A harmonic control method and system for frequency converter

Abstract

The invention discloses a harmonic control method and system for a frequency converter, the method comprising: obtaining an output voltage signal and a bus voltage signal of the frequency converter in a current time period; performing spectrum analysis on the output voltage signal to construct a first frequency distribution characteristic curve; correcting the first frequency distribution characteristic curve according to the bus voltage signal to obtain a second frequency distribution characteristic curve; if a frequency drift occurs in the output frequency signal of the frequency converter, tracking the frequency drift of the output frequency signal to obtain a frequency drift law; performing voltage waveform distortion analysis on the output voltage signal to obtain a total harmonic distortion rate; performing parameter optimization on a filter according to the total harmonic distortion rate and the frequency drift law to match a third frequency distribution characteristic of the filter with a second frequency distribution characteristic; and using the optimized filter to suppress harmonics of the frequency converter in the next time period. The invention can effectively improve the harmonic suppression effect of the frequency converter and improve the power supply quality of the power grid.

Description

Technical Field

[0001] The present invention relates to the technical field of frequency converters, and in particular to a harmonic control method and system for a frequency converter. Background Art

[0002] The inverter will generate a large number of harmonics during its actual operation. These harmonic pollution will reduce the working efficiency of the inverter and interfere with the surrounding electronic equipment, making it unable to work normally, thereby distorting the voltage and current waveforms in the power grid, or causing voltage fluctuations, flicker and three-phase imbalance, seriously affecting the power supply quality of the power grid. Therefore, how to effectively reduce the harmonic problem of the inverter has always been a research focus.

[0003] At present, the main way to control the harmonics of the inverter is filtering, such as using a filter reactor. The filter reactor has two spindles with winding coils, which filter out the harmonics by converting the absorbed harmonics into heat energy. However, the filter reactor can only filter out about 30% of the harmonics, and the filtering effect is relatively poor. Another example is using a series inductor for harmonic filtering, but increasing the inductance will affect the inverter output, and a small inductance will have limited filtering effect. Under the premise of not affecting the inverter output, the reactance rate of the filter inductor does not exceed 10%, so the harmonic content can only be reduced to about 70%, and the filtering effect is relatively general. Therefore, how to improve the harmonic suppression effect of the inverter is a technical problem that needs to be solved urgently in this field. Summary of the invention

[0004] In view of the problems existing in the prior art, an embodiment of the present invention provides a harmonic control method and system for a frequency converter, which can effectively improve the harmonic suppression effect of the frequency converter, thereby improving the power supply quality of the power grid.

[0005] In a first aspect, an embodiment of the present invention provides a harmonic control method for a frequency converter, comprising:

[0006] Obtain the output voltage signal and bus voltage signal of the inverter in the current time period;

[0007] Performing spectrum analysis on the output voltage signal to construct a first frequency distribution characteristic curve between the fundamental frequency and each harmonic frequency in the frequency domain;

[0008] According to the bus voltage signal, the first frequency distribution characteristic curve is corrected to obtain a second frequency distribution characteristic curve;

[0009] When the frequency drift occurs in the output frequency signal of the frequency converter, the frequency drift of the output frequency signal is tracked to obtain a frequency drift law;

[0010] Performing voltage waveform distortion analysis on the output voltage signal to obtain a total harmonic distortion rate;

[0011] Optimizing the parameters of the filter according to the total harmonic distortion rate and the frequency drift law, so that the third frequency distribution characteristic curve of the filter matches the second frequency distribution characteristic curve;

[0012] In the next time period, the optimized filter is used to suppress harmonics of the frequency converter.

[0013] As an improvement of the above solution, the frequency spectrum analysis of the output voltage signal is performed to construct a first frequency distribution characteristic curve between the fundamental frequency and each harmonic frequency in the frequency domain, including:

[0014] Performing windowing processing on the output voltage signal using a Hamming window function to obtain a first voltage signal;

[0015] Performing a fast Fourier transform on the first voltage signal to obtain frequency domain data;

[0016] Determine the fundamental frequency according to the frequency point with the largest amplitude in the frequency domain data;

[0017] Normalizing the frequency domain data according to the fundamental frequency to obtain a normalized spectrum;

[0018] Locating the frequencies in the normalized spectrum whose amplitudes exceed a first set threshold, and determining the harmonic order and the frequency sequence corresponding to each harmonic according to the relationship between the located frequencies and the multiples of the fundamental frequency;

[0019] Using a Butterworth filter to smooth the frequency sequence;

[0020] A first frequency distribution characteristic curve is established according to the ratio of the amplitude of the harmonic frequency corresponding to each harmonic in the smoothed frequency sequence to the amplitude corresponding to the fundamental frequency.

[0021] As an improvement of the above solution, the first frequency distribution characteristic curve is corrected according to the bus voltage signal to obtain a second frequency distribution characteristic curve, including:

[0022] Determining the sideband harmonic frequency displacement according to the bus voltage signal;

[0023] The first frequency distribution characteristic curve is corrected according to the sideband harmonic frequency displacement to obtain a second frequency distribution characteristic curve.

[0024] As an improvement of the above solution, determining the sideband harmonic frequency displacement according to the bus voltage signal includes:

[0025] A low-pass filter with a passband range within a preset Hertz range is used to low-pass filter the bus voltage signal to obtain a fluctuation amplitude sequence;

[0026] Performing envelope demodulation on the fluctuation amplitude sequence by using Hilbert transform to obtain an envelope demodulation signal;

[0027] Calculating the fluctuation frequency and fluctuation depth of the bus voltage signal according to the envelope demodulation signal;

[0028] The sideband harmonic frequency displacement is calculated according to the fluctuation frequency and the fluctuation depth.

[0029] As an improvement of the above solution, the first frequency distribution characteristic curve is corrected according to the sideband harmonic frequency displacement to obtain a second frequency distribution characteristic curve, including:

[0030] According to the sideband harmonic frequency displacement, a sideband frequency displacement comparison table is constructed; wherein the sideband frequency displacement comparison table includes the displacement between the fundamental frequency and each harmonic frequency;

[0031] Using the sideband frequency shift comparison table, modifying the first frequency distribution characteristic curve;

[0032] The modified first frequency distribution characteristic curve is divided into interpolation intervals, and each of the interpolation intervals is interpolated using the Lagrange interpolation method to obtain the second frequency distribution characteristic curve.

