A method and system for classifying power transmission line faults
By calculating the spectrogram characteristics and zero-sequence current of the three-phase current signal of the transmission line, a fault distinction vector is constructed and a negative selection algorithm is used for classification, which solves the complex problem of fault classification mechanism in the neural network and realizes the failure of transmission line Quick and accurate classification.
Patent Information
- Application Number
- CN202411206729.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-30
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2044-08-30
AI Technical Summary
In the prior art, the transmission line fault classification mechanism based on neural networks is complex and is prone to include irrelevant or redundant features, resulting in a decrease in classification speed and accuracy.
By collecting the three-phase current signals before and after the transmission line failure, calculating the total energy difference, Pearson correlation coefficient and spectrum shape similarity on the spectrum graph during calculation, the fault distinction vector is constructed, and the fault classification is used using a negative selection algorithm to avoid redundancy Problems of feature extraction.
Improve the distinction between transmission line faults, realize the accurate classification of three-phase short circuit, two-phase short circuit, two-phase short circuit and single-phase grounding faults, improve the classification speed and accuracy, and avoid the redundant data volume in traditional methods. Remaining the question.
Smart Images

Figure CN118734137B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of power transmission line fault classification, and in particular to a power transmission line fault classification method and system. Background Art
[0002] As an important part of the power system, transmission lines ensure the effective transmission and distribution of electric energy, which is vital to the operation of modern society. Transmission line faults will not only affect the stable operation of the power system, but also induce power outages, which will bring serious harm and impact to the national economic construction and people's daily life. Rapid and accurate transmission line fault classification is the prerequisite for locating transmission line faults and quickly restoring normal power supply, which is of great significance to ensuring the stability and safety of the power system.
[0003] With the development of neural network technology, artificial intelligence classifiers based on neural networks have better learning and generalization capabilities in the classification of transmission line faults. The extraction of fault features in transmission lines is the key to the classification of transmission line faults based on neural networks. The quality of the feature quantity directly affects the speed and accuracy of transmission line fault classification. However, the automatic extraction of fault features by neural networks will contain a large number of irrelevant or redundant features, which will cause the transmission line fault classification mechanism to become large and complex, thereby reducing the speed and accuracy of transmission line fault classification. Summary of the invention
[0004] In order to solve the above technical problems, the present application provides a method and system for classifying power transmission line faults. The technical solutions adopted are as follows:
[0005] In a first aspect, an embodiment of the present application provides a method for classifying power transmission line faults, the method comprising the following steps:
[0006] Collect the current signals of each phase in a preset time period before and after a transmission line fault occurs;
[0007] Obtain the total energy of each phase current signal on the time-frequency spectrum diagram; take the inverse of the difference in total energy between the phase current signals as the energy similarity between the phase current signals; calculate the Pearson correlation coefficient between the phase current signals, recorded as the first similarity, calculate the similarity of the spectrum shape of the spectrum diagram between the phase current signals, recorded as the second similarity, and take the product of the first similarity and the second similarity as the signal feature similarity between the phase current signals; take the product of the mean value of the energy similarity and the mean value of the signal feature similarity between all phase current signals as the first fault discrimination coefficient;
[0008] Based on the total energy difference of all phase current signals in time and frequency, all phase current signals are divided into two categories, and the difference in average total energy between the two categories of current signals in the time-frequency spectrum is used as the second fault discrimination coefficient;
[0009] The average value of the absolute value of the zero-sequence current of all phase current signals at all acquisition moments is taken as the third fault discrimination coefficient;
[0010] The first, second and third fault discrimination coefficients are combined into a fault discrimination vector; and the transmission line faults are classified according to the fault discrimination vector using a fault classification method based on a negative selection algorithm.
[0011] Preferably, the method of dividing all phase current signals into two categories based on the total energy difference of all phase current signals in time and frequency includes:
[0012] All phase current signals are divided into two categories according to the size of total energy; wherein, there are two-phase current signals in the first category and one-phase current signal in the second category; the two-phase current signals in the first category are the closest two-phase current signals among the three phases.
[0013] Preferably, the method for determining the second fault differentiation coefficient includes:
[0014] The mean value of the total energy of the two-phase current signals in the first category is used as the numerator, and the total energy of the current signals in the second category is used as the denominator;
[0015] The ratio of the numerator to the denominator is calculated as the second fault discrimination coefficient.
