Single-phase earth fault positioning method and system based on REMD-SD algorithm

Abstract

The invention relates to a single-phase earth fault positioning method and system based on an REMD-SD algorithm, and the method comprises the following steps: obtaining transient zero-sequence current of a plurality of detection points when a fault occurs, carrying out the robust empirical mode decomposition of the transient zero-sequence current, and extracting a first-order intrinsic mode function; calculating a second-order difference sequence of the first-order intrinsic mode function, and judging concavity and convexity of the first-order intrinsic mode function based on the second-order difference sequence; and based on the concavity and convexity comparison of the first-order intrinsic mode functions corresponding to the adjacent detection points, realizing the positioning of the single-phase earth fault. Compared with the prior art, the method has the advantages that the fault features are obvious, and then the fault positioning accuracy is improved.

Description

Technical Field

[0001] The present invention relates to the field of power system fault location, and particularly to a single-phase grounding fault location method and system based on the REMD-SD algorithm. Background Art

[0002] In a 10 kV power system, the ungrounded neutral system is a common grid grounding method. When a single-phase grounding fault occurs in an ungrounded neutral system, since the fault current is only the capacitive current of the power grid to the ground, which is relatively small, the system can continue to operate stably for a period of time, usually not exceeding 2 hours. This allows the power system to have time to locate the fault and perform repairs, reducing the power outage time and improving the continuity of power supply. Due to the ungrounded neutral, the system may generate a relatively high overvoltage during a single-phase grounding fault, especially during an intermittent arc grounding fault, which may cause an arc overvoltage up to 3.5 times the phase voltage, posing a threat to the equipment insulation. In an ungrounded neutral system, due to the small fault current and the inconspicuous fault characteristics, it is difficult for traditional grounding fault detection technologies to accurately locate the fault point, which poses a challenge to fault handling and system maintenance.

[0003] Li Weiguo et al. proposed a fault location method based on VMD decomposition and wave peak and valley algorithm in the literature "A Fault Section Location Method for Distribution Networks Based on the Concave-Convex Characteristics of Transient Zero-Sequence Current". This method is not affected by grounding resistance, fault initial angle and fault distance, has good applicability, and is easy to be programmed and implemented. However, this method is sensitive to noise. When the signal contains a high noise level, the decomposition result may be interfered by the noise, and this method requires the predefined mode number K, which is not applicable to all cases. Xia Yu et al. proposed a relative location method based on section zero-sequence energy in the literature "Research on a New Method for Locating and Isolating Single-Phase Grounding Fault Sections in Distribution Network Feeders". This method uses the characteristics that the zero-sequence energy function of the non-fault section is greater than zero and the zero-sequence energy function of the fault section is less than zero to determine the fault section. However, this method is easily affected by CT unbalanced current and long line distance, resulting in inaccurate results. Jia Junguo et al. proposed to monitor the zero-sequence current and zero-sequence voltage at each switch on a feeder, calculate the phasor sum of the zero-sequence current flowing into this section from the endpoints of the section (i.e., the zero-sequence current flowing into the section), to identify the fault section and judge the fault state of this feeder. This method constructs an amplitude criterion and a phase criterion according to the characteristics of the section zero-sequence current. However, when the grounding resistance is large, the distributed capacitance on the line will shunt the injected signal, which will interfere with this method. Wang Zheng et al. proposed a single-phase grounding fault location method for small current systems cooperating with FTU - the zero-sequence current increment method according to the active component direction protection method in the literature "Single-Phase Grounding Fault Location Method for Small Current Systems Cooperating with FTU". However, this method depends on the FTU device and may be affected when the FTU device fails to accurately measure or report data.

[0004] In summary, there are still some deficiencies in the existing grounding fault detection technologies, and it is necessary to further research and develop the single-phase grounding fault location technology for 10kV small current neutral ungrounded systems. Summary of the Invention

[0005] The purpose of the present invention is to overcome the above-mentioned defects existing in the prior art and provide a single-phase grounding fault location method and system based on the REMD-SD algorithm, which can make the fault characteristics obvious and thus improve the accuracy of fault location.