[0033] As an improvement of the above solution, when the frequency drift occurs in the output frequency signal of the frequency converter, the frequency drift of the output frequency signal is tracked to obtain the frequency drift law, including:

[0034] Obtaining the output frequency signal of the frequency converter in the current time period;

[0035] Performing analog-to-digital conversion on the output frequency signal to obtain a frequency sampling sequence;

[0036] Performing wavelet decomposition on the frequency sampling sequence to obtain a phase feature;

[0037] Tracking the rate of change of the phase characteristic to construct a frequency drift sequence;

[0038] Recursively processing the frequency drift sequence using a Kalman filter algorithm to obtain a slope of the frequency drift;

[0039] When the slope of the frequency drift exceeds the second set threshold, a mapping function between each harmonic frequency and the fundamental frequency is established as a frequency drift rule according to the slope of the frequency drift and the multiple relationship between each harmonic frequency and the fundamental frequency.

[0040] As an improvement of the above solution, the voltage waveform distortion analysis is performed on the output voltage signal to obtain the total harmonic distortion rate, including:

[0041] Performing Fourier transform on the output voltage signal to obtain a fundamental wave component and each harmonic component;

[0042] Calculating the distortion rate of each harmonic according to the fundamental wave component and each harmonic component;

[0043] The distortion rate of each harmonic is subjected to a square root operation to obtain a total harmonic distortion rate.

[0044] As an improvement of the above solution, the method of optimizing the filter parameters according to the total harmonic distortion rate and the frequency drift law so that the third frequency distribution characteristic curve of the filter matches the second frequency distribution characteristic curve includes:

[0045] According to the total harmonic distortion rate and the frequency drift law, a radial basis function neural network is used to iteratively optimize the filter parameters to obtain the filter parameters output in this round; the filter parameters include the center frequency and passband bandwidth of the filter;

[0046] According to the filter parameters outputted in this round, a Butterworth filter is constructed, and a frequency response test is performed on the Butterworth filter to obtain the test error of this round;

[0047] If the test error of this round is greater than the preset error threshold, the radial basis neural network is used again to iteratively optimize the parameters of the filter;

[0048] If the test error of this round is not greater than the preset error threshold, at the end of the current time period, the center frequency and passband bandwidth of the filter are updated using the center frequency and passband bandwidth in the filter parameters output in this round, so that the third frequency distribution characteristic curve of the filter matches the second frequency distribution characteristic curve.

[0049] As an improvement of the above solution, the calculation of the distortion rate of each harmonic according to the fundamental wave component and each harmonic component includes:

[0050] The amplitude ratio of each harmonic component to the fundamental component is calculated to obtain the distortion rate of each harmonic.

[0051] In a second aspect, an embodiment of the present invention provides a harmonic control system for a frequency converter, comprising: a controller, a filter connected to the controller, and a signal sampling module; wherein the filter and the signal sampling module are used to be connected to the frequency converter;

[0052] The signal sampling module is used to obtain the output voltage signal and bus voltage signal of the inverter in the current time period;

[0053] The controller comprises:

[0054] A spectrum analysis module, used for performing spectrum analysis on the output voltage signal, and constructing a first frequency distribution characteristic curve between the fundamental frequency and each harmonic frequency in the frequency domain;

[0055] A correction module, used for correcting the first frequency distribution characteristic curve according to the bus voltage signal to obtain a second frequency distribution characteristic curve;

[0056] A frequency drift tracking module, used to track the frequency drift of the output frequency signal of the frequency converter when the output frequency signal of the frequency converter drifts, so as to obtain a frequency drift law;

[0057] A waveform distortion analysis module, used to perform voltage waveform distortion analysis on the output voltage signal to obtain a total harmonic distortion rate;

[0058] A parameter optimization module, used for optimizing the parameters of the filter according to the total harmonic distortion rate and the frequency drift law, so that the third frequency distribution characteristic curve of the filter matches the second frequency distribution characteristic curve;

[0059] The harmonic suppression module is used to use the optimized filter to suppress harmonics of the frequency converter in the next time period.

[0060] Compared with the prior art, a harmonic control method and system of a frequency converter in an embodiment of the present invention obtains the output voltage signal and bus voltage signal of the frequency converter in the current time period; performs spectrum analysis on the output voltage signal, and constructs a first frequency distribution characteristic curve between the fundamental frequency and each harmonic frequency in the frequency domain; according to the bus voltage signal, the first frequency distribution characteristic curve is corrected to obtain a second frequency distribution characteristic curve; when the output frequency signal of the frequency converter has a frequency drift, the frequency drift of the output frequency signal is tracked to obtain the frequency drift law; the output voltage signal is subjected to voltage waveform distortion analysis to obtain the total harmonic distortion rate; the filter is parameterized according to the total harmonic distortion rate and the frequency drift law, so that the third frequency distribution characteristic of the filter matches the second frequency distribution characteristic; and the optimized filter is used to suppress the harmonics of the frequency converter in the next time period. The present invention can effectively improve the harmonic suppression effect of the frequency converter and improve the power supply quality of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] In order to more clearly illustrate the technical solution of the present invention, the drawings used in the implementation mode will be briefly introduced below. Obviously, the drawings described below are only some implementation modes of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0062] Figure 1 is a flow chart of a harmonic control method for a frequency converter provided by an embodiment of the present invention;

[0063] Figure 2 is a schematic diagram of a construction process of a first frequency distribution characteristic curve provided by an embodiment of the present invention;

[0064] Figure 3 It is a schematic diagram of the construction process of the frequency drift law provided by an embodiment of the present invention;

[0065] Figure 4 The invention is a structural block diagram of a harmonic control system of a frequency converter provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0066] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0067] It is understood that the various numbers involved in the embodiments of the present invention are only for the convenience of description and are not intended to limit the scope of the present application. The size of the sequence number of each process does not mean the order of execution, and the execution order of each process should be determined by its function and internal logic.

