DC power signal denoising method based on ENEMD-NWT

Through the ENEMD-NWT method, combined with Fourier transform and nonlinear wavelet threshold method, the problems of low denoising accuracy and poor robustness of DC signal are solved, and efficient denoising of DC power signals is achieved, which improves the denoising of power signals. Accuracy and efficiency.

CN119003995AInactive Publication Date: 2024-11-22STATE GRID SICHUAN ELECTRIC POWER CO MARKETING SERVICE CENT +1

Patent Information

Application Number
CN202411471954.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-22
Publication Date
2024-11-22
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing DC signal denoising methods have low accuracy, slow speed and poor robustness when processing non-stationary signals. Especially in DC microgrids, multi-source and multi-transform characteristics lead to ripple generation, affecting the power quality and billing accuracy.

Method used

The ENEMD-NWT-based method is adopted to obtain the main frequency of the noise signal through Fourier transform, and combine the integrated noise reconstruction empirical mode decomposition and nonlinear wavelet threshold method to denoise the eigenmodal function to achieve accurate decomposition of the signal and Refactor.

Benefits of technology

It improves the accuracy and efficiency of DC power signal denoising, avoids the problem of modal aliasing, adapts to signal extraction under strong noise interference conditions, and enhances the denoising ability of the power signal.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119003995A_ABST
    Figure CN119003995A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of electric power signal processing, and discloses a direct current electric power signal denoising method based on ENEMD-NWT, which comprises the following steps: performing Fourier transform on a collected direct current electric power signal to obtain the main frequency of a noise signal; decomposing the collected direct current power signal into an intrinsic mode function by adopting an integrated noise reconstruction empirical mode decomposition method; de-noising the intrinsic mode function by adopting a nonlinear wavelet threshold method according to the main frequency of the noise signal; and reconstructing the denoised intrinsic mode function to obtain a denoised direct current power signal. According to the invention, the extraction of the DC power useful signal under the strong noise interference condition can be realized, and the DC power signal denoising precision and efficiency are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of power signal processing, and in particular to a direct current power signal denoising method based on ENEMD-NWT. Background Art

[0002] In the process of development, the new power system has shown the characteristics of "double high", that is, a high proportion of new energy access and a high proportion of power electronic equipment. At the same time, in new energy scenarios such as distributed photovoltaics and charging piles, the application of DC technology is becoming more and more extensive. With the advantages of low transmission line cost and high efficiency, DC microgrid has become a rising star in the low-voltage power grid. However, the multi-source and multi-conversion characteristics in the DC microgrid will lead to various forms of ripple sources in the system. The presence of a large amount of ripple not only directly leads to the decline of the power quality of the DC microgrid, but also reduces the accuracy of electricity billing. It is crucial to achieve reliable power supply of DC microgrids, detection and filtering of noise such as ripple in DC power signals, and improve the accuracy of DC metering.

[0003] There are many existing DC signal denoising methods, such as fast Fourier transform (FFT), wavelet transform, empirical mode decomposition (EMD), variational mode decomposition (VMD) algorithm, etc. However, when performing signal detection or feature extraction, if one of the above algorithms is used alone, it may be limited by the shortcomings of a single algorithm. For example, FFT and its improved algorithms cannot adapt to non-stationary signals, the wavelet basis of the wavelet transform method is difficult to determine, the EMD algorithm has a mode mixing effect, and the determination of the optimal decomposition layer and penalty factor of the VMD algorithm is subjective.

[0004] Therefore, it is urgent to design a DC power signal denoising method with high accuracy, fast speed and high robustness. Summary of the invention

[0005] In view of the above-mentioned deficiencies in the prior art, the present invention provides a DC power signal denoising method based on ENEMD-NWT.

