Small current grounding detection method based on empirical wavelet transform and D-S evidence theory
By combining empirical wavelet transform and DS evidence theory, and using zero-sequence current and voltage to calculate wavelet energy and probability, the problem of accuracy and complexity in single-phase grounding fault detection in low-current grounding systems is solved, achieving efficient and low-cost fault identification.
Patent Information
- Application Number
- CN202411071609.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-06
- Publication Date
- 2024-11-19
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
In existing low-current grounding systems, the accuracy of single-phase grounding fault detection is low, and existing methods are computationally complex and costly, making it difficult to accurately identify faulty lines in complex environments.
By employing empirical wavelet transform and DS evidence theory, the energy and cumulative energy of the inherent empirical function of the wavelet are calculated by collecting the zero-sequence current and zero-sequence voltage of the line. Combined with DS theory, the orthogonality and probability are calculated to identify the faulty line.
It improves the accuracy of single-phase grounding fault detection, reduces computational complexity and equipment costs, has anti-interference capabilities, and can accurately identify faulty lines in complex environments.
Smart Images

Figure CN118980966A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical fields of power distribution network, small current grounding detection and the like, and in particular to a small current grounding detection method based on empirical wavelet transform and DS evidence theory. Background Art
[0002] According to statistics, in the low-current grounding system, single-phase grounding faults account for more than 80% of the distribution network faults. At present, the operation mode of the low-current grounding system is generally applied to the medium and low voltage distribution network. Among the faults that occur, the single-phase grounding fault has the highest frequency. The main characteristics of this type of fault are small current and complex grounding conditions when the fault occurs. Therefore, it is of great significance to the stability of power system operation to accurately identify the fault line when a fault occurs.
[0003] In the grounding detection of small current grounding system, there are zero-sequence current amplitude comparison method, zero-sequence admittance method, etc., which use the relationship between zero-sequence voltage and current as the judgment criterion. In these small current grounding detection methods, there are many electrical quantities to be detected and the calculation process is complicated. There are many inconveniences in the specific implementation and application process, and the single acquisition detection equipment requires a high additional cost. Summary of the invention
[0004] In view of the defects and shortcomings of the prior art, the purpose of the present invention is to provide a small current grounding detection method based on empirical wavelet transform and DS evidence theory to improve the accuracy of grounding fault detection.
[0005] Consider that DS theory (Dempster-Shafer) is a mathematical theory for dealing with uncertainty and incomplete information. Unlike Bayesian probability, DS theory does not require a priori probability to be set in advance, and can be directly reasoned based on current evidence. It can handle problems without complete information and allows uncertainty to be expressed by assigning trust. On this basis, the present invention proposes a small current grounding detection method based on empirical wavelet transform and DS evidence theory. It only needs to detect the zero-sequence current and zero-sequence voltage of the line, and calculate the IMF energy of each line in a suitable time window T0 and normalize it. In the same time window, the cumulative energy of each line is calculated and normalized. Finally, the orthogonal sum probability of the two indicators is calculated using DS theory, and the line with an orthogonal sum probability greater than 0.5 on each line is a fault line. The present invention improves the accuracy of fault identification by applying this technology to single-phase grounding detection of small current grounding systems in distribution networks.
[0006] This scheme uses a transformer to collect the zero-sequence current I0 and zero-sequence voltage U0 of the line; when the zero-sequence voltage U0 of the line is greater than the starting value U set, select a suitable time window T0 to perform wavelet empirical mode decomposition on the transient zero-sequence current of each line, decompose it into 5 layers of wavelet inherent empirical functions (IMF), calculate the energy value of the wavelet inherent empirical function of each line, and normalize the energy values of the wavelet inherent empirical functions of all lines; in the same time window, calculate the cumulative energy value of each line, and normalize the cumulative energy values of all lines; for each line, use DS theory to calculate the energy orthogonality and probability Y of the cumulative energy and IMF; when Y>Y set , the line is faulty, otherwise the line is normal. This method is less affected by noise and can accurately select the faulty line.
