A current digital filtering method for electric power equipment

Abstract

The present invention relates to the technical field of electric variable measurement, and proposes a current digital filtering method for electric power equipment, including: obtaining a current signal time series; obtaining a change sequence according to the current signal time series; obtaining a time domain steady-state convergence degree according to the change sequence; obtaining an energy decay sequence by discrete Fourier transform; obtaining an energy decay variation index according to the energy decay sequence; obtaining a step factor according to the time domain steady-state convergence degree and the energy decay variation index; obtaining a filtered current signal sequence based on the step factor using an LMS adaptive filtering algorithm; and obtaining a current digital measurement result based on the filtered current signal sequence using a digital signal conversion technology. The present invention improves the filtering effect of the current signal and the accuracy of the current measurement by improving the step factor in the LMS adaptive filtering algorithm.

Description

Technical Field

[0001] The invention relates to the technical field of electric variable measurement, and in particular to a current digital filtering method for electric power equipment. Background Art

[0002] With the development of new energy vehicles, charging piles are being used more and more widely, and the charging infrastructure for new energy vehicles is becoming more and more complete. In order to meet the needs of the development of new energy vehicles, a large number of charging piles need to be invested. However, current charging piles have many problems such as inaccurate real-time current measurement. Under the condition of random disturbance, the non-stationary signals generated by nonlinear loads such as impact and strong time-varying in the charging process of new energy vehicles lead to more noise in the actual measurement of charging pile current, resulting in poor accuracy of actual measurement of charging pile current.

[0003] In order to improve the accuracy of charging pile current measurement, the collected current signal needs to be filtered. Most of the current signals are filtered using filter algorithms to improve the accuracy of charging pile current measurement. Common filter algorithms, such as the adaptive filter algorithm (Least Mean Square, LMS), can adaptively adjust the filter parameters to filter the current signal. However, the selection of the step size factor in the traditional LMS adaptive algorithm is difficult, which easily leads to poor filtering effect of the current signal and cannot effectively improve the accuracy of charging pile current measurement. Summary of the invention

[0004] The present invention provides a current digital filtering method for electric power equipment to solve the problem that the filtering effect of current signal is poor and the current measurement accuracy of charging pile cannot be effectively improved. The technical scheme adopted is as follows:

[0005] An embodiment of the present invention provides a current digital filtering method for an electric power device, the method comprising the following steps:

[0006] Acquire the current signal, and use the current signal to acquire the current signal time series;

[0007] Obtaining a change sequence of each subsequence of the current signal time series according to the current signal time series; obtaining a time domain steady-state convergence degree of the current signal time series according to the change sequence of each subsequence of the current signal time series;

[0008] The attenuation uniformity sequence of each subsequence of the current signal time series is obtained by discrete Fourier transform; the energy attenuation variation index of the current signal time series is obtained according to the attenuation uniformity sequence of each subsequence of the current signal time series; the step size factor of the current signal time series is obtained according to the time domain steady-state convergence degree and the energy attenuation variation index of the current signal time series;

[0009] The LMS adaptive filtering algorithm is used to obtain a filtered current signal time series based on the step size factor, and a current digital measurement result is obtained according to the filtered current signal time series.

[0010] Preferably, the method for obtaining the change sequence of each subsequence of the current signal time series according to the current signal time series is:

[0011] A sequence sliding window of a preset scale is used to slide on the current signal time series, the sliding step length is a first preset parameter, and a sequence composed of current signals in each sliding position of the sequence sliding window in ascending time order is used as each subsequence of the current signal time series;

[0012] For each subsequence of the current signal time series, the mean of the amplitudes of all current signals in the subsequence is taken as the average index of the subsequence, the difference between the amplitude of the current signal at each moment in the subsequence and the average index of the subsequence is taken as the change index at each moment in the subsequence, and the sequence composed of the change indexes of all moments in the subsequence in ascending time order is taken as the change sequence of the subsequence.