[0006] In order to achieve the above-mentioned object of the invention, the technical solution adopted by the present invention is: A DC power signal denoising method based on ENEMD-NWT comprises the following steps: Performing Fourier transform on the collected DC power signal to obtain the main frequency of the noise signal; The collected DC power signal is decomposed into intrinsic mode functions using the integrated noise reconstruction empirical mode decomposition method; According to the main frequency of the noise signal, the nonlinear wavelet threshold method is used to denoise the intrinsic mode function; The denoised intrinsic mode function is reconstructed to obtain the denoised DC power signal.

[0007] Preferably, using an integrated noise reconstruction empirical mode decomposition method to decompose the collected DC power signal into an intrinsic mode function includes: Performing empirical mode decomposition on the collected DC power signal to obtain a first intrinsic mode function and a first residual component; Calculate the energy estimate for the first eigenmode function and calculate the confidence interval of the energy of the first eigenmode function at 95% and 99% confidence levels; The pure noise eigenmode function is selected based on the energy of the first eigenmode function and its confidence interval at 95% and 99% confidence levels; The pure noise intrinsic mode function is screened by using hard threshold to obtain the estimated noise; Construct new noise by randomly permuting the estimated noise, and reconstruct the observed signal based on the new noise; Performing empirical mode decomposition on the observed signal to obtain the second intrinsic mode function and the second residual component; The random permutation step is repeated until the preset stopping criterion of the error construction is reached, and the means of the second intrinsic mode function and the second residual component are calculated according to the number of repetitions to obtain the final intrinsic mode function and residual component.

[0008] Preferably, the calculation formula for calculating the energy estimation of the first eigenmode function is:

[0009] in, is the energy of the first eigenmode function, N is the length of the first eigenmode function component, is the sampling point of the first eigenmode function component, k is the component number, i is the sampling point number.

[0010] Preferably, the hard threshold is:

[0011] in, Set the threshold, To set the constant, is the energy of the pure noise intrinsic mode function component, l is the serial number of the pure noise intrinsic mode function component, N is the length of the first eigenmode function component.

[0012] Preferably, the observation signal is reconstructed according to the new noise as follows:

[0013] in, To observe the signal, is the estimated noise-free signal, For new noise, m is the noise number, t is the sampling time.

[0014] Preferably, the mean of the second intrinsic mode function and the second residual component is calculated according to the number of repetitions, and the final intrinsic mode function and residual component are obtained as follows:

[0015] in, is the final eigenmode function, is the final residual component, is the second eigenmode function component, is the second residual component, m is the noise number, n is the residual component number, r is the number of repetitions.

[0016] Preferably, according to the main frequency of the noise signal, the nonlinear wavelet threshold method is used to denoise the intrinsic mode function, which includes: Calculate the correlation coefficient and center frequency of each component of the eigenmode function; According to the correlation coefficient of each component of the eigenmode function and the absolute value of the deviation between the center frequency and the main frequency of the noise signal, the pure noise eigenmode function component, the pure eigenmode function component and the noisy eigenmode function component are determined; The nonlinear wavelet threshold method is used to denoise the noisy IMF components, remove the pure noise IMF components, and retain the pure IMF components.

[0017] Preferably, the nonlinear wavelet threshold method is used to denoise the noisy intrinsic mode function component, including: The noisy IMF components are transformed by using db4 wavelet and three-layer decomposition to obtain the wavelet coefficients corresponding to each decomposition layer.

[0018] Preferably, the nonlinear wavelet threshold method is used to denoise the noisy intrinsic mode function component, and the wavelet threshold function is specifically:

[0019] in, are the estimated coefficients for real signals, is the wavelet transform coefficient of the noisy observation signal, is the threshold value, j To decompose the scale, k Component number.

[0020] Preferably, reconstructing the denoised intrinsic mode function to obtain the denoised DC power signal includes: The denoised DC power signal is obtained by reconstructing the denoised noisy eigenmode function component and the pure eigenmode function component in the eigenmode function.