[0007] The present invention specifically adopts the following technical solutions:
[0008] A small current grounding detection method based on empirical wavelet transform and DS evidence theory uses a transformer to collect the zero-sequence current I0 and zero-sequence voltage U0 of the line; when the zero-sequence voltage U0 of the line is greater than the starting value U set When , it is judged that there is a line fault: the transient zero-sequence current of each line is decomposed by wavelet empirical mode decomposition using time window T0 to obtain a 5-layer wavelet intrinsic empirical function IMF, the energy value of the wavelet intrinsic empirical function of each line is calculated, and the energy values of the wavelet intrinsic empirical functions of all lines are normalized; within the same time window, the cumulative energy value of each line is calculated, and the cumulative energy values of all lines are normalized; for each line, the DS theory is used to calculate the energy orthogonality and probability Y of the cumulative energy and IMF; when Y is greater than the threshold Y set , the line is faulty, otherwise this line is normal.
[0009] Furthermore, the starting voltage U set The value is 0.1U N -0.2U N , U N is the rated voltage, T0 is a quarter of the cycle after the fault occurs, the length is 0.005s, and the threshold Y set The value of is 0.5.
[0010] Furthermore, when the zero-sequence voltage U0 < starting value U set When , it is determined that all lines are normal lines.
[0011] Furthermore, let any signal function be f(t), and the k-layer empirical mode intrinsic function IMF after the empirical mode function decomposition be f k (t), k = 1, 2, 3, ...;
[0012]
[0013] Among them, the symbol * is the convolution operation, is the approximation coefficient, is the detail coefficient, and ψ k (t) are respectively given by the empirical scaling function With the empirical wavelet function ψ k (ω) is obtained by inverse Fourier transform.
[0014] Furthermore, the approximation coefficient By empirical scaling function The inner product with the signal f(t) is generated, and the expression is:
[0015]
[0016] Detail coefficient From the empirical wavelet function ψ n The inner product with the signal f(t) is generated, and the expression is:
[0017]
[0018] F -1 (·) is the inverse Fourier transform.
[0019] Furthermore, for n>0, the empirical wavelet function ψ n (ω) and the empirical scaling function Respectively expressed as:
[0020]
[0021] In the formula,
[0022] λ(x)=x 4 (35-84x+70x 2 -20x 3 )
[0023] ω n :τ n =γω n
[0024]
[0025] in,
[0026] Perform Fourier transform on the original signal f(t), take [0,π] as the support interval of the Fourier spectrum, and the interval is ω n Divided into N consecutive segments, 2τ n ω n The interval length of the frequency band.
[0027] Furthermore, the zero-sequence voltage measured on line i is decomposed by j-layer wavelet empirical decomposition to obtain ωn The maximum value c of the interval length of the frequency band ij As follows:
[0028]
[0029] The energy calculation formula of the j-th layer inherent empirical function of the i-th line is:
[0030]
[0031] The total IMF energy of the 5-layer wavelet empirical decomposition of the i-th line is:
[0032]
[0033] Calculate the IMF energy sum of all K lines in the system:
[0034]
[0035] Normalize the IMF energy of the i-th line:
[0036]
[0037] The energy W accumulated by the i-th line in the time window T0 i for,
[0038] W i =P*T0=3U0I0cosθ1T0
[0039] Where θ1 is the initial phase angle of the fault, U0 is the zero-sequence voltage, and I0 is the zero-sequence current.
