Weak measurement current signal noise reduction processing method

Through adaptive morphological filtering technology and weighted superposition method, the problem of current signal noise influence in the small current grounded distribution network is solved, and weak current signal extraction and fault location under the background of complex noise are realized.

CN119024092APending Publication Date: 2024-11-26FUXIN POWER SUPPLY COMPANY STATE GRID LIAONING ELECTRIC POWER +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410689269.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-05-30
Publication Date
2024-11-26

AI Technical Summary

Technical Problem

In a small current grounded distribution network, the current signal is susceptible to noise, resulting in signal attenuation and difficulty in positioning faults. The existing noise reduction method is not suitable for fault location.

Method used

Adaptive morphological filtering technology is adopted to construct filtering functions and denoising signals by selecting appropriate structural element functions and weighting methods, adjust structural elements and weighting values ​​to obtain the best denoising signals, and combine open operations and closed operations to form an alternating mixture. Filter function, reduces peak and trough noise, filters the signal through weighted superposition to improve signal quality.

Benefits of technology

In the context of complex noise, weak measurement current signals are effectively extracted, fault locations are quickly identified, signal-to-noise ratio is improved, and fault locations are suitable for fault locations in distribution networks.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119024092A_ABST
    Figure CN119024092A_ABST
Patent Text Reader

Abstract

The invention discloses a weak measurement current signal noise reduction processing method, which adopts a weak signal noise reduction processing technology of self-adaptive morphological filtering, selects a proper structural element to carry out open operation on a signal to be processed so as to reduce peak noise in the signal, and carries out closed operation on the signal so as to reduce valley noise in the signal. The two alternating filtering functions are combined and averaged to form an alternating hybrid filtering function with a better filtering effect, and the problem of output offset can be well solved. Meanwhile, in the noise reduction process, a filtering signal and a noise signal are superposed in a weighting mode, and an optimal noise reduction signal is obtained by continuously adjusting the weight ratio of the filtered signal in a calculation result. By adopting the method provided by the invention, an optimal noise reduction signal can be obtained under a complex noise background, and a useful weak measurement current signal in a power distribution network in a fault positioning process can be quickly identified and extracted.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of current signal detection and processing, and in particular to a noise reduction processing method for a weak measured current signal. Background Art

[0002] Most of the 6-60kV distribution networks in my country adopt the neutral point non-effective grounding method, also known as the small current system. Its advantage is that when a single-phase grounding fault occurs, the grounding current is very small. In most cases, the arc at the grounding point can extinguish itself, the line insulation will not be damaged, and the line can be restored to normal without disconnecting the line, and the operation reliability is high. However, it is well known that there are two problems that need to be solved in the operation of the small current grounding power grid. One is that when permanent grounding occurs, the fault line needs to be selected, and the other is to locate the selected fault line in order to handle the fault. When locating the single-phase grounding fault of the distribution network system, it is first necessary to inject a detection current into the distribution network. The detection current signal will flow along the line to the fault point and flow into the earth through the fault point to form a loop. Since only the fault phase can detect the detection current, the operator can determine the fault location by detecting the current signal. In the process of detecting the current signal, the current signal is often affected by the noise of the distribution network. Therefore, it is necessary to improve the signal-to-noise ratio of the signal under the complex noise background to obtain the best noise elimination signal, quickly identify and extract the useful weak measurement current signal, and realize the identification and extraction of the weak detection current signal under the complex noise background. At present, the noise reduction processing methods for current signals are generally not suitable for the noise reduction of the injected detection current during the fault location process. At the same time, the transmission of the measurement signal in the single-phase grounding fault loop is affected by the distributed capacitance of the line, and the signal is constantly attenuated and distorted. Therefore, the current signal noise reduction method is difficult to apply in the fault location detection process. Summary of the invention

[0003] In order to solve the problems existing in the current signal noise reduction processing method, the present invention provides a weak measurement current signal noise reduction processing method, which is characterized by comprising the following steps:

[0004] S1, collects the original signal without noise;

[0005] S2, superimpose the white noise signal on the original signal to obtain the noise signal;

[0006] S3, select the structural element function and construct the filter function;

[0007] S4, inputting the noise signal into the filter function to obtain a filtered signal;

