Real-time harmonic suppression and compensation method for DAC output end

Abstract

The invention discloses a real-time harmonic suppression and compensation method for a DAC output end, and the method comprises the steps: collecting harmonic characteristic parameters of an output signal through a high-precision DAC calibration test platform, and carrying out the cooperative processing of a cascade notch dynamic harmonic cancellation model, a frequency band weight adaptive distribution model and a nonlinear even harmonic enhancement algorithm, harmonic feature extraction, model parameter configuration, weight distribution, even harmonic enhancement, dynamic offset and real-time feedback correction are sequentially completed. According to the method, model parameters are dynamically adjusted to adapt to a complex harmonic scene, the processing precision is improved by means of multi-model linkage and real-time feedback, the method has the advantages of being high in adaptability, accurate in processing and good in stability, DAC output signal distortion can be greatly reduced, the signal purity and reliability are improved, and the method is suitable for large-scale popularization and application. The method is suitable for communication, industrial control, test measurement and other environments with strict signal precision requirements.

Description

Technical Field

[0001] This invention relates to the field of harmonic suppression technology, and in particular to a method for real-time harmonic suppression and compensation at the output of a DAC. Background Technology

[0002] With the rapid development of electronic devices towards higher precision and higher bandwidth, the digital-to-analog converter (DAC), as a core device for converting analog signals to digital signals, directly affects the operational performance of the entire electronic system due to the quality of its output signal. In fields such as communications, industrial control, and test and measurement, which rely on high-precision signal transmission, the unavoidable harmonic components generated at the DAC output can lead to signal distortion, interfere with normal signal transmission, and severely restrict the measurement accuracy and data transmission reliability of the system. To solve this key technical problem, it is urgent to develop a technical solution capable of accurately processing DAC output harmonics. By integrating dedicated models and algorithms, real-time suppression and compensation of harmonic components can be achieved, meeting the application requirements of various fields for continuously improving the purity of DAC output signals.

[0003] Existing technologies have significant shortcomings in addressing harmonic issues at DAC outputs. On one hand, traditional harmonic suppression methods lack dynamic adaptability, making it difficult to adjust processing strategies based on the real-time frequency distribution, energy changes, and phase fluctuations of DAC output harmonics. They can only suppress harmonics in fixed frequency bands or specific orders, failing to comprehensively cover complex and varied harmonic scenarios. On the other hand, existing technologies have not achieved deep collaboration between harmonic suppression-related models, algorithms, and calibration test platforms. This results in insufficient accuracy in identifying and processing even-order harmonics, and the weight allocation lacks specificity, leading to a mismatch between harmonic cancellation strength and actual requirements. Furthermore, the lack of an effective real-time feedback correction mechanism makes it difficult to continuously optimize harmonic suppression effects and meet the stringent requirements of high-precision electronic systems for signal distortion control. Summary of the Invention

[0004] In order to overcome the shortcomings and deficiencies of the existing technology, the present invention provides a method for real-time harmonic suppression and compensation at the output of a DAC.

[0005] The technical solution adopted in this invention is a method for real-time harmonic suppression and compensation at the output of a DAC, comprising the following steps: S1, acquiring the original signal data of the DAC output through a high-precision DAC calibration test platform, and extracting the characteristic information of each harmonic component and the signal amplitude, frequency, and phase parameters included in the signal; S2, constructing a cascaded notch filter dynamic harmonic cancellation model, and setting the initial cascade stage number, notch filter center frequency range, and dynamic adjustment threshold based on the acquired harmonic characteristic parameters; S3, analyzing the signal energy distribution of different harmonic frequency bands using a frequency band weight adaptive allocation model, and based on... S4. The even-order harmonic components in the signal are specifically processed using a nonlinear even-order harmonic enhancement algorithm to strengthen their identifiable characteristics. S5. Dynamic cancellation of each harmonic component is performed based on a cascaded notch dynamic harmonic cancellation model, and the cancellation strength is adjusted by combining the frequency band weighting coefficients. S6. The output signal after cancellation processing is detected in real time using a high-precision DAC calibration test platform. The detection results are fed back to the cascaded notch dynamic harmonic cancellation model for dynamic correction of the model parameters, thus completing the real-time suppression and compensation of harmonics at the DAC output.

[0006] Furthermore, the expression for the frequency band weight adaptive allocation model is: ,in, For the first Weighting coefficients for each harmonic frequency band For the first Signal energy values ​​for each frequency band For the first Harmonic suppression priority coefficients for each frequency band The total number of harmonic frequency bands is divided. For the adaptive adjustment coefficient of the weight, For the first Harmonic content values ​​for each frequency band This represents the average harmonic content across all frequency bands.

[0007] Furthermore, the expression for the nonlinear even-order harmonic enhancement algorithm is as follows: ,in, For the enhanced Even-order harmonic signals For the original The amplitude of the even-order harmonics, This is a non-linear amplitude adjustment index. The fundamental angular frequency, For time variables, For the original The initial phase of the even-order harmonics The even-order harmonic enhancement factor is... The exponent of the nonlinear sine term, This is the correlation coefficient for the fundamental frequency.

[0008] Furthermore, the calibration model expression for the high-precision DAC calibration test platform is as follows: ,in, For DAC calibration coefficients, This is the actual output voltage of the DAC. For standard reference voltage, For the harmonic orders involved in the calibration, For the first Calibration weights for subharmonics For the first The content of subharmonics, To calibrate the platform system coefficients, To calibrate the sensitivity adjustment factor.

[0009] Furthermore, the expression for the real-time harmonic suppression feedback model at the DAC output is: ,in, To suppress feedback adjustment, This represents the change in harmonic content. To provide feedback and adjust the time interval, For feedback gain coefficient, This represents the average weighting coefficient for each frequency band. The initial harmonic content, This is the feedback attenuation coefficient.

[0010] Further, step S3 includes the following sub-steps: S31, dividing the original signal acquired by the high-precision DAC calibration test platform into frequency bands, determining the frequency range boundaries of each frequency band based on the harmonic frequency interval and energy distribution characteristics, ensuring that adjacent frequency bands do not overlap and cover all harmonic distribution intervals; S32, extracting the characteristic parameters of harmonic amplitude, frequency fluctuation range, and phase change trend in each frequency band, establishing a characteristic parameter database for each frequency band, and providing data support for weight allocation; S33, substituting the characteristic parameters of each frequency band into the frequency band weight adaptive allocation model, calculating the preliminary weight coefficients, and dynamically adjusting the preliminary weight coefficients in combination with the real-time harmonic content changes of the DAC output signal; S34, outputting the finally determined weight coefficients of each frequency band, synchronously transmitting them to the cascaded notch filter dynamic harmonic cancellation model, as the basis for adjusting the cancellation strength of the model.

[0011] Further, step S4 includes the following sub-steps: S41, separating even-order harmonic components from the acquired DAC output signal, removing fundamental components and odd-order harmonic interference signals mixed in the even-order harmonics using signal filtering technology, and retaining the pure even-order harmonic raw data; S42, determining the initial values ​​of various parameters in the nonlinear even-order harmonic enhancement algorithm, and setting the parameter adjustment range according to the amplitude, frequency, and phase characteristics of the even-order harmonics; S43, inputting the pure even-order harmonic raw data into the nonlinear even-order harmonic enhancement algorithm, and performing nonlinear processing on the amplitude and phase of the even-order harmonics according to the set parameters and adjustment range to enhance the difference characteristics between the even-order harmonics and other signal components; S44, outputting the enhanced even-order harmonic signal and transmitting it to the subsequent cascaded notch filter dynamic harmonic cancellation model to provide support for canceling even-order harmonics.