[0068] In embodiments of the present invention, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. The terms "include", "comprises", or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article, or device. In the absence of further restrictions, the elements defined by the statement "includes..." do not exclude the presence of other identical elements in the process, method, article, or device including the elements.

[0069] See also Figure 1 , Figure 1 1 is a flow chart of a harmonic control method for a frequency converter provided by an embodiment of the present invention. The harmonic control method for a frequency converter specifically includes:

[0070] S11: Obtain the output voltage signal and bus voltage signal of the inverter in the current time period;

[0071] S12: performing spectrum analysis on the output voltage signal, and constructing a first frequency distribution characteristic curve between the fundamental frequency and each harmonic frequency in the frequency domain;

[0072] S13: According to the bus voltage signal, modify the first frequency distribution characteristic curve to obtain a second frequency distribution characteristic curve;

[0073] S14: When the frequency drift occurs in the output frequency signal of the frequency converter, the frequency drift of the output frequency signal is tracked to obtain a frequency drift rule;

[0074] S15: performing voltage waveform distortion analysis on the output voltage signal to obtain a total harmonic distortion rate;

[0075] S16: Optimizing filter parameters according to the total harmonic distortion rate and the frequency drift law, so that the third frequency distribution characteristic curve of the filter matches the second frequency distribution characteristic curve;

[0076] S17: In the next time period, the optimized filter is used to suppress harmonics of the frequency converter.

[0077] In the embodiment of the present invention, the output voltage signal and bus voltage signal of the inverter can be collected in real time through voltage signal collection devices such as voltage sensors and voltmeters. At the same time, the output frequency signal of the inverter can be collected in real time through frequency collection devices. It should be understood that the voltage signal collection device and the frequency collection device can be independently set and connected to the inverter respectively, or integrated in the same signal collection module, which is not specifically limited in the embodiment of the present invention. The embodiment of the present invention obtains the output voltage signal and bus voltage signal of the inverter in the current time period from the collected signals; then, a spectrum analysis is performed on the output voltage signal of the inverter, and a first frequency distribution characteristic curve between the fundamental frequency and each harmonic frequency is constructed in the frequency domain; the bus voltage signal of the inverter is used to correct the first frequency distribution characteristic curve to obtain a second frequency distribution characteristic curve; thereafter, when the output frequency signal of the inverter has a frequency drift in the current time period, the frequency drift is tracked on the output frequency signal of the inverter in the current time period to obtain a frequency drift law; a voltage waveform distortion analysis is performed on the output voltage signal to obtain a total harmonic distortion rate; and according to the total harmonic distortion rate and the frequency drift law, the filter connected to the output end of the inverter is parameter optimized so that the third frequency distribution characteristic of the filter matches the second frequency distribution characteristic, so as to achieve the purpose of the filter suppressing the harmonics of the inverter; in the next time period, the optimized filter is used to suppress the harmonics of the inverter, which can effectively improve the harmonic suppression effect of the inverter, thereby improving the power supply quality of the power grid.

[0078] It should be noted that the third frequency distribution characteristic curve of the filter is constructed according to the voltage signal output by the filter. The specific construction process can refer to the construction process of the first frequency distribution characteristic curve of the inverter, which will not be repeated here.

[0079] In an optional embodiment, step S12: performing spectrum analysis on the output voltage signal to construct a first frequency distribution characteristic curve between the fundamental frequency and each harmonic frequency in the frequency domain, including:

[0080] S121: Performing windowing processing on the output voltage signal using a Hamming window function to obtain a first voltage signal;

[0081] S122: Performing a fast Fourier transform on the first voltage signal to obtain frequency domain data;

[0082] S123: determining a fundamental frequency according to a frequency point with a maximum amplitude in the frequency domain data;

[0083] S124: performing normalization processing on the frequency domain data according to the fundamental frequency to obtain a normalized spectrum;

[0084] S125: locating the frequencies in the normalized spectrum whose amplitudes exceed the first set threshold, and determining the harmonic order and the frequency sequence corresponding to each harmonic according to the relationship between the located frequencies and the multiples of the fundamental frequency;

[0085] S126: Using a Butterworth filter to smooth the frequency sequence;

[0086] S127: Establishing a first frequency distribution characteristic curve according to the ratio of the amplitude of the harmonic frequency corresponding to each harmonic in the smoothed frequency sequence to the amplitude corresponding to the fundamental frequency.

[0087] In an embodiment of the present invention, 32 power frequency cycles can be used as a time period to suppress harmonics of the frequency converter. For example, if the power frequency is the frequency of the mains electricity of 50Hz, then one power frequency cycle is 1 / 50=0.02, and 32 power frequency cycles are 0.02×32=0.64, that is, one time period is 0.64s. For example, in the current time period, the output voltage signal of the frequency converter is preprocessed, such as sampling the output voltage signal according to the set sampling frequency to obtain a sampling signal, wherein the set sampling frequency is a multiple relationship with the switching frequency of the frequency converter, for example, the set sampling frequency is equal to 20 times the switching frequency of the frequency converter; then the sampling signal is low-pass filtered to filter out high-order harmonic interference while retaining the power frequency fundamental component, and obtain a low-pass filtered signal.

[0088] For the signal after low-pass filtering, a Hamming window function is used for windowing processing to obtain a first voltage signal, and then the first voltage signal is subjected to a fast Fourier transform to obtain frequency domain data, and the frequency point with the largest amplitude in the frequency domain data is selected as the fundamental frequency; then each frequency point in the frequency domain data is normalized according to the fundamental frequency to obtain a normalized spectrum. It should be understood that the term "frequency point" here can also be described as frequency, and the two can be used interchangeably, and are not specifically limited in the embodiments of the present invention.

[0089] The frequencies in the normalized spectrum whose amplitudes exceed the first set threshold are located, and the harmonic order and the frequency sequence corresponding to each harmonic are determined based on the multiple relationship between the located frequencies and the fundamental frequency. Considering that the harmonic frequency and the fundamental frequency are in an integer multiple relationship, the number of types of frequencies in the located frequencies that are in an integer multiple relationship with the fundamental frequency is counted to determine the harmonic order, such as the second harmonic, third harmonic, fifth harmonic, seventh harmonic, etc., and then the frequency components of each harmonic are extracted from the normalized spectrum to obtain the frequency sequence of each harmonic (which can also be described as a harmonic frequency spectrum). It should be noted that the embodiment of the present invention does not specifically limit the value of the first set threshold, for example, the first set threshold is 0.05.