[0006] The purpose of the present invention can be achieved by the following technical solutions:

[0007] A single-phase grounding fault location method based on the REMD-SD algorithm, the method includes the following steps:

[0008] When a fault occurs, obtain the transient zero-sequence current of multiple detection points, perform robust empirical mode decomposition on the transient zero-sequence current, and extract the first-order intrinsic mode function;

[0009] Calculate the second-order difference sequence of the first-order intrinsic mode function, and determine the concavity and convexity of the first-order intrinsic mode function based on the second-order difference sequence;

[0010] Based on the comparison of the concavity and convexity of the first-order intrinsic mode function corresponding to adjacent detection points, the location of the single-phase grounding fault is realized.

[0011] Furthermore, the transient zero-sequence current of the detection point is the transient zero-sequence current within a set period before and after the fault occurrence time.

[0012] Furthermore, the steps of performing robust empirical mode decomposition and extracting the first-order intrinsic mode function include:

[0013] 101) Find the extreme points of the original signal x(t), calculate the upper and lower envelope lines through cubic spline function interpolation, obtain m(t) according to the upper and lower envelope lines, and subtract the mean value m(t) from the original signal to obtain the intermediate signal h(t);

[0014] 102) Verify whether the intermediate signal h(t) satisfies the IMF component conditions. If so, obtain the first-order intrinsic mode function c 1 (t) based on the intermediate signal h(t). If not, use the intermediate signal h(t) as a new signal and return to step 101);

[0015] 103) Subtract c 1 (t) from the original signal x(t) to obtain a new original signal d(t), return to step 101), and iterate the above process to obtain the remaining IMF components.

[0016] Furthermore, the stopping of the iteration includes one of the following situations:

[0017] The signal quantization value f of the k-th iteration k satisfies the conditions f k-2 <f k-1 and f k-1 <f k , then stop, save and return the result of the (k - 2)-th time;

[0018] When the number of iterations reaches the preset threshold, stop, save and return the value of the previous iteration process.

[0019] Furthermore, the signal quantization value f k is obtained based on the root mean square and excess kurtosis indexes of the intermediate signal.

[0020] Furthermore, calculating the second-order difference sequence of the first-order intrinsic mode function specifically includes the following steps:

[0021] 201) Calculate the first-order difference sequence H of the first-order intrinsic mode function 1 ;

[0022] 202) Perform a sign operation on the first-order difference sequence H 1 to obtain sequence F;

[0023] 202) Perform a first-order difference operation on sequence F again to obtain a second-order difference sequence H 2 .

[0024] Further, the calculation formula for sequence F is:

[0025]

[0026] where F(i) is the i-th value of sequence F.

[0027] Further, determining the concavity and convexity of the first-order intrinsic mode function based on the second-order difference sequence is specifically as follows:

[0028] The trough in the second-order difference sequence corresponds to the first-order intrinsic mode function being concave, and the peak in the second-order difference sequence corresponds to the first-order intrinsic mode function being convex.

[0029] Further, when performing single-phase ground fault location, the section where the first concavity and convexity of the first-order intrinsic mode functions at two detection points on the same line are opposite is determined as the fault section.

[0030] The present invention also provides a single-phase ground fault location system based on the REMD-SD algorithm, and the system includes:

[0031] A modal decomposition module, configured to obtain the transient zero-sequence current at multiple detection points when a fault occurs, perform robust empirical mode decomposition on the transient zero-sequence current, and extract the first-order intrinsic mode function;

[0032] A concavity and convexity discrimination module, configured to calculate the second-order difference sequence of the first-order intrinsic mode function, and determine the concavity and convexity of the first-order intrinsic mode function based on the second-order difference sequence;

[0033] A fault location module, configured to realize the location of single-phase ground faults based on the comparison of the concavity and convexity of the first-order intrinsic mode functions corresponding to adjacent detection points.