[0013] Preferably, the method for obtaining the time-domain steady-state convergence degree of the current signal time series according to the change sequence of each subsequence of the current signal time series is:

[0014] Obtaining a steady-state characteristic index of each subsequence of the current signal time series according to a numerical value in a change sequence of each subsequence of the current signal time series;

[0015] Obtaining the noise impedance index of each subsequence of the current signal time series according to the change of the value in the change sequence of each subsequence of the current signal time series;

[0016] The average value of the cumulative sum of the product of the steady-state characteristic index and the noise impedance index on the current signal time series is taken as the time-domain steady-state convergence degree of the current signal time series.

[0017] Preferably, the method for obtaining the steady-state characteristic index of each subsequence of the current signal time series according to the numerical value in the change sequence of each subsequence of the current signal time series is:

[0018] For the change sequence of each subsequence of the current signal time series, the negative mapping result with the natural constant as the base and the coefficient of variation of all values ​​in the change sequence as the exponent is used as the numerator;

[0019] Calculate the absolute value of the difference between each value in the change sequence and the previous value, and use the sum of the accumulated sum of the absolute values ​​in the change sequence and the second preset parameter as the denominator;

[0020] The ratio of the numerator to the denominator is taken as the steady-state characteristic index of the subsequence.

[0021] Preferably, the method for obtaining the noise impedance index of each subsequence of the current signal time series according to the change of the numerical value in the change sequence of each subsequence of the current signal time series is:

[0022] For the change sequence of each subsequence of the current signal time series, the information entropy and skewness of all the numerical values ​​in the change sequence are calculated respectively, and the product of the absolute value of the skewness and the information entropy is used as the first component factor, and the negative mapping result with the natural constant as the base and the first component factor as the exponent is used as the noise impedance index of the subsequence.

[0023] Preferably, the method of obtaining the attenuation uniformity sequence of each subsequence of the current signal time series by using discrete Fourier transform is:

[0024] Taking each subsequence of the current signal time series as the input of discrete Fourier transform, taking the output of discrete Fourier transform as all frequencies of each subsequence of the current signal time series and the frequency domain energy of all frequencies;

[0025] For each subsequence of the current signal time series, using MATLAB software to obtain an energy attenuation spectrum of the subsequence based on all frequencies of the subsequence and the frequency domain energy of all frequencies, taking the frequency domain energy of the fundamental frequency as the peak energy, and taking the frequency domain energy attenuated by the third preset parameter decibel as the energy threshold;

[0026] The frequency of the fundamental frequency in the energy attenuation spectrum of the subsequence and the set of frequencies corresponding to all frequency domain energies on the right side of the fundamental frequency that are higher than the energy threshold are taken as the effective frequency set, and the sequence composed of the frequency domain energies of all frequencies in the effective frequency set in ascending order of frequency is taken as the energy attenuation sequence of the subsequence;

[0027] Each difference between each element and the previous element in the energy attenuation sequence of the subsequence is calculated, and a sequence composed of all the differences is used as the attenuation uniformity sequence of the subsequence.

[0028] Preferably, the method for obtaining the energy attenuation variation index of the current signal time series according to the attenuation uniformity sequence of each subsequence of the current signal time series is:

[0029] Obtaining an energy attenuation influence index of each subsequence of the current signal time series according to an attenuation uniformity sequence of each subsequence of the current signal time series;

[0030] The average of the product of the standard deviation of all values ​​in the attenuation uniformity sequence of each subsequence of the current signal time series and the energy attenuation influence index accumulated on the current signal time series is taken as the energy attenuation variation index of the current signal time series.

[0031] Preferably, the method for obtaining the energy attenuation influence index of each subsequence of the current signal time series according to the attenuation uniformity sequence of each subsequence of the current signal time series is:

[0032] For each subsequence of the current signal time series, the negative mapping result with the natural constant as the base and the mean of all values ​​in the attenuation uniformity sequence of the subsequence as the exponent is taken as the energy attenuation influence index of the subsequence.

[0033] Preferably, the method for obtaining the step factor of the current signal time series according to the time domain steady-state convergence degree and energy decay variation index of the current signal time series is:

[0034] Calculating the maximum eigenvalue of the autocorrelation matrix of the current signal time series, and taking the reciprocal of the maximum eigenvalue as a first multiplication factor;

[0035] Calculate the product of a negative mapping result with a natural constant as the base and an energy attenuation variation index of the current signal time series as the exponent and the time domain steady-state convergence degree of the current signal time series, and use the reciprocal of the sum of the product and the second preset parameter as the second multiplication factor;

[0036] The product of the first multiplication factor and the second multiplication factor is used as the step factor of the current signal time series.