[0016] Preferably, the method of classifying the power transmission line fault using a fault classification method based on a negative selection algorithm according to the fault differentiation vector comprises:
[0017] All phase current signals collected each time in a preset time period before and after a transmission line fault occurs are taken as a sample, and all samples to be classified that have historically failed transmission lines are formed into a sample set;
[0018] The set consisting of the fault distinguishing vectors of all samples in the sample set is used as the self set in the fault classification method based on the negative selection algorithm;
[0019] A number of samples are obtained for all fault categories from historical samples of the transmission line, and a set of fault distinguishing vectors of all samples in each fault category is used as an initial fault detector set for each fault category in a fault classification method based on a negative selection algorithm;
[0020] A fault classification method based on negative selection algorithm is used to obtain the matching results of mature fault detectors for each fault distinguishing vector in the self-set; the fault type corresponding to the mature fault detector is used as the fault classification of the sample to which the corresponding fault distinguishing vector belongs in the self-set.
[0021] Preferably, the types of faults include: three-phase short circuit fault, two-phase short circuit fault, two-phase short circuit grounding fault, and single-phase grounding fault.
[0022] In a second aspect, an embodiment of the present application further provides a transmission line fault classification system, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein when the processor executes the computer program, the steps of any one of the above-mentioned transmission line fault classification methods are implemented.
[0023] This application has at least the following beneficial effects:
[0024] The present application constructs a fault discrimination vector by constructing the first, second and third fault discrimination coefficients of the time-frequency spectrum, spectrum diagram and zero-sequence current of the three-phase current signal of the transmission line. The beneficial effect is that the time-frequency energy, time-frequency domain distribution characteristics and zero-sequence current differences of the four faults of the transmission line, namely, three-phase short circuit fault, two-phase short circuit fault, two-phase short circuit grounding fault and single-phase grounding fault, are taken into account, thereby improving the discrimination between the four faults of the transmission line, namely, three-phase short circuit fault, two-phase short circuit fault, two-phase short circuit grounding fault and single-phase grounding fault. The present application completes the classification of transmission line faults based on the constructed fault discrimination vector and the fault classification method based on the negative selection algorithm. The beneficial effect lies in that the fault discrimination vector is used as the characteristic quantity in the transmission line fault classification method. While accurately and effectively classifying the four faults of the transmission line, namely, the three-phase short circuit fault, the two-phase short circuit fault, the two-phase short circuit grounding fault and the single-phase grounding fault, the application avoids the situation in which the redundant and huge amount of fault information data in the traditional method of using the neural network itself or the wavelet transform for fault feature extraction affects the speed of transmission line fault classification, thereby enabling the transmission line faults to be accurately classified more quickly. BRIEF DESCRIPTION OF THE DRAWINGS
[0025] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0026] Figure 1 A flow chart of a method for classifying power transmission line faults provided in one embodiment of the present application;
[0027] Figure 2 A flow chart of a method for determining energy similarity provided in one embodiment of the present application;
[0028] Figure 3 A flow chart of a method for determining signal feature similarity provided in one embodiment of the present application;
[0029] Figure 4 A flow chart of a method for determining a first fault differentiation coefficient provided in one embodiment of the present application;
[0030] Figure 5 A flow chart of a method for determining a second fault differentiation coefficient provided in one embodiment of the present application;
[0031] Figure 6 A flow chart of a method for determining a third fault differentiation coefficient provided in one embodiment of the present application;
[0032] Figure 7 A flow chart of a method for classifying power transmission line faults provided in one embodiment of the present application. DETAILED DESCRIPTION
[0033] In order to further explain the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the following is a detailed description of a method and system for classifying power transmission line faults proposed in the present application, its specific implementation, structure, features and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.
[0034] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.
[0035] The following is a detailed description of a method and system for classifying power transmission line faults provided by the present application in conjunction with the accompanying drawings.
[0036] An embodiment of the present application provides a method and system for classifying power transmission line faults.
[0037] Specifically, a method for classifying transmission line faults is provided as follows. Figure 1 , the method comprises the following steps:
[0038] The first step is to collect the current signals of each phase within a preset time period before and after a fault occurs on the transmission line.
