CN105811921B - A kind of method and wave filter for suppressing industrial frequency harmonic interference - Google Patents

Abstract

The invention provides a kind of method and wave filter for suppressing industrial frequency harmonic interference, it is related to field of signal processing, solves the problems, such as that the power frequency inhibition of existing wave filter is limited, the method includes：The time-domain information of time-domain information and default multiple harmonic reference signal according to acquired original signal, obtains the convolution nuclear information that each harmonic reference signal distinguishes corresponding industrial frequency harmonic interference signal；Frequency domain information and each harmonic reference signal according to each harmonic reference signal distinguish the convolution nuclear information of corresponding industrial frequency harmonic interference signal, obtain total frequency domain information of all industrial frequency harmonic interference signals；Total frequency domain information of frequency domain information and all industrial frequency harmonic interference signals according to acquired original signal, obtain the first estimation frequency domain information that acquired original signal removes the first signal after all industrial frequency harmonic interference signals, and estimate frequency domain information according to first, obtain the first estimation time-domain information of the first signal.The solution of the present invention effectively inhibits the interference of industrial frequency harmonic signal.

Description

Method and filter for inhibiting power frequency harmonic interference

Technical Field

The invention relates to the field of signal processing, in particular to a method and a filter for inhibiting power frequency harmonic interference.

Background

In the field of signal processing, such as electrocardiogram signals, image acquisition signals, electromagnetic detection signals and the like, power frequency interference generally exists. The existence of power frequency interference usually limits the signal bandwidth and detection effect of engineering application, and is a problem which is difficult to overcome. Especially for power frequency harmonic of 50Hz, a good narrow-band filter is difficult to design to overcome. Although some researchers have proposed optimal notch filters for spectral line enhancement, or improved such as IIR lattice adaptive notch filters, which are essentially narrow-band filters, these filters have limited power frequency rejection.

Disclosure of Invention

The invention aims to provide a method and a filter for inhibiting power frequency harmonic interference, and solves the problem of limited power frequency inhibition effect of the existing filter.

In order to solve the above technical problem, an embodiment of the present invention provides a method for suppressing power frequency harmonic interference, including:

acquiring convolution kernel information of power frequency harmonic interference signals corresponding to each harmonic reference signal according to time domain information of an original acquired signal and time domain information of a plurality of preset harmonic reference signals;

acquiring total frequency domain information of all power frequency harmonic interference signals according to the frequency domain information of each harmonic reference signal and convolution kernel information of the power frequency harmonic interference signals corresponding to each harmonic reference signal respectively;

according to the frequency domain information of the original collected signal and the total frequency domain information of all the power frequency harmonic interference signals, obtaining first estimated frequency domain information of a first signal of the original collected signal without all the power frequency harmonic interference signals, and according to the first estimated frequency domain information, obtaining first estimated time domain information of the first signal.

To solve the above technical problem, an embodiment of the present invention further provides a filter, including:

the first acquisition module is used for acquiring convolution kernel information of power frequency harmonic interference signals corresponding to each harmonic reference signal according to time domain information of an original acquisition signal and time domain information of a plurality of preset harmonic reference signals;

the second acquisition module is used for acquiring total frequency domain information of all power frequency harmonic interference signals according to the frequency domain information of each harmonic reference signal and convolution kernel information of the power frequency harmonic interference signals corresponding to each harmonic reference signal;

and the third acquisition module is used for acquiring first estimated frequency domain information of the first signal of the original acquisition signal without all the power frequency harmonic interference signals according to the frequency domain information of the original acquisition signal and the total frequency domain information of all the power frequency harmonic interference signals, and acquiring first estimated time domain information of the first signal according to the first estimated frequency domain information.

The technical scheme of the invention has the following beneficial effects:

according to the method for inhibiting the power frequency harmonic interference, firstly, convolution kernel information of power frequency harmonic interference signals corresponding to each harmonic reference signal is obtained according to time domain information of an original collected signal and time domain information of a plurality of preset harmonic reference signals; then, acquiring total frequency domain information of all power frequency harmonic interference signals according to convolution kernel information of the power frequency harmonic interference signals respectively corresponding to each harmonic reference signal; and finally, according to the frequency domain information of the original collected signal and the total frequency domain information of all power frequency harmonic interference signals, acquiring first estimated frequency domain information of the original collected signal without all power frequency harmonic interference signals, and according to the first estimated frequency domain information, acquiring first estimated time domain information of the first signal, so that the interference of the power frequency harmonic signals in the original collected signal is effectively eliminated. The method has great value in power frequency interference suppression, reduces the complexity of setting the wave trap, is simple and easy to implement, and can effectively suppress power frequency harmonic signals. Compared with the traditional wave trap, the method has the advantages of frequency adaptability, time delay estimation accuracy, good signal-to-noise ratio improvement degree, small signal distortion degree, suitability for common engineering application and good stability.

Drawings

FIG. 1 is a flow chart of a method for suppressing power frequency harmonic interference according to the present invention;

FIG. 2 is a flowchart of an embodiment of the method for suppressing power frequency harmonic interference according to the present invention;

FIG. 3 is a schematic diagram of signal time domain information that is not interfered by power frequency harmonics;

FIG. 4 is a schematic diagram of a signal amplitude spectrum without interference from power frequency harmonics;

FIG. 5 is a schematic diagram of a signal phase spectrum without interference from power frequency harmonics;

FIG. 6 is a schematic diagram of signal time domain information interfered by power frequency harmonics;

FIG. 7 is a schematic diagram of a signal amplitude spectrum disturbed by power frequency harmonics;

FIG. 8 is a schematic diagram of a signal phase spectrum disturbed by power frequency harmonics;

FIG. 9 is a schematic diagram of signal time domain information after power frequency harmonic interference is suppressed by applying the method of the present invention;

FIG. 10 is a schematic diagram of singular point elimination in the method for suppressing power frequency harmonic interference according to the present invention;

fig. 11 is a schematic structural diagram of the filter of the present invention.

