Low-current grounding system fault section positioning method based on transient zero-sequence current fuzzy entropy waveform similarity
By using transient zero-sequence current fuzzy entropy waveform similarity and CEEMDAN modal decomposition in the distribution network, combined with the fuzzy entropy waveform deletion characteristics, the problem of noise and grounding resistance interference in the small current grounding system is solved, and the fault is achieved The accurate positioning of the section is suitable for complex distribution network structures.
Patent Information
- Application Number
- CN202410963067.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-18
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2044-07-18
AI Technical Summary
The existing distribution network fault positioning method is susceptible to noise and grounding resistance in small current grounding systems, which affects positioning accuracy. Especially when multiple branch lines and ground capacitance increase, it is difficult to accurately locate the fault section. .
The fault segment positioning method of small current grounding system based on the similarity of the fuzzy entropy waveform of transient zero-sequence current is adopted. By building a distribution network model, transient zero-sequence current signals are collected, CEEMDAN modal decomposition is performed, and fuzzy entropy value and chaos are calculated. The degree, combined with the initial polarity characteristics of the fuzzy entropy waveform, double criterion analysis is performed to determine the fault segment.
This method can accurately identify fault segments under the influence of noise, sampling out-synchronization and grounding resistance. It is suitable for complex distribution network structures, improving the accuracy and adaptability of fault positioning.
Smart Images

Figure CN118937889A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of fault location of power distribution network lines, and in particular relates to a method for locating a fault section of a small current grounding system based on transient zero-sequence current fuzzy entropy waveform similarity. Background Art
[0002] As the end of the power system, the stability and reliability of the distribution network are crucial to the safe operation of the entire power system. In the distribution network, small current grounding fault is a common fault type, and its accurate and rapid location is of great significance for timely restoration of power supply and reduction of economic losses. However, the existing fault section location method has certain limitations.
[0003] At present, the fault section location methods for distribution network are mainly divided into matrix discrimination method based on fault overcurrent information, artificial intelligence method and grounding fault location method based on zero-sequence quantity characteristics.
[0004] However, the matrix discrimination method based on fault overcurrent information has high requirements on the number of detection devices and synchronization, has a large amount of calculation under complex topologies, and has poor adaptability; although the artificial intelligence method has low requirements on the number of detection devices, when the distribution network structure changes, the algorithm training time is long and the adaptability is poor; the waveform comparison based on zero-sequence current characteristics mainly includes entropy algorithm, polarity comparison method, frequency method, etc. Among them, the entropy algorithm has certain applications because it is suitable for processing nonlinear signals such as zero-sequence current, such as approximate entropy, energy relative entropy and sample entropy, but it will cause threshold failure and failure to locate the fault section under the influence of noise, grounding resistance, and the small difference in ground capacitance on both sides of the fault point.
[0005] In summary, the shortcomings of the existing technology are: 1. For a small current grounding system, the zero-sequence current signal is easily interfered by noise and grounding resistance, affecting the accuracy of fault section location. 2. When the proposed entropy algorithm is used to process the situation where there are multiple branch lines downstream of the fault point and the ground capacitance increases, the entropy value calculation results at both ends of the fault section may be close, making it impossible to accurately locate the fault section. Summary of the invention
[0006] In order to solve the above technical problems, the present invention proposes a method for locating the fault section of a small current grounding system based on the similarity of transient zero-sequence current fuzzy entropy waveforms to solve the problems existing in the above prior art.
[0007] To achieve the above object, the present invention provides a method for locating a fault section of a small current grounding system based on transient zero-sequence current fuzzy entropy waveform similarity, comprising:
[0008] Building a distribution network model of a small current grounding system, and collecting transient zero-sequence current signals of each section on the fault line in the distribution network model;
[0009] Performing CEEMDAN modal decomposition on the transient zero-sequence current signal to obtain a high-frequency imf1 component;
[0010] Calculate the fuzzy entropy value of the imf1 component, and define a chaos value as a first criterion for preliminary screening of fault sections, wherein the chaos value is used to measure the waveform similarity of transient zero-sequence current; based on the first criterion, obtain a first criterion result;
[0011] If the chaos ratio is less than a preset threshold, the second criterion based on the initial polarity of the fuzzy entropy waveform is used to analyze the waveform polarity to obtain a second criterion result;
[0012] When the first judgment criterion result is consistent with the second judgment criterion result, the fault section is located.