[0021] The present invention has the following beneficial effects: When denoising a DC power signal, the present invention first uses an integrated noise reconstruction empirical mode decomposition method to screen the integrated average intrinsic mode function by measuring the inherent noise components in the signal, eliminates unnecessary noise components in the signal, and realizes initial signal denoising. In this way, it not only inherits the adaptive multi-modal decomposition of the empirical mode decomposition method for non-stationary and nonlinear processes, but also avoids the modal aliasing problem of the traditional empirical mode decomposition method; and before signal reconstruction, the nonlinear wavelet threshold method is used to denoise the noise intrinsic mode function component, thereby realizing the extraction of useful DC power signals under strong noise interference conditions, and improving the denoising accuracy and efficiency of DC power signals. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 The figure is a flow chart of a DC power signal denoising method based on ENEMD-NWT. DETAILED DESCRIPTION

[0023] The specific implementation modes of the present invention are described below so that those skilled in the art can understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation modes. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the attached claims, these changes are obvious, and all inventions and creations utilizing the concept of the present invention are protected.

[0024] like Figure 1 As shown, the embodiment of the present invention provides a DC power signal denoising method based on ENEMD-NWT, comprising the following steps S1 to S4: S1. Perform Fourier transform on the collected DC power signal to obtain the main frequency of the noise signal; S2, decomposing the collected DC power signal into intrinsic mode functions using an integrated noise reconstruction empirical mode decomposition method; In an optional embodiment of the present invention, step S2 uses an integrated noise reconstruction empirical mode decomposition method to decompose the collected DC power signal into intrinsic mode functions, including: Performing empirical mode decomposition on the collected DC power signal to obtain a first intrinsic mode function and a first residual component; Calculate the energy estimate for the first eigenmode function and calculate the confidence interval of the energy of the first eigenmode function at 95% and 99% confidence levels; The pure noise eigenmode function is selected based on the energy of the first eigenmode function and its confidence interval at 95% and 99% confidence levels; The pure noise intrinsic mode function is screened by using hard threshold to obtain the estimated noise; Construct new noise by randomly permuting the estimated noise, and reconstruct the observed signal based on the new noise; Performing empirical mode decomposition on the observed signal to obtain the second intrinsic mode function and the second residual component; The random permutation step is repeated until the preset stopping criterion of the error construction is reached, and the means of the second intrinsic mode function and the second residual component are calculated according to the number of repetitions to obtain the final intrinsic mode function and residual component.

[0025] Exemplarily, in this embodiment, the original signal x ( t )= s ( t )+ n ( t ) is used for empirical mode decomposition (EMD), and the decomposed IMFs and residual components are recorded as and .in s ( t ) is the noise-free component, n ( t )for x ( t ) noise component.

[0026] This embodiment calculates the energy estimate of the first eigenmode function, and the calculation formula is:

[0027] in, is the energy of the first eigenmode function, N is the length of the first eigenmode function component, is the sampling point of the first eigenmode function component, k is the component number, i is the sampling point number.

[0028] This embodiment calculates the energy of the first eigenmode function The confidence intervals at the 95% and 99% confidence levels are expressed as: and .

[0029] In this embodiment, the energy of the first eigenmode function Instead of comparing with 95% and 99% IMFs, pure noise IMFs are selected and recorded as c l , that is, satisfying the following The corresponding c k ,in α Determined manually based on noise reduction requirements.

[0030]

[0031] The samples below the threshold are retained as the desired noise components. c l ( i )yes c l ( t ) i sampling points and defined as:

[0032] The threshold is manually specified Defined as , Set the threshold, To set the constant, is the energy of the pure noise intrinsic mode function component, l is the serial number of the pure noise intrinsic mode function component, N is the length of the first eigenmode function component.

[0033] Calculate the estimated noise for:

[0034] This embodiment estimates noise by random permutation Construct the mth noise signal , and then reconstruct the observed signal , specifically ;in represents the estimated noise-free signal.

[0035] This embodiment reconstructs the observed signal Perform EMD to obtain the decomposed IMFs and the residual component .