[0012] Further, step S5 includes the following sub-steps: S51, receiving the frequency band weighting coefficients and the enhanced even-order harmonic signal, and combining them with the initial parameters of the cascaded notch dynamic harmonic cancellation model set in step S2 to construct a complete harmonic cancellation calculation system; S52, for harmonic components of different frequency bands, adjusting the notch center frequency, cancellation gain, and bandwidth parameters of the model according to the corresponding weighting coefficients, so that the model has a stronger cancellation capability for high-weight frequency band harmonics; S53, inputting the processed DAC output signal into the adjusted cascaded notch dynamic harmonic cancellation model, performing targeted cancellation calculations on each harmonic component, and outputting the canceled signal in real time; S54, collecting the harmonic content data of the canceled signal, comparing it with the preset harmonic suppression standard, and fine-tuning the model parameters based on the comparison results to optimize the cancellation effect.

[0013] A method for real-time harmonic suppression and compensation at the output of a DAC is disclosed. This method is implemented through different units, including: a high-precision DAC output signal multi-dimensional acquisition unit, a cascaded notch filter dynamic harmonic cancellation model parameter configuration and calculation unit, a frequency band weight adaptive allocation coefficient real-time calculation unit, a nonlinear even-order harmonic feature enhancement processing unit, a DAC output signal harmonic content real-time detection and feedback unit, and a system parameter dynamic correction and collaborative control unit. The high-precision DAC output signal multi-dimensional acquisition unit is connected to the cascaded notch filter dynamic harmonic cancellation model parameter configuration and calculation unit, the frequency band weight adaptive allocation coefficient real-time calculation unit, and the nonlinear even-order harmonic feature enhancement processing unit, respectively, for synchronously transmitting the acquired signal data to each processing unit. The cascaded notch filter dynamic harmonic cancellation model parameter configuration and calculation unit, the frequency band weight adaptive allocation coefficient real-time calculation unit, and the nonlinear even-order harmonic feature enhancement processing unit are respectively connected to the cascaded notch filter dynamic harmonic cancellation model parameter configuration and calculation unit, the frequency band weight adaptive allocation coefficient real-time calculation unit, and the nonlinear even-order harmonic feature enhancement processing unit, for synchronously transmitting the acquired signal data to each processing unit. The harmonic cancellation model parameter configuration and calculation unit is connected to the nonlinear even-order harmonic characteristic enhancement processing unit and the DAC output signal harmonic content real-time detection and feedback unit. It receives the enhanced even-order harmonic signal and performs cancellation operation, while simultaneously receiving detection feedback data. The frequency band weight adaptive allocation coefficient real-time calculation unit is connected to the cascaded notch filter dynamic harmonic cancellation model parameter configuration and calculation unit, transmitting the weight coefficients. The DAC output signal harmonic content real-time detection and feedback unit is connected to the system parameter dynamic correction and collaborative control unit, feeding back detection data. The system parameter dynamic correction and collaborative control unit is connected to the other five units respectively, correcting the operating parameters of each unit according to the feedback data, and performing collaborative work among the units to complete the real-time suppression and compensation of harmonics at the DAC output.

[0014] Beneficial Effects: This invention proposes a real-time harmonic suppression and compensation method for DAC output. By integrating a cascaded notch dynamic harmonic cancellation model, a frequency band weight adaptive allocation model, a nonlinear even-order harmonic enhancement algorithm, and a high-precision DAC calibration and testing platform, a fully collaborative harmonic processing system is formed. By acquiring the harmonic characteristic parameters of the DAC output signal in real time, the cascaded notch order, the notch center frequency range, and the weight allocation coefficients are dynamically adjusted to achieve accurate adaptation to harmonic components with different frequency distributions, energy changes, and phase fluctuations, comprehensively covering complex and ever-changing harmonic scenarios. Addressing the lack of deep collaboration and real-time feedback mechanisms in existing technologies, this invention enhances the accuracy of even-order harmonic identification through a nonlinear even-order harmonic enhancement algorithm, achieves targeted cancellation intensity control through frequency band weight adaptive allocation, and combines real-time detection and feedback correction from a high-precision testing platform to construct a deep collaborative system of model, algorithm, and calibration platform, continuously optimizing the harmonic suppression effect. It significantly improves the purity of the DAC output signal, significantly reduces signal distortion caused by various harmonics, and enhances the stability and reliability of signal transmission; it does not rely on fixed processing parameters, adapts to harmonic characteristic changes under different operating conditions, and expands the applicability of the method; through multi-model collaboration and real-time feedback mechanisms, it improves the accuracy and timeliness of harmonic suppression, meeting the stringent requirements for high-precision signal output in fields such as communication, industrial control, and test and measurement. Attached Figure Description

[0015] Figure 1 This is a flowchart of the method steps of the present invention;

[0016] Figure 2 This is a diagram showing the unit composition for implementing the method of the present invention. Detailed Implementation

[0017] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. The application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0018] like Figure 1 As shown, a method for real-time harmonic suppression and compensation at the output of a DAC includes the following steps:

[0019] S1: The raw signal data of the DAC output terminal is acquired through a high-precision DAC calibration test platform, and the characteristic information of each harmonic component, as well as the signal amplitude, frequency, and phase parameters are extracted from the signal.

[0020] Specifically, step S1 involves signal acquisition using a high-precision DAC calibration test platform. This platform is equipped with a 24-bit resolution acquisition module, with the sampling frequency set to eight times the highest frequency of the DAC output signal. The acquisition duration is fixed at 60 consecutive signal cycles to ensure complete capture of the signal's periodic characteristics and harmonic distribution patterns. During the acquisition process, three core parameters—amplitude, frequency, and phase—are extracted simultaneously. The amplitude acquisition accuracy is controlled within 0.1% of the reference value, the frequency measurement resolution reaches 1Hz, and the phase detection error does not exceed 0.5 degrees. Fast Fourier Transform (FFT) technology is used to separate the fundamental and harmonic components, clearly defining the harmonic order as covering the 2nd to 20th harmonics. The amplitude percentage, frequency offset, and phase deviation data of each harmonic are recorded, forming a comprehensive dataset including the original signal waveform data and a harmonic characteristic parameter table. During implementation, a synchronous clock module ensures real-time synchronization between the acquired signal and the DAC output signal, with the synchronization error controlled within 10 nanoseconds. An 8-channel parallel acquisition method is used, with each channel's data transmission rate set to 1000 megabytes per second, ensuring the acquisition and storage of no less than 100,000 data sets within 60 signal cycles. This step provides accurate and comprehensive foundational data for subsequent model building, parameter configuration, and harmonic processing, avoiding deviations in subsequent processing results due to missing or distorted data. The acquisition parameters are all set to improve data integrity and accuracy, laying the data foundation for the entire harmonic suppression and compensation process.

[0021] S2, Construct a cascaded notch dynamic harmonic cancellation model, and set the initial cascade level, notch center frequency range and dynamic adjustment threshold based on the collected harmonic characteristic parameters.

