

Abstract

The invention discloses a high-precision voltage measurement method and system based on multi-parameter adaptive compensation, and relates to the technical field of intelligent voltage measurement. The high-precision voltage measurement method and system based on multi-parameter adaptive compensation comprises the steps that S1, voltage signal data and disturbance source characteristic data are collected, preprocessing is completed, and an analysis basis is constructed; s2, analyzing voltage excursion in combination with disturbance current characteristics, and dynamically adjusting a compensation path; s3, evaluating a compensation correction effect, and driving a compensation output strategy in real time; and S4, comprehensively analyzing the offset characteristics and the correction result, judging an abnormal section and triggering adaptive adjustment. The problems that disturbance components cannot be comprehensively extracted from multiple parameters such as PWM modulation frequency, wire harness impedance and working condition load change and a compensation function cannot be reversely constructed in an existing method, and a measured value does not have stability are solved.

Description

Technical Field

[0001] This invention relates to the field of intelligent voltage measurement technology, specifically to a high-precision voltage measurement method and system based on multi-parameter adaptive compensation. Background Technology

[0002] With the continuous improvement of automation levels in industrial settings, voltage measurement technology has been widely applied to the sensing of operating status, monitoring of power loads, and analysis of energy consumption in various mechatronic equipment. Current mainstream measurement methods typically employ high-speed analog-to-digital converters and high-precision voltage sensors for synchronous sampling, extracting instantaneous voltage values ​​at key nodes through standard sampling resistors, voltage dividers, or Hall effect voltage modules. To meet the adaptability requirements of non-steady-state operating conditions, some measurement devices also integrate basic signal processing logic such as digital filtering, level holding, and moving averages to improve the anti-interference capability and timing consistency of measurement data. In complex operating conditions, such as variable frequency speed control, pulse regulation, and nonlinear load operation scenarios, voltage measurement modules are often synchronously acquired along with multi-channel signals such as current, electromagnetic interference, and power load. These signals are then uniformly calculated and archived through a centralized processing unit, constructing a voltage data chain with temporal continuity and spatial distribution characteristics. This provides crucial electrical signal support for subsequent control logic judgments and equipment operation strategy adjustments.

[0003] For example, invention patent CN115308475B discloses a voltage measurement method, a voltage measurement circuit, and a voltage measurement device, relating to a voltage measurement method, a voltage measurement circuit, and a voltage measurement device. The method includes: controlling a first voltage detection circuit to detect the conductor under test, obtaining a first detection voltage; the conductor under test is any phase of a three-phase circuit; obtaining a first total circuit voltage based on a reference signal from the first voltage detection circuit and the first detection voltage; controlling a second voltage detection circuit to detect the conductor under test, obtaining a second detection voltage; the second voltage detection circuit has the same structure as the first voltage detection circuit, and the plates in the first and second voltage detection circuits are at a preset angle; obtaining a second total circuit voltage based on the reference signal from the second voltage detection circuit and the second detection voltage; and obtaining the voltage value of the conductor under test based on the first and second total circuit voltages. This method can effectively improve the accuracy of the voltage measurement results of the conductor under test.

[0004] For example, invention patent CN115219762B discloses a hydropower station safety measurement and control system and method based on digital twins. The voltage sensor includes: a sensing module for sensing the spatial coupling capacitance of the conductor under test through a differential probe and outputting a differential signal, wherein the conductor under test has a first voltage angular frequency; a reference signal module electrically connected to the sensing module for generating a reference voltage signal with a second voltage angular frequency, which is different from the first voltage angular frequency; and a signal processing module electrically connected to both the sensing module and the reference signal module for obtaining the differential signal of the conductor under test and the reference voltage differential signal from the differential signal through signal processing, and determining the voltage information of the conductor under test based on the differential signal of the conductor under test, the differential signal of the reference voltage differential signal, and the reference voltage signal. This invention introduces a reference voltage signal to process the differential signal detected by the sensing module, overcoming the problem that the measurement result is affected by the size of the coupling capacitance, and realizing the measurement of AC voltage amplitude.

[0005] However, existing voltage monitoring methods mostly use fixed sampling frequencies and uniform filtering parameters, lacking the analytical analysis and response modeling of the time-frequency structure of disturbance signals. This makes it impossible to identify and quantitatively model the source of interference based on multiple parameters such as PWM modulation frequency, harness impedance changes, and power load fluctuations. Furthermore, the interference filtering process often relies on static threshold settings, ignoring the dynamic response characteristics of the measurement circuit to disturbance signals under different operating conditions. This can easily lead to false filtering or insufficient compensation, thus affecting the reliability and effectiveness of voltage measurement results.

[0006] Therefore, there is an urgent need for a high-precision voltage measurement method and device based on multi-parameter adaptive compensation. Summary of the Invention

[0007] Technical problems to be solved

[0008] To address the shortcomings of existing technologies, this invention provides a high-precision voltage measurement method and system based on multi-parameter adaptive compensation. This solves the problem that existing methods cannot comprehensively extract disturbance components from multiple parameters such as PWM modulation frequency, harness impedance, and load changes, and construct compensation functions in reverse, resulting in unstable measured values.

[0009] Technical solution

[0010] To achieve the above objectives, the present invention provides the following technical solution: a high-precision voltage measurement method and system based on multi-parameter adaptive compensation, comprising: S1, acquiring voltage signal data and disturbance source characteristic data during the voltage detection process, and preprocessing the acquired voltage signal data and disturbance source characteristic data; S2, analyzing the disturbance offset characteristics in the measured voltage in conjunction with the disturbance current, and dynamically adjusting the compensation path structure based on the analysis results; S3, evaluating the voltage compensation correction in the measured voltage, and driving the compensation voltage output strategy in real time based on the evaluation results; S4, using the disturbance offset characteristic analysis results and the voltage compensation correction evaluation results as inputs, comprehensively analyzing the stability of the voltage signal, and dynamically determining abnormal sections and triggering local adaptive adjustment based on the analysis results.

[0011] Further, the specific steps for acquiring voltage signal data and disturbance source characteristic data during the voltage detection process are as follows: Acquire voltage signal data, including: instantaneous voltage value at the measurement point, voltage change lag time, voltage change rate, and current load current; simultaneously record the maximum voltage change in the current interference segment and the average voltage change within the current window; form an original measured voltage sequence from the acquired instantaneous voltage values ​​at the measurement points in chronological order, and calculate the voltage difference amplitude by comparing the current instantaneous voltage value with the previous instantaneous voltage value in each sampling period; use the original voltage measurement sequence as input to calculate the average voltage value of all measurement points within the sliding window, and then square and average the deviation between the instantaneous voltage value at each measurement point and the average voltage value to obtain the original voltage signal variance within the current window; acquire disturbance source characteristic data, including: real-time disturbance current value, number of high-frequency interferences, compensation voltage value, disturbance duration, and compensation voltage fluctuation amplitude; compare the ratio of the difference between the compensation voltage value and the instantaneous voltage value at the measurement point in real time, and record it as the relative error ratio.

[0012] Furthermore, the specific steps for preprocessing the acquired voltage signal data and disturbance source feature data are as follows: Standardization and anomaly suppression preprocessing are performed on the voltage signal data and disturbance source feature data; the voltage values ​​at each measurement point are uniformly normalized to improve the consistency of multi-measurement point analysis; a median filtering strategy is introduced into the original measured voltage sequence to smooth out local outlier fluctuations and remove transient spikes; the voltage change lag time is calibrated using interpolation reconstruction to calibrate the voltage response starting point, ensuring the accuracy of the lag parameter used for time-series compensation path adjustment; the disturbance feature data are uniformly constructed into a structured disturbance factor vector, which serves as the input variable for the disturbance response mapping function and inverse compensation function construction modules.