Detailed Description

In order to make the technical problems, technical solutions and advantages of the present invention more apparent, the following detailed description is given with reference to the accompanying drawings and specific embodiments.

The method for inhibiting power frequency harmonic interference of the embodiment of the invention utilizes the basic principle of power spectrum estimation to obtain convolution kernel information of the power frequency harmonic interference signal, and eliminates the power frequency harmonic interference signal in a frequency domain, thereby effectively inhibiting the interference of the power frequency harmonic without high-complexity computing resources; the method has the advantages of strong applicability to a small-sized computing platform, good signal-to-noise ratio improvement degree, small signal distortion degree, good real-time performance and good stability.

The first embodiment:

as shown in fig. 1, the method for suppressing power frequency harmonic interference according to the embodiment of the present invention includes:

step 101, obtaining convolution kernel information of power frequency harmonic interference signals corresponding to each harmonic reference signal respectively according to time domain information of an original collected signal and time domain information of a plurality of preset harmonic reference signals.

Here, the harmonic reference signals correspond to power frequency harmonic interference signals in the engineering signals, the number of the harmonic reference signals is the number of the power frequency harmonic interference signals in the frequency band of the engineering signals, and the frequency of the harmonic reference signals is equal to the frequency of the corresponding power frequency harmonic interference signals.

And 102, acquiring total frequency domain information of all power frequency harmonic interference signals according to the frequency domain information of each harmonic reference signal and convolution kernel information of the power frequency harmonic interference signals respectively corresponding to each harmonic reference signal.

The frequency domain information of the power frequency harmonic interference signal can be accurately obtained through the convolution kernel information of the power frequency harmonic interference signal, and then the interference of the power frequency harmonic interference signal on the originally acquired signal is removed in the frequency domain.

Step 103, obtaining first estimated frequency domain information of the first signal of the original collected signal without all power frequency harmonic interference signals according to the frequency domain information of the original collected signal and the total frequency domain information of all power frequency harmonic interference signals, and obtaining first estimated time domain information of the first signal according to the first estimated frequency domain information.

The total frequency domain information of all power frequency harmonic interference signals is subtracted from the frequency domain information of the original collected signal, so that the interference of the power frequency interference signals on the original collected signal can be removed in the frequency domain, and the first estimated time domain information of the first signal after all power frequency harmonic interference signals are removed from the original collected signal can be further obtained, so that the signal after the power frequency harmonic interference is effectively filtered is obtained.

In the embodiment, convolution kernel information of a power frequency harmonic interference signal corresponding to each harmonic reference signal is obtained according to time domain information of an original collected signal and time domain information of a plurality of preset harmonic reference signals; then, acquiring total frequency domain information of all power frequency harmonic interference signals according to convolution kernel information of the power frequency harmonic interference signals respectively corresponding to each harmonic reference signal; and finally, according to the frequency domain information of the original collected signal and the total frequency domain information of all power frequency harmonic interference signals, acquiring first estimated frequency domain information of the original collected signal without all power frequency harmonic interference signals, and according to the first estimated frequency domain information, acquiring first estimated time domain information of the first signal, so that the interference of the power frequency harmonic signals in the original collected signal is effectively eliminated. The method has great value in power frequency interference suppression, reduces the complexity of setting the wave trap, is simple and easy to implement, and can effectively suppress power frequency harmonic signals. Compared with the traditional wave trap, the method has the advantages of frequency adaptability, time delay estimation accuracy, good signal-to-noise ratio improvement degree, small signal distortion degree, suitability for common engineering application and good stability.

Second embodiment:

the method for inhibiting power frequency harmonic interference of the embodiment of the invention comprises the following steps:

step 1011, obtaining cross-power spectrum information of each harmonic reference signal and the originally acquired signal respectively according to the time domain information of the originally acquired signal and the time domain information of each harmonic reference signal.

Step 1012, respectively obtaining the self-power spectrum information of each harmonic reference signal according to the time domain information of each harmonic reference signal.

And 1013, obtaining convolution kernel information of the power frequency harmonic interference signals corresponding to each harmonic reference signal according to the self-power spectrum information and the corresponding cross-power spectrum information of each harmonic reference signal.

Here, the self-power spectrum information of each harmonic reference signal and the corresponding cross-power spectrum information are not related to each other, and the harmonic reference signals are also not related to each other. The convolution kernel information of the power frequency harmonic interference signals respectively corresponding to each harmonic reference signal can be accurately obtained according to the self-power spectrum information and the corresponding cross-power spectrum information of each harmonic reference signal.

And 102, acquiring total frequency domain information of all power frequency harmonic interference signals according to the frequency domain information of each harmonic reference signal and convolution kernel information of the power frequency harmonic interference signals respectively corresponding to each harmonic reference signal.

Step 103, obtaining first estimated frequency domain information of the first signal of the original collected signal without all power frequency harmonic interference signals according to the frequency domain information of the original collected signal and the total frequency domain information of all power frequency harmonic interference signals, and obtaining first estimated time domain information of the first signal according to the first estimated frequency domain information.

In this embodiment, the step 101 of the first embodiment is embodied as the step 1011-.

The original signals collected usually contain power frequency harmonic interference, and the original collected signals can be generally written as:

wherein x (t) is the time domain function of the original collected signalS (t) is the time domain function of the first signal of the original collected signal after all power frequency harmonic interference signals are removed, v (t) is the total time domain function of all power frequency harmonic interference signals, h_i(t) is the convolution kernel of the power frequency harmonic interference signal corresponding to the ith harmonic reference signal, v_i(t) is a time domain function of the ith harmonic reference signal, i is an integer between 1 and p, p is the number of power frequency harmonic interference signals in the engineering signal frequency band, and t is a time variable.