[0022] Specifically, step S2 focuses on the construction and initial parameter configuration of the cascaded notch dynamic harmonic cancellation model. The number of cascaded stages of the model is dynamically determined based on the harmonic order collected in S1. Eight stages are set for harmonic orders from 2nd to 10th, and 12 stages for harmonic orders from 11th to 20th. Each notch stage corresponds to a specific frequency band of harmonic components. The stages are connected in series to achieve step-by-step signal processing. The notch center frequency range is set according to the frequency distribution of each harmonic. The notch center frequency range corresponding to the 2nd harmonic is twice the fundamental frequency ± 5% of the fundamental frequency. The notch center frequency ranges corresponding to the 4th to 20th harmonics are set in the same proportion. The center frequency of each notch can be dynamically adjusted within its corresponding range. The dynamic adjustment threshold is set based on the proportion of harmonic amplitude. When the proportion of harmonic amplitude corresponding to a certain notch stage exceeds 3%, the parameter adjustment mechanism for that notch stage is triggered. The initial damping coefficient of the model was uniformly set to 0.2, the quality factor to 50, and the attenuation slope of each notch filter was controlled at 60 dB / decimal frequency to ensure that the notch filters have steep attenuation characteristics near the center frequency. Simultaneously, the signal passband loss was controlled within 0.5 dB to balance overall signal transmission efficiency. This step, by clarifying the model's hierarchical structure, connection method, and initial parameter boundaries, provides a stable and dynamically adjustable computational framework for subsequent harmonic cancellation. The parameter settings strictly match the actual harmonic characteristics acquired by S1, reserving 5% parameter adjustment space to meet subsequent real-time optimization needs, ensuring the model's adaptability to harmonic variations. Its implementation directly determines the basic efficiency and adjustment flexibility of subsequent harmonic cancellation.

[0023] S3 uses a frequency band weight adaptive allocation model to analyze the signal energy distribution of different harmonic frequency bands and allocates corresponding weight coefficients according to the harmonic frequency distribution density.

[0024] Specifically, step S3 employs a frequency band weight adaptive allocation model to calculate weight coefficients. First, the frequency range of the DAC output signal is divided into nine frequency bands, each with a bandwidth equal to one fundamental frequency. The first band covers 1.5 to 2.5 times the fundamental frequency (corresponding to the 2nd harmonic), the second band covers 2.5 to 3.5 times (corresponding to the 3rd to 4th harmonics), and subsequent bands are divided sequentially with the same bandwidth, ensuring that harmonics from the 2nd to the 20th harmonics are accurately assigned to their corresponding bands. The model calculates the total signal energy within each frequency band using a time-domain integration method. The integration time is set to 10 signal cycles, and the integration step size is 1 / 10 of the signal sampling period, ensuring the stability and accuracy of the energy values. The weight allocation is adjusted based on the harmonic frequency distribution density: for frequency bands with a density higher than the 3rd harmonic per Hz, the base value of the weight coefficient is increased by 20%; for frequency bands with a density lower than the 1st harmonic per Hz, the base value of the weight coefficient is decreased by 15%. Simultaneously, a secondary adjustment is made based on the amplitude proportion of harmonics in each frequency band. For frequency bands with an amplitude proportion exceeding 5%, the weighting coefficient is increased by 10% based on the baseline weighting coefficient; for frequency bands with an amplitude proportion below 1%, it is decreased by 8%. The final output weighting coefficient is strictly limited to the range of 0.1 to 0.9, and the sum of the weighting coefficients for all frequency bands is forcibly normalized to 1. The weighting coefficient is updated every 5 signal cycles. This step achieves precise weight allocation through multi-dimensional parameter calculation, enabling subsequent harmonic cancellation to focus on high-energy, high-distribution-density harmonic frequency bands, improving the targeting and efficiency of suppression. Its implementation process, through quantitative frequency band division and weight calculation rules, ensures the objectivity and real-time nature of weight allocation, providing key support for differentiated harmonic suppression.

[0025] S4. The even harmonic components in the signal are processed in a targeted manner using a nonlinear even harmonic enhancement algorithm to enhance the identifiable characteristics of even harmonics.

[0026] Specifically, step S4 utilizes a nonlinear even-harmonic enhancement algorithm to process even-harmonic components. First, bandpass filtering separates the 2nd, 4th, and 20th even harmonics. The filter's center frequency strictly corresponds to the frequency of each even harmonic, with a bandwidth set to 3% of the corresponding harmonic frequency and a filter order of 12 to ensure the removal of odd harmonics and fundamental interference. The signal-to-noise ratio of the filtered even-harmonic signal is increased to over 40 dB. The algorithm enhances the amplitude characteristics of even harmonics by adjusting the nonlinear gain coefficient. The gain coefficient is dynamically set based on the original amplitude of the even harmonic: for even harmonics with an original amplitude lower than 2% of the fundamental amplitude, the gain coefficient is set to 3.5; for those between 2% and 5%, it is set to 2.5; and for those higher than 5%, it is set to 1.5, avoiding excessive enhancement that could lead to signal distortion. Simultaneously, phase correction technology adjusts the phase deviation of the even harmonics, with a correction range of -10 degrees to 10 degrees and a correction step size of 0.1 degrees, ensuring that the phase characteristics of the even harmonics are more easily recognized by the subsequent cancellation model. The algorithm's nonlinear processing stage employs a third-order polynomial transformation with polynomial coefficients set to 0.08, -0.03, and 0.002. This transformation enhances the characteristic differences between even-order harmonics and the fundamental and odd-order harmonics, improving the amplitude recognition of even-order harmonics by over 30%. This step effectively addresses the issues of small amplitude and susceptibility to interference by specifically enhancing the identifiable characteristics of even-order harmonics, providing a clear target signal for subsequent accurate cancellation. All parameters are set based on the actual characteristics of even-order harmonics, strictly controlling the signal distortion rate to within 1% while balancing enhancement effects to ensure overall signal integrity.

[0027] S5, based on the cascaded notch dynamic harmonic cancellation model, performs dynamic cancellation operation on each harmonic component, and adjusts the cancellation strength by combining the frequency band weighting coefficient.

[0028] Specifically, step S5 implements harmonic cancellation based on the cascaded notch dynamic harmonic cancellation model. First, the signal processed in S4 is input into the model through a buffer module. The buffer duration of the buffer module is set to one signal cycle to ensure continuous signal transmission. The cancellation strength of each notch is adjusted according to the weighting coefficients assigned in S3: notches with weighting coefficients higher than 0.7 have a cancellation gain of 0.9; those between 0.4 and 0.7 have a cancellation gain of 0.7; and those below 0.4 have a cancellation gain of 0.5. During the cancellation process, the model uses a real-time monitoring module to detect the amplitude changes of each harmonic every two signal cycles. When the amplitude of a harmonic drops below 0.5% of the fundamental amplitude, the corresponding notch cancellation gain decreases by 20%; when it remains above 2% of the fundamental amplitude, the corresponding notch cancellation gain increases by 15%. Simultaneously, the center frequency of the notch filter is dynamically adjusted in step size of 0.1% of the corresponding harmonic frequency. Before each adjustment, the harmonic frequency offset direction is confirmed through a phase-locking module to ensure that the notch filter is always aligned with the harmonic frequency, avoiding cancellation failure due to frequency offset. Signal buffer modules are set between each level of notch filter in the model, with a buffer duration of one signal cycle to ensure the continuity and stability of signal processing at each level. This step, by dynamically adjusting the cancellation parameters in conjunction with weighting coefficients, achieves differentiated suppression of harmonics of different importance. The real-time adjustment mechanism ensures that the cancellation effect is unaffected by harmonic frequency offset or amplitude changes, significantly improving the accuracy and dynamic adaptability of harmonic suppression. Its implementation process involves multi-parameter linkage adjustment, enabling harmonic cancellation to accurately match the real-time changing harmonic characteristics.

[0029] S6 uses a high-precision DAC calibration test platform to detect the output signal after cancellation processing in real time, and feeds back the detection results to the cascaded notch dynamic harmonic cancellation model to dynamically correct the model parameters, thereby completing the real-time suppression and compensation of harmonics at the DAC output.