[0037] Preferably, the specific method of obtaining the filtered current signal time series based on the step size factor by using the LMS adaptive filtering algorithm and obtaining the current digital measurement result according to the filtered current signal time series is:

[0038] The current signal time series is used as the input of the LMS adaptive filtering algorithm, the step size factor of the current signal time series is used as the initial iteration step size of the LMS adaptive filtering algorithm, and the output of the LMS adaptive filtering algorithm is used as the filtered current signal time series;

[0039] The filtered current signal time series is used to perform digital signal conversion to obtain a current digital sequence, and the current digital sequence is used as a current digital measurement result.

[0040] The beneficial effects of the present invention are as follows: the present invention obtains a change sequence according to the change law of the numerical value in the time series, obtains the time domain steady-state convergence according to the change sequence, obtains the energy decay sequence by discrete Fourier transform, obtains the energy decay variation index according to the energy decay sequence, obtains the step factor according to the time domain steady-state convergence and the energy decay variation index, and obtains the filtered current signal time series based on the step factor by using an adaptive filtering algorithm. The beneficial effects are that the phenomenon of divergence due to poor steady-state error caused by an excessively large step factor is avoided, the filtering effect is improved, and the filtered current signal is made more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0042] Figure 1 A schematic diagram of a flow chart of a current digital filtering method for electric power equipment provided by an embodiment of the present invention;

[0043] Figure 2 A schematic diagram of a sequence sliding window of a current signal time series provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0044] 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.

[0045] See also Figure 1 , which shows a flow chart of a current digital filtering method for electric power equipment provided by an embodiment of the present invention, the method comprising the following steps:

[0046] Step S001, obtaining the current signal time series of the charging pile.

[0047] Since the charging pile needs to realize functions such as charging display, charging control, networking communication, and voltage and current data monitoring, the hardware system of the charging pile includes a hardware circuit for measuring the voltage and current of the charging pile and a USB communication circuit. The hardware circuit for measuring the voltage and current of the charging pile includes a sampling circuit. In the present invention, a current sensor is used to collect the current signal in the sampling circuit. The sampling time interval is 0.2s and the sampling time is 1h. The implementer can choose according to the actual situation. Furthermore, the sequence composed of the current signals at all sampling moments obtained by collection in ascending time order is used as the current signal time series.

[0048] At this point, the current signal time series is obtained.

[0049] Step S002, obtaining a subsequence of the current signal time series according to the current signal time series, obtaining a change sequence of the subsequence of the current signal time series according to the subsequence of the current signal time series, and obtaining a time domain steady-state convergence degree according to the change sequence of the subsequence of the current signal time series.

[0050] When using a charging pile to charge a new energy vehicle, external interference factors will cause abnormal fluctuations in the current in the charging pile. At this time, the collected current signal contains more noise components, which will greatly reduce the accuracy of current measurement. Therefore, in order to obtain more accurate digital current, it is necessary to filter the current signal time series.

[0051] The traditional LMS adaptive filtering algorithm is a filtering algorithm that automatically adjusts the filtering parameters, exploring the statistical characteristics of the current signal and noise components changing over time in order to achieve optimal filtering. However, the selection of the step size factor in the LMS adaptive filtering algorithm is difficult to determine. If the step size factor is too small, the convergence speed will be too slow; if the step size factor is too large, the steady-state error of the filter will be large, resulting in divergence of the convergence result, which greatly affects the filtering effect of the current signal. Therefore, it is necessary to adapt the step size factor in the LMS adaptive filtering algorithm so that the algorithm can have a good convergence speed while ensuring that the steady-state error is within a controllable range.

[0052] Specifically, set the current signal time series The sequence sliding window has a sliding step of 50. The schematic diagram of the sequence sliding window of the current signal time series is as follows Figure 2 Further, the sequence composed of the current signals in each sequence sliding window on the current signal time series in ascending time order is taken as each subsequence of the current signal time series.

