A method for fault detection in a flexible direct current power transmission system

Abstract

The application provides a flexible direct current power transmission system fault detection method, comprising the following steps: obtaining an original current signal; setting a search range for a parameter of a variational mode decomposition, thereby obtaining an original setting parameter; decomposing the original current signal into mode functions with different center frequencies; selecting a value corresponding to a minimum relative entropy as an optimal parameter; performing a final variational mode decomposition on the original current signal by using the optimal parameter, thereby obtaining final mode functions in an ordered arrangement; selecting a second final mode function from the final mode functions; performing a Hilbert-Huang transform on the second final mode function, thereby obtaining a Hilbert spectrum of the second final mode function; performing a synchronous compression transform on the Hilbert spectrum, thereby obtaining an instantaneous energy density of the Hilbert spectrum; and comparing an amplitude of the instantaneous energy density with a steady-state threshold, thereby realizing fault detection. The application can realize accurate and rapid fault detection on a flexible direct current power transmission system.

Description

Technical Field

[0001] This invention relates to the technical field of circuit system testing, and more specifically to a fault detection method for a flexible DC transmission system. Background Technology

[0002] Flexible DC transmission systems have advantages such as long transmission distance, large energy transmission capacity, low line loss, and flexible control. They are of great significance for the large-scale absorption of renewable energy and the rational allocation of resources, and therefore have received widespread attention from researchers in recent years.

[0003] However, flexible DC transmission systems are prone to faults during commissioning and operation, which can generate extremely large fault transient currents. These fault transient currents can reach their peak value within milliseconds, causing a huge impact on system equipment in a short period of time. This defect severely limits its development.

[0004] Because flexible DC transmission lines span long distances, operate in complex terrain environments, and have a high short-circuit fault rate, accurate and rapid fault detection is of great significance for improving the reliability of DC systems. Summary of the Invention

[0005] In view of this, the present invention provides a fault detection method for flexible DC transmission systems to achieve accurate and rapid fault detection in flexible DC transmission systems.

[0006] A fault detection method for a flexible DC transmission system includes:

[0007] Step S1: The DC line is sampled in real time by a current sensor to obtain the raw current signal, and the collected raw current signal is input into the window.

[0008] Step S2: Set the search range for the parameters of variational mode decomposition to obtain the original set parameters, including the number of modes. and penalty factor ;

[0009] Step S3: Apply variational mode decomposition to the original current signal using the original set parameters, thereby decomposing the original current signal into... Mode functions with different center frequencies;

[0010] Step S4, The reconstructed signal is obtained by adding the mode functions with different center frequencies. The relative entropy between the original current signal and the reconstructed signal is calculated based on kernel density estimation. Each pair within the search range is traversed. Value, select the value corresponding to the minimum relative entropy The value is used as the optimal parameter;

[0011] Step S5: Perform final variational mode decomposition on the original current signal using optimal parameters to obtain... Given a set of ordered final mode functions, select the second final mode function and perform a Hilbert-Huang transform on it to obtain its Hilbert spectrum.

[0012] Step S6: Perform synchronous compression transformation on the Hilbert spectrum of the second final mode function to remove the fuzzy energy caused by noise, harmonics or modulation sidebands, thereby obtaining the instantaneous energy density of the Hilbert spectrum. Compare the amplitude of the instantaneous energy density with the steady-state threshold to achieve fault detection.

[0013] The fault detection method for flexible DC transmission systems provided by the present invention has the following beneficial effects:

[0014] (1) In flexible DC transmission systems, the DC fault detection time should be limited to within 2ms to reliably protect the entire system and converter components. However, the training process of artificial intelligence algorithms is time-consuming, and model inference may involve complex calculations. This is an unbearable burden for the speed requirement of fault detection in flexible DC systems that emphasize millisecond-level response. This invention only requires fast variational mode decomposition, has low computational complexity, and can complete the mode number calculation in real time or near real time. The settings meet the speed requirements of fault detection.

[0015] (2) This invention can automatically adapt the most suitable decomposition parameters for different lines, different operating states, and different fault types, making it more versatile. It ensures that the difference between the reconstructed signal and the original signal is minimized, guaranteeing that the decomposed mode function can truly and completely retain all the features of the original signal, avoiding information loss or distortion. In addition, the second final mode function obtained by variational mode decomposition of the original current signal is used for fault detection. Since the second final mode function oscillates at the frequency of important characteristics, it can reduce noise interference to fault detection and improve the accuracy of fault detection.

[0016] (3) This invention introduces synchronous compression transformation, which recompresses and concentrates the diffused energy onto the true instantaneous frequency ridge, making the clear and brief impact energy band generated by the fault stand out, greatly improving the ability to identify weak fault features in complex noise environments. Moreover, this invention uses the amplitude of instantaneous energy density as the basis for fault detection, emphasizing the fault features within the selected frequency range, and can effectively suppress the influence of steady-state components. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the fault detection method for a flexible DC transmission system provided in an embodiment of the present invention.