A general harmonic reference signal can be written as:

v_i(t)＝Asin(2πf_it)；

wherein. A is a freely set constant, f_iThe frequency of the ith harmonic reference signal is equal to the frequency of the corresponding power frequency harmonic interference signal.

Based on the above theory, preferably, in step 1011, the cross-power spectral density function of each harmonic reference signal and the originally acquired signal can be obtained specifically through the following formula

In step 1012, the self-power spectral density function of each harmonic reference signal can be obtained by the following formula

Wherein v is_i(t + τ) is a time domain function of the i-th harmonic reference signal delay time τ, i being an integer between 1 and pAnd p is the number of harmonic reference signals (namely the number of power frequency harmonic interference signals in the engineering signal frequency band), t is a time variable, omega is a frequency variable, and j and tau are self-defined constants.

Since the harmonic reference signals are not correlated with each other, and s (t) and v (t) are also not correlated with each other, it can be obtained that:

therefore, in step 1013, the convolution kernel H of the power frequency harmonic interference signal corresponding to each harmonic reference signal can be obtained by the following formula_i(ω)：

Wherein,is a cross-power spectral density function of the ith harmonic reference signal and the original acquired signal,is the self-power spectral density function of the ith harmonic reference signal, H_i(omega) is h_i(t) Fourier transform function.

Further, the step 102 may specifically include:

step 1021, time domain function v of each harmonic reference signal_i(t) performing Fourier transform to obtain a frequency domain function V of each harmonic reference signal_i(ω)；

Step 1022, obtaining the total frequency domain function V (ω) of all power frequency harmonic interference signals by the following formula:

further, the step 103 may specifically include:

step 1031, performing Fourier transform on the time domain function of the original acquisition signal to obtain a frequency domain function X (omega) of the original acquisition signal;

step 1032, obtain a first estimated frequency domain function of the first signal by the following formula

Step 1033, for the first estimated frequency domain functionPerforming inverse Fourier transform to obtain a first estimated time domain function of the first signal

At this point, the cross-power spectral density function of each harmonic reference signal and the original collected signal is obtainedSelf-power spectral density function of each harmonic reference signalFurther obtaining convolution kernel function H of power frequency harmonic interference signal corresponding to each harmonic reference signal_i(ω)；And by a frequency domain function V of each harmonic reference signal_i(omega) and convolution kernel function H of each power frequency harmonic interference signal_i(omega), obtaining a total frequency domain function V (omega) of all power frequency harmonic interference signals; then, the frequency domain function X (omega) of the originally acquired signal is used for subtracting the total frequency domain function V (omega) of all power frequency harmonic interference signals to obtain a first estimated frequency domain function of the first signalFinally, inverse Fourier transform is carried out on the first estimated frequency domain function, and the first estimated time domain function of the first signal is accurately obtained

The third embodiment:

the method for inhibiting power frequency harmonic interference of the embodiment of the invention comprises the following steps:

step 101, obtaining convolution kernel information of power frequency harmonic interference signals corresponding to each harmonic reference signal respectively according to time domain information of an original collected signal and time domain information of a plurality of preset harmonic reference signals.

And 102, acquiring total frequency domain information of all power frequency harmonic interference signals according to the frequency domain information of each harmonic reference signal and convolution kernel information of the power frequency harmonic interference signals respectively corresponding to each harmonic reference signal.

Step 103, obtaining first estimated frequency domain information of the first signal of the original collected signal without all power frequency harmonic interference signals according to the frequency domain information of the original collected signal and the total frequency domain information of all power frequency harmonic interference signals.

And 104, acquiring a first estimated amplitude spectrum and a first estimated phase spectrum of the first signal according to the first estimated frequency domain information.

Here, the first estimated frequency domain function may be embodied asWritten separately in amplitude and phase:where ω is a frequency variable and j is a custom constant.

And 105, acquiring the amplitude spectrum of the original acquisition signal according to the frequency domain information of the original acquisition signal, and performing singular point elimination processing on the amplitude spectrum of the original acquisition signal to obtain a second estimated amplitude spectrum.

In the specific implementation, singular point elimination processing may be performed on the magnitude spectrum | X (ω) | of the frequency domain function X (ω) of the original acquired signal, so as to obtain a second estimated magnitude spectrum X' (ω) after elimination.

And 106, replacing the first estimated amplitude spectrum of the first estimated frequency domain information with a second estimated amplitude spectrum, and keeping the first estimated phase spectrum unchanged to obtain second estimated frequency domain information of the first signal.

Here, the first estimated frequency domain function may be embodied asThe amplitude spectrum is replaced by a second estimated amplitude spectrum X' (omega) to obtain a second estimated frequency domain function

And step 107, acquiring second estimated time domain information of the first signal according to the second estimated frequency domain information.

Here, the second estimated frequency domain function may be implemented specifically forFourier transform is carried out to obtain a second estimated time domain function of the first signal

In this embodiment, after the first estimated frequency domain information of the first signal is obtained, singular point elimination processing is performed on the magnitude spectrum of the originally acquired signal, and the magnitude spectrum of the first estimated frequency domain information is replaced with the eliminated magnitude spectrum, so as to obtain a second estimated magnitude spectrum, and further suppress interference of the power frequency harmonic signal on the originally acquired signal.

Preferably, as shown in fig. 10, the step 105 specifically includes:

and 1051, removing singular points of the amplitude spectrum of the original acquisition signal in a steady estimation mode, and reassigning the removed points in a smooth compensation mode to obtain a second estimated amplitude spectrum.