[0053] With the influence of external noise, the fluctuation characteristics of the current signal in the current signal time series often contain more noise components. In order to analyze the influence of the noise components on the current signal time series, the present invention combines the time domain and frequency domain characteristics for analysis. The noise components will affect the steady-state convergence of the current signal time series, so the influence of the noise components on the change of the current signal is analyzed by the steady-state convergence of the current signal time series.

[0054] Specifically, for each subsequence of the current signal time series, the mean of the amplitudes of all current signals in the subsequence is taken as the average index of the subsequence, the difference between the amplitude of the current signal at each moment in the subsequence and the average index of the subsequence is taken as the change index at each moment in the subsequence, and the sequence composed of the change indexes of all moments in the subsequence in ascending time order is taken as the change sequence of the subsequence.

[0055] Since the smaller the noise component contained in the current signal change, the better the steady-state convergence of the current signal time series, the smaller the change of the value in the change sequence of the subsequence and the stronger the concentration of the value; conversely, the larger the change of the value in the change sequence of the subsequence and the stronger the discreteness of the value.

[0056] Calculate the time domain steady-state convergence of the current signal time series:

[0057]

[0058]

[0059]

[0060] In the formula, represents the steady-state characteristic index of the ith subsequence of the current signal time series, represents an exponential function with a natural constant as base, Represents the coefficient of variation of all values ​​in the change sequence of the ith subsequence of the current signal time series, Represents the number of values ​​in the change sequence of the ith subsequence of the current signal time series, and They represent the jth and (j-1th) values ​​in the change sequence of the i-th subsequence of the current signal time series, respectively. is the error parameter. To avoid the denominator being 0, the empirical value of the error parameter is 1. represents the noise impedance index of the i-th subsequence of the current signal time series, The information entropy of all values ​​in the change sequence of the ith subsequence of the current signal time series is represented by: Represents the skewness of all values ​​in the change sequence of the i-th subsequence of the current signal time series; Indicates the time domain steady-state convergence of the current signal time series, It should be noted that the calculation of information entropy and skewness is a well-known technology and will not be described in detail.

[0061] The coefficient of variation of all values ​​in the change sequence of the ith subsequence of the current signal time series The larger the value, the greater the difference between the jth and (j-1th) values ​​in the change sequence of the i-th subsequence of the current signal time series. The larger the value, the smaller the steady-state characteristics reflected in the subsequence change sequence, that is, the smaller the concentration, the greater the influence of the noise component, and the smaller the steady-state characteristic index. In addition, the information entropy of all values ​​in the change sequence of the i-th subsequence of the current signal time series is The larger the value, the greater the absolute value of the skewness of all values ​​in the change sequence of the i-th subsequence of the current signal time series. The larger the first component factor The larger the value is, the richer the data is in the variation sequence, which means that the data varies to a certain extent. At the same time, the greater the asymmetry of the distribution is, the more complex the influence of the noise component is, and the smaller the noise impedance index is. Therefore, the steady-state characteristic index of the i-th subsequence of the current signal time series is The smaller the noise impedance index of the i-th subsequence of the current signal time series is, the The smaller it is, the worse the steady-state convergence of the current signal time series is. The current signal time series is greatly affected by noise, and the smaller the time domain steady-state convergence is.

[0062] At this point, the time domain steady-state convergence of the current signal time series is obtained.

[0063] Step S003, using discrete Fourier transform to obtain an energy attenuation sequence, obtaining an energy attenuation variation index according to the energy attenuation sequence, and obtaining a step size factor according to the time domain steady-state convergence degree and the energy attenuation variation index.