[0008] S5, superimposing the filtered signal and the noise signal in a weighted manner to obtain a noise reduction signal;

[0009] S6, adjusting the structure element function and the weight value, and obtaining different noise reduction signals according to the method described in S3 to S5;

[0010] S7, comparing the obtained different noise reduction signals with the original signal, the structure element function used by the noise reduction signal with the smallest difference from the original signal is the best structure element function, and the weighting value used by the noise reduction signal with the smallest difference from the original signal is the best weighting value;

[0011] S8, collecting the weak measurement current signal to be processed as the noise signal, taking the best structural element function obtained in S7 as the structural element function, taking the best weighting value obtained in S7 as the weighting value, and obtaining the noise reduction signal according to the method described in S3 to S5. The noise reduction signal obtained at this time is the signal of the weak measurement current signal after noise reduction processing.

[0012] Furthermore, the structural element function in S3 is one of a triangular function, a semicircular function or a sine-cosine function.

[0013] Furthermore, the method for constructing the filter function in S3 is:

[0014] L(n)=[(f)OC(g)+(f)CO(g)](n) / 2

[0015]

[0016] Among them, L(t) is the filter function, f(t) is the noise signal, and g(t) is the structure element function.

[0017] Furthermore, the method of superimposing the filtered signal and the noise signal in a weighted manner in S3 to obtain a noise reduction signal is:

[0018] J(t)=q×L(n)+(1-q)×f(n)

[0019] 0 <q<1

[0020] Among them, J(n) is the noise reduction signal, L(n) is the filter function, f(n) is the noise signal, and q is the weighting value.

[0021] Furthermore, the method for comparing the different noise reduction signals obtained in S3 with the original signal is to calculate the root mean square error between the noise reduction signal and the original signal. The smaller the root mean square error is, the smaller the difference between the noise reduction signal and the original signal is.

[0022] Furthermore, the root mean square error calculation method between the noise reduction signal and the original signal is:

[0023]

[0024] Among them, RAME is the root mean square error between the denoised signal and the original signal, f0(t) is the original signal, f1(t) is the denoised signal, and N is the number of data points of the denoised signal and the original signal in the sampling period.

[0025] The weak measurement current signal noise reduction processing method of the present invention adopts the weak signal noise reduction processing technology of adaptive morphological filtering. Selecting a suitable structural element to perform an opening operation on the signal to be processed can eliminate the peak noise in the signal, and performing a closing operation on the signal can eliminate the valley noise in the signal. Combining the two alternating filter functions and averaging them can form an alternating mixed filter function with better filtering effect, which can better solve the problem of output offset. At the same time, in the process of noise reduction, the filtered signal and the noise signal are superimposed in a weighted manner, and the weight proportion of the filtered signal in the calculation result is continuously adjusted to obtain the best noise reduction signal. The weak measurement current signal noise reduction processing method provided by the present invention can achieve the best noise reduction signal under a complex noise background, and can quickly identify and extract the useful weak measurement current signal in the distribution network during the fault location process. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The drawings in the specification, which constitute a part of this application, are used to provide a further understanding of the present invention. The illustrative embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0027] Figure 1 It is a schematic flow chart of the method for noise reduction of weak measurement current signal of the present invention. DETAILED DESCRIPTION

[0028] In order to deepen the understanding of the present invention, the present invention will be further described in detail below with reference to embodiments and drawings.

[0029] like Figure 1 As shown, the method for noise reduction of weak measured current signal in this embodiment includes the following steps:

[0030] S1, collects the original signal without noise;

[0031] S2, superimpose the white noise signal on the original signal to obtain the noise signal;

[0032] S3, select the structural element function and construct the filter function;

[0033] S4, inputting the noise signal into the filter function to obtain a filtered signal;

[0034] S5, superimposing the filtered signal and the noise signal in a weighted manner to obtain a noise reduction signal;

[0035] S6, adjusting the structure element function and the weight value, and obtaining different noise reduction signals according to the method described in S3 to S5;

[0036] S7, comparing the obtained different noise reduction signals with the original signal, the structure element function used by the noise reduction signal with the smallest difference from the original signal is the best structure element function, and the weighting value used by the noise reduction signal with the smallest difference from the original signal is the best weighting value;

[0037] S8, collecting the weak measurement current signal to be processed as the noise signal, taking the best structural element function obtained in S7 as the structural element function, taking the best weighting value obtained in S7 as the weighting value, and obtaining the noise reduction signal according to the method described in S3 to S5. The noise reduction signal obtained at this time is the signal of the weak measurement current signal after noise reduction processing.