[0034] Compared with the prior art, the present invention has the following beneficial effects:

[0035] 1. The present invention takes into account problems such as small fault current and unclear fault characteristics. By using REMD to decompose the transient zero-sequence current and extract the first-order intrinsic mode function, i.e., the IMF1 component, which is most similar to the original signal, for analysis. This makes the fault characteristics more obvious and enables more accurate analysis of the current.

[0036] 2. The present invention uses the data-driven REMD method, which greatly alleviates the problem of mode mixing. The decomposition result is more accurate, and when processing signals, it does not require frequency-domain analysis, nor does it need to pre-select basis functions. It can adaptively decompose signals based on the distribution of the extreme points of the signal itself.

[0037] 3. The present invention first decomposes the transient zero-sequence current through REMD and extracts the first-order modal component after decomposition. Then, it uses the second-order difference sequence to characterize the concavity and convexity of the IMF1 component, and the difference between the fault section and the non-fault section can be clearly seen, thereby enabling accurate fault location and accurately realizing single-phase grounding fault location in a 10 kV small-current neutral ungrounded system.

[0038] 4. The method of the present invention includes steps such as robust empirical mode decomposition, extraction of the first-order intrinsic mode function, and calculation of the second-order difference sequence in the data processing stage. It adopts a robust statistical method and has strong anti-noise ability. Theoretically, it is more suitable for analyzing the transient signal characteristics after a single-phase grounding fault. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 is a schematic flow chart of the present invention;

[0040] Figure 2 is the zero-sequence current of each detection point in the embodiment of the present invention;

[0041] Figure 3 is the IMF1 component of each detection point and its concavity and convexity in the embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0042] The present invention will be described in detail below with reference to the drawings and specific embodiments. This embodiment is implemented on the premise of the technical solution of the present invention, and detailed implementation methods and specific operation processes are given. However, the protection scope of the present invention is not limited to the following embodiments.

[0043] A method capable of realizing single-phase grounding fault location.

[0044] Embodiment 1

[0045] This embodiment provides a single-phase grounding fault location method based on the REMD-SD algorithm. The method includes the following steps: When a fault occurs, the transient zero-sequence current at each detection point is subjected to Robust Empirical Mode Decomposition (REMD), and its first-order Intrinsic Mode Functions (IMF), that is, the IMF1 component, is extracted. Then, the second difference sequence (SD) of the IMF1 component is calculated to characterize its concavity and convexity. Finally, the fault location is performed based on the initial concavity and convexity difference after the fault occurs.

[0046] As Figure 1 shown, the specific steps of the above single-phase grounding fault location method include:

[0047] S1. Determine whether a single-phase grounding fault has occurred. If so, execute step S2; if not, return to step S1.

[0048] S2. The substation terminal selects the faulty line and uploads the acquisition data of relevant fault indicators to the master station through the communication device.

[0049] S3. Obtain the transient zero-sequence current at multiple detection points. Specifically, the transient zero-sequence current at the detection point can be selected as the transient zero-sequence current within a set period before and after the fault occurrence time.

[0050] In this embodiment, the transient zero-sequence current data for a total of 1 / 5 cycle before and after the fault occurrence is selected for subsequent analysis. As Figure 2 shown, the detection points are I1 - I10.

[0051] S4. Perform REMD decomposition on the transient zero-sequence current data and extract the IMF1 component data.

[0052] The REMD decomposition process is as follows:

[0053] 1) Find the extreme points of the original signal x(t), and calculate the upper and lower envelope lines through cubic spline function interpolation, that is, the maximum envelope line x 1 (t) and the minimum envelope line x 2 (t), and obtain their mean value m(t) according to the upper and lower envelope lines.