[0030] Specifically, step S6 uses a high-precision DAC calibration test platform to perform real-time detection and feedback correction on the output signal after cancellation processing. The detection module uses the same 24-bit resolution acquisition unit as S1, and the sampling frequency remains consistent to ensure the accuracy of parameter comparison. Detection indicators include total harmonic distortion (THD), the percentage of each harmonic amplitude, and the stability of the fundamental amplitude. The detection accuracy of THD is controlled within 0.05%, the detection resolution of the percentage of each harmonic amplitude reaches 0.01%, and the stability of the fundamental amplitude is statistically analyzed on a 10-minute cycle, with fluctuations controlled within 0.2% of the baseline value. When the THD is detected to be higher than 0.3%, or the percentage of a certain harmonic amplitude is higher than 0.5%, the platform converts the detection data into a feedback signal and transmits it to the parameter adjustment module of the cascaded notch dynamic harmonic cancellation model. The feedback delay is controlled within 5 signal cycles. The feedback correction cycle is set to 5 signal cycles. Each correction adjusts the notch center frequency, cancellation gain, and frequency band weighting coefficient based on the detection data, with the adjustment amplitude not exceeding 10% of the current parameter value to avoid signal instability caused by parameter mutations. Simultaneously, the model parameters, corrected by feedback, are synchronized to the frequency band weight adaptive allocation model and the nonlinear even-order harmonic enhancement algorithm, achieving collaborative parameter optimization of the entire processing system. This step, through a closed-loop detection and feedback mechanism, continuously monitors the harmonic suppression effect, promptly corrects model parameter deviations, ensures the long-term stability and high purity of the DAC output signal, and guarantees that the harmonic suppression effect always meets the preset standards, forming a complete real-time processing closed loop. Its implementation process, through the combination of precise detection and robust correction, provides a guarantee for continuous optimization of the entire harmonic suppression and compensation method.

[0031] Preferably, the expression for the cascaded notch dynamic harmonic cancellation model is: ,in, This is the frequency response function of the cascaded notch dynamic harmonic cancellation model. For cascaded notch series, For the first The cancellation gain coefficient of the notch filter. For the first The center frequency of the notch filter. The input signal frequency, For the first The bandwidth adjustment factor for a notch filter. For the first Phase compensation coefficient of notch filter, This is a dynamic correction factor based on the harmonic frequency distribution of the DAC output.

[0032] Specifically, the cascaded notch dynamic harmonic cancellation model utilizes multiple cascaded notch units working collaboratively. The frequency response characteristics of each notch unit are determined by specific computational relationships, and the overall processing effect is obtained by sequentially superimposing the response results of each stage. The cascade number involved in the model ranges from 8 to 12 stages, with specific values ​​dynamically matched based on the harmonic orders collected in the early stages to ensure coverage of all harmonic frequency bands to be suppressed. The cancellation gain coefficient ranges from 0.5 to 0.9, with higher gain coefficients assigned to notch units corresponding to high-energy harmonics and lower gain coefficients assigned to notch units corresponding to low-energy harmonics. The notch center frequency strictly corresponds to the actual frequency of each harmonic, and the bandwidth adjustment factor is set from 0.1 to 0.3. This factor controls the frequency response bandwidth of the notch unit, ensuring accurate coverage of the target harmonics without affecting the fundamental signal. The phase compensation coefficient is adjusted from -10 degrees to 10 degrees to correct the phase deviation of the notch unit, ensuring precise alignment between the cancellation phase and the harmonic phase. The dynamic correction factor is updated every 5 signal cycles based on the real-time frequency distribution changes of the DAC output harmonics, and its value ranges from 0.8 to 1.2. The model is implemented by connecting notch filter units in series at each stage to process the input signal sequentially. Through multi-parameter coordinated adjustment, it achieves dynamic cancellation of harmonics of different frequency bands and energies, improving the accuracy and dynamic adaptability of harmonic cancellation and ensuring good suppression even when harmonic frequencies and energies change.

[0033] Preferably, the expression for the frequency band weight adaptive allocation model is: ,in, For the first Weighting coefficients for each harmonic frequency band For the first Signal energy values ​​for each frequency band For the first Harmonic suppression priority coefficients for each frequency band The total number of harmonic frequency bands is divided. For the adaptive adjustment coefficient of the weight, For the first Harmonic content values ​​for each frequency band This represents the average harmonic content across all frequency bands.

[0034] Specifically, the frequency band weight adaptive allocation model obtains the weight coefficients of each harmonic frequency band through multi-dimensional parameter calculations. The weight coefficients are calculated based on the signal energy of the frequency band and dynamically adjusted in conjunction with the harmonic suppression priority and the difference in harmonic content. The signal energy value of the frequency band is calculated using the time-domain integration method, with the integration time set to 10 signal cycles to ensure the stability of the energy calculation. The harmonic suppression priority coefficient is set according to the degree of influence of harmonics on signal distortion within the frequency band. The priority coefficient for high-impact frequency bands is 0.8 to 1.0, for medium-impact frequency bands it is 0.5 to 0.7, and for low-impact frequency bands it is 0.2 to 0.4. The total number of harmonic frequency bands is fixed at 9, covering the entire frequency range from the 2nd to the 20th harmonics. The adaptive weight adjustment coefficient is set to 0.3 to 0.5 to control the degree of influence of the difference in harmonic content on the weight coefficients. The harmonic content value of each frequency band is extracted through spectrum analysis technology, and the average harmonic content value is the arithmetic mean of the harmonic content of all frequency bands. The implementation process of this model involves first dividing the harmonic frequency bands and calculating the energy and harmonic content of each band. Then, the initial weighting coefficients are obtained by substituting them into the calculation rules. Finally, the coefficient values ​​are dynamically adjusted based on real-time changes in harmonic distribution. The weighting coefficients are strictly limited to a range of 0.1 to 0.9, and the sum of the coefficients for all frequency bands is normalized to 1, achieving precise weight allocation. This allows subsequent harmonic cancellation to focus on high-energy, high-impact harmonic frequency bands, improving overall suppression efficiency and targeting.

[0035] Preferably, the expression for the nonlinear even-order harmonic enhancement algorithm is: ,in, For the enhanced Even-order harmonic signals For the original The amplitude of the even-order harmonics, This is a non-linear amplitude adjustment index. The fundamental angular frequency, For time variables, For the original The initial phase of the even-order harmonics The even-order harmonic enhancement factor is... The exponent of the nonlinear sine term, This is the correlation coefficient for the fundamental frequency.

[0036] Specifically, the nonlinear even-order harmonic enhancement algorithm strengthens the characteristics of even-order harmonics through nonlinear transformation. The calculation process is based on the original amplitude, fundamental angular frequency, and initial phase of the even-order harmonics, combined with multiple nonlinear parameters to achieve feature enhancement. The original amplitude of the even-order harmonics is extracted after bandpass filtering. The nonlinear amplitude adjustment index is set to 1.5 to 3.0, which is dynamically selected according to the magnitude of the original amplitude of the even-order harmonics. The smaller the original amplitude, the larger the index, ensuring that weak even-order harmonics can be sufficiently enhanced. The product of the time variable and the fundamental angular frequency is used to determine the time-phase characteristics of the even-order harmonics, and the initial phase is kept consistent with the original even-order harmonics. The even-order harmonic enhancement coefficient is set to 0.2 to 0.5 to control the intensity of nonlinear enhancement and avoid excessive enhancement that could lead to signal distortion. The nonlinear sine term exponent is set to 2 to 4, and the fundamental frequency correlation coefficient is set to 1 to 3. These two parameters are used to construct the correlation between the even-order harmonics and the fundamental frequency, strengthening the unique characteristics of the even-order harmonics. The algorithm is implemented by first separating the pure even-order harmonic signal, then substituting it into the computational logic for nonlinear processing. During the processing, the signal distortion rate is monitored in real time to ensure that the distortion rate is controlled within 1%, which solves the problem that even-order harmonics have small amplitude and are easily masked by interference. By enhancing the identifiable characteristics of even-order harmonics, it provides a clear target signal for the accurate cancellation of subsequent cascaded notch filter models, thereby improving the suppression effect of even-order harmonics.