At the moment, singular points of the amplitude spectrum of the original acquisition signal are removed in a steady estimation mode, and the removed points are assigned again in a smooth compensation mode, so that the interference of the power frequency harmonic interference signal on the amplitude spectrum of the original acquisition signal can be effectively removed, the effect of inhibiting the power frequency interference is further improved, and the distortion of the signal is reduced.

The fourth embodiment:

the method for inhibiting power frequency harmonic interference of the embodiment of the invention comprises the following steps:

step 1011, according to the time domain information of the originally acquired signal and the time domain information of each harmonic reference signal, obtaining the cross-power spectral density function of each harmonic reference signal and the originally acquired signal respectively through the following formula

Step 1012, obtaining the self-power spectral density function of each harmonic reference signal according to the time domain information of each harmonic reference signal by the following formula

Where x (t) is a time domain function of the originally acquired signal, v_i(t) is a time domain function of the ith harmonic reference signal, v_i(t + τ) is a time domain function of the delay time τ of the ith harmonic reference signal, i is an integer between 1 and p, p is the number of the harmonic reference signals, t is a time variable, ω is a frequency variable, and j and τ are self-defined constants.

Step 1013, according to the self-power spectrum information and the corresponding cross-power spectrum information of each harmonic reference signal, obtaining a convolution kernel function H of the power frequency harmonic interference signal corresponding to each harmonic reference signal through the following formula_i(ω)：

Wherein,is a cross-power spectral density function of the ith harmonic reference signal and the original acquired signal,is a function of the self-power spectral density of the ith said harmonic reference signal.

Step 1021, time domain function v of each harmonic reference signal_i(t) carrying out Fourier transform to obtain a frequency domain function V of each harmonic reference signal_i(ω)。

Step 1022, obtaining the total frequency domain function V (ω) of all power frequency harmonic interference signals by the following formula:

and step 1031, performing Fourier transform on the time domain function of the original acquisition signal to obtain a frequency domain function X (omega) of the original acquisition signal.

Step 1032, obtain a first estimated frequency domain function of the first signal by the following formula

Step 1033, for the first estimated frequency domain functionPerforming inverse Fourier transform to obtain a first estimated time domain function of the first signal

And 104, acquiring a first estimated amplitude spectrum and a first estimated phase spectrum of the first signal according to the first estimated frequency domain information.

Here, the first estimated frequency domain function may be embodied asSplit writingIn the form of amplitude and phase:where ω is a frequency variable and j is a custom constant.

And 1051, removing singular points of the amplitude spectrum of the original acquisition signal in a steady estimation mode, and reassigning the removed points in a smooth compensation mode to obtain a second estimated amplitude spectrum.

In the specific implementation, singular point elimination processing may be performed on the magnitude spectrum | X (ω) | of the frequency domain function X (ω) of the original acquired signal, so as to obtain a second estimated magnitude spectrum X' (ω) after elimination.

And 106, replacing the first estimated amplitude spectrum of the first estimated frequency domain information with a second estimated amplitude spectrum, and keeping the first estimated phase spectrum unchanged to obtain second estimated frequency domain information of the first signal.

Here, the first estimated frequency domain function may be embodied asThe amplitude spectrum is replaced by a second estimated amplitude spectrum X' (omega) to obtain a second estimated frequency domain function

And step 107, acquiring second estimated time domain information of the first signal according to the second estimated frequency domain information.

Here, the second estimated frequency domain function may be implemented specifically forFourier transform is carried out to obtain a second estimated time domain function of the first signal

Generally, the collected original signals all contain power frequency harmonic interference, the time domain information, the amplitude spectrum and the phase spectrum of the signals are respectively shown in fig. 3-5 when the signals are not supposed to be interfered by the power frequency harmonic interference, and the time domain information, the amplitude spectrum and the phase spectrum of the signals are respectively shown in fig. 6-8 when the signals are supposed to be interfered by the power frequency harmonic interference.

By applying the method of the embodiment of the invention, as shown in FIG. 2, an original acquisition signal x (t) is obtained first, and a cross-power spectral density function of each harmonic reference signal and the original acquisition signal is obtained according to the original acquisition signal x (t) and the harmonic reference signal 1-pAnd the self-power spectral density function of each harmonic reference signalThen calculating convolution kernel H of each power frequency harmonic reference signal_i(ω) and time domain function v for each harmonic reference signal separately_i(t) carrying out Fourier transform to obtain a frequency domain function V of each harmonic reference signal_i(ω); convolution kernel H according to each power frequency harmonic reference signal_i(omega) and frequency domain function V of each harmonic reference signal_i(ω), obtaining a total frequency domain function V (ω) of all power frequency harmonic interference signals: fourier transformation is carried out on the time domain function X (t) of the original acquisition signal to obtain a frequency domain function X (omega) of the original acquisition signal, and the total frequency domain function V (omega) of all power frequency harmonic interference signals is subtracted from the frequency domain function X (omega) of the original acquisition signal to obtain a first estimated frequency domain functionAmplitude of frequency domain function X (omega) of original collected signalThe robust estimation of the power spectrum is carried out to obtain a second estimated amplitude spectrum X' (omega), the phase spectrum of the first estimated frequency domain function is obtained to be combined with the second estimated amplitude spectrum to obtain a second estimated frequency domain functionCarrying out Fourier inverse transformation on the second estimation frequency domain function to obtain a second estimation time domain functionFinally, the time domain information of the signal after the power frequency interference is filtered is shown in fig. 9.

As can be seen from fig. 9, the method of the embodiment of the present invention effectively suppresses interference of the power frequency harmonic signal in the originally acquired signal, so that the signal distortion is small, and the method has great value in power frequency interference suppression.