[0013] Further, the specific steps for analyzing the disturbance offset characteristics in the measured voltage in conjunction with the disturbance current are as follows: The level change sequence of the pulse width modulation control signal is directly acquired using a digital storage oscilloscope. The instantaneous phase extraction and time-frequency energy reconstruction of the level change sequence are performed using Hilbert transform and short-time Fourier analysis. Then, the time interval between consecutive rising edges is extracted using an adaptive threshold edge enhancement algorithm to form a periodic time sequence. The PWM modulation frequency is further calculated by calculating the average period of the periodic time sequence. The dynamic disturbance correction value is obtained by multiplying the real-time disturbance current value by the instantaneous voltage value at the measurement point and dividing by the voltage change lag time plus one. The trend response adjustment value is obtained by multiplying the PWM modulation frequency by the voltage difference value and dividing by the current load current, and finally multiplying by the corresponding frequency domain disturbance adjustment factor. The disturbance coupling gain value is obtained by multiplying the number of high-frequency interferences by the disturbance duration, adding one, and then taking the logarithm. The dynamic disturbance correction value, the trend response adjustment value, and the disturbance coupling gain value are added together to obtain the disturbance response mapping value.

[0014] Furthermore, the specific steps for dynamically adjusting the compensation path structure based on the analysis results are as follows: Real-time comparison of the disturbance response mapping value and the response mapping threshold of the current measurement point: When the disturbance response mapping value is less than or equal to the response mapping threshold, it is determined to be a controllable disturbance segment. The original sampling frequency, compensation intensity, and boundary screening window remain unchanged, and trend tracking and data caching continue to be performed based on the current voltage input. When the disturbance response mapping value is greater than the response mapping threshold, it is determined to be a significant interference segment. The local reverse correction process is immediately triggered: Short-time Fourier transform is used to mark the frequency band of the interference component, the abnormal response interval is locked, the stored steady-state samples are linked for difference correction, the current filter cutoff frequency is dynamically adjusted, and the corrected voltage value of the significant interference segment is reconstructed. Finally, the error-suppressed signal is output and marked as compensated.

[0015] Further, the specific steps for evaluating the voltage compensation correction in the measured voltage are as follows: A PXIe data acquisition card is used to continuously sample the original voltage signal in the target measurement channel at high frequency, forming an equally spaced time-series voltage sampling sequence. The sequence is then truncated into fixed-length data segments using a sliding window as the signal analysis window for the current time point. A fast Fourier transform is applied to the voltage sequence within the window to map the time-domain signal to the frequency domain, obtaining the corresponding spectral structure. In the spectral structure, the amplitude of the dominant frequency component with the largest energy within the current window is extracted and recorded as the frequency interference amplitude at the current moment. Simultaneously, the average of the frequency interference amplitudes over multiple time periods is calculated to obtain the average amplitude of the frequency band. The difference between the maximum voltage change in the current interference segment and the average voltage change within the current window is calculated, squared, and then 1 is added to obtain the natural logarithm, yielding the sudden change energy fluctuation value. The absolute value of the difference between the frequency interference amplitude and the average amplitude of the frequency band is calculated, multiplied by the frequency domain offset adjustment factor, and then the square root is obtained to obtain the frequency disturbance deviation value.

[0016] Divide the compensation smoothing adjustment factor by the value of the original voltage signal variance plus 1 within the current window to obtain the voltage residual correction value; subtract the frequency disturbance deviation value from the sudden energy fluctuation value and add the voltage residual correction value to obtain the reverse compensation output value.

[0017] Further, the specific steps of the real-time compensation voltage output strategy based on the evaluation results are as follows: The reverse compensation output value of each sampling period is calculated by difference through a sliding window to obtain the rate of change sequence within each period, and the average is calculated to obtain the average compensation change value; based on the trend of the reverse compensation output value change at the current measuring point, the voltage data stability state is dynamically determined and the following processing is performed: when the difference between the rate of change of the reverse compensation output value compared to the previous moment and the average compensation change value is less than the rate of change threshold, it is determined to be a sufficiently corrected segment, and the current compensation result is directly used as the final voltage value and written into the voltage sequence buffer in the monitoring link; when the rate of change of the reverse compensation output value within two consecutive sliding windows is one positive value and one negative value, and the change... If the rate of change of the compensation change is greater than the rate of change of the average compensation change, it is determined to be an unstable compensation section, and a fine-tuning feedback mechanism is immediately triggered: by narrowing the analysis window, increasing the weight of the frequency domain correction term, and simultaneously introducing the lateral comparison results of adjacent measurement points, a local weighted correction is performed to ensure stable output of the voltage transition section; when the difference between the rate of change of the reverse compensation output value and the average compensation change is greater than or equal to the rate of change threshold, and the sign of the rate of change is positive or negative for three consecutive sampling periods, it is determined to be a pulse disturbance evolution section, and a delayed confirmation strategy is entered: the compensation result of the current point is retained but not output immediately, and the smoothed and corrected voltage value is written to the main output channel after the compensation results of the next two sampling points confirm that the trend is consistent.

[0018] Furthermore, the specific steps for comprehensively analyzing the stability of the voltage signal using the disturbance offset characteristic analysis results and the voltage compensation correction evaluation results as input are as follows: Add one to the voltage change rate at the current sampling point, take the natural logarithm, and then multiply it by the squared value after subtracting the relative error ratio to obtain the comprehensive influence amplification value; square the disturbance response mapping value, divide it by the value of the compensated voltage fluctuation amplitude, add it to the reverse compensation output value, add one, and take the natural logarithm to obtain the compensation suppression driving value; divide the comprehensive influence amplification value by the compensation suppression driving value to obtain the stability verification value.

[0019] Furthermore, the specific steps for dynamically determining abnormal sections and triggering local adaptive adjustment based on the analysis results are as follows: Real-time comparison of the stability verification value and stability evaluation threshold of the current measurement point. The stability evaluation threshold is divided into a first stability threshold and a second stability threshold. When the stability verification value is less than or equal to the second stability threshold: the section is determined to be a high-disturbance, low-stability interference section. The disturbance source re-analysis process is automatically invoked, and the PWM modulation frequency change rate, wiring equivalent impedance difference, and load current response rate are re-extracted in the region. Local compensation reconstruction is implemented in conjunction with the error amplification trend of the current compensation function, and short-time multi-window average sliding suppression processing is performed to output a low-weight steady-state voltage. When the stability verification value is greater than the second stability threshold... When the value is less than or equal to the first stability threshold: the region is determined to be a disturbance transition fluctuation segment, and enters the dynamic compensation self-verification mechanism: the compensation residual changes of the two cycles before and after the current measurement point are loaded, the frequency domain suppression factor weight of the segment is reallocated, and frequency band reconstruction and intra-time window contrast enhancement are performed. When the self-verification trend is in a convergent state, it is included in the main output channel. When the residual continues to accumulate, it is upgraded to a buffer to be reconstructed, and the output mark is delayed. When the stability verification value is greater than the first stability threshold: the region is determined to be a high-confidence stable segment, and is automatically marked as the final correction area. The compensation voltage is directly used as the final output of the main measurement channel. At the same time, the disturbance response characteristics of the current segment, the compensation residual trajectory and the frequency domain suppression configuration are bound to generate structured compensation successful sample data.

[0020] The second aspect of this invention provides a high-precision voltage measurement system based on multi-parameter adaptive compensation, comprising: a multi-parameter disturbance source extraction module, which collects voltage signal data and disturbance source feature data during the voltage detection process and preprocesses the collected voltage signal data and disturbance source feature data; a disturbance response mapping calculation module, which analyzes the disturbance offset characteristics in the measured voltage in conjunction with the disturbance current and dynamically adjusts the compensation path structure based on the analysis results; a reverse compensation function construction module, which evaluates the voltage compensation correction in the measured voltage and drives the compensation voltage output strategy in real time based on the evaluation results; and a measurement stability verification and correction module, which takes the disturbance offset feature analysis results and the voltage compensation correction evaluation results as inputs, performs a comprehensive analysis of the stability of the voltage signal, and dynamically determines abnormal sections and triggers local adaptive adjustment based on the analysis results.

[0021] Beneficial effects

[0022] The present invention has the following beneficial effects:

[0023] (1) The high-precision voltage measurement method and system based on multi-parameter adaptive compensation realizes the joint perception of high-frequency switching interference, load change and PWM modulation factors by constructing disturbance response mapping value, thereby improving the identification accuracy and response sensitivity of interference sources.

[0024] (2) The high-precision voltage measurement method and system based on multi-parameter adaptive compensation enhances the ability of the measured value to maintain steady state under complex working conditions by constructing a logic to dynamically adjust the filtering parameters and correction amplitude through the reverse compensation function.