[0039] A fault recorder is used to collect A, B, and C three-phase current signals within a preset time period before and after a transmission line fault occurs. In the embodiment of the present application, the collection time period before and after the fault is set to the current signal within 50ms before the transmission line fault to 50ms after the fault occurs, and the collection frequency is set to 50kHz, where the collection time and collection frequency can be set by the implementer.
[0040] So far, the A-phase current signal, the B-phase current signal and the C-phase current signal are collected in a preset time period before and after the transmission line fault occurs according to the above method.
[0041] In the second step, the result of negative correlation mapping the difference between the total energy of each phase current signal in the time-frequency spectrum diagram is recorded as energy similarity; based on the similarity between each phase current signal and the similarity of the frequency spectrum diagram between each phase current signal, the similarity between each phase current signal is determined, which is recorded as signal feature similarity; the product of the mean energy similarity and the mean signal feature similarity between all phase current signals is recorded as the first fault discrimination coefficient.
[0042] Short circuit fault is a common fault in transmission lines, and short circuit faults usually include four types: three-phase short circuit fault, two-phase short circuit fault, two-phase short circuit grounding fault and single-phase grounding fault. For different transmission line short circuit faults, the three-phase current signal of the transmission line has different distribution characteristics.
[0043] In the first small step, the result of negative correlation mapping between the total energy of each phase current signal in the time-frequency spectrum is recorded as energy similarity.
[0044] Specifically, since the non-fault phase current signals in the transmission line have close time-frequency energy and similar time-frequency domain distribution characteristics, the fault phase current signals also have close time-frequency energy and similar time-frequency domain distribution characteristics, and the time-frequency domain distribution characteristics between the fault phase current signal and the non-fault phase current signal are quite different, if a three-phase short circuit fault occurs in the transmission line, the current signals of each phase have close time-frequency energy and similar time-frequency domain distribution characteristics. If a non-three-phase short circuit fault occurs in the transmission line, the time-frequency energy and time-frequency domain distribution characteristics between the current signals of each phase are quite different, where the non-three-phase short circuit fault refers to a single-phase grounding fault and a two-phase fault (a two-phase short circuit fault or a two-phase short circuit grounding fault).
[0045] Based on the above analysis, taking the A-phase current signal as an example, the time-frequency spectrum TA of the A-phase current signal is extracted using short-time Fourier transform, and the total energy of the time-frequency spectrum TA is used as the time-frequency energy Ta of the A-phase current signal, wherein the extraction of the time-frequency spectrum and the calculation of the total energy are well-known technologies, and the specific process is not repeated here. Using the same method as the time-frequency energy Ta, the time-frequency energy Tb of the B-phase current signal and the time-frequency energy Tc of the C-phase current signal are obtained respectively.
[0046] Taking the A-phase current signal and the B-phase current signal as an example, the energy similarity J(A, B) between the A-phase current signal and the B-phase current signal is calculated to characterize the closeness of the time-frequency energy between the A-phase current signal and the B-phase current signal. Specifically, the difference in time-frequency energy between the A-phase current signal and the B-phase current signal is negatively correlated to determine the energy similarity between the current signals of each phase. In one implementation of the embodiment of the present application, the specific expression is: , where Ta and Tb represent the time-frequency energy of phase A current signal and phase B current signal respectively; is a parameter adjustment constant. In order to prevent the denominator from being 0, in this embodiment The value is 1.
[0047] In one implementation of the embodiments of the present application, the time-frequency energy difference between the A-phase current signal and the B-phase current signal is negatively correlated in a reciprocal form. In other embodiments of the present application, negative correlation mapping can also be achieved through a negative exponential function. In other embodiments of the present application, other basic mathematical operations can also be used to achieve negative correlation mapping, which is not limited or elaborated here.
[0048] It should be understood that the closer the time-frequency energy between the A-phase current signal and the B-phase current signal in the transmission line is, The smaller it is, the closer the time-frequency energy between the A-phase current signal and the B-phase current signal is, that is, the greater the energy similarity J(A,B).
[0049] The same method as the energy similarity J(A,B) is used to calculate the energy similarity between each pair of current signals in the A-phase current signal, the B-phase current signal, and the C-phase current signal.
[0050] In the embodiment of the present application, the flow chart of the method for determining energy similarity is as shown in the attached Figure 2 shown.
[0051] At this point, the energy similarity between the current signals of each phase can be obtained by the above method.