[0040] Furthermore, the cumulative energy of all K lines in the system is calculated as:
[0041]
[0042] Normalize the energy of the i-th line:
[0043]
[0044] The occurrence of a single-phase grounding fault in the DS theory is recorded as event Θ, which refers to the occurrence of a single-phase grounding fault in the Kth line, and is recorded as Θ = {L(1), L(2), .., L(K)};
[0045] The basic trust assignment function m defined on the recognition framework corresponding to event Θ is a 2 Θ →[0,1], satisfies the following conditions:
[0046]
[0047] In the formula, m(C) is the trust value of the sub-event, The trust value of the empty event;
[0048] Assume that m1 and m2 are trust values proposed based on different reliability criteria, and the sub-events in the calculation process of m1 are {A1, A2, ..., A n}, the sub-events in the calculation process of m2 are {B1,B2,...,B n}; Different reliability is calculated based on the joint orthogonality and probability Y:
[0049]
[0050] Where K is the conflict coefficient, It requires that the reliabilities not required for the empty set be assigned to the non-empty set to satisfy the reliability assignment function.
[0051] Furthermore, the orthogonal sum probability Y of the two evaluation indicators of IMF energy and cumulative energy is calculated on each line; when Y>Y set When it is detected, it is a faulty line; otherwise, it is a normal line.
[0052] And, a small current grounding system adopts the above method to perform small current grounding detection.
[0053] The present invention and its preferred solution can realize reliable identification of ground fault lines by simply collecting zero-sequence voltage and zero-sequence current, performing simple processing, and combining DS theoretical calculation, and have strong anti-interference capability, and can accurately identify fault lines in complex environments. BRIEF DESCRIPTION OF THE DRAWINGS
[0054] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments:
[0055] Figure 1 It is a schematic diagram of the overall process of an embodiment of the present invention;
[0056] Figure 2 Schematic diagram of the PSCAD / EMTDC simulation model according to an embodiment of the present invention. DETAILED DESCRIPTION
[0057] Hereinafter, the specific embodiments of the present application will be described in detail with reference to the accompanying drawings. According to these detailed descriptions, those skilled in the art can clearly understand the present application and can implement the present application. Without violating the principles of the present application, the features in the various embodiments can be combined to obtain new implementations, or certain features in certain embodiments can be replaced to obtain other preferred implementations.
[0058] The technical solution of the present invention is described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations on the technical solution of the present invention. The embodiments of the present invention and the technical features in the embodiments may be combined with each other unless there is a conflict.
[0059] It should be understood that when used in this specification and the appended claims, the terms "include" and "comprises" indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.
[0060] It should also be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.
[0061] The term "and / or" is only a description of the association relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " generally indicates that the related objects are in an "or" relationship.
[0062] In order to make the features and advantages of this patent more obvious and easy to understand, the following embodiments are specifically described in detail with the accompanying drawings as follows:
[0063] Example 1
[0064] This embodiment introduces a small current grounding detection method based on empirical wavelet transform and DS evidence theory. Its most basic design includes:
[0065] Use transformers to collect the zero-sequence current I0 and zero-sequence voltage U0 of the line;
[0066] When the zero-sequence voltage U0 of the line is greater than the starting value U set , determine whether there is a line fault, and for each line, use the DS theory to calculate the cumulative energy and the energy orthogonal to the IMF and the probability Y. When Y>Y set , the line is faulty, otherwise this line is normal.
[0067] Example 2
[0068] Based on Example 1, this example provides the following preferred specific design:
[0069] When the zero-sequence voltage U0 of the line is less than the starting value U set, the line is a normal line.
[0070] Comparison K r And protection start setting K set Size, when K r ≤K set , the line is a normal line.
[0071] Select the time window as T0, perform wavelet empirical mode decomposition on the zero-sequence current of each line, and decompose it into 5 layers of wavelet intrinsic empirical functions. set The value is 0.1U N -0.2U N , U N is the rated voltage, T0 is a quarter of the cycle after the fault occurs and the length is 0.005s, Y set =0.5.
[0072] Assume that any signal function is f(t), and the k-layer empirical mode intrinsic function IMF after empirical mode function decomposition is f k (t)(k=1,2,3,...).
[0073]
[0074] Among them, the symbol * is the convolution operation, is the approximation coefficient, is the detail coefficient, and ψ k (t) is With ψ k (ω) is obtained by inverse Fourier transform.