[0018] Figure 2 Here is an example waveform of a raw current signal;

[0019] Figure 3 Here is an example of the instantaneous energy density response of a faulty circuit;

[0020] Figure 4 Here is an example of the instantaneous energy density response of a non-faulty circuit;

[0021] Figure 5 The instantaneous energy density response of the second final mode function after adding Gaussian white noise with a signal-to-noise ratio of 25dB is shown.

[0022] Figure 6 The instantaneous energy density response of the second final mode function after adding Gaussian white noise with a signal-to-noise ratio of 55dB is shown. Detailed Implementation

[0023] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain embodiments of the present invention, and should not be construed as limiting the present invention.

[0024] Please see Figure 1 The present invention provides a fault detection method for a flexible DC transmission system, comprising steps S1 to S6:

[0025] Step S1: The DC line is sampled in real time by a current sensor to obtain the raw current signal, and the collected raw current signal is input into the window.

[0026] Specifically, an exemplary waveform of the original current signal is as follows: Figure 2 As shown, in this embodiment, step S1 satisfies the following formula:

[0027]

[0028] in, Indicates length is One-dimensional discrete-time signal sequence, This represents the first sampling point of the sampled discrete-time signal of the current. This represents the second sampling point of the sampled discrete-time signal of the current. The first digit of the sampled discrete-time current signal represents the... One sampling point.

[0029] Step S2: Set the search range for the parameters of variational mode decomposition to obtain the original set parameters, including the number of modes. and penalty factor .

[0030] For example, based on prior knowledge, the modal number The search range is 2 to 10, with a penalty factor. The search range is 1000 to 5000.

[0031] Step S3: Apply variational mode decomposition to the original current signal using the original set parameters, thereby decomposing the original current signal into... Mode functions with different center frequencies.

[0032] Specifically, step S3 includes:

[0033] Step S3.1, based on the original current signal, obtain the length as... A one-dimensional discrete-time signal sequence, using the original set parameters, is used to process a length of... Variational mode decomposition is performed on a one-dimensional discrete-time signal sequence, and a mode bandwidth weighting factor is introduced to obtain a constrained variational problem;

[0034] In step S3.1, the expression of the constrained variational problem for:

[0035]

[0036] in, This indicates taking the minimum value. The first one to be decomposed One modal function, It is the first The center frequency of each modal function This represents the modal bandwidth weighting factor. Indicates time The partial derivatives, Represents the Dirac function, Represents the imaginary unit. This represents the convolution operation. This represents the square of the L2 norm.

[0037] Step S3.2: Obtain the extended Lagrange expression by introducing the Lagrange penalty operator, and transform the constrained variational problem into an unconstrained variational problem based on the extended Lagrange expression;

[0038] In step S3.2, the extended Lagrange expression is:

[0039]

[0040] in, It is an extended Lagrange expression. It is the Lagrange penalty operator. It is the original current signal. Indicates the inner product.

[0041] Step S3.3: Perform iterative calculations on the unconstrained variational problem to obtain the optimal solution. Update the mode function based on the optimal solution to output the solution. Mode functions with different center frequencies.

[0042] In step S3.3, when updating the mode function, a relaxation factor is introduced to predict the changing trend of the Lagrange penalty operator through extrapolation, thereby accelerating the update of the Lagrange penalty operator. The expression is:

[0043]

[0044] in, It is the updated Lagrange penalty operator. It is the Lagrange penalty operator before the update. It is a relaxation factor. This represents the noise tolerance parameter. It is the Fourier transform of the original current signal. Indicates the updated number Fourier transform of the modal functions.

[0045] Step S4, The reconstructed signal is obtained by adding the mode functions with different center frequencies. The relative entropy between the original current signal and the reconstructed signal is calculated based on kernel density estimation. Each pair within the search range is traversed. Value, select the value corresponding to the minimum relative entropy The value is used as the optimal parameter.

[0046] Among them, the reconstructed signal Satisfy the following formula:

[0047]

[0048] in, , , , These are the first mode function, the second mode function, and the third mode function, respectively. The modal function, the first One modal function.

[0049] The formula for calculating relative entropy is:

[0050]

[0051] in, The relative entropy of the original current signal and the reconstructed signal. This represents the kernel density estimate of the original current signal. This represents the kernel density estimate of the reconstructed signal. Indicates using Distribution differences Indicates using Distribution The differences.

[0052] Step S5: Perform final variational mode decomposition on the original current signal using optimal parameters to obtain... Given a set of ordered final mode functions, select the second final mode function and perform a Hilbert-Huang transform on it to obtain its Hilbert spectrum.