The method of the embodiment of the invention is based on Fourier transform, does not need high-complexity computing resources, and has strong applicability to small-sized computing platforms; the implementation of the method of the embodiment of the invention has no iterative relationship, the main part is carried out based on Fourier transform, and the real-time performance is good; the method of the embodiment of the invention can obviously improve the signal-to-noise ratio, can almost extract the signal which is the same as the original signal, and fully reduces the distortion caused by power frequency interference.

Fifth embodiment:

as shown in fig. 11, an embodiment of the present invention also provides a filter including:

the first obtaining module 1101 is configured to obtain convolution kernel information of power frequency harmonic interference signals corresponding to each harmonic reference signal according to time domain information of an original acquired signal and time domain information of a plurality of preset harmonic reference signals.

A second obtaining module 1102, configured to obtain total frequency domain information of all power frequency harmonic interference signals according to the frequency domain information of each harmonic reference signal and convolution kernel information of the power frequency harmonic interference signal corresponding to each harmonic reference signal.

A third obtaining module 1103, configured to obtain, according to the frequency domain information of the original collected signal and the total frequency domain information of all the power frequency harmonic interference signals, first estimated frequency domain information of a first signal of the original collected signal from which all the power frequency harmonic interference signals are removed, and obtain, according to the first estimated frequency domain information, first estimated time domain information of the first signal.

The filter provided by the embodiment of the invention has great value in power frequency interference suppression, reduces the complexity of setting the wave trap, is simple and easy to implement, and can effectively suppress power frequency harmonic signals. Compared with the traditional wave trap, the method has the advantages of frequency adaptability, time delay estimation accuracy, good signal-to-noise ratio improvement degree, small signal distortion degree, suitability for common engineering application and good stability.

Sixth embodiment:

an embodiment of the present invention further provides a filter, including:

a first obtaining unit 11011, configured to obtain, according to the time domain information of the original acquired signal and the time domain information of each harmonic reference signal, cross-power spectrum information of each harmonic reference signal and the original acquired signal respectively.

A second obtaining unit 11012, configured to obtain self-power spectrum information of each harmonic reference signal according to the time domain information of each harmonic reference signal, respectively.

A third obtaining unit 11013, configured to obtain convolution kernel information of the power frequency harmonic interference signal corresponding to each harmonic reference signal according to the self-power spectrum information of each harmonic reference signal and the corresponding cross-power spectrum information.

A second obtaining module 1102, configured to obtain total frequency domain information of all power frequency harmonic interference signals according to the frequency domain information of each harmonic reference signal and convolution kernel information of the power frequency harmonic interference signal corresponding to each harmonic reference signal.

A third obtaining module 1103, configured to obtain, according to the frequency domain information of the original collected signal and the total frequency domain information of all the power frequency harmonic interference signals, first estimated frequency domain information of a first signal of the original collected signal from which all the power frequency harmonic interference signals are removed, and obtain, according to the first estimated frequency domain information, first estimated time domain information of the first signal.

Preferably, the first obtaining unit 11011 specifically includes:

a first obtaining subunit 110111, configured to obtain a cross-power spectral density function of each of the harmonic reference signals and the originally acquired signal according to the following formula

Where x (t) is a time domain function of the originally acquired signal, v_i(t + τ) is a time domain function of the delay time τ of the ith harmonic reference signal, i is an integer between 1 and p, p is the number of the harmonic reference signals, t is a time variable, ω is a frequency variable, and j and τ are self-defined constants.

Preferably, the second obtaining unit 11012 specifically includes:

a second obtaining subunit 110121, configured to obtain a self-power spectral density function of each of the harmonic reference signals by the following formula

Wherein v is_i(t) is a time domain function of the ith harmonic reference signal, v_i(t + τ) is a time domain function of the delay time τ of the ith harmonic reference signal, i is an integer between 1 and p, p is the number of the harmonic reference signals, t is a time variable, ω is a frequency variable, and j and τ are self-defined constants.

Preferably, the third obtaining unit 11013 specifically includes:

a third obtaining subunit 110131, configured to obtain a convolution kernel H of the power frequency harmonic interference signal corresponding to each of the harmonic reference signals according to the following formula_i(ω)：

Wherein,is a cross-power spectral density function of the ith harmonic reference signal and the original acquired signal,is the self-power spectrum density function of the ith harmonic reference signal, i is an integer between 1 and p, p is the number of the harmonic reference signals, and omega is the frequency variable.

Preferably, the second obtaining module 1102 specifically includes:

a first transforming unit 11021, configured to perform fourier transform on the time domain function of each harmonic reference signal respectively to obtain a frequency domain function V of each harmonic reference signal_i(ω)；

A fourth obtaining unit 11022, configured to obtain a total frequency domain function V (ω) of all the power frequency harmonic interference signals according to the following formula:

wherein H_i(ω) is the convolution kernel function, V, of the power frequency harmonic interference signal corresponding to the ith harmonic reference signal_i(ω) is a frequency domain function of the ith harmonic reference signal, i is an integer between 1 and p, p is the number of harmonic reference signals, and ω is a frequency variable.

Preferably, the third obtaining module 1103 specifically includes:

a second transforming unit 11031, configured to perform fourier transform on the time domain function of the original acquired signal to obtain a frequency domain function X (ω) of the original acquired signal;

a fifth obtaining unit 11032, configured to obtain a first estimated frequency domain function of the first signal by the following formula

V (omega) is a total frequency domain function of all the power frequency harmonic interference signals, and omega is a frequency variable;

a third transforming unit 11033, configured to perform inverse fourier transform on the first estimated frequency domain function, so as to obtain a first estimated time domain function of the first signal.