[0075] Approximation coefficient By empirical scaling function The inner product with the signal f(t) is generated, and the expression is:
[0076]
[0077] Detail coefficient From the empirical wavelet function ψ n The inner product with the signal f(t) is generated, and the expression is:
[0078]
[0079] F -1 (·) is the inverse Fourier transform.
[0080] For n>0, the empirical wavelet function ψ n (ω) and the scaling function It is expressed as:
[0081]
[0082]
[0083] In the formula,
[0084] λ(x)=x 4 (35-84x+70x 2 -20x 3 )
[0085] ω n :τ n =γω n
[0086]
[0087] in,
[0088] Perform Fourier transform on the original signal f(t), take [0,π] as the support interval of the Fourier spectrum, and the interval is ω n Divided into N consecutive segments (N∈N + ), 2τ n ω n The interval length of the frequency band.
[0089] For the zero-sequence voltage measured on line i, perform j-layer wavelet empirical decomposition to obtain ω n The maximum value c of the interval length of the frequency band ij As follows:
[0090]
[0091] The energy calculation formula of the j-th layer inherent empirical function of the i-th line is:
[0092]
[0093] The total IMF energy of the ith line is (5-layer wavelet empirical decomposition):
[0094]
[0095] Calculate the IMF energy sum of all lines in the system (K lines):
[0096]
[0097] Normalize the IMF energy of the i-th line:
[0098]
[0099] The energy W accumulated by the i-th line in the time window T0 i for,
[0100] W i =P*T0=3U0I0cosθ1T0
[0101] Where θ1 is the initial phase angle of the fault, U0 is the zero-sequence voltage, and I0 is the zero-sequence current.
[0102] Calculate the cumulative energy of all lines in the system (K lines):
[0103]
[0104] Normalize the energy of the i-th line:
[0105]
[0106] Example 3
[0107] Based on the calculation of the IMF energy index and the cumulative energy and index in Example 2, this embodiment provides a specific solution for calculating the energy orthogonality of the cumulative energy and the IMF and the probability Y for fault discrimination. First, the occurrence of a line single-phase grounding fault in the DS theory is recorded as event Θ, and its content is that the occurrence of a single-phase grounding fault on the Kth line is recorded as Θ = {L(1), L(2), .., L(K)}.
[0108] The basic trust assignment function m defined on the recognition framework Θ is a 2 Θ →[0,1], the function satisfies the following conditions:
[0109]
[0110] In the formula, m(C) is the trust value of the sub-event, The confidence value of the null event.
[0111] Assume that m1 and m2 are trust values proposed based on different reliability criteria, and the sub-events in the calculation process of m1 are {A1, A2, ..., A n}, the sub-events in the calculation process of m2 are {B1,B2,...,B n}. Then the different reliability is calculated based on the joint orthogonal sum probability Y:
[0112]
[0113] Where K is the conflict coefficient, It requires that the reliabilities not required for the empty set be assigned to the non-empty set to satisfy the reliability assignment function.
[0114] like Figure 1As shown, a typical implementation process of the solution of the present invention is provided by combining the above embodiments 1-3 as follows: collecting the zero-sequence current I0 and zero-sequence voltage U0 of the line;
[0115] Step (1): When the zero-sequence voltage U0 of the line is detected to be greater than the starting value U set When the fault occurs, it is determined that a line fault has occurred, and the fault line selection algorithm steps (2) to (6) are entered;
[0116] Step (2): Select the time window as T0, perform wavelet empirical mode decomposition on the zero-sequence current of each line, and decompose it into 5 layers of wavelet intrinsic empirical functions;
[0117] Step (3): Calculate the energy value of the wavelet intrinsic empirical function of each line, and normalize the IMF energy values of the wavelet intrinsic empirical functions of all lines;
[0118] Step (4): within the same time window T0, calculate the cumulative energy value of each line, and normalize the cumulative energy values of all lines;
[0119] Step (5): For each line, use DS theory to calculate the energy orthogonality and probability Y of the cumulative energy and IMF;
[0120] Step (6): When Y>0.5, the line is faulty, otherwise the line is normal.