[0053] In this embodiment, the second final mode function is selected because the second final mode function oscillates at a frequency with important characteristics, which can reduce noise interference in fault detection and thus improve the accuracy of fault detection.

[0054] Step S6: Perform synchronous compression transformation on the Hilbert spectrum of the second final mode function to remove the fuzzy energy caused by noise, harmonics or modulation sidebands, thereby obtaining the instantaneous energy density of the Hilbert spectrum. Compare the amplitude of the instantaneous energy density with the steady-state threshold to achieve fault detection.

[0055] Specifically, step S6 includes:

[0056] Step S6.1: Perform synchronous compression transformation on the Hilbert spectrum of the second final mode function, using the Dirac function as a selector to compress the Hilbert spectrum into a time-frequency energy ridge, and obtain a reconstructed spectrum.

[0057] Wherein, step S6.1 satisfies the following formula:

[0058]

[0059]

[0060] in, For a single spectrum reconstruction, The normalization constant is For the Hilbert spectrum of the second final mode function, To preset the center frequency range, The compressed instantaneous angular frequency The differential, For instantaneous amplitude, The instantaneous angular frequency, Indicates time The differential.

[0061] Step S6.2: Remove the fuzzy energy caused by noise, harmonics, or modulation sidebands from the first reconstructed spectrum to obtain the second reconstructed spectrum;

[0062] Wherein, step S6.2 satisfies the following formula:

[0063]

[0064] in, For secondary reconstruction of the spectrum, It is an instantaneous frequency function. This represents the integral bandwidth.

[0065] Step S6.3: Calculate the instantaneous energy density based on the secondary reconstructed spectrum, and compare the amplitude of the instantaneous energy density with the steady-state threshold. If the amplitude of the instantaneous energy density exceeds the steady-state threshold, it is determined to be a fault condition.

[0066] Specifically, the instantaneous energy density response results of an exemplary faulty line and the instantaneous energy density response results of a non-faulty line are respectively as follows: Figure 3 and Figure 4 As shown, from Figure 3 and Figure 4 It can be seen that instantaneous energy density can effectively detect whether a fault condition has occurred.

[0067] Furthermore, since background noise affects the spectral distribution, existing frequency-domain-based fault detection methods are susceptible to noise interference. This invention evaluates the robustness of the proposed method to noise by adding Gaussian white noise with different signal-to-noise ratios to the DC line current signal under varying load conditions. After adding Gaussian white noise with a signal-to-noise ratio of 25 dB, the instantaneous energy density of the second final mode function is as follows: Figure 5 As shown, after adding Gaussian white noise with a signal-to-noise ratio of 55dB, the instantaneous energy density of the second final mode function is as follows: Figure 6 As shown, from Figure 5 and Figure 6 It can be seen that the instantaneous energy density response of the method proposed in this invention remains below the steady-state threshold and is not affected by noise, thus exhibiting good robustness.

[0068] In summary, the fault detection method for flexible DC transmission systems according to the above embodiments has the following beneficial effects:

[0069] (1) In flexible DC transmission systems, the DC fault detection time should be limited to within 2ms to reliably protect the entire system and converter components. However, the training process of artificial intelligence algorithms is time-consuming, and model inference may involve complex calculations. This is an unbearable burden for the speed requirement of fault detection in flexible DC systems that emphasize millisecond-level response. This invention only requires fast variational mode decomposition, has low computational complexity, and can complete the mode number calculation in real time or near real time. The settings meet the speed requirements of fault detection.

[0070] (2) This invention can automatically adapt the most suitable decomposition parameters for different lines, different operating states, and different fault types, making it more versatile. It ensures that the difference between the reconstructed signal and the original signal is minimized, guaranteeing that the decomposed mode function can truly and completely retain all the features of the original signal, avoiding information loss or distortion. In addition, the second final mode function obtained by variational mode decomposition of the original current signal is used for fault detection. Since the second final mode function oscillates at the frequency of important characteristics, it can reduce noise interference to fault detection and improve the accuracy of fault detection.

[0071] (3) This invention introduces synchronous compression transformation, which recompresses and concentrates the diffused energy onto the true instantaneous frequency ridge, making the clear and brief impact energy band generated by the fault stand out, greatly improving the ability to identify weak fault features in complex noise environments. Moreover, this invention uses the amplitude of instantaneous energy density as the basis for fault detection, emphasizing the fault features within the selected frequency range, and can effectively suppress the influence of steady-state components.

[0072] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this invention patent should be determined by the appended claims.