[0090] Finally, the Butterworth filter is used to smooth the frequency sequence of each harmonic; for the frequency sequence of each harmonic after smoothing, the amplitude of the harmonic frequency in each frequency sequence is calculated, and the ratio of the amplitude of each harmonic frequency to the amplitude of the fundamental frequency is calculated. The calculated ratio is used as the value of the data point as the ordinate and the corresponding harmonic frequency as the abscissa, and a curve fitting is performed to obtain the first frequency distribution characteristic curve.

[0091] In an optional embodiment, step S13: correcting the first frequency distribution characteristic curve according to the bus voltage signal to obtain a second frequency distribution characteristic curve includes:

[0092] Determining the sideband harmonic frequency displacement according to the bus voltage signal;

[0093] The first frequency distribution characteristic curve is corrected according to the sideband harmonic frequency displacement to obtain a second frequency distribution characteristic curve.

[0094] Specifically, determining the sideband harmonic frequency displacement according to the bus voltage signal includes:

[0095] A low-pass filter with a passband range within a preset Hertz range is used to low-pass filter the bus voltage signal to obtain a fluctuation amplitude sequence;

[0096] Performing envelope demodulation on the fluctuation amplitude sequence by using Hilbert transform to obtain an envelope demodulation signal;

[0097] Calculating the fluctuation frequency and fluctuation depth of the bus voltage signal according to the envelope demodulation signal;

[0098] The sideband harmonic frequency displacement is calculated according to the fluctuation frequency and the fluctuation depth.

[0099] Specifically, the first frequency distribution characteristic curve is corrected according to the sideband harmonic frequency displacement to obtain the second frequency distribution characteristic curve, including:

[0100] According to the sideband harmonic frequency displacement, a sideband frequency displacement comparison table is constructed; wherein the sideband frequency displacement comparison table includes the displacement between the fundamental frequency and each harmonic frequency;

[0101] Using the sideband frequency shift comparison table, modifying the first frequency distribution characteristic curve;

[0102] The modified first frequency distribution characteristic curve is divided into interpolation intervals, and each of the interpolation intervals is interpolated using the Lagrange interpolation method to obtain the second frequency distribution characteristic curve.

[0103] In an embodiment of the present invention, the bus voltage signal of the inverter in the current time period (i.e., the input voltage signal of the inverter) is also sampled according to the set sampling frequency, and the sampled bus voltage signal is low-pass filtered using a low-pass filter with a passband range within a preset Hertz range to filter out high-order harmonic interference while retaining the power frequency fundamental component, and the bus voltage signal after low-pass filtering is Fourier decomposed to obtain a fundamental component, and then the amplitude corresponding to each sampling point in the fundamental component (i.e., the sampling point corresponding to the sampling according to the set sampling frequency) is extracted, and the difference between the amplitude corresponding to each sampling point and the steady-state amplitude is curve fitted to obtain a fluctuation amplitude sequence; the fluctuation amplitude sequence can also be described as an amplitude fluctuation curve of the fundamental component, which reflects the amplitude fluctuation characteristics of the fundamental component of the bus voltage signal (i.e., the difference between the amplitude corresponding to each sampling point and the steady-state amplitude), for example, the amplitude fluctuation curve records the degree to which the amplitude of the fundamental component corresponding to each sampling point deviates from the steady-state amplitude.

[0104] It should be noted that the embodiment of the present invention does not specifically limit the value of the preset hertz range. For example, the preset hertz range is 45 Hz to 55 Hz.

[0105] Then, the Hilbert transform is used to perform envelope demodulation on the above-mentioned fluctuation amplitude sequence to obtain an envelope demodulation signal; assuming that the function expression of the fluctuation amplitude sequence is x(t), t represents the time of signal sampling; the Hilbert transform is performed on the fluctuation amplitude sequence x(t) to obtain a demodulated signal z(t).

[0106] The modulus of the demodulated signal |z(t)| is the envelope demodulated signal E(t), z(t)=x(t)+jH[x(t)], j represents the imaginary unit, H represents the Hilbert transform operator, .

[0107] In the embodiment of the present invention, the ratio of the difference between the maximum value and the minimum value of the envelope demodulation signal to the average value of the envelope demodulation signal is the fluctuation depth.

[0108] Phase of the envelope demodulated signal is the instantaneous phase, . Instantaneous phase The derivative d / dt is the instantaneous frequency f(t), and the average value of the instantaneous frequency f(t) is the fluctuation frequency.

[0109] It should be understood that the fluctuation frequency is generally in the range of 0 to 10 Hz, corresponding to voltage fluctuations caused by disturbance sources such as mechanical load changes and motor start and stop. The frequency displacement of the sideband harmonics is affected by both the fluctuation frequency and the fluctuation depth. Assuming that under the typical working conditions of the fundamental frequency of 50 Hz, the fluctuation frequency of 5 Hz, and the fluctuation depth of 0.1, the first harmonic produces a frequency displacement of ±5 Hz on both sides of the fundamental, and the second harmonic displacement is ±10 Hz. For example, through the formula f m =±m×(h×f)×(f0 / f) to calculate the displacement of each harmonic, where m represents the harmonic order, h represents the fluctuation depth, f represents the fluctuation frequency, and f0 represents the fundamental frequency. f m Indicates the mth harmonic displacement.

[0110] Based on the fluctuation frequency and fluctuation depth, the frequency displacement of each harmonic is calculated as the sideband harmonic frequency displacement, and then a sideband frequency displacement comparison table is constructed. The sideband frequency displacement comparison table contains the frequency offset of each harmonic, that is, the displacement between each harmonic frequency and the fundamental frequency.