[0013] Preferably, the distribution network model includes a Y-connected transformer and an arc-extinguishing coil with a compensation degree, and the fault line in the distribution network model is divided into sections by a detection device to obtain a plurality of sections.
[0014] Preferably, the calculation process of calculating the fuzzy entropy value of the imf1 component includes:
[0015] Step 1: Select the input high-frequency transient zero-sequence current signal as follows:
[0016] {i(t s ),s=1,2,···,N};
[0017] Step 2: Using a sequence of length m as the time window, divide the transient zero-sequence current signal into k = N-m + 1 sequences
[0018]
[0019] in, I(t s ) m The mean value of
[0020]
[0021] Step 3: Calculate the vector I(t i ) m and I(t j ) m The distance is the Chebyshev distance, that is, the maximum absolute value of the numerical difference of each element, that is:
[0022]
[0023] Step 4: Introduce fuzzy membership n metric I(ti ) m , I(t j ) m The similarity of , for 1≤j≤Nm,j≠i, is:
[0024]
[0025] Where n and r are the boundary gradient and width;
[0026] Step 5: Average all memberships except itself:
[0027]
[0028] Step 6: Increase the dimension to m+1 and repeat steps 2 to 5;
[0029] Step 7: Calculate the fuzzy entropy FuzzyEn(m,r,n) of the signal sequence as:
[0030] FuzzyEn(m,r,n)=lnφ m,r,n -lnφ m+1,r,n .
[0031] Preferably, the formula for calculating the fuzzy entropy disorder value of the imf1 component is:
[0032]
[0033] Among them, F(n) is the chaos degree of the nth measurement point, N is the number of sampling points, and f i is the fuzzy entropy value of the i-th sampling point.
[0034] Preferably, the process of obtaining the first criterion result includes:
[0035] Calculate the chaos values of the measuring points at both ends of the line segment respectively, and compare the chaos values of the measuring points at both ends to obtain the chaos ratio;
[0036] The chaos ratio is judged. If the chaos ratio is greater than or equal to the preset threshold, the output is a fault section, which is the result of the first judgment; if the chaos ratio is less than the preset threshold, the second judgment is used for analysis.
[0037] Preferably, the formula of the first criterion is:
[0038]
[0039] Among them, F (i) With F (j) For line section L ij The chaos value of the measuring points at both ends, point i is the measuring point close to one end of the busbar, and point j is the measuring point close to the end of the line.
[0040] Preferably, the process of the second criterion includes:
[0041] The polarity characteristics of the high-frequency part IMF1 of the transient zero-sequence current are extracted using the symmetrical Sigmoid function to obtain the zero-sequence current waveform with polarity characteristics;
[0042] The fuzzy entropy values of the sampling points are weighted to process the fuzzy entropy waveform, so as to obtain a fuzzy entropy waveform curve reflecting the initial polarity characteristics of the high-frequency part IMF1.
[0043] Preferably, the formula of the second criterion includes:
[0044] Diff(i)=I(i+1)-I(i),i∈1,2,…,N-1;
[0045]
[0046] Diff2(i)=Trend(i+1)-Trend(i);
[0047] Among them, I is the fuzzy entropy value sequence, DiffI is the first-order difference vector, and Diff2(i) is the second-order difference sequence.
[0048] The present invention also provides an electronic device, comprising: a processor and a memory storing computer program instructions; when the processor executes the computer program instructions, the method for locating a fault section of a small current grounding system based on fuzzy entropy dual criteria is implemented.
[0049] Compared with the prior art, the present invention has the following advantages and technical effects:
[0050] The present invention provides a method for locating a fault section of a small current grounding system based on transient zero-sequence current fuzzy entropy waveform similarity, comprising: building a distribution network model of a small current grounding system, collecting transient zero-sequence current signals of various sections on a fault line in the distribution network model; performing CEEMDAN modal decomposition on the transient zero-sequence current signal to obtain an imf1 component of a high-frequency part; calculating a fuzzy entropy value of the imf1 component, and defining a chaos value as a first criterion for preliminarily screening a fault section; obtaining a first criterion result based on the first criterion; if the chaos ratio is less than a preset threshold, analyzing the waveform polarity using a second criterion based on an initial polarity of a fuzzy entropy waveform to obtain a second criterion result; and obtaining the location of the fault section when the first criterion result is consistent with the second criterion result.