[0025] (3) The high-precision voltage measurement method and system based on multi-parameter adaptive compensation achieves accurate compensation for abnormal segments through stability verification and local correction mechanism, thereby improving the continuity and robustness of the measurement process.

[0026] (4) The high-precision voltage measurement method and system based on multi-parameter adaptive compensation avoids misjudgment and response delay caused by static limitation through threshold-free anomaly identification and adaptive path optimization, thereby enhancing the system's adaptability and scalability.

[0027] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0028] Figure 1 This is a flowchart of the high-precision voltage measurement method based on multi-parameter adaptive compensation according to the present invention;

[0029] Figure 2 This is a structural diagram of the high-precision voltage measurement system based on multi-parameter adaptive compensation according to the present invention;

[0030] Figure 3 This is a line graph of the stability verification values ​​involved in this invention;

[0031] Figure 4 This is a flowchart illustrating the dynamic processing based on the reverse compensation output value involved in this invention. Detailed Implementation

[0032] 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.

[0033] Please see Figures 1-4 This invention provides a technical solution: a high-precision voltage measurement method and system based on multi-parameter adaptive compensation, comprising: S1, acquiring voltage signal data and disturbance source characteristic data during voltage detection, and preprocessing the acquired voltage signal data and disturbance source characteristic data; S2, analyzing the disturbance offset characteristics in the measured voltage in conjunction with the disturbance current, and dynamically adjusting the compensation path structure based on the analysis results; S3, evaluating the voltage compensation correction in the measured voltage, and driving the compensation voltage output strategy in real time based on the evaluation results; S4, using the disturbance offset characteristic analysis results and the voltage compensation correction evaluation results as inputs, comprehensively analyzing the stability of the voltage signal, and dynamically determining abnormal sections and triggering local adaptive adjustment based on the analysis results.

[0034] Specifically, the steps for acquiring voltage signal data and disturbance source characteristic data during voltage detection are as follows: In the voltage signal data acquisition stage, during the actual operation of the industrial mechatronics equipment, the instantaneous voltage value collected at the measurement point needs to be retrieved, and the voltage change lag time and voltage change rate should be recorded simultaneously to construct a dynamic evaluation basis for voltage response characteristics. Simultaneously, the current load current within the corresponding sampling period is also acquired as a background reference parameter for subsequent interference attribution analysis. During this process, fluctuation characteristics need to be extracted from local segments under high-frequency disturbances, and the maximum voltage change in the current disturbance segment is statistically analyzed segment by segment. Combined with the difference values ​​of continuous points within the sampling sliding window, the average voltage change within the current window is calculated to support the stability judgment of voltage fluctuation trends.

[0035] The acquired instantaneous voltage values ​​at measurement points are reconstructed into an original measured voltage sequence in chronological order. Within each fixed sampling period, the difference between the current instantaneous voltage and the previous instantaneous voltage is continuously calculated, and the voltage difference amplitude is extracted as a basic indicator of the transient intensity of the interference. Subsequently, using the original voltage measurement sequence as input, a sliding window of fixed length is set, and the average value of all voltage points within the window is calculated. Then, the squares of the differences between the instantaneous voltage at each measurement point and the average value are summed and averaged to obtain the variance of the original voltage signal within the current window, which is used to quantitatively characterize the dispersion of local fluctuation amplitudes in the measurement channel.

[0036] During the disturbance source characteristic data acquisition phase, real-time disturbance current values ​​are obtained to capture the transient current peak at the time of disturbance triggering. The number of high-frequency disturbances is extracted as a statistical reference for abnormal pulse behavior, and the compensation voltage value formed under disturbance intervention is also collected. To accurately characterize the disturbance effect segment, the duration of each disturbance response in the measurement channel and the amplitude of the compensation voltage fluctuation it causes are also recorded. Finally, the ratio of the difference between the compensation voltage value and the instantaneous voltage value at the measurement point is calculated, and the ratio is output in real time as the relative error ratio.

[0037] In this implementation plan, a high-precision voltage measurement dataset under interference is constructed to form a mapping basis between dynamic fluctuation characteristics and disturbance response characteristics, providing quantifiable input for subsequent compensation path construction and stability correction. By jointly acquiring instantaneous voltage values, voltage difference amplitudes, voltage change lag times, voltage change rates, and current load current at measurement points, the time response and fluctuation characteristics of the voltage signal under disturbance are comprehensively reflected. The variance of the original voltage signal constructed based on the original measured voltage sequence can effectively identify local discreteness and unstable sections in the measurement process. By comprehensively extracting real-time disturbance current values, high-frequency interference frequency, disturbance duration, and compensation voltage fluctuation amplitude from the disturbance source characteristic data, the strength of voltage offset and disturbance coupling is quantified. The real-time calculated relative error ratio serves as a key disturbance significance indicator, providing reliable support for subsequent compensation accuracy determination and dynamic strategy adjustment.

[0038] Specifically, the preprocessing steps for the acquired voltage signal data and disturbance source feature data are as follows: During the standardization and anomaly suppression preprocessing of the voltage signal data and disturbance source feature data, the instantaneous voltage values ​​at the measurement points are first normalized according to the minimum and maximum values ​​of each measurement channel. This unifies the data distribution across different measurement points to the same scale, improving the relative consistency and comparability of subsequent feature analysis of the voltage change rate and voltage difference amplitude at multiple measurement points. Simultaneously, the constructed original measured voltage sequence is introduced into a median filtering mechanism to smoothly remove abnormal spike fluctuations caused by short-term disturbances, avoiding local extrema that could lead to biases in signal trend judgment. For the voltage change lag time, an interpolation reconstruction method based on sampling time sequence is used, combined with the high-frequency response change trend, to finely calibrate the voltage change starting point, ensuring the accuracy of the lag characteristics in disturbance response modeling and time-series compensation strategy adjustment. Subsequently, all disturbance source characteristic data, including real-time disturbance current value, number of high-frequency interferences, disturbance duration, compensation voltage value, and compensation voltage fluctuation amplitude, are restructured and standardized and encapsulated to form a structured disturbance factor vector. This vector serves as the core input in the subsequent disturbance response mapping function and reverse compensation function construction modules, supporting the generation of dynamic compensation paths and the calculation of local adjustments in abnormal segments.

[0039] In this implementation plan, the collected voltage signal data and disturbance source feature data are uniformly normalized, filtered and structured preprocessed to improve the comparability of data between multiple measurement points, reduce the interference of local outliers on the overall trend analysis, and ensure that various disturbance parameters have a stable and standardized input form in the subsequent mapping modeling and compensation function construction, thereby providing a reliable feature basis and data support for the generation of dynamic compensation paths and high-precision voltage correction process.

[0040] Specifically, the analysis of the disturbance offset characteristics in the measured voltage in conjunction with the disturbance current involves the following steps: The level change sequence of the pulse width modulation control signal is directly acquired using a digital storage oscilloscope. The instantaneous phase extraction and time-frequency energy reconstruction of the level change sequence are performed using Hilbert transform and short-time Fourier analysis. Then, the time interval between consecutive rising edges is extracted using an adaptive threshold edge enhancement algorithm to form a periodic time series. The PWM modulation frequency is further calculated by calculating the average period of the periodic time series. The dynamic disturbance correction value is obtained by multiplying the real-time disturbance current value by the instantaneous voltage value at the measurement point and dividing by the voltage change lag time plus one. The trend response adjustment value is obtained by multiplying the PWM modulation frequency by the voltage difference value and dividing by the current load current, and finally multiplying by the corresponding frequency domain disturbance adjustment factor. The disturbance coupling gain value is obtained by multiplying the number of high-frequency interferences by the disturbance duration, adding one, and then taking the logarithm. The dynamic disturbance correction value, the trend response adjustment value, and the disturbance coupling gain value are added together to obtain the disturbance response mapping value.