[0064] Further, the frequency domain attenuation information of the current signal time series is obtained in combination with the frequency domain characteristics of each subsequence. Each subsequence of the current signal time series is used as the input of the discrete Fourier transform, and the output of the discrete Fourier transform is used as all frequencies of each subsequence of the current signal time series and the frequency domain energy of all frequencies. The discrete Fourier transform is a well-known technology and is not described in detail. For all frequencies of each subsequence of the current signal time series and the frequency domain energy of all frequencies, the energy attenuation spectrum of the subsequence in the frequency domain is drawn based on all frequencies and the frequency domain energy of all frequencies using MATLAB software. Since the infinity of frequency in the frequency domain space is not conducive to subsequent analysis, in order to facilitate subsequent analysis, the frequency domain energy of the base frequency is used as the peak energy, and the frequency domain energy under the peak energy attenuation of 10dB is used as the energy threshold, that is, the energy threshold is one tenth of the peak energy, and the frequency corresponding to the base frequency and all the frequency domain energies above the energy threshold on the right side of the base frequency on the energy attenuation spectrum is used as the effective frequency set, and the frequency domain energy of all frequencies in the effective frequency set is used as the sequence composed of the frequency domain energy of all frequencies in the frequency ascending order as the energy attenuation sequence of the subsequence.

[0065] When the current signal time series contains more noise components, the energy decay characteristics of the subsequence are slower, and it contains more frequency components, and the attenuation uniformity changes greatly; conversely, when the current signal time series contains fewer noise components, the energy decay characteristics of the subsequence are faster, and it contains fewer frequency components, and the attenuation uniformity changes less.

[0066] Furthermore, for the energy attenuation sequence of each subsequence of the current signal time series, each difference between each element in the energy attenuation sequence and the previous element is calculated, and the sequence composed of all the differences is used as the attenuation uniformity sequence of the subsequence. It should be noted that the first element in the energy attenuation sequence does not have a previous element and cannot reflect the change in energy attenuation uniformity, so the difference is calculated starting from the second element in the energy attenuation sequence, that is, the difference between the second element and the first element in the energy attenuation sequence is calculated, the difference between the third element and the second element, and so on, the difference between the last element and the previous element.

[0067] Calculate the energy decay variation index of the current signal time series:

[0068]

[0069]

[0070] In the formula, Represents the energy decay influence index of the i-th subsequence of the current signal time series, represents an exponential function with a natural constant as base, Represents the number of values ​​in the attenuation uniformity sequence of the i-th subsequence of the current signal time series, Represents the bth value in the attenuation uniformity sequence of the i-th subsequence of the current signal time series; Represents the energy decay variation index of the current signal time series, represents the number of subsequences in the current signal time series, Represents the standard deviation of all values ​​in the attenuation uniformity sequence of the i-th subsequence of the current signal time series.

[0071] The mean value of all values ​​in the attenuation uniformity sequence of the i-th subsequence of the current signal time series The larger the value is, the greater the frequency domain energy attenuation of the subsequence is, and the faster the attenuation speed is. At this time, the noise signal component contained in the current signal time series is more likely to be less, that is, the smaller the influence of noise on the current is, the smaller the energy attenuation influence index is. In addition, the energy attenuation influence index of the i-th subsequence of the current signal time series is The smaller the value, the standard deviation of all values ​​in the attenuation uniformity sequence of the i-th subsequence of the current signal time series The smaller it is, the smaller the influence of noise on current is, and the more concentrated the numerical distribution of the attenuation uniformity sequence is. At this time, the noise signal component contained in the current signal time series is more likely to be less, that is, the smaller the influence of noise on current is, the smaller the energy attenuation variation index is.

[0072] At this point, the time domain steady-state convergence degree and energy attenuation variation index are obtained based on the characteristics of the time domain and frequency domain.

[0073] Furthermore, the step size factor is obtained according to the time domain steady-state convergence degree and the energy attenuation variation index. The value range of the step size parameter in the traditional LMS adaptive filtering algorithm is ,in Represents the largest eigenvalue of the autocorrelation matrix of the input signal.

[0074] Calculate the step factor of the current signal time series:

[0075]

[0076] In the formula, represents the step factor of the current signal time series, represents the maximum eigenvalue of the autocorrelation matrix of the current signal time series, Indicates the time domain steady-state convergence of the current signal time series, represents an exponential function with a natural constant as base, Represents the energy decay variation index of the current signal time series, It should be noted that the calculation of the maximum eigenvalue of the autocorrelation matrix of the current signal time series is a well-known technology and will not be described in detail.