[0111] Then, the frequency offsets of each harmonic recorded in the sideband frequency shift comparison table are used to correct the data points representing the ratio of the amplitude of each harmonic frequency to the amplitude of the fundamental frequency in the first frequency distribution characteristic curve. The specific correction process is as follows:

[0112] Each harmonic frequency is corrected by using its frequency offset. For example, at the third harmonic 150Hz, the third harmonic displacement is ±15Hz, and the corrected third harmonic frequency is between 135Hz and 165Hz. The amplitude of the corrected third harmonic frequency can be obtained by calculating the mean amplitude at 135Hz-165Hz, and then curve fitting is performed based on the ratio of the amplitude of each corrected harmonic frequency to the amplitude of the fundamental frequency as a new data point to obtain the corrected first frequency distribution curve.

[0113] The modified first frequency distribution characteristic curve is divided into interpolation intervals. For example, two connected data points are used as two end points of the interpolation interval, and each interpolation interval is interpolated using the Lagrange interpolation method. The second frequency distribution characteristic curve is generated by curve fitting the data points in the modified first frequency distribution curve (i.e., the above-mentioned new data points) and the interpolated data points.

[0114] The embodiment of the present invention corrects the first frequency distribution curve by each harmonic frequency, and then interpolates the corrected curve to generate a second frequency distribution characteristic curve, so that the finally generated second frequency distribution characteristic curve can reflect the actual harmonic frequency distribution characteristics after the sideband modulation effect.

[0115] In an optional embodiment, step S14: when the output frequency signal of the frequency converter has frequency drift, tracking the frequency drift of the output frequency signal to obtain a frequency drift law includes:

[0116] S141: Obtaining the output frequency signal of the frequency converter in the current time period;

[0117] S142: Perform analog-to-digital conversion on the output frequency signal to obtain a frequency sampling sequence;

[0118] S143: performing wavelet decomposition on the frequency sampling sequence to obtain a phase feature;

[0119] S144: Tracking the rate of change of the phase feature to construct a frequency drift sequence;

[0120] S145: recursively process the frequency drift sequence using a Kalman filter algorithm to obtain a slope of the frequency drift;

[0121] S146: When the slope of the frequency drift exceeds a second set threshold, a mapping function between each harmonic frequency and the fundamental frequency is established as a frequency drift rule according to the slope of the frequency drift and the multiple relationship between each harmonic frequency and the fundamental frequency.

[0122] When acquiring and sampling the output voltage signal and bus voltage signal of the frequency converter, the embodiment of the present invention will also synchronously acquire and sample the output frequency signal of the frequency converter to perform frequency drift determination. The specific frequency drift determination process is as follows:

[0123] The output frequency signal collected in the current time period is converted from an analog signal to a digital signal to obtain a frequency sampling sequence f s (t), used to reflect the inverter t s The output frequency at the moment, where f s The subscript s is used to identify the output frequency signal;

[0124] Perform wavelet decomposition on the frequency sampling sequence to obtain wavelet coefficients, use the phase angle of the wavelet coefficients as the phase feature, track the rate of change of the phase feature, and construct a frequency drift sequence;

[0125] For example, the phase characteristics ,in, Represents wavelet coefficients The real part of Represents wavelet coefficients The imaginary part of a represents the expansion factor and b represents the scale factor. Frequency drift sequence .

[0126] For the above frequency drift sequence The Kalman filter is used for recursive processing. The specific process is as follows:

[0127] The state equation of the Kalman filter is ,in, represents the state vector at time K, represents the state vector at time k−1, F represents the state transfer matrix, F= , Indicates the update time interval of the state vector; represents the process noise at time k; is the noise that obeys the Gaussian distribution and is the noise of the prediction process, which corresponds to The noise of each component in is Gaussian white noise with an expected value of 0 and a covariance of Q. ~N(0,Q), Q is the process excitation noise covariance. It should be understood that the process excitation noise covariance Q (i.e. the covariance of the system process) is used to represent the error between the state transition matrix and the actual process. Since the process signal cannot be directly observed, the value of Q is difficult to determine. It is the state variable used by the Kalman filter to estimate the discrete time process, also known as the noise brought by the prediction model itself.

[0128] The observation equation of the Kalman filter is , where H represents the observation matrix, H= ; represents the observation noise at time k, Obeying Gaussian distribution, ~N(0, R), where R is the measurement noise covariance. It should be understood that when the Kalman filter is actually implemented, the measurement noise covariance R can generally be observed and is a known condition of the Kalman filter.

[0129] By shifting the frequency The values ​​of at each moment are used as the state vector, substituted into the state equation and observation equation of the Kalman filter algorithm for recursive processing, and the state estimate and covariance estimate are continuously updated to obtain the final state vector estimate, such as .

[0130] Extract the second element in the state transition matrix , by calculating the state vector estimate and The ratio of , and the slope c of the frequency drift is obtained.

[0131] When the slope of the frequency drift exceeds the second set threshold, it is determined that the frequency drift occurs, and then a mapping function between each harmonic frequency and the fundamental frequency is established according to the slope c of the frequency drift and the multiple relationship between each harmonic frequency and the fundamental frequency. =c×m× ×m× , as the frequency drift law. Among them, represents the mth harmonic frequency, Indicates the fundamental frequency.

[0132] It should be noted that the embodiment of the present invention does not specifically limit the value of the second set threshold. For example, the second set threshold may be 0.2.

[0133] In an optional embodiment, step S15: performing voltage waveform distortion analysis on the output voltage signal to obtain a total harmonic distortion rate includes:

[0134] Performing Fourier transform on the output voltage signal to obtain a fundamental wave component and each harmonic component;

[0135] Calculating the distortion rate of each harmonic according to the fundamental wave component and each harmonic component;

[0136] The distortion rate of each harmonic is subjected to a square root operation to obtain a total harmonic distortion rate.

[0137] Specifically, the distortion rate of each harmonic can be obtained by calculating the amplitude ratio of each harmonic component to the fundamental component, and then the distortion rate of each harmonic is squared and summed, and the sum is squared to obtain the total harmonic distortion rate.