[0054]

[0055] 2) Subtract the mean value m(t) from the original signal to obtain an intermediate signal h(t), that is:

[0056] h(t) = x(t) - m(t) (2)

[0057] 3) Verify whether the intermediate signal h(t) satisfies the two conditions of the IMF component. If not, treat h(t) as a new signal and repeat the above steps until the conditions are met, obtaining the first-order IMF component c 1 (t).

[0058] 4) Among them, the two conditions of the IMF component include the local extreme value condition and the envelope mean value condition. The local extreme value condition is: the sum of the number of local maxima and the number of local minima must be equal to the number of zero crossings, or at most differ by 1. That is to say, a zero crossing must immediately follow an extreme value. This ensures that the waveform of the IMF component is locally symmetric about the zero average value. The envelope mean value condition is: at any time point, the average value of the upper envelope defined by the local maximum and the lower envelope defined by the local minimum needs to be close to zero. This means that the waveform of the IMF component is balanced in each local part without significant offset.

[0059] 5) Subtract the first-order IMF component c 1 (t) from the original signal x(t) to obtain a new original signal d(t), that is:

[0060] d(t) = x(t) - c 1 (t) (3)

[0061] 6) Iterate the above process to obtain the remaining IMF components until the number of iterations is less than the preset threshold and the decomposition stops.

[0062] During the REMD decomposition process, it is also necessary to consider the stopping condition of the iterative screening. The screening process is as follows:

[0063] 1) During the EMD decomposition process, the energy of each point and the total energy of the mean signal g(n) of the intermediate signal h(t) approach zero, which is impossible to achieve in practice. Therefore, the REMD decomposition uses the root mean square (RMS) to represent the total energy of the signal to achieve practical quantization. The formula for RMS is as follows:

[0064]

[0065] 2) In the formula, N sis the total number of samples, k is the number of iterations in the decomposition process, and n is the sampling point. As can be seen from the above, the value of RMS should approach zero. However, there is a situation where most of the values of the signal are very small or even zero while a small part of the values are very high, and the RMS value in this case will also approach zero. To avoid the occurrence of the above situation, REMD uses the excess kurtosis index, which can evaluate the kurtosis of a signal. The excess kurtosis index is defined as follows:

[0066]

[0067] Similarly, the smaller the value of EK, the better. Through the above two indicators, the signal can be quantitatively analyzed, that is:

[0068] f k = RMS K + |EK k | (8)

[0069] The REMD decomposition uses a new adaptive heuristic mechanism that can automatically stop the screening process when the number of screening times reaches the optimal number of iterations. The specific process is to calculate the value of f in equation (8) during each screening process and compare it with the previous values until the conditions f k < f k-2 and f k-1 < f k-1 < f k are satisfied. The screening stops, and the result of the (k - 2)th time is saved and returned. When the number of iterations reaches the preset threshold, the screening process stops, and the value of the previous iteration process is saved and returned. This adaptive screening stop criterion can enable the REMD decomposition to obtain better signal decomposition and demodulation performance.

[0070] S5. Calculate the second-order difference sequence of the first-order intrinsic mode function, and determine the concavity and convexity of the first-order intrinsic mode function based on the second-order difference sequence.

[0071] Since the transient current sequence collected in practice is a discrete sequence, the concavity and convexity determination theorem of the function is not applicable. Therefore, in this embodiment, the second-order difference sequence is used to describe the peaks and valleys of the signal to describe the concavity and convexity of the IMF1 component. The concavity and convexity discrimination process is as follows:

[0072] 1) Extract the IMF 1 component data X = [X 1 , X 2 ...X N of the transient current signal after REMD decomposition, and calculate its first-order difference sequence H 1 .

[0073] H 1 (i) = X(i + 1) - X(i) (9)

[0074] where \(i\in[1, 2,\cdots,N - 1]\).

[0075] 2) Perform a sign operation on the first-order difference sequence \(H\) 1 to obtain the sequence \(F\).

[0076]

[0077] 3) Perform a first-order difference operation on the sequence \(F\) obtained by the sign operation again to obtain the second-order difference sequence \(H\) 2 .