[0077] The maximum eigenvalue of the autocorrelation matrix of the current signal time series The larger the first multiplication factor The smaller the value, the more stable the time domain of the current signal time series is. The larger the energy decay variation index of the current signal time series is, the The smaller the first multiplication factor The smaller it is, the less the current signal time series is affected by the noise component, and the better the steady-state convergence of the current signal time series is. The smaller the step factor is, the better the steady-state convergence is maintained. Conversely, the larger the step factor is, the faster the convergence speed is.

[0078] At this point, the step factor of the current signal time series is obtained.

[0079] Step S004: using an LMS adaptive filtering algorithm to obtain a filtered current signal sequence based on a step size factor, and using a digital signal conversion technique to obtain a current digital measurement result based on the filtered current signal sequence.

[0080] Furthermore, the current signal time series is used as the input of the LMS adaptive filtering algorithm, the step size factor of the current signal time series is used as the initial iteration step size of the LMS adaptive filtering algorithm, and the output of the LMS adaptive filtering algorithm is used as the filtered current signal time series. The LMS adaptive filtering algorithm is a well-known technology and will not be described in detail.

[0081] The filtered current signal time series is used to perform digital signal conversion to obtain a more accurate current digital sequence, and the current digital sequence is used as the current digital measurement result of the charging pile.

[0082] At this point, a current digital filtering method for power equipment is completed.

[0083] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modification, equivalent replacement, improvement, etc. made within the principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A current digital filtering method for power equipment, characterized in that: The method comprises the following steps: Acquire the current signal, and use the current signal to acquire the current signal time series; Obtaining a change sequence of each subsequence of the current signal time series according to the current signal time series; obtaining a time domain steady-state convergence degree of the current signal time series according to the change sequence of each subsequence of the current signal time series; The attenuation uniformity sequence of each subsequence of the current signal time series is obtained by discrete Fourier transform; the energy attenuation variation index of the current signal time series is obtained according to the attenuation uniformity sequence of each subsequence of the current signal time series; the step size factor of the current signal time series is obtained according to the time domain steady-state convergence degree and the energy attenuation variation index of the current signal time series; Using an LMS adaptive filtering algorithm to obtain a filtered current signal time series based on the step size factor, and obtaining a current digital measurement result according to the filtered current signal time series; The method of using discrete Fourier transform to obtain the attenuation uniformity sequence of each subsequence of the current signal time series is: Taking each subsequence of the current signal time series as the input of discrete Fourier transform, taking the output of discrete Fourier transform as all frequencies of each subsequence of the current signal time series and the frequency domain energy of all frequencies; For each subsequence of the current signal time series, using MATLAB software to obtain an energy attenuation spectrum of the subsequence based on all frequencies of the subsequence and the frequency domain energy of all frequencies, taking the frequency domain energy of the fundamental frequency as the peak energy, and taking the frequency domain energy attenuated by the third preset parameter decibel as the energy threshold; The frequency of the fundamental frequency in the energy attenuation spectrum of the subsequence and the set of frequencies corresponding to all frequency domain energies on the right side of the fundamental frequency that are higher than the energy threshold are taken as the effective frequency set, and the sequence composed of the frequency domain energies of all frequencies in the effective frequency set in ascending order of frequency is taken as the energy attenuation sequence of the subsequence; Each difference between each element and the previous element in the energy attenuation sequence of the subsequence is calculated, and a sequence composed of all the differences is used as the attenuation uniformity sequence of the subsequence.

2. A current digital filtering method for electric power equipment according to claim 1, characterized in that: The method for obtaining the change sequence of each subsequence of the current signal time series according to the current signal time series is: A sequence sliding window of a preset scale is used to slide on the current signal time series, the sliding step length is a first preset parameter, and a sequence composed of current signals in each sliding position of the sequence sliding window in ascending time order is used as each subsequence of the current signal time series; For each subsequence of the current signal time series, the mean of the amplitudes of all current signals in the subsequence is taken as the average index of the subsequence, the difference between the amplitude of the current signal at each moment in the subsequence and the average index of the subsequence is taken as the change index at each moment in the subsequence, and the sequence composed of the change indexes of all moments in the subsequence in ascending time order is taken as the change sequence of the subsequence.