[0121] Example 4
[0122] This embodiment uses PSCAD / EMTDC to simulate the system and build a simulation model. Figure 2 As shown. The simulation model has five feeders with lengths of 8km, 12km, 20km, 15km and 16km respectively. The remaining parameters are set as follows: the system power supply voltage is 110kV, the frequency is 50Hz, the power supply grounding resistance is 10Ω, the bus neutral point is grounded through a 0.3H inductor, the grounding fault is a single-phase grounding fault of phase A, the short-circuit point is 6km away from the bus at feeder 1, and the grounding resistance R of the short-circuit point is 10Ω. The line is an overhead line and cable, and its setting parameters are shown in Table 1.
[0123] By the methods of Examples 1 to 3, the interval is taken as ω n It is divided into 50 continuous segments, and the IMF energy values of feeders 1 to 5 are calculated. The accumulated energy and orthogonal sum probability Y are calculated. The faulty line is identified as feeder 1 through calculation. The calculation results are shown in Table 2, which verifies the feasibility of the detection method.
[0124] The results of the algorithm calculation are shown in the following table:
[0125] Table 1 Feeder parameters
[0126]
[0127] Table 2 Fault line identification results
[0128] Feeder No. Energy accumulation and IMF Energy Orthogonality and probability Y 1 0.8362 0.8541 0.9729 2 0.0495 0.0574 0.0043 3 0.0846 0.0861 0.0057 4 0.0840 0.0715 0..0153 5 0.0756 0.0697 0.0029
[0129] It can be seen from Table 2 that when the grounding resistance in the small current grounding system is 10Ω, the algorithm calculates the orthogonal probability and Y of each line. The orthogonal probability and Y of feeder 1>Yset, and the algorithm reliably and accurately identifies the faulty line.
[0130] In summary, the present invention proposes a new protection criterion based on the relationship between IMF energy and energy accumulation, which can reliably and accurately identify the faulty line.
[0131] 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.
[0132] The above is a preferred embodiment of the present invention. It should be pointed out that, for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.
[0133] This patent is not limited to the above-mentioned optimal implementation mode. Anyone can derive various other forms of a small current grounding detection method based on empirical wavelet transform and DS evidence theory under the inspiration of this patent. All equal changes and modifications made according to the scope of the patent application of the present invention should be covered by this patent.
Claims
1. A small current grounding detection method based on empirical wavelet transform and DS evidence theory, characterized by: Use transformers to collect the zero-sequence current I0 and zero-sequence voltage U0 of the line; when the zero-sequence voltage U0 of the line is greater than the starting value U set When , it is judged that there is a line fault: the transient zero-sequence current of each line is decomposed by wavelet empirical mode decomposition using time window T0 to obtain a 5-layer wavelet intrinsic empirical function IMF, the energy value of the wavelet intrinsic empirical function of each line is calculated, and the energy values of the wavelet intrinsic empirical functions of all lines are normalized; within the same time window, the cumulative energy value of each line is calculated, and the cumulative energy values of all lines are normalized; for each line, the DS theory is used to calculate the energy orthogonality and probability Y of the cumulative energy and IMF; when Y is greater than the threshold Y set , the line is faulty, otherwise this line is normal.
2. The small current grounding detection method based on empirical wavelet transform and DS evidence theory according to claim 1 is characterized in that: Starting voltage U set The value is 0.1U N -0.2U N , U N is the rated voltage, T0 is a quarter of the cycle after the fault occurs, the length is 0.005s, and the threshold Y set The value of is 0.
5.
3. The small current grounding detection method based on empirical wavelet transform and DS evidence theory according to claim 1 is characterized in that: When the zero sequence voltage U0 < starting value U set When , it is determined that all lines are normal lines.