[0052] In the second small step, based on the similarity between the current signals of each phase and the similarity of the frequency spectrum between the current signals of each phase, the similarity between the current signals of each phase is determined, which is recorded as the signal feature similarity.
[0053] The A-phase, B-phase, and C-phase current signals are respectively taken as inputs, and the frequency spectra PA, PB, and PC corresponding to the A-phase, B-phase, and C-phase current signals are obtained by Fourier transform, which are used for subsequent analysis of the frequency domain distribution characteristics of the A-phase current signal. The acquisition of the frequency spectra is a well-known technology, and the specific process will not be repeated here.
[0054] Taking the A-phase current signal and the B-phase current signal as an example, the signal feature similarity S(A, B) between the A-phase current signal and the B-phase current signal is constructed to characterize the similarity of the time-frequency domain distribution characteristics between the A-phase current signal and the B-phase current signal. In the embodiment of the present application, the specific expression is: S(A, B)=s1(A, B)*s2(A, B), where s1(A, B) represents the similarity between the A-phase current signal and the B-phase current signal, which is recorded as the first similarity. The similarity is calculated in this embodiment using the Pearson correlation coefficient. The calculation process of the correlation coefficient is a well-known technology, and the specific process will not be repeated here. In other implementations of the embodiments of the present application, the negative correlation mapping of the DTW distance between the current signals can also be used to calculate the similarity; s2(A, B) represents the similarity between the spectrum graph PA of the A-phase current signal and the spectrum graph PB of the B-phase current signal, which is recorded as the second similarity. When calculating the similarity between the spectrum graphs, this embodiment uses the spectral shape similarity between the spectrum graphs as the similarity between them, wherein the calculation of the spectral shape similarity is a well-known technology, and the specific process will not be repeated here.
[0055] In other implementations of the embodiments of the present application, the specific expression may also be: S(A, B)=s1(A, B)+s2(A, B), which is set by the implementer.
[0056] It should be understood that the more similar the time domain distribution characteristics are between the A-phase current signal and the B-phase current signal in the transmission line, the greater the similarity s1(A, B) between the A-phase current signal and the B-phase current signal. At the same time, the more similar the frequency domain distribution characteristics are between the A-phase current signal and the B-phase current signal in the transmission line, the greater the spectral shape similarity s2(A, B) of the spectrum diagram between the A-phase current signal and the B-phase current signal. Then, the greater the similarity of the time and frequency domain distribution characteristics between the A-phase current signal and the B-phase current signal, that is, the greater the signal feature similarity S(A, B).
[0057] Using the same method as the signal feature similarity S(A,B), the signal feature similarities between each pair of current signals in the A-phase current signal, the B-phase current signal, and the C-phase current signal are calculated respectively.
[0058] In the embodiment of the present application, the flow chart of the method for determining the signal feature similarity is as shown in the attached Figure 3 shown.
[0059] At this point, the signal feature similarity between the current signals of each phase can be obtained.
[0060] In the third small step, the product of the mean energy similarity and the mean signal feature similarity between all phase current signals is recorded as the first fault discrimination coefficient.
[0061] Further, according to the mean energy similarity and signal feature similarity between all phase current signals, the first fault discrimination coefficient of the transmission line is obtained: 1. It is used to distinguish between three-phase short-circuit faults and non-three-phase short-circuit faults occurring in the transmission line. The specific implementation method is as follows: the energy similarity mean and the signal feature similarity mean between all phase current signals are merged to obtain the first fault distinction coefficient.
[0062] In the embodiment of the present application, the specific expression of the first fault differentiation coefficient is: , where represents the first fault discrimination coefficient, J represents the mean value of the energy similarity between the two current signals in the A-phase current signal, the B-phase current signal, and the C-phase current signal; S represents the mean value of the signal feature similarity between the two current signals in the A-phase current signal, the B-phase current signal, and the C-phase current signal. In other implementations of the present application, it is also possible to calculate , and obtain the first fault discrimination coefficient.