Seventh embodiment:

an embodiment of the present invention further provides a filter, including:

the first obtaining module 1101 is configured to obtain convolution kernel information of power frequency harmonic interference signals corresponding to each harmonic reference signal according to time domain information of an original acquired signal and time domain information of a plurality of preset harmonic reference signals.

A second obtaining module 1102, configured to obtain total frequency domain information of all power frequency harmonic interference signals according to the frequency domain information of each harmonic reference signal and convolution kernel information of the power frequency harmonic interference signal corresponding to each harmonic reference signal.

A third obtaining module 1103, configured to obtain, according to the frequency domain information of the original collected signal and the total frequency domain information of all the power frequency harmonic interference signals, first estimated frequency domain information of the first signal, from which all the power frequency harmonic interference signals are removed, of the original collected signal.

A fourth obtaining module 1104, configured to obtain a first estimated magnitude spectrum and a first estimated phase spectrum of the first signal according to the first estimated frequency domain information.

The eliminating module 1105 is configured to obtain the magnitude spectrum of the original acquired signal according to the frequency domain information of the original acquired signal, and perform singular point eliminating processing on the magnitude spectrum of the original acquired signal to obtain a second estimated magnitude spectrum.

A replacing module 1106, configured to replace the first estimated magnitude spectrum of the first estimated frequency domain information with the second estimated magnitude spectrum, and keep the first estimated phase spectrum unchanged, so as to obtain second estimated frequency domain information of the first signal.

A sixth obtaining module 1107, configured to obtain the second estimated time domain information of the first signal according to the second estimated frequency domain information.

The eliminating module 1105 specifically includes:

and the rejecting unit 11051 is configured to perform singular point rejection on the amplitude spectrum of the original acquired signal in a robust estimation manner, and reassign the rejected points in a smooth compensation manner to obtain a second estimated amplitude spectrum.

The filter provided by the embodiment of the invention has great value in power frequency interference suppression, reduces the complexity of setting the wave trap, is simple and easy to implement, and can effectively suppress power frequency harmonic signals. Compared with the traditional wave trap, the method has the advantages of frequency adaptability, time delay estimation accuracy, good signal-to-noise ratio improvement degree, small signal distortion degree, suitability for common engineering application and good stability.

It should be noted that in the drawings or description, the same drawing reference numerals are used for similar or identical parts. Implementations not depicted or described in the drawings are of a form known to those of ordinary skill in the art. Additionally, while exemplifications of parameters including particular values may be provided herein, it is to be understood that the parameters need not be exactly equal to the respective values, but may be approximated to the respective values within acceptable error margins or design constraints. Directional phrases used in the embodiments, such as "upper," "lower," "front," "rear," "left," "right," and the like, refer only to the orientation of the figure. Accordingly, the directional terminology used is intended to be in the nature of words of description rather than of limitation.

Further, the above definitions of the various elements and methods are not limited to the various specific structures, shapes or arrangements of parts mentioned in the examples, which may be easily modified or substituted by those of ordinary skill in the art. For example: the phase setting of the harmonic reference signal is not necessarily zero phase, and the constant A can be freely set; the parameter settings for robust estimation can be set on a case-by-case basis, and usually do not cause too large a bias.

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), or the like.

While the foregoing is directed to the preferred embodiment of the present invention, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention as defined in the appended claims.

Claims (18)

1. A method for suppressing power frequency harmonic interference is characterized by comprising the following steps:

acquiring convolution kernel information of power frequency harmonic interference signals corresponding to each harmonic reference signal according to time domain information of an original acquired signal and time domain information of a plurality of preset harmonic reference signals;

acquiring total frequency domain information of all power frequency harmonic interference signals according to the frequency domain information of each harmonic reference signal and convolution kernel information of the power frequency harmonic interference signals corresponding to each harmonic reference signal respectively;

according to the frequency domain information of the original collected signal and the total frequency domain information of all the power frequency harmonic interference signals, obtaining first estimated frequency domain information of a first signal of the original collected signal without all the power frequency harmonic interference signals, and according to the first estimated frequency domain information, obtaining first estimated time domain information of the first signal.

2. The method according to claim 1, wherein the obtaining convolution kernel information of the power frequency harmonic interference signal corresponding to each harmonic reference signal according to the time domain information of the originally acquired signal and the time domain information of a plurality of preset harmonic reference signals specifically comprises:

acquiring cross-power spectrum information of each harmonic reference signal and the original acquisition signal according to the time domain information of the original acquisition signal and the time domain information of each harmonic reference signal;

respectively acquiring the self-power spectrum information of each harmonic reference signal according to the time domain information of each harmonic reference signal;

and acquiring convolution kernel information of power frequency harmonic interference signals corresponding to each harmonic reference signal according to the self-power spectrum information and the corresponding cross-power spectrum information of each harmonic reference signal.

3. The method according to claim 2, wherein the obtaining cross-power spectrum information of each of the harmonic reference signals and the original acquired signal according to the time domain information of the original acquired signal and the time domain information of each of the harmonic reference signals comprises:

obtaining a cross-power spectral density function of each harmonic reference signal and the original acquired signal respectively through the following formula

$S_{{\mathrm{xv}}_i}\left(\mathrm{\ω}\right)={\∫}_{-\mathrm{\∞}}^{+\mathrm{\∞}}{\[{\∫}_{-\mathrm{\∞}}^{+\mathrm{\∞}}x\left(t\right)v_i(t+\mathrm{\τ})dt\]}^{-j\mathrm{\ω}\mathrm{\τ}}dt;$

Where x (t) is a time domain function of the originally acquired signal, v_i(t + τ) is a time domain function of the delay time τ of the ith harmonic reference signal, i is an integer between 1 and p, p is the number of the harmonic reference signals, t is a time variable, ω is a frequency variable, and j and τ are self-defined constants.