[0037] Preferably, the calibration model expression of the high-precision DAC calibration test platform is: ,in, For DAC calibration coefficients, This is the actual output voltage of the DAC. For standard reference voltage, For the harmonic orders involved in the calibration, For the first Calibration weights for subharmonics For the first The content of subharmonics, To calibrate the platform system coefficients, To calibrate the sensitivity adjustment factor.

[0038] Specifically, the calibration model of the high-precision DAC calibration test platform calculates calibration coefficients by combining the difference between the actual DAC output voltage and the standard reference voltage, along with harmonic content and calibration parameters, providing a precise calibration benchmark for the entire harmonic suppression system. Both the actual DAC output voltage and the standard reference voltage are acquired using a 24-bit resolution acquisition module, with acquisition accuracy controlled within 0.05% of the benchmark value. The harmonic orders involved in the calibration cover from the 2nd to the 20th, ensuring that all harmonics affecting signal quality are included in the calibration range. The calibration weight of each harmonic is set according to its influence on calibration accuracy, with high-impact harmonics having a calibration weight of 0.8 to 1.0 and low-impact harmonics having a weight of 0.2 to 0.4. The calibration platform system coefficient is a fixed value of 0.95, used to correct for the platform's own system errors. The calibration sensitivity adjustment factor is set to 0.3 to 0.6, which adjusts the calibration model's response sensitivity to voltage differences and harmonic content. The implementation process of this model involves first acquiring the actual output voltage of the DAC and the standard reference voltage, extracting the content of each harmonic, and then substituting them into the calculation rules to calculate the calibration coefficients. The calibration coefficients are recalculated every 10 minutes to ensure the long-term stability of calibration accuracy. The calibration coefficients range from 0.8 to 1.2, providing a precise calibration benchmark for the cascaded notch filter model and the weighted allocation model, correcting system errors, and improving the overall accuracy of the entire harmonic suppression and compensation method.

[0039] Preferably, the expression for the real-time harmonic suppression feedback model at the DAC output is: ,in, To suppress feedback adjustment, This represents the change in harmonic content. To provide feedback and adjust the time interval, For feedback gain coefficient, This represents the average weighting coefficient for each frequency band. The initial harmonic content, This is the feedback attenuation coefficient.

[0040] Specifically, the real-time harmonic suppression feedback model at the DAC output obtains the suppression feedback adjustment amount through the coordinated calculation of parameters such as harmonic content change, feedback adjustment time interval, and average weighting coefficient, providing a basis for model parameter correction. The harmonic content change is obtained by comparing the current harmonic content with the harmonic content of the previous cycle, with the calculation period set to 5 signal cycles. The feedback adjustment time interval is fixed at 2 signal cycles to ensure the real-time nature of the feedback. The feedback gain coefficient is set to 0.4 to 0.6 to control the magnitude of the feedback adjustment amount and avoid over-adjustment that could lead to system instability. The average weighting coefficient for each frequency band is the arithmetic mean of the weighting coefficients for all frequency bands, and the initial harmonic content is the harmonic content data collected when the system starts up. The feedback attenuation coefficient is set to 0.1 to 0.2 to slow down the attenuation rate of the feedback adjustment amount and ensure the continuity of the adjustment effect. The implementation of this model involves real-time acquisition of harmonic content data, calculation of the content change, substitution into the calculation logic to obtain the feedback adjustment amount, and then transmitting the adjustment amount to the cascaded notch dynamic harmonic cancellation model to correct parameters such as the notch center frequency and cancellation gain of the model. The feedback adjustment ranges from -0.2 to 0.2, constructing a real-time feedback closed loop. By continuously monitoring the harmonic suppression effect and dynamically correcting the model parameters, the parameter drift problem in the harmonic suppression process is solved, ensuring the long-term stable operation of the system and maintaining a good harmonic suppression effect.

[0041] Preferably, step S3 includes the following sub-steps: S31, dividing the original signal acquired by the high-precision DAC calibration test platform into frequency bands, determining the frequency range boundaries of each frequency band based on the harmonic frequency interval and energy distribution characteristics, ensuring that adjacent frequency bands do not overlap and cover all harmonic distribution intervals; S32, extracting the characteristic parameters of harmonic amplitude, frequency fluctuation range, and phase change trend in each frequency band, establishing a characteristic parameter database for each frequency band, and providing data support for weight allocation; S33, substituting the characteristic parameters of each frequency band into the frequency band weight adaptive allocation model, calculating the preliminary weight coefficients, and dynamically adjusting the preliminary weight coefficients in combination with the real-time harmonic content changes of the DAC output signal; S34, outputting the finally determined weight coefficients of each frequency band, and synchronously transmitting them to the cascaded notch filter dynamic harmonic cancellation model as the basis for adjusting the cancellation strength of the model.

[0042] Specifically, step S3 involves the following steps: S31 first, the raw signal acquired by the high-precision DAC calibration test platform is divided into frequency bands. Based on the harmonic frequency intervals and energy distribution characteristics, the frequency range boundaries of each band are determined. The signal is divided into 9 bands with a bandwidth equal to one fundamental frequency. The first band covers 1.5 to 2.5 times the fundamental frequency, and subsequent bands increase by one bandwidth sequentially, ensuring no overlap between adjacent bands and complete coverage of the 2nd to 20th harmonic distribution range. The frequency band boundary error is controlled within 0.5% of the fundamental frequency. S32 extracts characteristic parameters such as the amplitude, frequency fluctuation range, and phase change trend of the harmonics within each band. The amplitude extraction accuracy is 0.1% of the reference value, the frequency fluctuation range measurement resolution reaches 1 Hz, and the phase change trend is obtained by fitting data from 30 consecutive signal cycles. Simultaneously, a database of characteristic parameters for each frequency band is established, with an update cycle of 5 signal cycles to ensure data timeliness. S33 substitutes the characteristic parameters of each frequency band into the frequency band weight adaptive allocation model to calculate the initial weight coefficients. Then, the coefficients are adjusted according to the real-time harmonic content changes of the DAC output signal, with an adjustment step size of 0.05. After each adjustment, the sum of the coefficients is verified to be 1. If it deviates, normalization correction is performed. S34 outputs the final determined weight coefficients for each frequency band, with the coefficient values ​​strictly limited to the range of 0.1 to 0.9. This is simultaneously transmitted to the cascaded notch filter dynamic harmonic cancellation model through a high-speed data transmission channel, with the transmission delay controlled within 2 signal cycles. This step-by-step implementation method achieves orderly connection between frequency band division, parameter extraction, weight calculation, and coefficient transmission through layered processing, providing accurate weight basis for subsequent harmonic cancellation, ensuring that high-energy, high-distribution-density frequency bands are suppressed in a focused manner, and improving the overall processing targeting and efficiency.