[0063] It should be understood that the closer the time-frequency energy between the current signals of each phase in the transmission line is, the larger J is. At the same time, the more similar the time-frequency distribution characteristics between the current signals of each phase in the transmission line are, the larger S is, and the more likely the fault in the transmission line is a three-phase short circuit fault, that is, the first fault discrimination coefficient is The larger the value, the less similar the time-frequency energy between the current signals of each phase in the transmission line is, the smaller J is, and the less similar the time-frequency distribution characteristics between the current signals of each phase in the transmission line are, the smaller S is, and the more likely the fault in the transmission line is a non-three-phase short-circuit fault, that is, the first fault discrimination coefficient The smaller.
[0064] In the embodiment of the present application, the flow chart of the method for determining the first fault differentiation coefficient is as shown in the attached figure. Figure 4 shown.
[0065] At this point, the first fault differentiation coefficient can be obtained through the above method.
[0066] In the third step, all phase current signals are divided into two categories based on the total energy difference of all phase current signals in time and frequency, and the difference in average total energy between the two types of current signals in the time-frequency spectrum is used as the second fault discrimination coefficient.
[0067] Secondly, since the time-frequency energy of the fault phase current signal in the three-phase current signal of the transmission line is much greater than that of the non-fault phase current signal, the non-fault phase current signals have similar time-frequency energy, and the fault phase current signals also have similar time-frequency energy. Therefore, if a single-phase grounding fault occurs in the transmission line, the time-frequency energy of two phase current signals in the three-phase current signal is approximately equal, and the time-frequency energy of these two phase current signals is much smaller than the time-frequency energy of the third phase current signal. If a two-phase fault occurs in the transmission line (a two-phase short circuit fault or a two-phase short circuit grounding fault), the time-frequency energy of two phase current signals in the three-phase current signal is approximately equal, and the time-frequency energy of these two phase current signals is much greater than the time-frequency energy of the third phase current signal. If a three-phase short circuit fault occurs in the transmission line, the time-frequency energy of the three-phase current signals is approximately equal.
[0068] Based on the above analysis, the two current signals with the closest time-frequency energy are selected from the A-phase current signal, the B-phase current signal, and the C-phase current signal as the first category. The mean value T1 of the time-frequency energy of the two-phase current signals in the first category is calculated. The remaining one-phase current signal is taken as the second category, and the time-frequency energy T2 of the current signal in the second category is obtained. According to the difference between the time-frequency energy of the first category and the second category, the second fault discrimination coefficient of the transmission line is obtained. , which is used to distinguish between a single-phase ground fault and a two-phase fault occurring in a transmission line. The specific expression in the embodiment of the present application is:
[0069] , where represents the second fault distinction coefficient, T1 represents the mean of the time-frequency energy of the two-phase current signals in the first category, and T2 represents the time-frequency energy of the current signal in the second category.
[0070] In other implementations of the embodiments of the present application, the expression may also be set as: , which is set by the implementer.
[0071] It should be understood that if the time-frequency energy of the two-phase current signal in the transmission line is much smaller than the time-frequency energy of the third-phase current signal, that is, the greater the degree to which T1 is smaller than T2, the more likely the fault in the transmission line is a single-phase grounding fault, that is, the second fault discrimination coefficient 2 is smaller; if the time-frequency energy of the two-phase current signal in the transmission line is much greater than the time-frequency energy of the third-phase current signal, the greater the degree to which T1 is greater than T2, the more likely the fault in the transmission line is a two-phase fault, that is, the second fault discrimination coefficient 2 is larger.
[0072] In the embodiment of the present application, the flow chart of the method for determining the second fault differentiation coefficient is as shown in the attached Figure 5 shown.
[0073] At this point, the second fault differentiation coefficient can be obtained through the above method.
[0074] In the fourth step, the average value of the absolute value of the zero-sequence current of all phase current signals at all acquisition moments is taken as the third fault discrimination coefficient.
[0075] Secondly, if the fault in the transmission line is a grounding fault, the zero-sequence current of the three-phase current signal in the transmission line is very large. If the fault is a non-grounding fault, the zero-sequence current is very small and close to zero.