[0036] This embodiment repeats the random permutation step until the preset stop criterion of the error construction is reached, and the total number of repetitions is recorded as r. The final result is output by calculating the ensemble mean of IMFs and residuals:

[0037] in, is the final eigenmode function, is the final residual component, is the second eigenmode function component, is the second residual component, m is the noise number, n is the residual component number, r is the number of repetitions.

[0038] S3, according to the main frequency of the noise signal, the eigenmode function is denoised using a nonlinear wavelet threshold method; In an optional embodiment of the present invention, step S3 calculates the mean of the second intrinsic mode function and the second residual component according to the number of repetitions, and obtains the final intrinsic mode function and residual component as follows: Preferably, according to the main frequency of the noise signal, the nonlinear wavelet threshold method is used to denoise the intrinsic mode function, which includes: Calculate the correlation coefficient and center frequency of each component of the eigenmode function; According to the correlation coefficient of each component of the eigenmode function and the absolute value of the deviation between the center frequency and the main frequency of the noise signal, the pure noise eigenmode function component, the pure eigenmode function component and the noisy eigenmode function component are determined; The nonlinear wavelet threshold method is used to denoise the noisy IMF components, remove the pure noise IMF components, and retain the pure IMF components.

[0039] Exemplarily, this embodiment calculates the correlation coefficient and center frequency of each IMF component , according to the correlation coefficient of each IMF component and the center frequency The absolute value of the deviation Determine the pure noise IMF component, pure IMF component and noisy IMF component. The value of the correlation coefficient represents the correlation between the IMF component and the signal. The more noise contained in the component, the lower the correlation. The correlation value of the noise component is 0. The IMF component with a slight correlation is defined as the noisy IMF component. The minimum value is taken. That is, the IMF component whose center frequency is closest to the main frequency is regarded as the pure IMF component.

[0040] This embodiment retains the pure IMF component, removes the pure noise IMF component, and uses the nonlinear wavelet threshold method (NWT) to perform overall denoising on the noisy IMF component, specifically: The noisy IMF components are transformed by using db4 wavelet and three-layer decomposition to obtain the wavelet coefficients corresponding to each decomposition layer.

[0041] Select the threshold function. The wavelet threshold function used by NWT is:

[0042] in, are the estimated coefficients for real signals, is the wavelet transform coefficient of the noisy observation signal, is the threshold value, j To decompose the scale, k Component number.

[0043] The wavelet coefficients smaller than the threshold are set to zero, and the wavelet coefficients larger than the threshold are retained, and the IMF is retained and the wavelet coefficients are set to zero. .

[0044] S4. Reconstruct the denoised intrinsic mode function to obtain a denoised DC power signal.

[0045] In an optional embodiment of the present invention, step S4 reconstructs the denoised intrinsic mode function to obtain a denoised DC power signal including: The inverse wavelet transform method is used to reconstruct the denoised noisy intrinsic mode function components and pure intrinsic mode function components in the intrinsic mode function to obtain the denoised DC power signal.

[0046] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0047] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0048] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0049] The present invention uses specific embodiments to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

[0050] Those skilled in the art will appreciate that the embodiments described herein are intended to help readers understand the principles of the present invention, and should be understood that the protection scope of the present invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific variations and combinations that do not deviate from the essence of the present invention based on the technical revelations disclosed by the present invention, and these variations and combinations are still within the protection scope of the present invention.

Claims

1. A DC power signal denoising method based on ENEMD-NWT, characterized in that: The following steps are involved: Performing Fourier transform on the collected DC power signal to obtain the main frequency of the noise signal; The collected DC power signal is decomposed into intrinsic mode functions using the integrated noise reconstruction empirical mode decomposition method; According to the main frequency of the noise signal, the nonlinear wavelet threshold method is used to denoise the intrinsic mode function; The denoised intrinsic mode function is reconstructed to obtain the denoised DC power signal.