[0051] The present invention utilizes the characteristics of zero-sequence current when a single-phase grounding fault occurs in a small current grounding system of a distribution network, and effectively extracts the high-frequency part containing the zero-sequence current characteristics through CEEMDAN, thereby overcoming the problem of modal aliasing in traditional EMD, and being able to provide accurate IMF1 high-frequency part, reducing the influence of arc suppression coil on transient zero-mode current; secondly, a method for locating the fault section of a small current grounding system based on the similarity of transient zero-sequence current fuzzy entropy waveforms, first uses the first criterion to quickly determine the fault section, and then uses the second criterion to further determine the situation when the capacitance to the ground on both sides is similar, which is suitable for distribution network structures with complex downstream branch lines and a small number of outgoing lines. The present invention can accurately locate the fault section. In addition, the algorithm of the present invention is less affected by sampling asynchrony, noise, and grounding resistance, and can better adapt to actual operating conditions. BRIEF DESCRIPTION OF THE DRAWINGS
[0052] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:
[0053] Figure 1 A schematic diagram of a single-phase grounding model of a power distribution network according to an embodiment of the present invention;
[0054] Figure 2 This is a flow chart of fault section location according to an embodiment of the present invention;
[0055] Figure 3 A schematic diagram of a fuzzy entropy waveform according to an embodiment of the present invention;
[0056] Figure 4 This is a schematic diagram of a transient zero-sequence current waveform of a fault with a delay of 10ms according to an embodiment of the present invention;
[0057] Figure 5 It is a schematic diagram of the chaos ratio between the fault section and the non-fault section at different grounding resistances according to an embodiment of the present invention. DETAILED DESCRIPTION
[0058] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.
[0059] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.
[0060] Embodiment 1
[0061] This embodiment provides a method for locating a fault section of a small current grounding system based on transient zero-sequence current fuzzy entropy waveform similarity, including:
[0062] S1. Build a distribution network model of a small current grounding system, and collect zero-sequence current signals of each section on the fault line in the distribution network model;
[0063] Specifically, PSCAD / EMTDC was used to build the following Figure 1 In the distribution network model of the small current grounding system shown in the figure, the transformer adopts the YΔ connection mode, the arc suppression coil compensation degree adopts p=8%, and the arc suppression coil inductance is calculated by formula (1).
[0064]
[0065] Among them C Σ is the total capacitance of one-phase line to ground in the system, and the arc suppression coil inductance value L is obtained. k =16.6864H,S i (i=1,…,11) is each detection device FTU on the fault line. The line is divided into sections by the detection device. The sampling frequency is set to 200kHz. The line parameters are shown in Table 1 below.
[0066] Table 1
[0067]
[0068]
[0069] S2, performing CEEMDAN modal decomposition on the transient zero-sequence current signal to obtain a high-frequency imf1 component;
[0070] The decomposition process of the transient zero-sequence current signal includes the following specific calculation steps:
[0071] (1) Add positive and negative white noise (-1) to the original transient zero-sequence current signal s(t) q a0n i (t) get the new signal: s(t)+(-1) q a0n i (t), perform N EMD decompositions on the new signal to obtain N first-order intrinsic mode components imf1 i (t):
[0072] s(t)+(-1) q a0n i (t) = imf1 i (t)+r1 i (t) (2)
[0073] Among them, ni (t) is a Gaussian white noise signal that satisfies the normal distribution, i=1,2,···,N, N is the number of times white noise is added, a0 is the amplitude of white noise, q=1,2, r1 i (t) is the first-order residual.
[0074] (2) The N generated modal components are averaged to obtain the first intrinsic modal component of CEEMDAN decomposition.
[0075]
[0076] Calculate the remaining components after removing the first modal component:
[0077]
[0078] (3) Add positive and negative paired Gaussian white noise to r1(t) to obtain a new signal, and use the new signal as the carrier to perform EMD decomposition to obtain N second-order IMF components
[0079]
[0080] Similarly, for N Taking the mean, we get the second eigenmode component:
[0081]
[0082] The remaining components after removing the second modal component:
[0083]
[0084] (4) Repeat (2) to (3) above until the remaining component r k (t) is a monotonic function and cannot be further decomposed. Finally, K IMF components are obtained. Then the original zero-sequence current signal s(t) can be expressed as:
[0085]
[0086] Where K is the number of eigenmode decompositions, and R(t) is the final remaining residual.