[0076] Based on the above analysis, based on the A-phase current signal, the B-phase current signal and the C-phase current signal, the zero-sequence current of the three-phase current signal of the transmission line at each acquisition time of the current signal is calculated respectively. For example, at a certain acquisition time of the current signal, the amplitudes of the A-phase current signal, the B-phase current signal and the C-phase current signal are Ia, Ib and Ic respectively. Then the zero-sequence current I0 of the three-phase current of the transmission line at this acquisition time is Ia+Ib+Ic / 3, and the average of the absolute values of all the zero-sequence currents calculated is recorded as the third fault discrimination coefficient of the transmission line. , which is used to distinguish between grounding faults (single-phase grounding faults and two-phase short-circuit grounding faults) and non-grounding faults (three-phase short-circuit faults and two-phase short-circuit faults) in the transmission line. The closer the value of the zero-sequence current of the three-phase current of the transmission line at all acquisition times is to 0, that is, The smaller it is, the more likely the fault in the transmission line is a non-grounding fault. The larger it is, the more likely the fault in the transmission line is a ground fault.
[0077] In the embodiment of the present application, the flow chart of the method for determining the third fault differentiation coefficient is as shown in the attached figure. Figure 6 shown.
[0078] At this point, the third fault differentiation coefficient can be obtained through the above method.
[0079] The fifth step is to form a fault discrimination vector by combining the first, second and third fault discrimination coefficients; and classify the transmission line faults according to the fault discrimination vector using a fault classification method based on a negative selection algorithm.
[0080] Furthermore, a fault differentiation vector of the transmission line is constructed according to the first, second and third fault differentiation coefficients: , used to distinguish between three-phase short circuit fault, two-phase short circuit fault, two-phase short circuit grounding fault and single-phase grounding fault in the transmission line, specifically: , where represents the fault distinguishing vector, , , They represent the first, second and third fault discrimination coefficients respectively.
[0081] It should be noted that The larger it is, the more likely the fault in the transmission line is a three-phase short-circuit fault; The smaller, The smaller it is, the more likely the fault in the transmission line is a single-phase grounding fault; The smaller, The bigger, The larger it is, the more likely the fault in the transmission line is a two-phase short-circuit grounding fault; The smaller, The bigger, The smaller it is, the more likely the fault in the transmission line is a two-phase short circuit fault.
[0082] In the embodiment of the present application, the three-phase current signal collected between 50 ms before and 50 ms after the transmission line fault occurs is taken as a sample.
[0083] A sample set consisting of all samples to be classified that have historically failed in the transmission line is obtained, and all samples in the sample set are classified into fault classes using a fault classification method based on a negative selection algorithm, wherein a set consisting of fault distinguishing vectors of all samples in the sample set is used as a self-set in the fault classification method based on a negative selection algorithm.
[0084] From the historical three-phase current signals of the transmission line, 50 samples are obtained for each of the four types of faults, namely, three-phase short circuit fault, two-phase short circuit fault, two-phase short circuit grounding fault and single-phase grounding fault. The set of fault discrimination vectors of all samples in each type of fault is used as the initial fault detector set for each type of fault in the fault classification method based on the negative selection algorithm.
[0085] Among them, the number of four mature fault detectors corresponding to the four types of faults, namely three-phase short circuit fault, two-phase short circuit fault, two-phase short circuit grounding fault and single-phase grounding fault, in the fault classification method based on the negative selection algorithm is set to 150. The self-set matching threshold, feature matching threshold and suppression threshold in the fault classification method based on the negative selection algorithm are set to 0.1, 0.01 and 0.99995 respectively. The matching thresholds of the mature fault detectors for the four types of faults are set to 0.833, 0.92593, 0.84746 and 0.95238 respectively. The fault classification method based on the negative selection algorithm is used to obtain the matching results of the mature fault detector for each fault discrimination vector in the self-set, and the fault type corresponding to the mature fault detector is used as the fault classification of the sample to which the corresponding fault discrimination vector in the self-set belongs.
[0086] Among them, the fault types include three-phase short circuit fault, two-phase short circuit fault, two-phase short circuit grounding fault, and single-phase grounding fault. The fault classification method based on the negative selection algorithm is a well-known technology, and the values set therein can be set by themselves in other implementations of the embodiments of the present application, and the specific process will not be repeated.
[0087] In the embodiment of the present application, a flow chart of a method for classifying transmission line faults is shown in the attached Figure 7 shown.
[0088] The fault type of each sample in the sample set is obtained respectively, and the fault classification of all three-phase current signals to be classified that have historically had faults in the transmission line is performed.
[0089] Based on the same inventive concept as the above method, an embodiment of the present application also provides a transmission line fault classification system, including a memory, a processor, and a computer program stored in the memory and running on the processor, and when the processor executes the computer program, the steps of a transmission line fault classification method described in any one of the above methods are implemented.