2. A DC power signal denoising method based on ENEMD-NWT according to claim 1, characterized in that: The integrated noise reconstruction empirical mode decomposition method is used to decompose the collected DC power signal into intrinsic mode functions including: Performing empirical mode decomposition on the collected DC power signal to obtain a first intrinsic mode function and a first residual component; Calculate the energy estimate for the first eigenmode function and calculate the confidence interval of the energy of the first eigenmode function at 95% and 99% confidence levels; The pure noise eigenmode function is selected based on the energy of the first eigenmode function and its confidence interval at 95% and 99% confidence levels; The pure noise intrinsic mode function is screened by using hard threshold to obtain the estimated noise; Construct new noise by randomly permuting the estimated noise, and reconstruct the observed signal based on the new noise; Performing empirical mode decomposition on the observed signal to obtain the second intrinsic mode function and the second residual component; The random permutation step is repeated until the preset stopping criterion of the error construction is reached, and the means of the second intrinsic mode function and the second residual component are calculated according to the number of repetitions to obtain the final intrinsic mode function and residual component.

3. A DC power signal denoising method based on ENEMD-NWT according to claim 2, characterized in that: The calculation formula for calculating the energy estimate of the first eigenmode function is: in, is the energy of the first eigenmode function, N is the length of the first eigenmode function component, is the sampling point of the first eigenmode function component, k is the component number, i is the sampling point number.

4. A DC power signal denoising method based on ENEMD-NWT according to claim 2, characterized in that: The specific hard threshold is: in, Set the threshold, To set the constant, is the energy of the pure noise intrinsic mode function component, l is the serial number of the pure noise intrinsic mode function component, N is the length of the first eigenmode function component.

5. The method for denoising a DC power signal based on ENEMD-NWT according to claim 2, characterized in that: Reconstructing the observed signal based on the new noise is specifically: in, To observe the signal, is the estimated noise-free signal, For new noise, m is the noise number, t is the sampling time.

6. A DC power signal denoising method based on ENEMD-NWT according to claim 2, characterized in that: The mean of the second eigenmode function and the second residual component is calculated according to the number of repetitions, and the final eigenmode function and residual component are obtained as follows: in, is the final eigenmode function, is the final residual component, is the second eigenmode function component, is the second residual component, m is the noise number, n is the residual component number, r is the number of repetitions.

7. A DC power signal denoising method based on ENEMD-NWT according to claim 1, characterized in that: According to the main frequency of the noise signal, the nonlinear wavelet threshold method is used to denoise the intrinsic mode function, including: Calculate the correlation coefficient and center frequency of each component of the eigenmode function; According to the correlation coefficient of each component of the eigenmode function and the absolute value of the deviation between the center frequency and the main frequency of the noise signal, the pure noise eigenmode function component, the pure eigenmode function component and the noisy eigenmode function component are determined; The nonlinear wavelet threshold method is used to denoise the noisy IMF components, remove the pure noise IMF components, and retain the pure IMF components.

8. A DC power signal denoising method based on ENEMD-NWT according to claim 7, characterized in that: The nonlinear wavelet threshold method is used to denoise the noisy intrinsic mode function components, including: The noisy IMF components are transformed by using db4 wavelet and three-layer decomposition to obtain the wavelet coefficients corresponding to each decomposition layer.

9. The method for denoising a DC power signal based on ENEMD-NWT according to claim 7, characterized in that: The nonlinear wavelet threshold method is used to denoise the noisy intrinsic mode function component. The specific wavelet threshold function is: in, are the estimated coefficients for real signals, is the wavelet transform coefficient of the noisy observation signal, is the threshold value, j To decompose the scale, k Component number.

10. The method for denoising a DC power signal based on ENEMD-NWT according to claim 1, characterized in that: The denoised intrinsic mode function is reconstructed to obtain the denoised DC power signal including: The denoised DC power signal is obtained by reconstructing the denoised noisy eigenmode function component and the pure eigenmode function component in the eigenmode function.

Citation Information

Patent Citations

  • Ensemble noise-reconstructed EMD (empirical mode decomposition) method for early and compound faults of machinery

    CN103226649A

Cited By

  • Geological disaster intelligent monitoring system and method based on multiple sensors

    CN119625965A