[0087] S3, calculate the fuzzy entropy value of the imf1 component. The calculation process of the fuzzy entropy value is as follows:
[0088] From the above analysis, it can be seen that the transient zero-sequence current can be regarded as a nonlinear and non-stationary signal. Fuzzy entropy (FurryEn) as a method to describe the complexity of the signal time series can reflect the resonance characteristics of the zero-sequence current signal. Its principle is to introduce a fuzzy membership function to describe the degree to which the zero-sequence current signal belongs to a fuzzy set at a certain sampling moment. The value range of the membership function is between 0 and 1. The entropy of the fuzzy set reflects the degree of confusion of the information contained in the signal. The higher the resonance frequency of the zero-sequence current signal, the greater the amplitude change, and the faster the attenuation degree changes, the greater the corresponding entropy value. The specific zero-sequence current entropy value calculation process is as follows:
[0089] (1) The high-frequency transient zero-sequence current signal selected as input is as follows:
[0090] {i(t s ),s=1,2,···,N} (9)
[0091] (2) Using a sequence of length m as the time window, the transient zero-sequence current signal is divided into k = N - m + 1 sequences
[0092]
[0093] in I(t s ) m The mean value of
[0094]
[0095] (3) Calculate the vector I(t i ) m and I(t j ) m The distance is the Chebyshev distance, that is, the maximum absolute value of the numerical difference of each element, that is:
[0096]
[0097] (4) Introducing the fuzzy membership n metric I(t i ) m , I(t j ) m Similarity. For 1≤j≤Nm,j≠i, we have:
[0098]
[0099] Where n and r are the boundary gradient and width.
[0100] (5) Average all memberships except itself:
[0101]
[0102] (6) Increase the dimension to m+1 and repeat steps (2) to (5);
[0103] (7) Calculate the fuzzy entropy FuzzyEn(m,r,n) of the signal sequence as:
[0104] FuzzyEn(m,r,n)=lnφ m,r,n -lnφ m+1,r,n (15)
[0105] S4, defining a chaos value as a first criterion for preliminary screening of fault sections; and obtaining a first criterion result based on the first criterion;
[0106] Specifically, the signal complexity of zero-sequence current on the fault line is close on the same side of the fault point, while the complexity on both sides of the fault point is quite different. The high-frequency part of the transient zero-sequence current is input, the fuzzy entropy value of each sampling point is calculated, and then the fuzzy entropy average value is calculated using formula (16) to obtain the chaos value reflecting the signal complexity characteristics. The calculation formula is as follows:
[0107]
[0108] Where: F(n) is the chaos degree of the nth measurement point, N is the number of sampling points, f i is the fuzzy entropy value of the i-th sampling point.
[0109] The chaos degree criterion can be used to indirectly measure the waveform similarity on both sides of the line section, that is, the chaos degree is close in the non-fault section, while the chaos degree on both sides of the fault section is quite different.
[0110]
[0111] Among them, F (i) With F (j) For line section L ij The chaos values of the measuring points at both ends, where point i is the measuring point close to one end of the busbar and point j is the measuring point close to the end of the line. Considering the error in calculating the fuzzy entropy value, a 10% margin is set.
[0112] S5. If the chaos ratio is less than a preset threshold, a second criterion based on the initial polarity of the fuzzy entropy waveform is used for analysis to obtain a second criterion result;
[0113] Specifically, when the number of outgoing lines in the distribution network is small, there are multiple branch lines downstream of the fault point, and the fault location is close to the bus, the equivalent capacitance and inductance values of the upstream and downstream parts on both sides of the fault point are relatively close to each other. At this time, the transient zero-sequence resonant frequency and attenuation speed on both sides of the fault section are also relatively close. Using the first criterion fuzzy entropy value chaos to reflect the characteristics of the fault section may cause the criterion to fail and lead to failure in section positioning. However, the initial polarity of the high-frequency component of the transient zero-sequence current is not affected by this and can still be used as a theoretical basis for fault section positioning.
[0114] Sigmoid function is a common S-shaped function, which is often used as the activation function of neural networks. The specific function expression is:
[0115]
[0116] This function can map real numbers to (0, 1) and is used for binary classification. It works better when the feature differences are complex or not very large. In addition, as an activation function, the sigmoid function has the advantage of being smooth and easy to derive.