[0043] Preferably, step S4 includes the following sub-steps: S41, separating even-order harmonic components from the acquired DAC output signal, removing fundamental components and odd-order harmonic interference signals mixed in the even-order harmonics using signal filtering technology, and retaining pure even-order harmonic raw data; S42, determining the initial values ​​of various parameters in the nonlinear even-order harmonic enhancement algorithm, and setting the parameter adjustment range according to the amplitude, frequency, and phase characteristics of the even-order harmonics; S43, inputting the pure even-order harmonic raw data into the nonlinear even-order harmonic enhancement algorithm, and performing nonlinear processing on the amplitude and phase of the even-order harmonics according to the set parameters and adjustment range to enhance the difference characteristics between the even-order harmonics and other signal components; S44, outputting the enhanced even-order harmonic signal and transmitting it to the subsequent cascaded notch filter dynamic harmonic cancellation model to provide support for canceling even-order harmonics.

[0044] Specifically, in step S4, S41 first separates the even-order harmonic components from the acquired DAC output signal. A 12th-order bandpass filter is used to remove the fundamental frequency component and odd-order harmonic interference signals mixed in the even-order harmonics. The filter's center frequency corresponds to each even-order harmonic frequency, and the bandwidth is set to 3% of the corresponding harmonic frequency. After filtering, the signal-to-noise ratio of the even-order harmonic signal is increased to over 40 dB, ensuring the preservation of pure original even-order harmonic data. S42 determines the initial values ​​of various parameters for the nonlinear even-order harmonic enhancement algorithm. The nonlinear amplitude adjustment index is initially set to 2.0, the even-order harmonic enhancement coefficient is initially set to 0.3, the nonlinear sine term index is initially set to 3, and the fundamental frequency correlation coefficient is initially set to 2. Simultaneously, the adjustment range of each parameter is set according to the even-order harmonic amplitude, frequency, and phase characteristics. The amplitude adjustment index can vary between 1.5 and 3.0, and the enhancement coefficient can vary between 0.2 and 0.5. S43 inputs the clean even-order harmonic raw data into the nonlinear even-order harmonic enhancement algorithm. According to the set parameters and adjustment range, the amplitude and phase of the even-order harmonics are nonlinearly processed. The amplitude adjustment uses a stepped gain, and the phase adjustment step is 0.1 degrees. This processing enhances the difference characteristics between even-order harmonics and other signal components. During processing, the signal distortion rate is monitored in real time to ensure that the distortion rate does not exceed 1%. S44 outputs the enhanced even-order harmonic signal. After the signal is temporarily stored for one signal cycle through a buffer module, it is transmitted to the subsequent cascaded notch filter dynamic harmonic cancellation model. During transmission, the signal amplitude fluctuation is controlled within 0.5%. This step-by-step implementation method effectively solves the problem of the difficulty in even-order harmonic identification by coherently executing filtering, parameter configuration, enhancement processing, and signal transmission, providing a clear target signal for accurate cancellation.

[0045] Preferably, step S5 includes the following sub-steps: S51, receiving the frequency band weighting coefficients and the enhanced even-order harmonic signal, and combining them with the initial parameters of the cascaded notch dynamic harmonic cancellation model set in step S2 to construct a complete harmonic cancellation calculation system; S52, adjusting the notch center frequency, cancellation gain, and bandwidth parameters of the model according to the corresponding weighting coefficients for harmonic components of different frequency bands, so that the model has a stronger cancellation capability for high-weight frequency band harmonics; S53, inputting the processed DAC output signal into the adjusted cascaded notch dynamic harmonic cancellation model, performing targeted cancellation calculations on each harmonic component, and outputting the canceled signal in real time; S54, collecting the harmonic content data of the canceled signal, comparing it with the preset harmonic suppression standard, and fine-tuning the model parameters according to the comparison results to optimize the cancellation effect.

[0046] Specifically, step S5 is divided into several steps. In S51, the frequency band weighting coefficient and the enhanced even harmonic signal are first received. The weighting coefficient is verified for completeness and validity by the data verification module. The even harmonic signal is first processed by amplitude normalization. After processing, the amplitude is uniformly mapped to the range of 0.1 to 0.9. Then, combined with the initial parameters of the cascaded notch dynamic harmonic cancellation model set in step S2, including 8 to 12 cascade levels, an initial damping coefficient of 0.2, and a quality factor of 50, a complete harmonic cancellation calculation system is constructed. After the system is started, an initial self-check is performed to ensure that the connection of each module is normal. S52 adjusts the notch center frequency, cancellation gain, and bandwidth parameters of the model according to the corresponding weighting coefficients for harmonic components in different frequency bands. For frequency bands with weighting coefficients higher than 0.7, the notch center frequency accuracy is improved to 0.1%, the cancellation gain is set to 0.9, and the bandwidth is narrowed to 2% of the harmonic frequency. For frequency bands with weighting coefficients between 0.4 and 0.7, the cancellation gain is set to 0.7 and the bandwidth is set to 3%. For frequency bands with weighting coefficients lower than 0.4, the cancellation gain is set to 0.5 and the bandwidth is set to 4%, giving the model a stronger cancellation capability for harmonics in high-weight frequency bands. S53 inputs the processed DAC output signal into the adjusted cascaded notch dynamic harmonic cancellation model, and performs targeted cancellation operations on each harmonic component in a cascaded order. The processing time for each notch stage is controlled within one signal cycle, and the canceled signal is output in real time, with the sampling frequency of the output signal consistent with that of the input. The S54 collects the harmonic content data of the signal after cancellation processing and compares it with the preset harmonic suppression standard. The preset standard is that the total harmonic distortion is less than 0.3% and the single harmonic amplitude ratio is less than 0.5%. Based on the comparison results, the model parameters are fine-tuned, and the fine-tuning range does not exceed 10% of the current parameters. This step-by-step implementation method significantly improves the accuracy and dynamic adaptability of harmonic suppression through closed-loop execution of system construction, parameter adjustment, cancellation calculation and effect optimization, ensuring that the ideal cancellation effect can be achieved under different working conditions.

[0047] The cascaded notch dynamic harmonic cancellation model is a harmonic processing model consisting of 8 to 12 series notch units. Each notch unit corresponds to a specific frequency band harmonic, and the precise cancellation of the 2nd to 20th harmonics is achieved through multi-parameter coordinated adjustment. The implementation process of this model is as follows: First, the number of cascade stages is determined based on the harmonic orders collected in the early stage. Stages are set to 8 for harmonic orders 2 to 10, and to 12 for harmonic orders 11 to 20. Then, the center frequencies of each notch filter are set within a range of 2 ± 5% of the fundamental frequency, with an initial damping coefficient of 0.2, a quality factor of 50, and an attenuation slope of 60 dB / decimal. During operation, the harmonic amplitude change is detected every two signal cycles. When the harmonic amplitude is lower than 0.5% of the fundamental amplitude, the cancellation gain is reduced by 20%, and when it is higher than 2%, the cancellation gain is increased by 15%. Simultaneously, the center frequency is dynamically adjusted in steps of 0.1% of the harmonic frequency. Through multi-stage cascade processing and dynamic parameter adjustment, harmonic components of different frequency bands and energies are specifically canceled, avoiding fundamental signal loss exceeding 0.5 dB. This model breaks through the limitations of traditional fixed-parameter notch filters, achieving real-time adaptation to harmonic frequency shifts and energy fluctuations. It improves harmonic suppression accuracy to less than 0.3% total distortion, providing core assurance for the purity of DAC output signals and meeting the stringent requirements of high-precision electronic systems for signal stability.