3. A current digital filtering method for electric power equipment according to claim 1, characterized in that: The method for obtaining the time domain steady-state convergence degree of the current signal time series according to the change sequence of each subsequence of the current signal time series is: Obtaining a steady-state characteristic index of each subsequence of the current signal time series according to a numerical value in a change sequence of each subsequence of the current signal time series; Obtaining the noise impedance index of each subsequence of the current signal time series according to the change of the value in the change sequence of each subsequence of the current signal time series; The average value of the cumulative sum of the product of the steady-state characteristic index and the noise impedance index on the current signal time series is taken as the time-domain steady-state convergence degree of the current signal time series.

4. A current digital filtering method for electric power equipment according to claim 3, characterized in that: The method for obtaining the steady-state characteristic index of each subsequence of the current signal time series according to the numerical value in the change sequence of each subsequence of the current signal time series is: For the change sequence of each subsequence of the current signal time series, the negative mapping result with the natural constant as the base and the coefficient of variation of all values ​​in the change sequence as the exponent is used as the numerator; Calculate the absolute value of the difference between each value in the change sequence and the previous value, and use the sum of the accumulated sum of the absolute values ​​in the change sequence and the second preset parameter as the denominator; The ratio of the numerator to the denominator is taken as the steady-state characteristic index of the subsequence.

5. A current digital filtering method for electric power equipment according to claim 3, characterized in that: The method for obtaining the noise impedance index of each subsequence of the current signal time series according to the change of the value in the change sequence of each subsequence of the current signal time series is: For the change sequence of each subsequence of the current signal time series, the information entropy and skewness of all the numerical values ​​in the change sequence are calculated respectively, and the product of the absolute value of the skewness and the information entropy is used as the first component factor, and the negative mapping result with the natural constant as the base and the first component factor as the exponent is used as the noise impedance index of the subsequence.

6. A current digital filtering method for electric power equipment according to claim 1, characterized in that: The method for obtaining the energy attenuation variation index of the current signal time series according to the attenuation uniformity sequence of each subsequence of the current signal time series is: Obtaining an energy attenuation influence index of each subsequence of the current signal time series according to an attenuation uniformity sequence of each subsequence of the current signal time series; The mean of the product of the standard deviation of all values ​​in the attenuation uniformity sequence of each subsequence of the current signal time series and the energy attenuation influence index accumulated on the current signal time series is used as the energy attenuation variation index of the current signal time series; The method for obtaining the energy attenuation influence index of each subsequence of the current signal time series according to the attenuation uniformity sequence of each subsequence of the current signal time series is: For each subsequence of the current signal time series, the negative mapping result with the natural constant as the base and the mean of all values ​​in the attenuation uniformity sequence of the subsequence as the exponent is taken as the energy attenuation influence index of the subsequence.

7. A current digital filtering method for electric power equipment according to claim 1, characterized in that: The method for obtaining the step factor of the current signal time series according to the time domain steady-state convergence degree and energy attenuation variation index of the current signal time series is: Calculating the maximum eigenvalue of the autocorrelation matrix of the current signal time series, and taking the reciprocal of the maximum eigenvalue as a first multiplication factor; Calculate the product of a negative mapping result with a natural constant as the base and an energy decay variation index of the current signal time series as the exponent and the time domain steady-state convergence degree of the current signal time series, and use the reciprocal of the sum of the product and the second preset parameter as the second multiplication factor; The product of the first multiplication factor and the second multiplication factor is used as the step factor of the current signal time series.

8. A current digital filtering method for electric power equipment according to claim 1, characterized in that: The specific method of using the LMS adaptive filtering algorithm to obtain the filtered current signal time series based on the step size factor and obtaining the current digital measurement result according to the filtered current signal time series is: The current signal time series is used as the input of the LMS adaptive filtering algorithm, the step size factor of the current signal time series is used as the initial iteration step size of the LMS adaptive filtering algorithm, and the output of the LMS adaptive filtering algorithm is used as the filtered current signal time series; The filtered current signal time series is used to perform digital signal conversion to obtain a current digital sequence, and the current digital sequence is used as a current digital measurement result.

Images