[0078] \(H\) 2 (i)=F(i + 1)-F(i) (11)

[0079] As can be seen from the above steps, when the value of the second-order difference sequence \(H\) 2 is 2, it is a trough, and when it is -2, it is a peak. Furthermore, the concavity and convexity of the IMF 1 component can be discriminated (a trough represents concavity, and a peak represents convexity).

[0080] Such as Figure 3 and Tables 1 - 4 show the IMF1 components, concavity and convexity of each detection point, and the initial concavity and convexity of each detection point under different conditions obtained in this embodiment. It is possible to conveniently locate the fault interval according to the difference in the initial concavity and convexity, indicating the availability of the method of this embodiment.

[0081] Table 1 Initial concavity and convexity of each detection point under different fault positions

[0082]

[0083] Table 2 Initial concavity and convexity of each detection point at different fault occurrence times

[0084]

[0085] Table 3 Initial concavity and convexity of each detection point under different grounding resistance values

[0086]

[0087] Table 4 Concavity and convexity of each detection point under noise interference

[0088]

[0089] S6. Based on the comparison of the concavity and convexity of the first-order intrinsic mode functions corresponding to adjacent detection points, the single-phase grounding fault is located.

[0090] In this embodiment, the initial (first) concavity and convexity features of the IMF1 components at each detection point after the fault occurs are summarized, and the following results are obtained: 1) The concavity and convexity of the non-faulty paths on the non-faulty line and the faulty line are the same; 2) The concavity and convexity of the faulty path and the non-faulty path are opposite. From the above results, it can be known that the fault section criterion is: after a single-phase grounding fault occurs, the section where the initial concavity and convexity of the IMF1 components at two detection points on the same line are opposite is the fault section.

[0091] Aiming at the problems in the ungrounded neutral system, such as small fault current, unclear fault characteristics, and difficult accurate fault location of traditional grounding fault detection technologies, the above method proposes a method combining REMD decomposition and second-order difference sequence to judge the fault interval, making the fault characteristics obvious and enabling more accurate determination of the fault interval.

[0092] If the above method is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0093] Embodiment 2

[0094] This embodiment provides a single-phase grounding fault location system based on the REMD-SD algorithm. The system includes a modal decomposition module, a modal decomposition module, and a fault location module. Among them, the modal decomposition module is used to obtain the transient zero-sequence current at multiple detection points when a fault occurs, perform robust empirical mode decomposition on the transient zero-sequence current, and extract the first-order intrinsic mode function; the concavity and convexity discrimination module is used to calculate the second-order difference sequence of the first-order intrinsic mode function and discriminate the concavity and convexity of the first-order intrinsic mode function based on the second-order difference sequence; the fault location module is used to realize the location of the single-phase grounding fault based on the comparison of the concavity and convexity of the first-order intrinsic mode functions corresponding to adjacent detection points.

[0095] The rest is the same as in Embodiment 1.

[0096] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) that contain computer-usable program code. The solutions in the embodiments of the present invention can be implemented in various computer languages. For example, object-oriented programming languages such as Java and interpreted scripting languages such as JavaScript, etc.

[0097] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be realized by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for realizing the functions specified in one Figure 1 one flow or multiple flows and / or blocks Figure 1 means for realizing the functions specified in one block or multiple blocks.

[0098] The preferred specific embodiments of the present invention have been described in detail above. It should be understood that those of ordinary skill in the art can make many modifications and variations based on the concept of the present invention without creative efforts. Therefore, all technical solutions that can be obtained by those skilled in the art in the technical field of the present invention through logical analysis, reasoning, or limited experiments based on the concept of the present invention on the basis of the prior art should fall within the protection scope determined by the claims.