[0041] The formula for calculating the disturbance response mapping value is:

[0042] ;

[0043] In the formula, It represents the real-time disturbance current value, which reflects the intensity of the current change at the moment the disturbance occurs. It is the peak value of the current signal collected in real time in the power supply circuit, which comes from the sampling results of the current sensor. It represents the instantaneous voltage value at the measuring point, used to capture the immediate response level of the voltage under disturbance. It is the raw voltage reading collected by the measuring end at the instant of disturbance, and comes from the sampling data of a high-precision voltage sensor. It represents the voltage change lag time, which measures the degree of delay in the voltage response to current disturbances. It is the time interval between the peak of the disturbance current and the start of the voltage change, and is derived from the time synchronization calibration of the voltage and current channels. This represents the PWM modulation frequency, used to characterize the excitation intensity and periodic characteristics of high-frequency interference sources. It is the frequency of the PWM control signal of the internal switching power supply. The level change sequence of the pulse width modulation control signal is directly acquired through a digital storage oscilloscope. The instantaneous phase extraction and time-frequency energy reconstruction of the level change sequence are performed using Hilbert transform and short-time Fourier analysis. Then, the time interval between consecutive rising edges is extracted by combining the adaptive threshold edge enhancement algorithm to form a periodic time series. The PWM modulation frequency is further obtained by calculating the average period of the periodic time series. It represents the voltage difference amplitude, which reflects the fluctuation intensity of voltage changes within the interference period. It is the absolute difference between voltage readings between adjacent sampling points and is derived from the sliding window comparison results of the time-series voltage sampling sequence. It represents the current load current and is used to measure the actual operating condition of the current carried by the measurement point. It is the stable output current level of the equipment under operating conditions. This represents the number of high-frequency interferences, used to characterize the frequency of high-frequency disturbance behaviors identified per unit time. It is a statistical value of the number of abnormal pulses appearing in the voltage waveform, derived from the threshold comparison and edge detection results of the sampled data by the anomaly identification module. This indicates the duration of the disturbance, used to assess the impact period of a single disturbance. It is the total length of time that a single interference event continues to exceed the fluctuation threshold in the voltage channel, and is derived from the time window analysis results of each pulse event by the interference identification module. The frequency domain disturbance adjustment factor, ranging from 0.8 to 2.5, measures the excitation characteristics and response adaptation of disturbance behavior in the frequency domain. It originates from the frequency domain coupling degree between the spectral reconstruction value and the PWM modulation frequency within the current measurement point's frequency band. In the specific calculation process, firstly, the disturbance voltage response sequence of the compensation channel is used as input. Short-time Fourier transform and Hilbert spectral analysis are employed to perform local frequency deconstruction, extracting the main energy accumulation frequency bands and abrupt frequency segments. Then, the frequency reconstruction distribution curve within the frequency band is calculated, and the alignment and coupling rate of the PWM modulation frequency corresponding to the current channel are evaluated to form a frequency domain disturbance sensitivity map of the measurement point. Next, the frequency domain fluctuation stability index matrix is ​​constructed by combining the number of high-frequency interferences and the duration of the disturbance, characterizing the persistence and density of the current disturbance behavior in the frequency domain. Finally, the spectral coupling strength, the peak distribution of disturbance energy, and the frequency domain disturbance rhythm matching degree are weighted and normalized to comprehensively generate the frequency domain disturbance adjustment factor α. When the frequency response of the measurement point is highly coupled with the PWM modulation frequency and the disturbance energy distribution is concentrated, the α value is increased to enhance the frequency domain stripping effect; when the spectral response is discrete or the high-frequency disturbance rhythm is unstable, the α value is decreased to reduce the frequency domain adjustment intensity and ensure the frequency domain adaptability and dynamic suppression accuracy of the disturbance stripping process.

[0044] This implementation scheme comprehensively evaluates the impact of disturbance sources on voltage measurement accuracy. Specifically, it analyzes key factors including disturbance excitation intensity, response hysteresis characteristics, frequency domain excitation matching, and disturbance duration. First, by multiplying the real-time disturbance current value and the instantaneous voltage value at the measurement point, and normalizing the result with the voltage change hysteresis time, it characterizes the instantaneous voltage response intensity and delay caused by the current disturbance, reflecting the direct impact effect of the disturbance in the time domain. Second, it proportionally calculates the PWM modulation frequency, voltage difference amplitude, and standard period parameter, and introduces a frequency domain disturbance adjustment factor to weight and adjust the spectral excitation intensity, capturing the periodic effect of high-frequency disturbances on voltage signal fluctuations. Finally, it integrates the two parameters of high-frequency interference frequency and disturbance duration, and uses a logarithmic function to jointly harmonicize them, measuring the repeatability and impact period of the disturbance event, thereby characterizing the structural interference risk of the disturbance. Through the above multi-parameter interactive calculation, the final disturbance response mapping value can comprehensively reflect the response sensitivity of the measurement channel under complex disturbance environments, providing a quantitative basis for the subsequent dynamic adjustment of the compensation path and adaptive control of the correction range.

[0045] Specifically, the steps for dynamically adjusting the compensation path structure based on the analysis results are as follows:

[0046] The disturbance response mapping value of the current measurement point is compared with the response mapping threshold in real time to determine whether the current channel is in a disturbance-dominated state, which serves as the triggering basis for compensation path adjustment.

[0047] When the disturbance response mapping value is less than or equal to the response mapping threshold, it indicates that the disturbance characteristics of the current measurement point are within a controllable range, and the signal fluctuation has not shown a significant non-steady-state shift. At this point, it is determined to be a controllable disturbance segment. The original sampling frequency is maintained unchanged, and the current compensation intensity and boundary screening window parameters are retained to avoid introducing unnecessary intervention adjustments. At the same time, the current instantaneous voltage value of the measurement point continues to be used as the reference signal for trend tracking, and normal fluctuation monitoring, data caching, and buffer queue writing tasks are performed to maintain the stability of the signal processing path.

[0048] When the disturbance response mapping value is greater than the response mapping threshold, it indicates that the current voltage channel is significantly affected by the disturbance source and has obvious response distortion. At this time, it is determined to be a significant interference segment, and the regional-level reverse correction process is immediately initiated: First, the short-time Fourier transform is called to perform frequency domain analysis on the sampled data of the current channel to identify the main frequency range and energy concentration area of ​​the high-frequency disturbance, and the boundary range of the abnormal response interval is located based on the frequency band marker; then, the previously cached steady-state samples are linked to complete the construction of the reference state through difference calculation, and the voltage data of the disturbed area is fitted and corrected; at the same time, the cutoff frequency parameter of the filter is dynamically adjusted according to the disturbance intensity to enhance the suppression effect on the target interference frequency band and avoid insufficient compensation and over-correction; finally, the voltage reconstruction and error suppression processing of the interference segment is completed, and the output corrected voltage signal is marked and recorded as the compensated state.

[0049] In this implementation scheme, based on the determination result of the disturbance response mapping value, dynamic identification and compensation trigger control of the current measurement channel disturbance state are realized, thereby effectively distinguishing between controllable disturbances and significant interference scenarios, and deciding whether to activate the local reverse correction process accordingly, improving the pertinence of compensation intervention and the stability of voltage signal processing. By comparing the disturbance response mapping value with the response mapping threshold in real time, it is determined whether the section where the measurement point is located is significantly affected by high-frequency PWM modulation, load switching, and electromagnetic coupling; when it is determined to be a significantly interfered section, a short-time Fourier transform is immediately driven to extract the interference frequency band, completing the reverse reconstruction and error suppression of the abnormal section; when it is determined to be a controllable disturbance section, the original compensation strategy remains unchanged, keeping the signal link running stably, thereby achieving robust control of the high-precision measurement path without introducing redundant compensation.

[0050] Specifically, the evaluation steps for voltage compensation correction in voltage measurement are as follows: A PXIe data acquisition card is used to continuously sample the original voltage signal in the target measurement channel at high frequency, forming an equally spaced time-series voltage sampling sequence. The sequence is then truncated into fixed-length data segments using a sliding window as the signal analysis window for the current time point. A fast Fourier transform is applied to the voltage sequence within the window to map the time-domain signal to the frequency domain, obtaining the corresponding spectral structure. In the spectral structure, the amplitude of the dominant frequency component with the highest energy within the current window is extracted and recorded as the frequency interference amplitude at the current moment. Simultaneously, multiple time-series frequency... The average value of the frequency interference amplitude is calculated to obtain the average amplitude of the frequency band. The difference between the maximum voltage change in the current interference segment and the average voltage change in the current window is squared, and then 1 is added and the natural logarithm is taken to obtain the sudden energy fluctuation value. The absolute value of the difference between the frequency interference amplitude and the average amplitude of the frequency band is multiplied by the frequency domain offset adjustment factor and then the square root is taken to obtain the frequency disturbance deviation value. The compensation smoothing adjustment factor is divided by the variance of the original voltage signal in the current window plus 1 to obtain the voltage residual correction value. The sudden energy fluctuation value is subtracted from the frequency disturbance deviation value and then the voltage residual correction value is added to obtain the reverse compensation output value.