4. The small current grounding detection method based on empirical wavelet transform and DS evidence theory according to claim 1 is characterized in that: Assume that any signal function is f(t), and the k-layer empirical mode intrinsic function IMF after the empirical mode function decomposition is f k (t), k = 1, 2, 3, ...; Among them, the symbol * is the convolution operation, is the approximation coefficient, is the detail coefficient, and ψ k (t) are respectively given by the empirical scaling function With the empirical wavelet function ψ k (ω) is obtained by inverse Fourier transform.
5. The small current grounding detection method based on empirical wavelet transform and DS evidence theory according to claim 4 is characterized in that: The approximation coefficient By empirical scaling function The inner product with the signal f(t) is generated, and the expression is: Detail coefficient From the empirical wavelet function ψ n The inner product with the signal f(t) is generated, and the expression is: F -1 (·) is the inverse Fourier transform.
6. The small current grounding detection method based on empirical wavelet transform and DS evidence theory according to claim 4 is characterized in that: For n>0, the empirical wavelet function ψ n (ω) and the empirical scaling function Respectively expressed as: In the formula, λ(x)=x 4 (35-84x+70x 2 -20x 3 ) oh n :t n =I n in, Perform Fourier transform on the original signal f(t), take [0,π] as the support interval of the Fourier spectrum, and the interval is ω n Divided into N consecutive segments, 2τ n ω n The interval length of the frequency band.
7. The small current grounding detection method based on empirical wavelet transform and DS evidence theory according to claim 4 is characterized by: Perform j-layer wavelet empirical decomposition on the zero-sequence voltage measured on line i and obtain ω n The maximum value c of the interval length of the frequency band ij As follows: The energy calculation formula of the j-th layer inherent empirical function of the i-th line is: The total IMF energy of the 5-layer wavelet empirical decomposition of the i-th line is: Calculate the IMF energy sum of all K lines in the system: Normalize the IMF energy of the i-th line: The energy W accumulated by the i-th line in the time window T0 i for, W i =P*T0=3U0I0cosθ1T0 Where θ1 is the initial phase angle of the fault, U0 is the zero-sequence voltage, and I0 is the zero-sequence current.
8. The small current grounding detection method based on empirical wavelet transform and DS evidence theory according to claim 7 is characterized in that: The cumulative energy of all K lines in the calculation system is: Normalize the energy of the i-th line: The occurrence of a single-phase grounding fault in the DS theory is recorded as event Θ, which refers to the occurrence of a single-phase grounding fault in the Kth line, and is recorded as Θ = {L(1), L(2), .., L(K)}; The basic trust assignment function m defined on the recognition framework corresponding to event Θ is a 2 Θ →[0,1], satisfies the following conditions: In the formula, m(C) is the trust value of the sub-event, The trust value of the empty event; Assume that m1 and m2 are trust values proposed based on different reliability criteria, and the sub-events in the calculation process of m1 are {A1, A2, ..., A n }, the sub-events in the calculation process of m2 are {B1,B2,...,B n }; Different reliability is calculated based on the joint orthogonality and probability Y: Where K is the conflict coefficient, It requires that the reliabilities not required for the empty set be assigned to the non-empty set to satisfy the reliability assignment function.
9. The small current grounding detection method based on empirical wavelet transform and DS evidence theory according to claim 8 is characterized in that: Calculate the orthogonal sum probability Y of the two evaluation indicators of IMF energy and cumulative energy on each line; when Y>Y set When it is detected, it is a faulty line; otherwise, it is a normal line.
10. A low current grounding system, characterized in that: A small current grounding detection is performed using the method according to any one of claims 1 to 9.
Citation Information
Patent Citations
Distribution network line selection method without tuning parameter
CN109188192A
Single-phase-to-ground fault line selection method for small current
CN109884464A
Small current grounding fault line selection method based on stable transient information of evidence combination
CN113640615A
Method for locating distribution network circuit fault based on full waveform information
US20160061873A1