[0138] In an optional embodiment, the step of optimizing the parameters of the filter according to the total harmonic distortion rate and the frequency drift law so that the third frequency distribution characteristic curve of the filter matches the second frequency distribution characteristic curve includes:

[0139] According to the total harmonic distortion rate and the frequency drift law, a radial basis function neural network is used to iteratively optimize the filter parameters to obtain the filter parameters output in this round; the filter parameters include the center frequency and passband bandwidth of the filter;

[0140] According to the filter parameters outputted in this round, a Butterworth filter is constructed, and a frequency response test is performed on the Butterworth filter to obtain the test error of this round;

[0141] If the test error of this round is greater than the preset error threshold, the radial basis neural network is used again to iteratively optimize the parameters of the filter;

[0142] If the test error of this round is not greater than the preset error threshold, the center frequency and passband bandwidth of the filter are updated using the center frequency and passband bandwidth in the filter parameters output in this round at the end of the current time period, so that the third frequency distribution characteristic curve of the filter matches the second frequency distribution characteristic curve.

[0143] In the embodiment of the present invention, after calculating the total harmonic distortion rate, a harmonic frequency data table is constructed based on the above-mentioned frequency drift law, and the table records the harmonic frequencies of each sub-harmonic in the next time period, for example, by mapping function (t+1) Estimate the harmonic frequencies at each moment of the next time period. Then use the total harmonic distortion rate and harmonic frequency data table, and estimate the harmonic frequencies at each moment of the next time period as input items, and input them into the pre-trained radial basis neural network for filter parameter prediction to obtain the center frequency and passband bandwidth of the filter.

[0144] It should be noted that the training process of the radial basis neural network belongs to the existing technology and will not be described in detail here. For example, the center frequency and passband bandwidth of the filter in the historical working process are obtained as the label data of the radial basis neural network, and the total harmonic distortion rate and each harmonic frequency of the corresponding inverter in the historical working process are used as training samples of the radial basis neural network. The radial basis neural network is trained until the accuracy of the radial basis neural network meets the preset accuracy requirements, and the network training is completed.

[0145] According to the center frequency and passband bandwidth currently output by the network, a Butterworth filter is constructed, and the frequency response test of the Butterworth filter is performed, and the test error is obtained; wherein, the frequency response characteristics of the Butterworth filter are obtained by frequency scanning, the scanning frequency is from 0 to 800Hz, the step interval is 1Hz, and the mean square error between the actual response and the target response is the test error, which reflects the filter parameter optimization effect. If the test error is greater than the preset error threshold, it means that the filter parameter optimization effect is poor, and the total harmonic distortion rate and harmonic frequency data table are re-entered into the radial basis neural network to perform filter parameter iterative optimization; if the test error is not greater than the preset error threshold, it means that the filter parameter optimization effect is good, and at the end of the current time period, the center frequency and passband bandwidth in the filter parameters output in this round are used to update the center frequency and passband bandwidth of the filter.

[0146] It should be noted that the embodiment of the present invention does not specifically limit the value of the preset error threshold. For example, the preset error threshold may be 0.05.

[0147] In addition, it is also necessary to verify whether the third frequency distribution characteristic of the optimized filter matches the second frequency distribution characteristic.

[0148] When the test error is not greater than the preset error threshold and the third frequency distribution characteristic curve of the optimized filter matches the second frequency distribution characteristic curve (that is, the two curves are exactly the same or the similarity of the two curves is greater than the set similarity threshold), it means that the filter parameters meet the requirements, otherwise the total harmonic distortion rate and harmonic frequency data table are re-input into the radial basis neural network for iterative parameter optimization of the filter.

[0149] The change in sideband harmonic characteristics caused by frequency drift shows different degrees of frequency shift and distortion at each harmonic frequency. Based on the frequency drift law and total harmonic distortion rate, the center frequency and passband bandwidth of the filter connected to the output end of the inverter are adjusted, so that the center frequency of the filter is gradually adjusted according to the frequency drift law, and dynamic tracking of the sideband harmonic frequency is achieved. The automatic adjustment of the passband bandwidth of the filter ensures that the sideband harmonics are always within the passband bandwidth, so that the third frequency distribution characteristics of the filter match the second frequency distribution characteristics, thereby achieving the effect of inverter harmonic suppression.

[0150] See also Figure 4 , Figure 4 The embodiment of the present invention provides a structural block diagram of a harmonic control system of a frequency converter, wherein the harmonic control system of the frequency converter comprises: a controller 11, a filter 12 connected to the controller 11, and a signal sampling module 13; the filter 12 and the signal sampling module 13 are used to be connected to the frequency converter 100;

[0151] The signal sampling module 13 is used to obtain the output voltage signal and bus voltage signal of the inverter 100 in the current time period.

[0152] The controller 11 includes:

[0153] A spectrum analysis module, used for performing spectrum analysis on the output voltage signal, and constructing a first frequency distribution characteristic curve between the fundamental frequency and each harmonic frequency in the frequency domain;

[0154] A correction module, used for correcting the first frequency distribution characteristic curve according to the bus voltage signal to obtain a second frequency distribution characteristic curve;

[0155] A frequency drift tracking module, used to track the frequency drift of the output frequency signal of the frequency converter when the output frequency signal of the frequency converter drifts, so as to obtain a frequency drift law;

[0156] A waveform distortion analysis module, used to perform voltage waveform distortion analysis on the output voltage signal to obtain a total harmonic distortion rate;

[0157] A parameter optimization module, used for optimizing the parameters of the filter according to the total harmonic distortion rate and the frequency drift law, so that the third frequency distribution characteristic curve of the filter matches the second frequency distribution characteristic curve;

[0158] The harmonic suppression module is used to use the optimized filter to suppress harmonics of the frequency converter in the next time period.

[0159] In an optional embodiment, the spectrum analysis module includes:

[0160] A windowing processing unit, configured to perform windowing processing on the output voltage signal using a Hamming window function to obtain a first voltage signal;

[0161] A fast Fourier transform unit, used for performing a fast Fourier transform on the first voltage signal to obtain frequency domain data;

[0162] A fundamental frequency determination unit, used to determine the fundamental frequency according to the frequency point with the largest amplitude in the frequency domain data;

[0163] A normalization processing unit, used for performing normalization processing on the frequency domain data according to the fundamental frequency to obtain a normalized spectrum;

[0164] A frequency sequence determination unit, used to locate the frequencies in the normalized spectrum whose amplitudes exceed a first set threshold, and determine the harmonic order and the frequency sequence corresponding to each harmonic according to the multiple relationship between the located frequencies and the fundamental frequency;

[0165] A smoothing processing unit, used for smoothing the frequency sequence using a Butterworth filter;

[0166] The first frequency distribution characteristic curve establishing unit is used to establish the first frequency distribution characteristic curve according to the ratio of the amplitude of the harmonic frequency corresponding to each harmonic in the smoothed frequency sequence to the amplitude corresponding to the fundamental frequency.