[0051] The formula for calculating the reverse compensation output value is:

[0052] ;

[0053] In the formula, This represents the maximum voltage change in the current interference segment, used to assess the peak value of a transient pulse disturbance, which originates from the range in a short-time voltage sequence; This represents the average value of the voltage change within the current window, used to reflect the general level of voltage fluctuations, and is derived from the differential averaging of multiple consecutive sampling points. The frequency interference amplitude at the current moment is used to assess the interference intensity in the frequency band where the sampling point is located. It is derived from: using a PXIe data acquisition card to continuously sample the original voltage signal in the target measurement channel at high frequency to form an equally spaced time-series voltage sampling sequence. The sequence is then cut into a fixed-length data segment of 512 points in a sliding window manner as the signal analysis window at the current time point. A fast Fourier transform is applied to the voltage sequence within the window to map the time domain signal to the frequency domain, obtaining the corresponding spectrum structure. In the spectrum structure, the amplitude of the dominant frequency component with the largest energy within the current window is extracted, which is the frequency interference amplitude for the time period. The average amplitude of the frequency band is used to check whether the current frequency band disturbance is too strong. It is derived from the average of the frequency interference amplitudes over multiple time periods. U represents the original measured voltage sequence, which is a continuous voltage signal without filtering and compensation. It is formed from the instantaneous voltage values ​​of the collected measurement points in chronological order. This represents the variance of the original voltage signal within the current window, used to evaluate signal volatility. The voltage difference amplitude is obtained by calculating the difference between the current instantaneous voltage value and the instantaneous voltage value at the previous moment in each sampling period. The original voltage measurement sequence is used as input, and the average voltage value of all measurement points is calculated within a fixed-length window. The deviation between the instantaneous voltage value of each measurement point and the average voltage value is then squared and averaged. The compensation smoothing adjustment factor, ranging from 0.2 to 1, is used to adjust the proportion of the smoothing term in the reverse compensation output value, controlling the stability and abrupt change suppression capability of the compensation curve during the disturbance response process. It originates from the continuity characteristics of the compensation output sequence within the current time window. In the specific calculation, firstly, the compensation voltage value change sequence within the sliding window is extracted, an instantaneous slope sequence is constructed, and its rate of change, reversal frequency, and mean square error are statistically analyzed to describe the fluctuation characteristics of the current compensation path. Then, it is matched with the calibrated stable output characteristic curve in the historical steady-state samples. Multi-dimensional alignment is performed based on slope balance, relative fluctuation amplitude, and peak duration indicators, and the deviation level between the compensation change within the current window and the steady-state curve is calculated accordingly. Finally, the value of the compensation smoothing adjustment factor is generated. When the compensation path fluctuates drastically and the waveform changes discontinuously, the compensation smoothing adjustment factor increases, strengthening the restriction of the smoothing term on the compensation output. When the compensation value fluctuates moderately and the trend is clear and continuous, the compensation smoothing adjustment factor decreases to improve compensation accuracy and response speed, enhancing the adaptability to continuous disturbance scenarios. The compensation smoothing adjustment factor, ranging from 0.5 to 2, is used to control the degree of frequency component shift in the inverse compensation output. It measures the deviation between the current spectral structure and the steady-state frequency distribution, originating from the main frequency drift amplitude after the current voltage sequence undergoes frequency domain transformation. In the specific calculation, a Fast Fourier Transform is first performed on the voltage signal within the current sliding window to extract the energy center location and distribution range of the main frequency band. The difference between the current window's main frequency center frequency and the historical steady-state main frequency is calculated, forming the frequency shift. Then, the shift value is corrected by combining the spectral change rate, energy concentration, and frequency band shape to eliminate spectral drift errors caused by short-term disturbances, ultimately forming the numerical output of the compensation smoothing adjustment factor. When there is a persistent shift in the spectral center and a significant change in the main frequency mode, the compensation smoothing adjustment factor value increases, increasing the response weight to the frequency shift effect. Conversely, when the frequency domain energy structure remains highly consistent with the steady state, the compensation smoothing adjustment factor value decreases, weakening the disturbance of the shift term on the compensation path, thus maintaining the stability and selective sensitivity of the compensation strategy in the frequency dimension.

[0054] In this implementation scheme, the significance and response complexity of voltage fluctuations at the current measurement point within the interference segment are comprehensively quantified. By coupling the deviation between the maximum voltage change and the average fluctuation level, the current frequency disturbance intensity, and the joint characteristics between dynamic disturbance estimation and compensation error, a reverse compensation output value that can be used to drive the local reverse correction process is constructed. This enables adaptive identification and fine correction of voltage anomalies in high-frequency disturbance segments, improving the steady-state recovery capability and dynamic interference filtering accuracy of the measurement data.

[0055] Specifically, the real-time compensation voltage output strategy based on the evaluation results involves the following steps: using a sliding window mechanism, a difference calculation operation is performed on the reverse compensation output value in each sampling period to obtain the change amplitude during each period and construct a change rate sequence; then, the sequence is processed by moving average to obtain the compensation change average value reflecting the evolution trend of the compensation amplitude within the current window, providing a benchmark reference for subsequent stability state determination.

[0056] Based on the trend of the reverse compensation output value change at the current measurement point, the state category of the voltage sequence is dynamically identified and the differentiated processing logic is executed accordingly.

[0057] When the difference between the rate of change of the reverse compensation output value from the previous moment and the average compensation change is less than the rate of change threshold, it indicates that the current compensation amplitude fluctuation is controllable and the trend tends to be stable. It is determined to be a section that has been fully corrected. The current compensation result is directly written as the final voltage value into the voltage sequence buffer in the monitoring link for subsequent steady-state analysis and trend tracking.

[0058] When the rate of change of the reverse compensation output value is one positive and one negative within two consecutive sliding windows, and the rate of change is greater than the rate of change of the average compensation value, it indicates that the voltage response is under the influence of high-frequency unstable disturbances. This is determined to be an unstable compensation section, and a fine-tuning feedback mechanism is immediately triggered: on the one hand, the analysis window range is tightened to improve the sensitivity of anomaly detection; on the other hand, the response weight of the frequency domain correction term in the compensation path is enhanced to improve the interference suppression efficiency. At the same time, the compensation trend and voltage drift direction of adjacent measurement points are retrieved as lateral references, and cross-point weighted fusion correction is performed to ensure the continuous stability of the voltage compensation path in the high-incidence disturbance section.

[0059] When the difference between the rate of change of the compensated output value and the average value of the compensated change is greater than or equal to the rate of change threshold, and the sign of the rate of change is consistent in three consecutive sampling periods (i.e., all are positive or all are negative), it indicates that the current compensation process is showing a continuous upward or downward trend, and is determined to be a pulse disturbance evolution segment. At this time, a delayed confirmation strategy is initiated: the current compensation result is not output immediately, but is retained in the buffer state. When the compensation change trend of the next two consecutive sampling points is consistent with the direction of the current point, the smoothed and corrected voltage value is written to the main output channel. This avoids abnormal fluctuations caused by short-term pulse signals and ensures the stability and robustness of the compensation output result during the disturbance transition process.