[0117] The Sigmoid function can optimize the fuzzy entropy value input polarity feature when there are discontinuities. In order not to lose the polarity feature, this paper uses the symmetric Sigmoid function, the expression is
[0118]
[0119] The function value range is (-1, 1), which can successfully reflect the concave and convex characteristics of the fuzzy entropy waveform and make the image smooth and easy to observe.
[0120] The specific steps of this criterion are: first use the symmetric Sigmoid function to extract the polarity characteristics of the high-frequency part IMF1 of the transient zero-sequence current, then use the fuzzy entropy value of the sampling point to weight the waveform, and finally obtain a fuzzy entropy waveform curve that reflects the initial polarity characteristics of the high-frequency part IMF1.
[0121] The process of extracting the initial polarity features of the fuzzy entropy waveform is as follows:
[0122] (1) Calculate the first difference vector DiffI of the fuzzy entropy value sequence I, as shown in formula (20):
[0123] Diff(i)=I(i+1)-I(i),i∈1,2,…,N-1 (20)
[0124] (2) Perform traversal operation on the first-order difference vector DiffI using the symbolic function formula (21) to obtain the vector TrendI;
[0125]
[0126] (3) Calculate the second-order difference of the fuzzy entropy sequence, perform a secondary difference operation on the vectorized first-order difference vector TrendI obtained above, and obtain a second-order difference sequence reflecting the polarity of the initial fuzzy entropy waveform, as shown in formula (22):
[0127] Diff2(i)=Trend(i+1)-Trend(i) (22)
[0128] S6. When the first criterion result is consistent with the second criterion result, the fault section is located.
[0129] Specifically, this embodiment takes advantage of the fuzzy entropy algorithm and proposes a method for locating the fault section of a small current grounding system. The flow chart is as follows: Figure 2 As shown, the specific steps are as follows:
[0130] (1) Setting the starting criterion: The instantaneous value of the neutral point voltage U0 is greater than the maximum value of the rated phase voltage U m 15% of U0, that is, U0>0.15U m When the FTU device is started, the three-phase transient current is collected;
[0131] (2) The collected zero-sequence current of the fault line is subjected to CEEMDAN modal decomposition to obtain the imf components of different central frequency bands, and then the high-frequency part imf1 is selected as the input for fuzzy entropy value calculation;
[0132] (3) Calculate the fuzzy entropy value of the high-frequency part imf1, and then substitute it into formula (16) to obtain the chaos degree of each measuring point;
[0133] (4) Using the first criterion to preliminarily screen out possible fault sections;
[0134] (5) For the screening results obtained in step (4), if the chaos ratio is greater than or equal to 1.7, it is directly output as a fault section; if it is less than 1.7, the second criterion is used for further analysis;
[0135] (6) When the output results of the two judgment criteria are consistent, the fault section is confirmed and the positioning is completed.
[0136] In order to verify the effectiveness of the algorithm proposed in this embodiment, a single-phase grounding fault is set between sections S5 and S6, the fault phase is phase A, and the initial phase angle of the fault is Transition resistance R f=10Ω, switch K is closed, and the neutral point is grounded through the arc suppression coil. The imf1 component obtained by CEEMDAN decomposition is input into the calculation of the fuzzy entropy chaos of the detection points along the fault line, and the entropy value ratio of the two detection points before and after each section is calculated based on the line section divided by each detection point. The calculation results are shown in the following table:
[0137] The calculation results of the fuzzy entropy chaos degree of the fault line are shown in Table 2. It can be seen from Table 2 that the fuzzy entropy values of the measuring points at both ends of the fault section are larger than those of the healthy section, exceeding the threshold set by the first criterion. This reflects that there is a large difference in the transient zero-sequence current signals at both ends of the fault section, while the difference in the healthy section is small.
[0138] Table 2
[0139]
[0140]
[0141] From the calculation results of the chaos value in Table 2, it can be seen that when the fault occurs far away from the bus, the chaos ratio of the fault section calculated by the first criterion is much greater than the threshold value of 1.7, and the fault section can be directly determined without starting the second criterion.
[0142] Embodiment 2
[0143] Assume that the fault occurs in the S2~S3 section. At this time, the capacitance of the upstream part of the fault point to ground is C Σs =0.1754μF (excluding the fault section), C Σx =0.1540μF (excluding the fault section). At this time, the capacitance values of the upstream and downstream parts to the ground are relatively close. The fuzzy entropy values of the measuring points at both ends of the fault section are calculated, and the degree of confusion is calculated to obtain the results shown in Table 3 below.