[0038] Furthermore, the structural element function in S3 is one of a triangular function, a semicircular function or a sine-cosine function.

[0039] Furthermore, the method for constructing the filter function in S3 is:

[0040] L(n)=[(f)OC(g)+(f)CO(g)](n) / 2

[0041]

[0042]

[0043] Among them, L(t) is the filter function, f(t) is the noise signal, and g(t) is the structure element function.

[0044] Furthermore, the method of superimposing the filtered signal and the noise signal in a weighted manner in S3 to obtain a noise reduction signal is:

[0045] J(t)=q×L(n)+(1-q)×f(n)

[0046] 0 <q<1

[0047] Among them, J(n) is the noise reduction signal, L(n) is the filter function, f(n) is the noise signal, and q is the weighting value.

[0048] Furthermore, the method for comparing the different noise reduction signals obtained in S3 with the original signal is to calculate the root mean square error between the noise reduction signal and the original signal. The smaller the root mean square error is, the smaller the difference between the noise reduction signal and the original signal is.

[0049] Furthermore, the root mean square error calculation method between the noise reduction signal and the original signal is:

[0050]

[0051] Among them, RAME is the root mean square error between the denoised signal and the original signal, f0(t) is the original signal, f1(t) is the denoised signal, and N is the number of data points of the denoised signal and the original signal in the sampling period.

[0052] The above are only specific embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for noise reduction of weak measurement current signal, It is characterized in that The following steps are included: S1, collects the original signal without noise; S2, superimpose the white noise signal on the original signal to obtain the noise signal; S3, select the structural element function and construct the filter function; S4, inputting the noise signal into the filter function to obtain a filtered signal; S5, superimposing the filtered signal and the noise signal in a weighted manner to obtain a noise reduction signal; S6, adjusting the structure element function and the weight value, and obtaining different noise reduction signals according to the method described in S3 to S5; S7, comparing the obtained different noise reduction signals with the original signal, the structure element function used by the noise reduction signal with the smallest difference from the original signal is the best structure element function, and the weighting value used by the noise reduction signal with the smallest difference from the original signal is the best weighting value; S8, collecting the weak measurement current signal to be processed as the noise signal, taking the best structural element function obtained in S7 as the structural element function, taking the best weighting value obtained in S7 as the weighting value, and obtaining the noise reduction signal according to the method described in S3 to S5. The noise reduction signal obtained at this time is the signal of the weak measurement current signal after noise reduction processing.

2. The method for noise reduction of weak measurement current signal according to claim 1, It is characterized in that The structural element function in S3 is one of a triangular function, a semicircular function or a sine-cosine function.

3. The method for noise reduction of weak measurement current signal according to claim 1, It is characterized in that The method for constructing the filter function in S3 is L(n)=[(f)OC(g)+(f)CO(g)](n) / 2 Among them, L(t) is the filter function, f(t) is the noise signal, and g(t) is the structure element function.

4. The method for noise reduction of weak measurement current signal according to claim 1, It is characterized in that The method of superimposing the filtered signal and the noise signal in a weighted manner in S3 to obtain the noise reduction signal is J(t)=q×L(n)+(1-q)×f(n) 0<q<1 Among them, J(n) is the noise reduction signal, L(n) is the filter function, f(n) is the noise signal, and q is the weighting value.

5. The method for noise reduction of weak measurement current signal according to claim 1, It is characterized in that The method for comparing the different noise reduction signals obtained in S3 with the original signal is to calculate the root mean square error between the noise reduction signal and the original signal. The smaller the root mean square error is, the smaller the difference between the noise reduction signal and the original signal is.

6. The method for noise reduction of weak measurement current signal according to claim 5, It is characterized in that The root mean square error calculation method between the noise reduction signal and the original signal is: Among them, RAME is the root mean square error between the denoised signal and the original signal, f 0 (t) is the original signal, f 1 (t) is the denoised signal, and N is the number of data points of the denoised signal and the original signal in the sampling period.