[0060] like Figure 4 The diagram shows a dynamic processing flowchart based on the reverse compensation output value provided in this application example. It begins by acquiring the current reverse compensation output value, then calculates the rate of change of the reverse compensation output value and compares it with the average compensation change within the sliding window to determine if it is in a stable state. When the trend is determined to be stable and the fluctuation amplitude is close to the sliding average, it is considered a stable segment, and the current compensation value is directly output. When the trend is unstable, it further determines whether the fluctuation amplitude has significantly increased and analyzes whether the current direction of change is opposite to the previous cycle. When the direction is opposite, it indicates a disturbed and unstable segment, immediately triggering a fine-tuning mechanism to narrow the analysis window, increase the weight of the frequency domain correction term, and output the fine-tuned correction compensation value. When the directions are the same, it is determined to be a trend transition segment, triggering a delayed confirmation mechanism. The confirmed correction compensation value is output only after the subsequent sliding window data trends are consistent and tend to smooth. Through the above multi-branch judgment and strategy linkage, a highly sensitive and strongly adaptive reverse compensation control logic is constructed based on the dynamic processing flowchart of the reverse compensation output value, enabling dynamic and stable correction and intelligent voltage output under different disturbance scenarios.

[0061] In this implementation scheme, the fluctuation patterns in the compensation output process are dynamically identified, and a sequence of the rate of change of the reverse compensation output value is constructed through a sliding window. The average value of the compensation change is extracted as a stability benchmark. Combined with the reverse compensation output trend of the current measurement point, the voltage signal is determined to be in a stable correction, abnormal disturbance, or pulse evolution state, and different processing mechanisms are triggered accordingly: including directly caching the stable compensation result, fine-tuning the compensation weight of the unstable section, and delaying the confirmation of suspected pulse disturbances. This enables accurate identification and adaptive output control of the disturbance-sensitive section, ensuring the stability and reliability of the compensation result.

[0062] Specifically, using the disturbance offset characteristic analysis results and voltage compensation correction evaluation results as inputs, the comprehensive analysis of voltage signal stability is performed as follows: The voltage change rate at the current sampling point is incremented by one, and the natural logarithm is taken. This logarithm is then multiplied by the square of the value minus the relative error ratio to obtain the comprehensive influence amplification value. The disturbance response mapping value is squared, divided by the value of the compensation voltage fluctuation amplitude, and then added to the reverse compensation output value. This squared value is incremented by one, and the natural logarithm is taken to obtain the compensation suppression driving value. The comprehensive influence amplification value is divided by the compensation suppression driving value to obtain the stability verification value.

[0063] The formula for calculating the stability check value is:

[0064] ;

[0065] In the formula, It represents the rate of change of voltage at the current sampling point, and represents the voltage difference between two adjacent sampling points divided by the sampling interval, derived from the differential calculation of the original voltage; It represents the relative error ratio between the compensated voltage value and the original voltage value, and the percentage difference between the compensated voltage and the actual acquired voltage. It is derived from the percentage difference between the compensated voltage value and the instantaneous voltage value at the measuring point in real time. Indicates the reverse compensation output value; This represents the disturbance response mapping value. This indicates the compensation voltage fluctuation amplitude within the current window, and the voltage variance within the sliding window, reflecting whether the current measurement is in a state of severe fluctuation.

[0066] In this implementation example, the voltage change rate of Example 1 is set to 0.42, the relative error ratio is 0.08, the reverse compensation output value is 0.33, the disturbance response mapping value is 0.21, and the compensation voltage fluctuation amplitude is 0.11.

[0067] In Example 2, the voltage change rate is set to 0.66, the relative error ratio is 0.15, the reverse compensation output value is 0.48, the disturbance response mapping value is 0.38, and the compensation voltage fluctuation amplitude is 0.19.

[0068] In Example 3, the voltage change rate is set to 0.23, the relative error ratio is 0.06, the reverse compensation output value is 0.25, the disturbance response mapping value is 0.17, and the compensation voltage fluctuation amplitude is 0.09.

[0069] In Example 4, the voltage change rate is set to 0.78, the relative error ratio is 0.24, the reverse compensation output value is 0.67, the disturbance response mapping value is 0.44, and the compensation voltage fluctuation amplitude is 0.22.

[0070] In Example 5, the voltage change rate is set to 0.51, the relative error ratio is 0.12, the reverse compensation output value is 0.39, the disturbance response mapping value is 0.29, and the compensation voltage fluctuation amplitude is 0.13.

[0071] In Example 6, the voltage change rate is set to 0.34, the relative error ratio is 0.09, the reverse compensation output value is 0.31, the disturbance response mapping value is 0.19, and the compensation voltage fluctuation amplitude is 0.10.

[0072] In Example 7, the voltage change rate is set to 0.59, the relative error ratio is 0.18, the reverse compensation output value is 0.52, the disturbance response mapping value is 0.36, and the compensation voltage fluctuation amplitude is 0.17. The stability check values ​​for each example are calculated, as shown in Table 1.

[0073] Table 1 Stability Verification Value Data Table

[0074] Example number Voltage change rate relative error ratio Inverse compensation output value Disturbance response mapping value Compensation voltage fluctuation amplitude Stability check value Example 1 0.42 0.08 0.33 0.21 0.11 0.504 Example 2 0.66 0.15 0.48 0.38 0.19 0.317 Example 3 0.23 0.06 0.25 0.17 0.09 0.684 Example 4 0.78 0.24 0.67 0.44 0.22 0.248 Example 5 0.51 0.12 0.39 0.29 0.13 0.429 Example 6 0.34 0.09 0.31 0.19 0.10 0.589 Example 7 0.59 0.18 0.52 0.36 0.17 0.362

[0075] like Figure 3 As shown in Table 1, this is a line graph of the stability verification values ​​provided in this application example. Figure 3 As can be seen, Example 3 has the highest stability verification value of 0.684, indicating that the voltage change rate at the current measurement point is low, the compensation voltage fluctuation amplitude is small, and the disturbance response is stable. This comprehensively reflects that the voltage fluctuation at the measurement point is in a low-interference, low-distortion state, and its stability is high. It can be used as a voltage reference sampling point in the current batch for trend benchmark verification and synchronous correction of smoothing strategies. In contrast, Example 4 has the lowest stability verification value of only 0.248. Although its reverse compensation output value is large, the voltage change rate is high, the relative error ratio is significant, and the disturbance response amplitude is drastic. This indicates that the measurement point is currently in a high-fluctuation interference environment, with poor stability. It is not suitable as a direct basis for voltage output and interference compensation and error suppression should be strengthened. Overall, the stability verification value line graph clearly presents the distribution of voltage sampling reliability for different measurement points in the current batch. A higher evaluation value indicates that the measurement point data is more robust and more suitable for use as a control anchor point and correction reference. A lower evaluation value suggests the need to be wary of abnormal fluctuations and sampling error risks, and to promptly activate the local correction mechanism to improve measurement reliability.

[0076] Specifically, the steps for dynamically determining abnormal sections and triggering local adaptive adjustments based on the analysis results are as follows: Real-time comparison of the stability verification value and stability evaluation threshold of the current measurement point. The stability evaluation threshold is refined into two levels: the first stability threshold and the second stability threshold, corresponding to the reliability grading control strategy in the voltage signal output process.

[0077] When the stability check value is less than or equal to the second stability threshold, the current segment is determined to be a high-disturbance, low-stability interference segment, indicating that the voltage fluctuation is severe and the error is difficult to suppress using conventional compensation strategies. At this time, the disturbance source re-analysis process is immediately triggered to re-extract and model the key factors such as high-frequency interference components, PWM modulation frequency change rate, equivalent impedance difference in the wiring path, and load current response rate in the current measurement point area; linked to the error amplification trend of the current compensation function, a local compensation reconstruction path is dynamically constructed; subsequently, a short-time multi-window average sliding strategy is introduced to suppress and smooth abnormal signals, and the final output is a steady-state voltage with a low-weight label, so as to reduce the interference effect of segment signals on the main channel output in the calculation.

[0078] When the stability check value is greater than the second stability threshold but less than or equal to the first stability threshold, the current segment is determined to be a disturbance transition fluctuation segment. Such regions are often at the critical boundary between interference response and compensation correction. A dynamic compensation self-checking mechanism is introduced: First, the compensation residual change sequence within two sampling periods before and after the current measurement point is loaded, and its change trend is evaluated to determine whether it has convergence characteristics. Then, the frequency domain suppression factor weights within the current segment are reallocated, and the frequency band reconstruction strategy is dynamically adjusted. Contrast enhancement processing is performed within the time window to improve the separation capability between noise and residuals. When the self-checking trend shows a convergent and stable state, it indicates that the compensation is effective and can be included in the main output channel. If the residual continues to accumulate fluctuations, it is marked as a buffer to be reconstructed, the output is temporarily suspended, and the correction is prioritized.