[0048] The frequency band weight adaptive allocation model is a weight control model based on multi-dimensional parameter calculations. It allocates weight coefficients by quantitatively analyzing harmonic distribution characteristics to achieve differentiated harmonic suppression. The implementation process of this model is as follows: First, the signal frequency is divided into 9 frequency bands with a bandwidth of 1 times the fundamental frequency, covering the 2nd to 20th harmonics. Then, the energy of each frequency band is calculated by time-domain integration over 10 signal cycles. The baseline weight is determined by combining the frequency distribution density of the 1st to 3rd harmonics per hertz and the amplitude proportion of 1% to 5%. Subsequently, the weights are corrected with an adaptive adjustment coefficient of 0.3 to 0.5 to ensure that the coefficient value is between 0.1 and 0.9 and the sum is normalized to 1. The coefficient is updated every 5 signal cycles, providing an accurate weight basis for the cascaded notch filter model. This allows high-energy, high-distribution-density frequency bands to receive higher cancellation priority. For example, frequency bands with a weight coefficient higher than 0.7 correspond to a cancellation gain of 0.9, and frequency bands with a weight coefficient lower than 0.4 correspond to a cancellation gain of 0.5. This model addresses the problem of low efficiency in traditional uniform weight suppression. By dynamically allocating weights, it improves the targeted nature of harmonic suppression by more than 30%, reducing the system's computational load while ensuring suppression effectiveness, thus providing support for efficient processing in complex harmonic scenarios.

[0049] The nonlinear even-order harmonic enhancement algorithm is a feature enhancement processing algorithm for 2nd, 4th to 20th even-order harmonics. It improves the identifiability of even-order harmonics through a combination of filtering, gain adjustment and phase correction. The implementation process of this algorithm is as follows: First, a 12th-order bandpass filter is used to separate even-order harmonics. The filter bandwidth is set to 3% of the corresponding harmonic frequency, so that the signal-to-noise ratio after filtering is improved to more than 40 dB. Then, a nonlinear gain coefficient of 1.5 to 3.5 is set according to the original amplitude of the even-order harmonics. When the original amplitude is less than 2% of the fundamental frequency, 3.5 is used, and when it is more than 5% of the fundamental frequency, 1.5 is used. Then, the phase is corrected in a step of 0.1 degrees within the range of -10 degrees to 10 degrees. The difference characteristics between even-order harmonics and other components are enhanced through third-order polynomial transformation, which solves the problem of small amplitude of even-order harmonics and easy masking by interference. This improves the identification of even-order harmonic amplitude by more than 30%, while controlling the signal distortion rate to within 1%. This algorithm provides a clear target signal for the cascaded notch filter model, significantly improves the cancellation accuracy of even harmonics, avoids signal distortion caused by the failure to effectively suppress even harmonics, and ensures that the overall purity of the DAC output signal meets the application requirements of high-precision fields such as communication and industrial control.

[0050] The high-precision DAC calibration and testing platform is a comprehensive support platform with signal acquisition, detection, calibration, and feedback functions. Equipped with a 24-bit resolution acquisition unit, it provides the data foundation and accuracy guarantee for the entire harmonic suppression system. The platform's implementation process is as follows: In the acquisition phase, 60 signal cycles are continuously acquired at a sampling rate 8 times the DAC's highest output frequency, simultaneously extracting amplitude, frequency, and phase parameters. The amplitude accuracy reaches a reference value of 0.1%, and the phase error does not exceed 0.5 degrees. In the calibration phase, calibration coefficients are calculated every 10 minutes. Based on the difference between the actual output voltage and the standard reference voltage, combined with the 2nd to 20th harmonic content and calibration weights of 0.2 to 1.0, the calibration coefficient values ​​range from 0.8 to 1.2. In the detection phase, the processed signal is monitored in real time, with a total harmonic distortion detection accuracy of 0.05%. Every 5 signal cycles, the detection results are fed back to the core model, providing accurate raw data to correct system errors, providing real-time feedback on the suppression effect, and ensuring the timeliness and accuracy of parameter adjustments for each module. The platform establishes a full-process data support and closed-loop feedback mechanism, which improves the long-term stability of the entire harmonic suppression system, controls the fundamental amplitude fluctuation within 0.2%, provides reliable hardware and data support for multi-model collaborative work, and is the key foundation for realizing real-time harmonic suppression and compensation.

[0051] like Figure 2As shown, a method for real-time harmonic suppression and compensation at the output of a DAC is implemented through different units, including: a high-precision DAC output signal multi-dimensional acquisition unit, a cascaded notch filter dynamic harmonic cancellation model parameter configuration and calculation unit, a frequency band weight adaptive allocation coefficient real-time calculation unit, a nonlinear even-order harmonic feature enhancement processing unit, a DAC output signal harmonic content real-time detection and feedback unit, and a system parameter dynamic correction and collaborative control unit; the high-precision DAC output signal multi-dimensional acquisition unit is connected to the cascaded notch filter dynamic harmonic cancellation model parameter configuration and calculation unit, the frequency band weight adaptive allocation coefficient real-time calculation unit, and the nonlinear even-order harmonic feature enhancement processing unit, respectively, for synchronously transmitting the acquired signal data to each processing unit; the cascaded notch filter... The dynamic harmonic cancellation model parameter configuration and calculation unit is connected to the nonlinear even-order harmonic characteristic enhancement processing unit and the DAC output signal harmonic content real-time detection and feedback unit. It receives the enhanced even-order harmonic signal and performs cancellation operation, while simultaneously receiving detection feedback data. The frequency band weight adaptive allocation coefficient real-time calculation unit is connected to the cascaded notch filter dynamic harmonic cancellation model parameter configuration and calculation unit, transmitting the weight coefficients. The DAC output signal harmonic content real-time detection and feedback unit is connected to the system parameter dynamic correction and collaborative control unit, feeding back detection data. The system parameter dynamic correction and collaborative control unit is connected to the other five units respectively, correcting the operating parameters of each unit according to the feedback data, and performing collaborative work among the units to complete the real-time suppression and compensation of harmonics at the DAC output.

[0052] A real-time harmonic suppression and compensation method for DAC output addresses the lack of dynamic adaptability in traditional methods. It leverages a high-precision testing platform to capture the frequency distribution, energy variation, and phase fluctuation characteristics of harmonics at the DAC output in real time. By dynamically adjusting the order, center frequency range, and weighting coefficients of the cascaded notch filter, it overcomes the limitations of fixed parameter processing, achieving comprehensive coverage of complex and variable harmonic scenarios and ensuring targeted processing of harmonic components under different operating conditions. To address the lack of deep collaboration and real-time feedback mechanisms in existing technologies, a nonlinear even-order harmonic enhancement algorithm is used to strengthen the identifiability of even-order harmonics. Adaptive frequency band weighting optimizes the cancellation intensity ratio, and real-time detection data from the high-precision testing platform is used to correct model parameters. This constructs a closed-loop processing system with deep linkage between the model, algorithm, and calibration platform, completely resolving the problems of insufficient processing accuracy and unstable results in traditional techniques.

[0053] This method is highly adaptable, requiring no fixed processing parameters. It can dynamically adjust the processing strategy based on the real-time characteristics of the DAC output harmonics, meeting the harmonic suppression needs of DAC devices under different types and operating conditions. It boasts high processing accuracy, improving the target harmonic identification accuracy through an even-order harmonic enhancement algorithm. Combined with adaptive weight allocation and cascaded notch filtering for dynamic cancellation, it achieves precise suppression of each harmonic, significantly reducing signal distortion. Furthermore, it exhibits good stability, relying on real-time detection and feedback correction mechanisms to continuously optimize processing parameters, ensuring long-term stability of harmonic suppression effects. This effectively improves the purity and reliability of the DAC output signal, providing stable signal assurance for high-precision fields such as communication, industrial control, and testing and measurement.