[0090] The various embodiments in the present application are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.
[0091] It should be noted that, unless otherwise specified and limited, terms such as "include", "comprises" or any other variants thereof are intended to cover non-exclusive inclusion, so that a circuit structure, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such article or device. In the absence of further restrictions, an element defined by the sentence "including one..." does not exclude the existence of other identical elements in the article or device including the element. In addition, the term "and\or" used herein includes any and all combinations of one or more related listed items.
[0092] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention herein. The present application is intended to cover any variations, uses or adaptations of the present application, which follow the general principles of the present application and include common knowledge or customary technical means in the art that are not invented by the present application.
[0093] It should be understood that the present application is not limited to the exact construction that has been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof.
Claims
1. A method for classifying power transmission line faults, characterized in that: The method comprises the following steps: Collect the current signals of each phase in a preset time period before and after a fault occurs on the transmission line; Obtain the total energy of each phase current signal on the time-frequency spectrum diagram; take the inverse of the difference in total energy between the phase current signals as the energy similarity between the phase current signals; calculate the Pearson correlation coefficient between the phase current signals, recorded as the first similarity, calculate the similarity of the spectrum shape of the spectrum diagram between the phase current signals, recorded as the second similarity, and take the product of the first similarity and the second similarity as the signal feature similarity between the phase current signals; take the product of the mean value of the energy similarity and the mean value of the signal feature similarity between all phase current signals as the first fault discrimination coefficient; Based on the total energy difference of all phase current signals in time and frequency, all phase current signals are divided into two categories, and the difference in average total energy between the two categories of current signals in the time-frequency spectrum is used as the second fault discrimination coefficient; The average value of the absolute value of the zero-sequence current of all phase current signals at all acquisition moments is taken as the third fault discrimination coefficient; The first, second and third fault discrimination coefficients are combined into a fault discrimination vector; and the transmission line faults are classified according to the fault discrimination vector using a fault classification method based on a negative selection algorithm; The method divides all phase current signals into two categories based on the total energy difference of all phase current signals in time and frequency, including: All phase current signals are divided into two categories according to the total energy; wherein the first category has two-phase current signals and the second category has one-phase current signal; the two-phase current signals in the first category are the two-phase current signals closest to each other among the three phases; The method for determining the second fault differentiation coefficient includes: The mean value of the total energy of the two-phase current signals in the first category is used as the numerator, and the total energy of the current signals in the second category is used as the denominator; The ratio of the numerator to the denominator is calculated as the second fault discrimination coefficient.
2. A method for classifying power transmission line faults according to claim 1, characterized in that: The method of classifying the transmission line faults according to the fault differentiation vector using a fault classification method based on a negative selection algorithm comprises: All phase current signals collected each time in a preset time period before and after a transmission line fault occurs are taken as a sample, and all samples to be classified that have historically failed transmission lines are formed into a sample set; The set consisting of the fault distinguishing vectors of all samples in the sample set is used as the self set in the fault classification method based on the negative selection algorithm; A number of samples are obtained for all fault categories from historical samples of the transmission line, and a set of fault distinguishing vectors of all samples in each fault category is used as an initial fault detector set for each fault category in a fault classification method based on a negative selection algorithm; A fault classification method based on negative selection algorithm is used to obtain the matching results of mature fault detectors for each fault distinguishing vector in the self-set; the fault type corresponding to the mature fault detector is used as the fault classification of the sample to which the corresponding fault distinguishing vector belongs in the self-set.
3. A method for classifying power transmission line faults according to claim 1, characterized in that: The types of faults include: three-phase short circuit fault, two-phase short circuit fault, two-phase short circuit grounding fault, and single-phase grounding fault.
4. A transmission line fault classification system, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of a power transmission line fault classification method as described in any one of claims 1 to 3 are implemented.
Citation Information
Patent Citations
Line fault judgment method based on wavelet analysis and support vector machine
CN105093066A
Lightning stroke fault recognition method for EHV (extra-high voltage) DC transmission line based energy spectrum similarity
CN107255743A
Method for detecting similarity of audio signals
CN108597537A
Power transmission line fault identification method based on deep learning
CN112147462A
Permanent magnet synchronous motor bearing fault diagnosis method based on motor current analysis method
CN116400215A