[0079] When the stability check value exceeds the first stability threshold, the current segment is determined to be a high-confidence stable segment, meaning that the sampled data fluctuates little, the compensation residual is controllable, and the frequency domain response is stable. It is directly marked as the final correction zone, and the compensation voltage does not require further processing, directly serving as the final output voltage value in the main measurement channel. Simultaneously, to improve the compensation strategy transfer capability, the data is bound to the current disturbance response characteristics, the compensation residual trajectory, and the frequency domain suppression configuration parameters to form structured successful compensation samples. These serve as benchmark samples for the dynamic control module, supporting strategy optimization and response acceleration under similar disturbance environments.

[0080] In this implementation scheme, voltage data segments are managed hierarchically and dynamically processed based on stability verification values, enabling accurate identification and differentiated response control of high-disturbance segments, fluctuating transition segments, and stable segments. By comparing stability verification values ​​with dual-layer stability thresholds in real time, the stable state of the voltage signal in the current segment under disturbance conditions is identified. For high-disturbance segments, compensation path reconstruction and signal smoothing are triggered; for fluctuating segments, a self-verification mechanism and frequency domain weighting correction are introduced; and for stable segments, the final voltage value is directly output and compensation samples are generated, ensuring that the overall measurement process has robustness, adaptability, and accurate response capabilities under dynamic disturbance environments.

[0081] The second aspect of the present invention provides a high-precision voltage measurement system based on multi-parameter adaptive compensation, comprising: a multi-parameter disturbance source extraction module, used to collect voltage signal data and disturbance source feature data during the voltage detection process, wherein the voltage signal data includes the instantaneous voltage value at the measurement point, the voltage change lag time, the voltage change rate and the current load current, and the disturbance source feature data includes the real-time disturbance current value, the number of high-frequency interferences, the compensation voltage fluctuation amplitude and the disturbance duration, and performs standardization and normalization processing, local anomaly smoothing suppression and response lag calibration preprocessing on the collected voltage signal data and disturbance source feature data to construct a structured input sequence;

[0082] The disturbance response mapping calculation module is used to perform joint frequency domain and time domain analysis on the disturbance offset characteristics in the measured voltage by combining the disturbance current, extract the disturbance response mapping value, and dynamically adjust the compensation path structure based on the analysis results to adapt to the response model switching requirements under different disturbance scenarios, ensuring that the compensation path has adaptive adjustment capability under the dominant disturbance direction.

[0083] The reverse compensation function construction module is used to perform multi-cycle sliding window evaluation on the voltage compensation correction in the measured voltage, generate reverse compensation output value, identify the stability and abnormal jump points of the changing trend, and drive the compensation voltage output strategy in real time based on the evaluation results, and realize dynamic voltage compensation control by linking the frequency domain offset adjustment factor and the compensation smoothing adjustment factor.

[0084] The measurement stability verification and correction module is used to perform multi-source comparison risk assessment and multi-threshold classification judgment on the stability of the voltage signal in the current measurement section, taking the disturbance offset feature analysis results and voltage compensation correction evaluation results as inputs. Based on the analysis results, it dynamically determines abnormal sections and triggers a local adaptive adjustment process, including disturbance source reanalysis, frequency band suppression reconstruction and compensation function reinitialization, to ensure the continuity and reliability of measurement output under different disturbance environments.

[0085] In this implementation scheme, the multi-parameter disturbance source extraction module constructs a structured input feature set. By collecting the instantaneous voltage value, voltage change lag time, voltage change rate and load current of the measurement point during the voltage detection process, as well as the real-time disturbance current, high-frequency interference frequency and compensation voltage fluctuation amplitude of the disturbance source, it performs preprocessing operations such as standardization and normalization, outlier removal and lag time calibration, providing a data foundation for subsequent disturbance mapping modeling and compensation logic.

[0086] The disturbance response mapping calculation module calculates the degree of influence of disturbance sources on the measured voltage. By extracting disturbance offset features and establishing disturbance response mapping values, it quantifies the impact of different disturbance components on voltage stability within the current window and dynamically selects an appropriate compensation path structure accordingly, thereby improving the flexibility and accuracy in dealing with the diversity of disturbances.

[0087] The reverse compensation function construction module generates stable and traceable reverse compensation output results. It evaluates the changing trend, fluctuation mode and residual behavior of voltage compensation correction through a sliding window method. Combined with frequency domain offset adjustment factor and compensation smoothing adjustment factor, it drives the implementation of voltage compensation strategy and dynamically outputs reverse correction voltage value to ensure the consistency of compensation results and error control capability.

[0088] The measurement stability verification and correction module determines the reliable stability state of the voltage measurement results. By fusing disturbance offset features and compensation correction trajectories, it calculates the stability verification value and compares it with multi-level stability thresholds. It then classifies the current section into high-risk disturbance sections, transitional fluctuation sections, and high-confidence stable sections, and performs reconstruction correction, self-verification confirmation, and final output operations in a coordinated manner to ensure the robustness and reliability of the measurement data under disturbance conditions.

[0089] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0090] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A high-precision voltage measurement method based on multi-parameter adaptive compensation, characterized in that: include: S1, collect voltage signal data and disturbance source characteristic data during the voltage detection process, and preprocess the collected voltage signal data and disturbance source characteristic data; S2, combined with the disturbance current, analyzes the disturbance offset characteristics in the measured voltage, and dynamically adjusts the compensation path structure based on the analysis results; S3 evaluates the voltage compensation correction in the measured voltage and drives the compensation voltage output strategy in real time based on the evaluation results. S4 takes the disturbance offset characteristic analysis results and voltage compensation correction evaluation results as inputs to comprehensively analyze the stability of the voltage signal, and dynamically determines abnormal sections and triggers local adaptive adjustment based on the analysis results.

2. The high-precision voltage measurement method based on multi-parameter adaptive compensation according to claim 1, characterized in that: The specific steps for acquiring voltage signal data and disturbance source characteristic data during the voltage detection process are as follows: Collect voltage signal data, including: instantaneous voltage value at the measuring point, voltage change lag time, voltage change rate, current load current, and record the maximum voltage change in the current interference segment and the average voltage change within the current window. The original measured voltage sequence is formed from the instantaneous voltage values ​​of the collected measurement points in chronological order. The voltage difference amplitude is obtained by calculating the difference between the current instantaneous voltage value and the instantaneous voltage value at the previous moment in each sampling period. The original voltage measurement sequence is used as input to calculate the average voltage value of all measurement points within the sliding window. The deviation between the instantaneous voltage value of each measurement point and the average voltage value is squared and averaged to obtain the variance of the original voltage signal within the current window. Collect disturbance source characteristic data, including: real-time disturbance current value, number of high-frequency interferences, compensation voltage value, disturbance duration and compensation voltage fluctuation amplitude. Compare the ratio of the difference between the compensation voltage value and the instantaneous voltage value at the measuring point in real time, and record it as the relative error ratio.

3. The high-precision voltage measurement method based on multi-parameter adaptive compensation according to claim 1, characterized in that: The specific steps for preprocessing the acquired voltage signal data and disturbance source characteristic data are as follows: The voltage signal data and disturbance source characteristic data are standardized and anomaly suppression preprocessed. The voltage values ​​at the measurement points are uniformly normalized to improve the consistency of multi-measurement point analysis. The original measured voltage sequence is smoothed by introducing a median filtering strategy to remove local outlier fluctuations and transient spikes. The voltage change lag time is calibrated by interpolation reconstruction to ensure the accuracy of the lag parameter for timing compensation path adjustment. The perturbation feature data are uniformly constructed into a structured perturbation factor vector, which serves as the input variable for the construction modules of the perturbation response mapping function and the inverse compensation function.