4. The method according to claim 2, wherein the obtaining self-power spectrum information of each harmonic reference signal according to the time domain information of each harmonic reference signal respectively comprises:

by the following formulaObtaining a self-power spectral density function of each harmonic reference signal

$S_{v_i{v}_i}\left(\mathrm{\ω}\right)={\∫}_{-\mathrm{\∞}}^{+\mathrm{\∞}}\[{\∫}_{-\mathrm{\∞}}^{+\mathrm{\∞}}{v}_i\left(t\right)v_i(t+\mathrm{\τ})dt\]e^{-j\mathrm{\ω}\mathrm{\τ}}dt;$

Wherein v is_i(t) is a time domain function of the ith harmonic reference signal, v_i(t + τ) is a time domain function of the delay time τ of the ith harmonic reference signal, i is an integer between 1 and p, p is the number of the harmonic reference signals, t is a time variable, ω is a frequency variable, and j and τ are self-defined constants.

5. The method according to claim 2, wherein the obtaining convolution kernel information of the power frequency harmonic interference signal corresponding to each of the harmonic reference signals according to the self-power spectrum information and the cross-power spectrum information of each of the harmonic reference signals includes:

obtaining convolution kernel function H of power frequency harmonic interference signal corresponding to each harmonic reference signal through the following formula_i(ω)：

$H_i\left(\mathrm{\ω}\right)=\frac{S_{{\mathrm{xv}}_i}\left(\mathrm{\ω}\right)}{S_{v_i{v}_i}\left(\mathrm{\ω}\right)};$

Wherein,is a cross-power spectral density function of the ith harmonic reference signal and the original acquired signal,is the self-power spectrum density function of the ith harmonic reference signal, i is an integer between 1 and p, p is the number of the harmonic reference signals, and omega is the frequency variable.

6. The method according to claim 1, wherein the obtaining total frequency domain information of all the power frequency harmonic interference signals according to the frequency domain information of each harmonic reference signal and convolution kernel information of the power frequency harmonic interference signal corresponding to each harmonic reference signal respectively comprises:

respectively carrying out Fourier transform on the time domain function of each harmonic reference signal to obtain the frequency domain function V of each harmonic reference signal_i(ω)；

Obtaining a total frequency domain function V (omega) of all the power frequency harmonic interference signals by the following formula:

$V\left(\mathrm{\ω}\right)=\underset{i=1}{\overset{P}{\Σ}}{H}_i\left(\mathrm{\ω}\right)V_i\left(\mathrm{\ω}\right);$

wherein H_i(ω) is the convolution kernel function, V, of the power frequency harmonic interference signal corresponding to the ith harmonic reference signal_i(ω) is a frequency domain function of the ith harmonic reference signal, i is an integer between 1 and p, p is the number of harmonic reference signals, and ω is a frequency variable.

7. The method according to claim 1, wherein the obtaining, according to the frequency domain information of the original collected signal and the total frequency domain information of all the power frequency harmonic interference signals, first estimated frequency domain information of a first signal of the original collected signal from which all the power frequency harmonic interference signals are removed, and obtaining, according to the first estimated frequency domain information, first estimated time domain information of the first signal specifically includes:

performing Fourier transform on the time domain function of the original acquisition signal to obtain a frequency domain function X (omega) of the original acquisition signal;

obtaining a first estimated frequency domain function of the first signal by

V (omega) is a total frequency domain function of all the power frequency harmonic interference signals, and omega is a frequency variable;

and carrying out inverse Fourier transform on the first estimation frequency domain function to obtain a first estimation time domain function of the first signal.

8. The method of claim 1, wherein after obtaining the first estimated frequency domain information of the first signal of the original collected signal without all the power frequency harmonic interference signals according to the frequency domain information of the original collected signal and the total frequency domain information of all the power frequency harmonic interference signals, the method further comprises:

acquiring a first estimated amplitude spectrum and a first estimated phase spectrum of the first signal according to the first estimated frequency domain information;

acquiring the amplitude spectrum of the original acquisition signal according to the frequency domain information of the original acquisition signal, and performing singular point elimination processing on the amplitude spectrum of the original acquisition signal to obtain a second estimated amplitude spectrum;

replacing the first estimated amplitude spectrum of the first estimated frequency domain information with the second estimated amplitude spectrum, and keeping the first estimated phase spectrum unchanged to obtain second estimated frequency domain information of the first signal;

and acquiring second estimated time domain information of the first signal according to the second estimated frequency domain information.

9. The method according to claim 8, wherein the singular point elimination processing is performed on the amplitude spectrum of the original acquired signal to obtain a second estimated amplitude spectrum, specifically comprising:

and performing singular point elimination on the amplitude spectrum of the original acquisition signal in a steady estimation mode, and performing reassignment on the eliminated points in a smooth compensation mode to obtain a second estimated amplitude spectrum.

10. A filter, comprising:

the first acquisition module is used for acquiring convolution kernel information of power frequency harmonic interference signals corresponding to each harmonic reference signal according to time domain information of an original acquisition signal and time domain information of a plurality of preset harmonic reference signals;

the second acquisition module is used for acquiring total frequency domain information of all power frequency harmonic interference signals according to the frequency domain information of each harmonic reference signal and convolution kernel information of the power frequency harmonic interference signals corresponding to each harmonic reference signal;

and the third acquisition module is used for acquiring first estimated frequency domain information of the first signal of the original acquisition signal without all the power frequency harmonic interference signals according to the frequency domain information of the original acquisition signal and the total frequency domain information of all the power frequency harmonic interference signals, and acquiring first estimated time domain information of the first signal according to the first estimated frequency domain information.