[0144] Table 3
[0145]
[0146] It can be seen from this that when the fault section is close to the bus side, the chaos values of the measuring points at both ends of the fault section are relatively close, and the ratio is close to the threshold, which may cause the failure of locating the fault section. Therefore, the second criterion is activated for further judgment.
[0147] The preliminary fault section obtained by the first criterion is analyzed. This paper selects the sampling points from 600 to 1200 in the fault section to calculate the fuzzy entropy value. The fuzzy entropy waveform obtained is as follows: Figure 3 shown.
[0148] Embodiment 3
[0149] (1) Impact of sampling synchronization
[0150] In the actual process of collecting transient current, FTU may be affected by the asynchronous sampling time. When the f1 fault occurs, a unified time window of 20ms is set. The sampling start time of the fixed measuring point S5 is 0ms after the fault. The S6 measuring point is sampled at different delayed times to obtain the fuzzy entropy value and calculate its chaos degree. The calculation results of the fuzzy entropy chaos degree of the fault section at different starting times are shown in Table 4:
[0151] Table 4
[0152]
[0153] It can be seen that the fuzzy entropy chaos value of S6 measurement point does not change much, the error range is small, and its chaos ratio with S5 exceeds the set threshold. Sampling asynchrony has almost no effect on the positioning result of the fault section. This is because the sampling time asynchrony will only cause the waveform to shift and will not affect its resonant frequency and attenuation speed. Figure 4 The transient zero-sequence current waveform of measuring point S6 when delayed by 10ms.
[0154] (2) Influence of grounding resistance
[0155] When the f1 fault occurs, different transition resistances are set from 0 to 700Ω, and S4 to S7 are taken as representative measuring points, where S4 to S5 and S6 to S7 are non-fault sections, and S5 to S6 is the fault section. The fuzzy entropy values at both ends of the line section are calculated, and the chaos degree is calculated as follows: Figure 5 shown.
[0156] Depend on Figure 5 It can be seen that no matter it is a low-resistance fault or a high-resistance fault, the chaos ratio of the fault section is much greater than that of the non-fault section, and the fault section can be accurately located. Although the chaos ratio of the fault section will decrease with the increase of transition resistance, its downward trend will continue to slow down with the increase of grounding resistance, and it can still meet the impact of actual grounding resistance.
[0157] (3) Impact of noise
[0158] In actual operation, the zero-sequence current will be affected by noise when it is collected, causing the zero-sequence current signal waveform to change. Therefore, Gaussian white noise with a signal-to-noise ratio of 20db is added to the zero-sequence current signal collected by each FTU on the fault line. The other conditions are set the same as those under the f1 fault. The fuzzy entropy confusion values of the S4~S7 measuring points before and after adding noise are calculated, as shown in Table 5:
[0159] Table 5
[0160]
[0161] After adding noise, the chaos ratios of S4~S5, S5~S6, and S6~S7 sections are calculated to be 0.96, 4.13, and 1.03, respectively. The faulty section can be directly determined to be S5~S6. It can be seen that the faulty section can still be accurately located under the influence of noise.
[0162] The above are only preferred specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.
Claims
1. A method for locating fault sections in a small current grounding system based on transient zero-sequence current fuzzy entropy waveform similarity, characterized in that: The following steps are involved: Building a distribution network model of a small current grounding system, and collecting transient zero-sequence current signals of each section on the fault line in the distribution network model; Performing CEEMDAN modal decomposition on the transient zero-sequence current signal to obtain a high-frequency imf1 component; Calculate the fuzzy entropy value of the imf1 component and define a chaos value as the first criterion for preliminary screening of fault sections, wherein the chaos value is used to measure the waveform similarity of transient zero-sequence current; Based on the first criterion, obtaining a first criterion result; If the chaos ratio is less than a preset threshold, the second criterion based on the initial polarity of the fuzzy entropy waveform is used to analyze the waveform polarity to obtain a second criterion result; When the first judgment criterion result is consistent with the second judgment criterion result, the fault section is located.
2. The method for locating the fault section of a small current grounding system based on the transient zero-sequence current fuzzy entropy waveform similarity according to claim 1 is characterized in that: The distribution network model includes a Y-connected transformer and an arc-extinguishing coil with a compensation degree. The fault line in the distribution network model is divided into sections by a detection device to obtain a plurality of sections.