[0167] In an optional embodiment, the correction module includes:

[0168] A frequency displacement determination unit, used to determine the sideband harmonic frequency displacement according to the bus voltage signal;

[0169] The curve correction unit is used to correct the first frequency distribution characteristic curve according to the sideband harmonic frequency displacement to obtain a second frequency distribution characteristic curve.

[0170] In an optional embodiment, the frequency shift determination unit includes:

[0171] A low-pass filtering subunit, configured to perform low-pass filtering on the bus voltage signal using a low-pass filter having a passband range within a preset Hertz range, to obtain a fluctuation amplitude sequence;

[0172] An envelope demodulation subunit, used to perform envelope demodulation on the fluctuation amplitude sequence using Hilbert transform to obtain an envelope demodulation signal;

[0173] A fluctuation data calculation subunit, used for calculating the fluctuation frequency and fluctuation depth of the bus voltage signal according to the envelope demodulation signal;

[0174] The sideband harmonic frequency displacement calculation subunit is used to calculate the sideband harmonic frequency displacement according to the fluctuation frequency and the fluctuation depth.

[0175] In an optional embodiment, the curve correction unit includes:

[0176] A sideband frequency displacement comparison table construction subunit is used to construct a sideband frequency displacement comparison table according to the sideband harmonic frequency displacement; wherein the sideband frequency displacement comparison table includes the displacement between the fundamental frequency and each harmonic frequency;

[0177] A frequency distribution curve correction subunit, configured to correct the first frequency distribution characteristic curve using the sideband frequency shift comparison table;

[0178] The curve interpolation subunit is used to divide the modified first frequency distribution characteristic curve into interpolation intervals, and interpolate each of the interpolation intervals using the Lagrange interpolation method to obtain the second frequency distribution characteristic curve.

[0179] In an optional embodiment, the signal sampling module is further used to obtain the output frequency signal of the frequency converter in the current time period;

[0180] The frequency drift tracking module comprises:

[0181] An analog-to-digital conversion unit, used to perform analog-to-digital conversion on the output frequency signal to obtain a frequency sampling sequence;

[0182] A wavelet decomposition unit, used for performing wavelet decomposition on the frequency sampling sequence to obtain a phase feature;

[0183] A frequency drift sequence construction unit, used to track the rate of change of the phase characteristic and construct a frequency drift sequence;

[0184] A frequency drift slope calculation unit, used for recursively processing the frequency drift sequence using a Kalman filter algorithm to obtain a slope of the frequency drift;

[0185] The frequency drift law determination unit is used to establish a mapping function between each harmonic frequency and the fundamental frequency as a frequency drift law according to the slope of the frequency drift and the multiple relationship between each harmonic frequency and the fundamental frequency when the slope of the frequency drift exceeds a second set threshold.

[0186] In an optional embodiment, the distortion analysis module includes:

[0187] A Fourier transform unit, used for performing Fourier transform on the output voltage signal to obtain a fundamental wave component and each harmonic component;

[0188] A distortion rate calculation unit, used to calculate the distortion rate of each harmonic according to the fundamental wave component and each harmonic component;

[0189] Specifically, the distortion rate of each harmonic can be obtained by calculating the amplitude ratio of each harmonic component to the fundamental wave component.

[0190] The total harmonic distortion rate calculation unit is used to perform a square root operation on the distortion rate of each harmonic to obtain the total harmonic distortion rate.

[0191] In an optional embodiment, the parameter optimization module includes:

[0192] A first iterative optimization unit is used to perform iterative optimization of the filter parameters by using a radial basis neural network according to the total harmonic distortion rate and the frequency drift law, so as to obtain the filter parameters output in this round; the filter parameters include the center frequency and passband bandwidth of the filter;

[0193] A testing unit, used to construct a Butterworth filter according to the filter parameters output in this round, and perform a frequency response test on the Butterworth filter to obtain a test error in this round;

[0194] A second iterative optimization unit is used to re-adopt a radial basis neural network to iteratively optimize the parameters of the filter if the test error of this round is greater than a preset error threshold;

[0195] A parameter updating unit is used to update the center frequency and passband bandwidth of the filter using the center frequency and passband bandwidth in the filter parameters output in this round at the end of the current time period if the test error of this round is not greater than a preset error threshold, so that the third frequency distribution characteristic curve of the filter matches the second frequency distribution characteristic curve.

[0196] It should be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the accompanying drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art may understand and implement it without paying any creative effort.

[0197] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, many improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A harmonic control method for a frequency converter, characterized in that: include: Obtain the output voltage signal and bus voltage signal of the inverter in the current time period; Performing spectrum analysis on the output voltage signal to construct a first frequency distribution characteristic curve between the fundamental frequency and each harmonic frequency in the frequency domain; According to the bus voltage signal, the first frequency distribution characteristic curve is corrected to obtain a second frequency distribution characteristic curve; When the frequency drift occurs in the output frequency signal of the frequency converter, the frequency drift of the output frequency signal is tracked to obtain a frequency drift law; Performing voltage waveform distortion analysis on the output voltage signal to obtain a total harmonic distortion rate; Optimizing the parameters of the filter according to the total harmonic distortion rate and the frequency drift law, so that the third frequency distribution characteristic curve of the filter matches the second frequency distribution characteristic curve; In the next time period, the optimized filter is used to suppress harmonics of the frequency converter.