11. The filter of claim 10, wherein the first obtaining module specifically comprises:

the first acquisition unit is used for acquiring cross-power spectrum information of each harmonic reference signal and the original acquisition signal according to the time domain information of the original acquisition signal and the time domain information of each harmonic reference signal;

the second acquisition unit is used for respectively acquiring the self-power spectrum information of each harmonic reference signal according to the time domain information of each harmonic reference signal;

and the third obtaining unit is used for obtaining convolution kernel information of the power frequency harmonic interference signals corresponding to each harmonic reference signal according to the self-power spectrum information and the corresponding cross-power spectrum information of each harmonic reference signal.

12. The filter according to claim 11, wherein the first obtaining unit specifically includes:

a first obtaining subunit, configured to obtain a cross-power spectral density function of each of the harmonic reference signals and the original collected signal according to the following formula

$S_{{\mathrm{xv}}_i}\left(\mathrm{\ω}\right)={\∫}_{-\mathrm{\∞}}^{+\mathrm{\∞}}\[{\∫}_{-\mathrm{\∞}}^{+\mathrm{\∞}}x\left(t\right)v_i(t+\mathrm{\τ})dt\]e^{-j\mathrm{\ω}\mathrm{\τ}}dt;$

Where x (t) is a time domain function of the originally acquired signal, v_i(t + τ) is a time domain function of the delay time τ of the ith harmonic reference signal, i is an integer between 1 and p, p is the number of the harmonic reference signals, t is a time variable, ω is a frequency variable, and j and τ are self-defined constants.

13. The filter according to claim 11, wherein the second obtaining unit specifically includes:

a second obtaining subunit, configured to obtain a self-power spectral density function of each of the harmonic reference signals according to the following formula

$S_{v_i{v}_i}\left(\mathrm{\ω}\right)={\∫}_{-\mathrm{\∞}}^{+\mathrm{\∞}}\[{\∫}_{-\mathrm{\∞}}^{+\mathrm{\∞}}{v}_i\left(t\right)v_i(t+\mathrm{\τ})dt\]e^{-j\mathrm{\ω}\mathrm{\τ}}dt;$

Wherein v is_i(t) is a time domain function of the ith harmonic reference signal, v_i(t + τ) is a time domain function of the delay time τ of the ith harmonic reference signal, i is an integer between 1 and p, p is the number of the harmonic reference signals, t is a time variable, ω is a frequency variable, and j and τ are self-defined constants.

14. The filter according to claim 11, wherein the third obtaining unit specifically includes:

a third obtaining subunit, configured to obtain a convolution kernel function H of the power frequency harmonic interference signal corresponding to each of the harmonic reference signals according to the following formula_i(ω)：

$H_i\left(\mathrm{\ω}\right)=\frac{S_{{\mathrm{xv}}_i}\left(\mathrm{\ω}\right)}{S_{v_i{v}_i}\left(\mathrm{\ω}\right)};$

Wherein,is a cross-power spectral density function of the ith harmonic reference signal and the original acquired signal,is the self-power spectrum density function of the ith harmonic reference signal, i is an integer between 1 and p, p is the number of the harmonic reference signals, and omega is the frequency variable.

15. The filter of claim 10, wherein the second obtaining module specifically comprises:

a first transform unit, configured to perform fourier transform on the time domain function of each harmonic reference signal to obtain a frequency domain function V of each harmonic reference signal_i(ω)；

A fourth obtaining unit, configured to obtain a total frequency domain function V (ω) of all the power frequency harmonic interference signals according to the following formula:

$V\left(\mathrm{\ω}\right)=\underset{i=1}{\overset{P}{\Σ}}{H}_i\left(\mathrm{\ω}\right)V_i\left(\mathrm{\ω}\right);$

wherein H_i(omega) is a convolution kernel function of a power frequency harmonic interference signal corresponding to the ith harmonic reference signal，V_i(ω) is a frequency domain function of the ith harmonic reference signal, i is an integer between 1 and p, p is the number of harmonic reference signals, and ω is a frequency variable.

16. The filter of claim 10, wherein the third obtaining module specifically comprises:

the second transformation unit is used for carrying out Fourier transformation on the time domain function of the original acquisition signal to obtain a frequency domain function X (omega) of the original acquisition signal;

a fifth obtaining unit, configured to obtain a first estimated frequency domain function of the first signal according to the following formula

${\hat{S}}^{\′}\left(\mathrm{\ω}\right)=X\left(\mathrm{\ω}\right)-V\left(\mathrm{\ω}\right);$

V (omega) is a total frequency domain function of all the power frequency harmonic interference signals, and omega is a frequency variable;

and the third transformation unit is used for performing inverse Fourier transformation on the first estimation frequency domain function to obtain a first estimation time domain function of the first signal.

17. The filter of claim 10, further comprising:

a fourth obtaining module, configured to obtain a first estimated magnitude spectrum and a first estimated phase spectrum of the first signal according to the first estimated frequency domain information;

the elimination module is used for acquiring the amplitude spectrum of the original acquisition signal according to the frequency domain information of the original acquisition signal, and performing singular point elimination processing on the amplitude spectrum of the original acquisition signal to obtain a second estimated amplitude spectrum;

a replacing module, configured to replace the first estimated magnitude spectrum of the first estimated frequency domain information with the second estimated magnitude spectrum, and keep the first estimated phase spectrum unchanged, to obtain second estimated frequency domain information of the first signal;

a sixth obtaining module, configured to obtain second estimated time domain information of the first signal according to the second estimated frequency domain information.

18. The filter according to claim 17, wherein the culling module specifically comprises:

and the eliminating unit is used for eliminating singular points of the amplitude spectrum of the original acquisition signal in a steady estimation mode, and evaluating the eliminated points again in a smooth compensation mode to obtain a second estimated amplitude spectrum.