3. The method for locating the fault section of a small current grounding system based on the transient zero-sequence current fuzzy entropy waveform similarity according to claim 1 is characterized in that: The process of calculating the fuzzy entropy value of the imf1 component includes: Step 1: Select the input high-frequency transient zero-sequence current signal as follows: {i(t s ),s=1,2,···,N}; Step 2: Using a sequence of length m as a time window, divide the transient zero-sequence current signal into k=N-m+1 sequences; in, is I(t s ) m The mean value of Step 3: Calculate the vector I(t i ) m and I(t j ) m The distance is the Chebyshev distance, that is, the maximum absolute value of the difference between the numerical values of each element, that is: Step 4: Introduce fuzzy membership n metric I(t i ) m , I(t j ) m The similarity of , for 1≤j≤Nm,j≠i, is: Where n and r are the boundary gradient and width; Step 5: Average all memberships except itself: Step 6: Increase the dimension to m+1 and repeat steps 2 to 5; Step 7: Calculate the fuzzy entropy FuzzyEn(m,r,n) of the signal sequence as: FuzzyEn(m,r,n)=lnφ m,r,n -lnφ m+1,r,n 。 4. The method for locating a fault section of a small current grounding system based on transient zero-sequence current fuzzy entropy waveform similarity according to claim 1 is characterized in that: The formula for calculating the fuzzy entropy value chaos of the imf1 component is: Among them, F(n) is the chaos degree of the nth measurement point, N is the number of sampling points, and f i is the fuzzy entropy value of the i-th sampling point.
5. The method for locating the fault section of a small current grounding system based on the transient zero-sequence current fuzzy entropy waveform similarity according to claim 1 is characterized in that: The process of obtaining the first criterion result includes: Calculate the chaos values of the measuring points at both ends of the line segment respectively, and compare the chaos values of the measuring points at both ends to obtain the chaos ratio; The chaos ratio is judged. If the chaos ratio is greater than or equal to the preset threshold, the output is a fault section, which is the result of the first judgment; if the chaos ratio is less than the preset threshold, the second judgment is used for analysis.
6. The method for locating the fault section of a small current grounding system based on transient zero-sequence current fuzzy entropy waveform similarity according to claim 1 is characterized in that: The formula of the first criterion is: Among them, F (i) With F (j) For line section L ij The chaos value of the measuring points at both ends, point i is the measuring point close to one end of the busbar, and point j is the measuring point close to the end of the line.
7. The method for locating the fault section of a small current grounding system based on transient zero-sequence current fuzzy entropy waveform similarity according to claim 1 is characterized in that: The second criterion process includes: The polarity characteristics of the high-frequency part IMF1 of the transient zero-sequence current are extracted using the symmetrical Sigmoid function, and the zero-sequence current waveform with polarity characteristics is obtained. The fuzzy entropy value of the sampling point is weighted on the waveform extracted by Sigmoid to obtain a fuzzy entropy waveform curve that reflects the initial polarity characteristics of the high-frequency part IMF1.
8. The method for locating the fault section of a small current grounding system based on transient zero-sequence current fuzzy entropy waveform similarity according to claim 1 is characterized in that: The formula of the second criterion includes: Diff(i)=I(i+1)-I(i),i∈1,2,…,N-1; Diff2(i)=Trend(i+1)-Trend(i); Among them, I is the fuzzy entropy value sequence, DiffI is the first-order difference vector, and Diff 2(i) is the second-order difference sequence.
9. An electronic device, characterized in that: The device includes: a processor and a memory storing computer program instructions; when the processor executes the computer program instructions, it implements the method for locating the fault section of a small current grounding system based on fuzzy entropy dual criteria as described in any one of claims 1-8.
Citation Information
Patent Citations
Single phase ground fault section locating method in small current grounding system
CN102621449A
Single-phase ground fault section positioning method of small-current ground system
CN102788926A
Fault voting line selection method by using Prony relative entropy
CN103135037A
Single-phase earth fault section positioning method of low-current grounding system
CN112485592A
Fault detection method suitable for source network load storage complex power grid
CN113820564A
Cited By
Ground fault identification method based on zero-sequence voltage periodic differential energy
CN119644046A
Battery connector real-time current monitoring method and system
CN119667559A
Waveform complexity measurement method, device and equipment based on waveform feature vector
CN119807710A
Method for measuring frequency of power frequency voltage
CN120314645A
Adaptive identification and extraction method for transient traveling wave signals in complex environment
CN121479284A