2. The harmonic control method of the frequency converter according to claim 1, characterized in that: The performing spectrum analysis on the output voltage signal to construct a first frequency distribution characteristic curve between the fundamental frequency and each harmonic frequency in the frequency domain includes: Performing windowing processing on the output voltage signal using a Hamming window function to obtain a first voltage signal; Performing a fast Fourier transform on the first voltage signal to obtain frequency domain data; Determine the fundamental frequency according to the frequency point with the largest amplitude in the frequency domain data; Normalizing the frequency domain data according to the fundamental frequency to obtain a normalized spectrum; Locating the frequencies in the normalized spectrum whose amplitudes exceed a first set threshold, and determining the harmonic order and the frequency sequence corresponding to each harmonic according to the relationship between the located frequencies and the multiples of the fundamental frequency; Using a Butterworth filter to smooth the frequency sequence; A first frequency distribution characteristic curve is established according to the ratio of the amplitude of the harmonic frequency corresponding to each harmonic in the smoothed frequency sequence to the amplitude corresponding to the fundamental frequency.

3. The harmonic control method of the frequency converter according to claim 1, characterized in that: The step of correcting the first frequency distribution characteristic curve according to the bus voltage signal to obtain a second frequency distribution characteristic curve comprises: Determining the sideband harmonic frequency displacement according to the bus voltage signal; The first frequency distribution characteristic curve is corrected according to the sideband harmonic frequency displacement to obtain a second frequency distribution characteristic curve.

4. The harmonic control method of the frequency converter according to claim 3, characterized in that: Determining the sideband harmonic frequency displacement according to the bus voltage signal includes: A low-pass filter with a passband range within a preset Hertz range is used to low-pass filter the bus voltage signal to obtain a fluctuation amplitude sequence; Performing envelope demodulation on the fluctuation amplitude sequence by using Hilbert transform to obtain an envelope demodulation signal; Calculating the fluctuation frequency and fluctuation depth of the bus voltage signal according to the envelope demodulation signal; The sideband harmonic frequency displacement is calculated according to the fluctuation frequency and the fluctuation depth.

5. The harmonic control method of the frequency converter according to claim 4, characterized in that: The first frequency distribution characteristic curve is corrected according to the sideband harmonic frequency displacement to obtain a second frequency distribution characteristic curve, including: According to the sideband harmonic frequency displacement, a sideband frequency displacement comparison table is constructed; wherein the sideband frequency displacement comparison table includes the displacement between the fundamental frequency and each harmonic frequency; Using the sideband frequency shift comparison table, modifying the first frequency distribution characteristic curve; The modified first frequency distribution characteristic curve is divided into interpolation intervals, and each of the interpolation intervals is interpolated using the Lagrange interpolation method to obtain the second frequency distribution characteristic curve.

6. The harmonic control method of the frequency converter according to claim 1, characterized in that: When the frequency drift occurs in the output frequency signal of the frequency converter, tracking the frequency drift of the output frequency signal to obtain a frequency drift rule includes: Obtaining the output frequency signal of the frequency converter in the current time period; Performing analog-to-digital conversion on the output frequency signal to obtain a frequency sampling sequence; Performing wavelet decomposition on the frequency sampling sequence to obtain a phase feature; Tracking the rate of change of the phase characteristic to construct a frequency drift sequence; Recursively processing the frequency drift sequence using a Kalman filter algorithm to obtain a slope of the frequency drift; When the slope of the frequency drift exceeds the second set threshold, a mapping function between each harmonic frequency and the fundamental frequency is established as a frequency drift rule according to the slope of the frequency drift and the multiple relationship between each harmonic frequency and the fundamental frequency.

7. The harmonic control method of the frequency converter according to claim 1, characterized in that: The performing voltage waveform distortion analysis on the output voltage signal to obtain a total harmonic distortion rate includes: Performing Fourier transform on the output voltage signal to obtain a fundamental wave component and each harmonic component; Calculating the distortion rate of each harmonic according to the fundamental wave component and each harmonic component; The distortion rate of each harmonic is subjected to a square root operation to obtain a total harmonic distortion rate.

8. The harmonic control method of the frequency converter according to claim 1, characterized in that: The step of optimizing the parameters of the filter according to the total harmonic distortion rate and the frequency drift law so that the third frequency distribution characteristic curve of the filter matches the second frequency distribution characteristic curve includes: According to the total harmonic distortion rate and the frequency drift law, a radial basis function neural network is used to iteratively optimize the filter parameters to obtain the filter parameters output in this round; the filter parameters include the center frequency and passband bandwidth of the filter; According to the filter parameters outputted in this round, a Butterworth filter is constructed, and a frequency response test is performed on the Butterworth filter to obtain the test error of this round; If the test error of this round is greater than the preset error threshold, the radial basis neural network is used again to iteratively optimize the parameters of the filter; If the test error of this round is not greater than the preset error threshold, the center frequency and passband bandwidth of the filter are updated using the center frequency and passband bandwidth in the filter parameters output in this round at the end of the current time period, so that the third frequency distribution characteristic curve of the filter matches the second frequency distribution characteristic curve.

9. The harmonic control method of the frequency converter according to claim 7, characterized in that: The step of calculating the distortion rate of each harmonic according to the fundamental wave component and each harmonic component comprises: The amplitude ratio of each harmonic component to the fundamental component is calculated to obtain the distortion rate of each harmonic.

10. A harmonic control system for a frequency converter, characterized in that: include: A controller, a filter connected to the controller, and a signal sampling module; wherein the filter and the signal sampling module are used to be connected to a frequency converter; The signal sampling module is used to obtain the output voltage signal and bus voltage signal of the inverter in the current time period; The controller comprises: A spectrum analysis module, used for performing spectrum analysis on the output voltage signal, and constructing a first frequency distribution characteristic curve between the fundamental frequency and each harmonic frequency in the frequency domain; A correction module, used for correcting the first frequency distribution characteristic curve according to the bus voltage signal to obtain a second frequency distribution characteristic curve; A frequency drift tracking module, used to track the frequency drift of the output frequency signal of the frequency converter when the output frequency signal of the frequency converter drifts, so as to obtain a frequency drift law; A waveform distortion analysis module, used to perform voltage waveform distortion analysis on the output voltage signal to obtain a total harmonic distortion rate; A parameter optimization module, used for optimizing the parameters of the filter according to the total harmonic distortion rate and the frequency drift law, so that the third frequency distribution characteristic curve of the filter matches the second frequency distribution characteristic curve; The harmonic suppression module is used to use the optimized filter to suppress harmonics of the frequency converter in the next time period.

Images