[0054] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," "connect," "link," and "fix" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal communication between two components. Those skilled in the art will understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0055] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various equivalent changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for harmonic real-time suppression and compensation of a DAC output, characterized by, Comprise the following steps: S1, the original signal data of the output end of the DAC is collected by the high-precision DAC calibration test platform, the harmonic component characteristic information included in the signal and the signal amplitude, frequency and phase parameters are extracted; S2, a cascaded notch dynamic harmonic cancellation model is constructed, the initial cascade number, the notch center frequency interval and the dynamic adjustment threshold of the model are set based on the collected harmonic characteristic parameters; S3, the signal energy distribution of different harmonic frequency bands is analyzed by using a frequency band weight adaptive allocation model, and the corresponding weight coefficient is allocated according to the harmonic frequency distribution density; S4, the even harmonic components in the signal are processed by using a nonlinear even harmonic enhancement algorithm, and the identifiable characteristics of the even harmonics are strengthened; S5, the cascaded notch dynamic harmonic cancellation model is used to implement dynamic cancellation operation on each harmonic component, and the cancellation strength is adjusted in combination with the frequency band weight coefficient; S6, the output signal after cancellation processing is detected in real time by the high-precision DAC calibration test platform, the detection result is fed back to the cascaded notch dynamic harmonic cancellation model, the dynamic correction of the model parameters is performed, and the real-time suppression and compensation of the harmonics of the DAC output end are completed.

2. The method of claim 1, wherein, The expression of the frequency band weight adaptive allocation model is: wherein, is the weight coefficient of the th harmonic frequency band, is the signal energy value of the th frequency band, is the harmonic suppression priority coefficient of the th frequency band, is the total number of harmonic frequency band division, is the weight adaptive adjustment coefficient, is the harmonic content value of the th frequency band, is the average harmonic content value of all frequency bands.

3. The method of claim 1, wherein the DAC output harmonic real-time suppression and compensation method is characterized by, The expression of the non-linear even-harmonic enhancement algorithm is: wherein, is the enhanced even-harmonic signal, is the original even-harmonic amplitude, is the non-linear amplitude adjustment exponent, is the fundamental angular frequency, is the time variable, is the initial phase of the original even-harmonic, is the even-harmonic enhancement coefficient, is the non-linear sinusoidal term exponent, is the fundamental frequency correlation coefficient.

4. The method of claim 1, wherein, The calibration model expression for the high-precision DAC calibration test platform is: ,in, For DAC calibration coefficients, This is the actual output voltage of the DAC. For standard reference voltage, For the harmonic orders involved in the calibration, For the first Calibration weights for subharmonics For the first The content of subharmonics, To calibrate the platform system coefficients, To calibrate the sensitivity adjustment factor.

5. The method of claim 1, wherein, The harmonic real-time suppression feedback model expression of the DAC output end is: Wherein, is the feedback adjustment amount, is the harmonic content change amount, is the feedback adjustment time interval, is the feedback gain coefficient, is the average weight coefficient of each frequency band, is the initial harmonic content, is the feedback attenuation coefficient.

6. The method of claim 1, wherein, The S3 comprises the following steps: S31, the original signal collected by the high-precision DAC calibration test platform is divided into frequency bands, the frequency range boundary of each frequency band is determined according to the harmonic frequency interval and the energy distribution characteristics, and it is ensured that the adjacent frequency bands have no overlap and cover all the harmonic distribution intervals; S32, the harmonic amplitude, frequency fluctuation range and phase change trend characteristic parameters in each frequency band are extracted, and a characteristic parameter database of each frequency band is established to provide data support for weight allocation; S33, the characteristic parameters of each frequency band are substituted into the frequency band weight adaptive allocation model, the preliminary weight coefficient is calculated, and the preliminary weight coefficient is dynamically adjusted in combination with the real-time harmonic content change of the DAC output signal; S34, the finally determined weight coefficient of each frequency band is output and transmitted to the cascaded notch dynamic harmonic cancellation model as the basis for adjusting the cancellation strength of the model.

7. The method of claim 1, wherein, The S4 comprises the following steps: S41, the even harmonic components are separated from the collected DAC output signal, the fundamental component and odd harmonic interference signal mixed in the even harmonic are removed by signal filtering technology, and the pure even harmonic original data is retained; S42, the initial values of each parameter in the nonlinear even harmonic enhancement algorithm are determined, and the parameter adjustment range is set according to the amplitude, frequency and phase characteristics of the even harmonic; S43, the pure even harmonic original data is input into the nonlinear even harmonic enhancement algorithm, the amplitude and phase of the even harmonic are nonlinearly processed according to the set parameters and adjustment range, and the difference characteristics of the even harmonic and other signal components are strengthened; S44, the enhanced even harmonic signal is output and transmitted to the subsequent cascaded notch dynamic harmonic cancellation model to support the cancellation of the even harmonic.

8. The method of claim 1, wherein, The S5 comprises the following steps: S51, receiving the frequency band weight coefficient and the enhanced even harmonic signal, combining the initial parameters of the cascaded notch dynamic harmonic cancellation model set in step S2 to build a complete harmonic cancellation operation system; S52, for the harmonic components of different frequency bands, adjusting the notch center frequency, cancellation gain and bandwidth parameters of the model according to the corresponding weight coefficient, so that the model has stronger cancellation ability for high weight frequency band harmonics; S53, inputting the processed DAC output signal into the adjusted cascaded notch dynamic harmonic cancellation model, performing targeted cancellation operation on each harmonic component, and outputting the signal after cancellation processing in real time; S54, collecting the harmonic content data of the signal after cancellation processing, comparing with the preset harmonic suppression standard, and adjusting the model parameters according to the comparison result to optimize the cancellation effect.

9. The method of harmonic real-time suppression and compensation of a DAC output according to any one of claims 1-8, characterized in that, The method is realized by different units, including: a high-precision DAC output signal multi-dimensional acquisition unit, a cascaded notch dynamic harmonic cancellation model parameter configuration and operation unit, a frequency band weight adaptive allocation coefficient real-time calculation unit, a nonlinear even harmonic feature enhancement processing unit, a DAC output signal harmonic content real-time detection and feedback unit, and a system parameter dynamic correction and cooperative control unit; the high-precision DAC output signal multi-dimensional acquisition unit is connected with the cascaded notch dynamic harmonic cancellation model parameter configuration and operation unit, the frequency band weight adaptive allocation coefficient real-time calculation unit and the nonlinear even harmonic feature enhancement processing unit respectively, and is used for synchronously transmitting the collected signal data to each processing unit; the cascaded notch dynamic harmonic cancellation model parameter configuration and operation unit is connected with the nonlinear even harmonic feature enhancement processing unit and the DAC output signal harmonic content real-time detection and feedback unit, receives the enhanced even harmonic signal and performs cancellation operation, and receives detection feedback data at the same time; the frequency band weight adaptive allocation coefficient real-time calculation unit is connected with the cascaded notch dynamic harmonic cancellation model parameter configuration and operation unit, and transmits the weight coefficient; the DAC output signal harmonic content real-time detection and feedback unit is connected with the system parameter dynamic correction and cooperative control unit, and feeds back the detection data; the system parameter dynamic correction and cooperative control unit is connected with the other five units respectively, corrects the operation parameters of each unit according to the feedback data, cooperatively works with each unit, and completes the real-time suppression and compensation of the DAC output end harmonic.

Images