Claims

1. A single-phase grounding fault location method based on REMD-SD algorithm, characterized in that: The method comprises the following steps: When a fault occurs, the transient zero-sequence current of multiple detection points is obtained, the transient zero-sequence current is subjected to robust empirical mode decomposition, and the first-order intrinsic mode function is extracted; Calculating a second-order difference sequence of the first-order intrinsic mode function, and determining the concavity and convexity of the first-order intrinsic mode function based on the second-order difference sequence; Based on the comparison of the concavity and convexity of the first-order natural mode functions corresponding to adjacent detection points, the single-phase grounding fault is located.

2. The single-phase grounding fault location method based on REMD-SD algorithm according to claim 1 is characterized in that: The transient zero-sequence current at the detection point is the transient zero-sequence current within a set period before and after the fault occurs.

3. The single-phase grounding fault location method based on REMD-SD algorithm according to claim 1 is characterized in that: The step of performing robust empirical mode decomposition and extracting the first-order intrinsic mode function comprises: 101) Find the extreme points of the original signal x(t), calculate the upper and lower envelopes by cubic spline function interpolation, obtain m(t) based on the upper and lower envelopes, and subtract the mean m(t) from the original signal to obtain the intermediate signal h(t); 102) Verify whether the intermediate signal h(t) satisfies the IMF component condition. If so, obtain the first-order intrinsic mode function c1(t) based on the intermediate signal h(t). If not, take the intermediate signal h(t) as a new signal and return to step 101); 103) Subtract c1(t) from the original signal x(t) to obtain a new original signal d(t), return to step 101), and iterate the above process to obtain the remaining IMF components.

4. The single-phase grounding fault location method based on REMD-SD algorithm according to claim 3 is characterized in that: The stopping of the iteration includes one of the following situations: The signal quantization value f of the kth iteration k Satisfy the condition f k-2 <f k-1 and f k-1 <f k , then stop, save and return the result of the k-2th time; When the number of iterations reaches the preset threshold, the process stops, saves and returns the value of the previous iteration.

5. The single-phase grounding fault location method based on REMD-SD algorithm according to claim 4 is characterized in that: The signal quantization value f k The RMS and excess kurtosis indicators are obtained based on the intermediate signal.

6. The single-phase grounding fault location method based on REMD-SD algorithm according to claim 1, characterized in that: Calculating the second-order difference sequence of the first-order intrinsic mode function specifically includes the following steps: 201) calculating a first-order difference sequence H1 of the first-order intrinsic mode function; 202) performing a sign operation on the first-order difference sequence H1 to obtain a sequence F; 202) Performing a first-order difference operation on the sequence F again to obtain a second-order difference sequence H2.

7. The single-phase grounding fault location method based on REMD-SD algorithm according to claim 6 is characterized in that: The calculation formula of the sequence F is: Among them, F(i) is the i-th value of sequence F.

8. The single-phase grounding fault location method based on REMD-SD algorithm according to claim 1, characterized in that: The method for determining the concavity and convexity of the first-order intrinsic mode function based on the second-order difference sequence is as follows: A trough in the second-order difference sequence corresponds to that the first-order intrinsic mode function is concave, and a peak in the second-order difference sequence corresponds to that the first-order intrinsic mode function is convex.

9. The single-phase grounding fault location method based on REMD-SD algorithm according to claim 1, characterized in that: When locating a single-phase grounding fault, the first section with opposite concavity and convexity of the first-order natural mode function of two detection points on the same line is determined as the fault section.

10. A single-phase grounding fault location system based on REMD-SD algorithm, characterized in that: The system includes: A modal decomposition module is used to obtain transient zero-sequence currents of multiple detection points when a fault occurs, perform robust empirical modal decomposition on the transient zero-sequence current, and extract the first-order intrinsic modal function; a concavity and convexity determination module, used for calculating a second-order difference sequence of the first-order intrinsic mode function, and determining the concavity and convexity of the first-order intrinsic mode function based on the second-order difference sequence; The fault location module is used to locate the single-phase grounding fault based on the comparison of the concavity and convexity of the first-order natural mode functions corresponding to adjacent detection points.

Images