Claims

1. A fault detection method for a flexible DC transmission system, characterized in that, include: Step S1: The DC line is sampled in real time by a current sensor to obtain the raw current signal, and the collected raw current signal is input into the window. Step S2: Set the search range for the parameters of variational mode decomposition to obtain the original set parameters, including the number of modes. and penalty factor ; Step S3: Apply variational mode decomposition to the original current signal using the original set parameters, thereby decomposing the original current signal into... Mode functions with different center frequencies; Step S4, will The reconstructed signal is obtained by adding the mode functions with different center frequencies. The relative entropy between the original current signal and the reconstructed signal is calculated based on kernel density estimation. Each pair within the search range is traversed. Value, select the value corresponding to the minimum relative entropy The value is used as the optimal parameter; Step S5: Perform final variational mode decomposition on the original current signal using optimal parameters to obtain... Given a set of ordered final mode functions, select the second final mode function and perform a Hilbert-Huang transform on it to obtain its Hilbert spectrum. Step S6: Perform synchronous compression transformation on the Hilbert spectrum of the second final mode function to remove the fuzzy energy caused by noise, harmonics or modulation sidebands, thereby obtaining the instantaneous energy density of the Hilbert spectrum. Compare the amplitude of the instantaneous energy density with the steady-state threshold to achieve fault detection.

2. The fault detection method for flexible DC transmission systems according to claim 1, characterized in that, Step S3 specifically includes: Step S3.1, based on the original current signal, obtain the length as... A one-dimensional discrete-time signal sequence, using the original set parameters, is used to process a length of... Variational mode decomposition is performed on a one-dimensional discrete-time signal sequence, and a mode bandwidth weighting factor is introduced to obtain a constrained variational problem; Step S3.2: Obtain the extended Lagrange expression by introducing the Lagrange penalty operator, and transform the constrained variational problem into an unconstrained variational problem based on the extended Lagrange expression; Step S3.3: Perform iterative calculations on the unconstrained variational problem to obtain the optimal solution. Update the mode function based on the optimal solution to output the solution. Mode functions with different center frequencies.

3. The fault detection method for flexible DC transmission systems according to claim 2, characterized in that, In step S3.1, the expression of the constrained variational problem for: in, This indicates taking the minimum value. The first one to be decomposed One modal function, It is the first The center frequency of each modal function This represents the modal bandwidth weighting factor. Indicates time The partial derivatives, Represents the Dirac function, Represents the imaginary unit. This represents the convolution operation. This represents the square of the L2 norm.

4. The fault detection method for flexible DC transmission systems according to claim 3, characterized in that, In step S3.2, the extended Lagrange expression is: in, It is an extended Lagrange expression. It is a Lagrange penalty operator. It is the original current signal. Indicates the inner product.

5. The fault detection method for a flexible DC transmission system according to claim 4, characterized in that, In step S3.3, when updating the mode function, a relaxation factor is introduced to predict the changing trend of the Lagrange penalty operator through extrapolation, thereby accelerating the update of the Lagrange penalty operator. The expression is: in, It is the updated Lagrange penalty operator. It is the Lagrange penalty operator before the update. It is a relaxation factor. This represents the noise tolerance parameter. It is the Fourier transform of the original current signal. Indicates the updated number Fourier transform of the modal functions.

6. The fault detection method for a flexible DC transmission system according to claim 5, characterized in that, In step S4, the formula for calculating relative entropy is: in, The relative entropy of the original current signal and the reconstructed signal. This represents the kernel density estimate of the original current signal. This represents the kernel density estimate of the reconstructed signal. Indicates using Distribution differences Indicates using Distribution The differences.

7. The fault detection method for a flexible DC transmission system according to claim 6, characterized in that, Step S6 specifically includes: Step S6.1: Perform synchronous compression transformation on the Hilbert spectrum of the second final mode function, using the Dirac function as a selector to compress the Hilbert spectrum into a time-frequency energy ridge, and obtain a reconstructed spectrum. Step S6.2: Remove the fuzzy energy caused by noise, harmonics, or modulation sidebands from the first reconstructed spectrum to obtain the second reconstructed spectrum; Step S6.3: Calculate the instantaneous energy density based on the secondary reconstructed spectrum, and compare the amplitude of the instantaneous energy density with the steady-state threshold. If the amplitude of the instantaneous energy density exceeds the steady-state threshold, it is determined to be a fault condition.

8. The fault detection method for a flexible DC transmission system according to claim 7, characterized in that, Step S6.1 satisfies the following formula: in, For a single spectrum reconstruction, The normalization constant is For the Hilbert spectrum of the second final mode function, To preset the center frequency range, The compressed instantaneous angular frequency The differential, For instantaneous amplitude, The instantaneous angular frequency, Indicates time The differential.

9. The fault detection method for a flexible DC transmission system according to claim 8, characterized in that, Step S6.2 satisfies the following formula: in, For secondary reconstruction of the spectrum, It is an instantaneous frequency function. This represents the integral bandwidth.

Images