4. The high-precision voltage measurement method based on multi-parameter adaptive compensation according to claim 1, characterized in that: The specific steps for analyzing the disturbance offset characteristics in the measured voltage by combining the disturbance current are as follows: The level change sequence of the pulse width modulation control signal is directly acquired by a digital storage oscilloscope. The instantaneous phase extraction and time-frequency energy reconstruction of the level change sequence are performed by Hilbert transform and short-time Fourier analysis. The time interval between consecutive rising edges is extracted by an adaptive threshold edge enhancement algorithm to form a periodic time sequence. The PWM modulation frequency is further calculated by calculating the average period of the periodic time sequence. The dynamic disturbance correction value is obtained by multiplying the real-time disturbance current value by the instantaneous voltage value at the measuring point and then dividing by the voltage change lag time plus one. Multiply the PWM modulation frequency by the voltage difference and divide by the current load current, then multiply by the corresponding frequency domain disturbance adjustment factor to obtain the trend response adjustment value; The disturbance coupling gain is obtained by multiplying the number of high-frequency interferences by the duration of the disturbance, adding one, and then taking the logarithm. The dynamic disturbance correction value, the trend response adjustment value, and the disturbance coupling gain value are added together to obtain the disturbance response mapping value.

5. The high-precision voltage measurement method based on multi-parameter adaptive compensation according to claim 1, characterized in that: The specific steps for dynamically adjusting the compensation path structure based on the analysis results are as follows: Real-time comparison of the disturbance response mapping value and the response mapping threshold at the current measurement point: When the disturbance response mapping value is less than or equal to the response mapping threshold, it is determined to be a disturbance controllable section. The original sampling frequency, compensation intensity and boundary screening window are kept unchanged, and trend tracking and data caching are continued based on the current voltage input. When the disturbance response mapping value is greater than the response mapping threshold, it is determined to be a significant interference segment, and the local reverse correction process is immediately triggered: short-time Fourier transform is used to mark the frequency band of the interference component, the abnormal response interval is locked, the stored steady-state samples are linked to perform difference correction, the current filter cutoff frequency is dynamically adjusted, and the corrected voltage value of the significant interference segment is reconstructed. Finally, the error-suppressed signal is output and marked as compensated.

6. The high-precision voltage measurement method based on multi-parameter adaptive compensation according to claim 1, characterized in that: The specific steps for evaluating the voltage compensation correction in the measured voltage are as follows: The PXIe data acquisition card is used to continuously sample the raw voltage signal in the target measurement channel at high frequency, forming an equally spaced time-series voltage sampling sequence. The sequence is then truncated into fixed-length data segments using a sliding window method, which serve as the signal analysis window for the current time point. A fast Fourier transform is applied to the voltage sequence within the window to map the time-domain signal to the frequency domain, obtaining the corresponding spectral structure. In the spectral structure, the amplitude of the dominant frequency component with the largest energy within the current window is extracted and recorded as the frequency interference amplitude at the current moment. Simultaneously, the average value of the frequency interference amplitudes over multiple time periods is calculated to obtain the average amplitude of the frequency band. The difference between the maximum voltage change in the current interference segment and the average voltage change in the current window is calculated, squared, and then 1 is added to obtain the natural logarithm of the sudden change energy fluctuation value. The absolute value of the difference between the frequency interference amplitude and the mean amplitude of the frequency band is calculated, multiplied by the frequency domain offset adjustment factor, and then the square root is obtained to get the frequency disturbance deviation value. Divide the compensation smoothing adjustment factor by the value of the original voltage signal variance plus 1 within the current window to obtain the voltage residual correction value; The reverse compensation output value is obtained by subtracting the frequency disturbance deviation value from the sudden energy fluctuation value and then adding the voltage residual correction value.

7. The high-precision voltage measurement method based on multi-parameter adaptive compensation according to claim 1, characterized in that: The specific steps of the strategy for real-time driving of compensation voltage output based on evaluation results are as follows: By performing differential calculations on the reverse compensation output values ​​of each sampling period through a sliding window, the rate of change sequence within each period is obtained and averaged to obtain the mean compensation change. Based on the current trend of the reverse compensation output value at the measurement point, the stable state of the voltage data is dynamically determined and the following processing is performed: When the difference between the rate of change of the reverse compensation output value from the previous moment and the average compensation change value is less than the rate of change threshold, it is determined to be a section that has been fully corrected. The current compensation result is directly used as the final voltage value and written into the voltage sequence buffer in the monitoring link. When the rate of change of the reverse compensation output value is one positive and one negative within two consecutive sliding windows, and the rate of change is greater than the rate of change of the average compensation value, it is determined to be an unstable compensation section, and the fine-tuning feedback mechanism is immediately triggered: by narrowing the analysis window, increasing the weight of the frequency domain correction term, and simultaneously introducing the lateral comparison results of adjacent measurement points, local weighted correction is performed to ensure stable output of the voltage transition section. When the difference between the rate of change of the reverse compensation output value from the previous moment and the average compensation change is greater than or equal to the rate of change threshold, and the sign of the rate of change is positive or negative for three consecutive sampling periods, it is determined to be a pulse disturbance evolution segment and enters the delayed confirmation strategy: retain the compensation result of the current point but do not output it immediately, wait for the compensation results of the next two sampling points to confirm that the trend is consistent, and then write the smoothed and corrected voltage value to the main output channel.

8. The high-precision voltage measurement method based on multi-parameter adaptive compensation according to claim 1, characterized in that: The specific steps for comprehensively analyzing the stability of the voltage signal, using the disturbance offset characteristic analysis results and the voltage compensation correction evaluation results as inputs, are as follows: Add one to the voltage change rate at the current sampling point, take the natural logarithm, and then multiply it by the square of one minus the relative error ratio to obtain the comprehensive effect amplification value. The square of the disturbance response mapping value is divided by the value of the compensation voltage fluctuation amplitude, then added to the reverse compensation output value, and the natural logarithm is added to obtain the compensation suppression drive value. Divide the overall amplification value by the compensation suppression driving value to obtain the stability check value.

9. The high-precision voltage measurement method based on multi-parameter adaptive compensation according to claim 1, characterized in that: The specific steps for dynamically determining abnormal segments and triggering local adaptive adjustments based on the analysis results are as follows: Real-time comparison of the stability check value of the current measuring point with the stability evaluation threshold. The stability evaluation threshold is divided into a first stability threshold and a second stability threshold. When the stability check value is less than or equal to the second stability threshold: the segment is determined to be a high-disturbance, low-stability interference segment, the disturbance source re-analysis process is automatically invoked, the PWM modulation frequency change rate, wiring equivalent impedance difference and load current response rate are re-extracted in the region, local compensation reconstruction is implemented in combination with the error amplification trend of the current compensation function, and short-time multi-window average sliding suppression processing is performed to output low-weight steady-state voltage. When the stability check value is greater than the second stability threshold and less than or equal to the first stability threshold: the region is determined to be a disturbance transition fluctuation segment, and enters the dynamic compensation self-checking mechanism: load the compensation residual changes of the two cycles before and after the current measurement point, reallocate the frequency domain suppression factor weight of the segment and perform frequency band reconstruction and time window contrast enhancement. When the self-checking trend is in a convergent state, it is included in the main output channel. When the residual continues to accumulate, it is upgraded to a buffer to be reconstructed and the output mark is delayed. When the stability check value is greater than the first stability threshold: the region is determined to be a high-confidence stable segment, automatically marked as the final correction region, and the compensation voltage is directly used as the final output of the main measurement channel. At the same time, the disturbance response characteristics of the current segment, the compensation residual trajectory and the frequency domain suppression configuration are bound to generate structured compensation successful sample data.

10. The high-precision voltage measurement system based on multi-parameter adaptive compensation as described in any one of claims 1-9, characterized in that: include: The multi-parameter disturbance source extraction module collects voltage signal data and disturbance source feature data during the voltage detection process, and preprocesses the collected voltage signal data and disturbance source feature data. The disturbance response mapping calculation module analyzes the disturbance offset characteristics in the measured voltage in conjunction with the disturbance current, and dynamically adjusts the compensation path structure based on the analysis results. The reverse compensation function construction module evaluates the voltage compensation correction in the measured voltage and drives the compensation voltage output strategy in real time based on the evaluation results. The measurement stability verification and correction module takes the disturbance offset characteristic analysis results and voltage compensation correction evaluation results as inputs to comprehensively analyze the stability of the voltage signal, and dynamically determines abnormal sections and triggers local adaptive adjustment based on the analysis results.

Images