CN116413553A - Rapid line selection method for small-current ground fault - Google Patents
Rapid line selection method for small-current ground fault Download PDFInfo
- Publication number
- CN116413553A CN116413553A CN202310386910.7A CN202310386910A CN116413553A CN 116413553 A CN116413553 A CN 116413553A CN 202310386910 A CN202310386910 A CN 202310386910A CN 116413553 A CN116413553 A CN 116413553A
- Authority
- CN
- China
- Prior art keywords
- line
- fault
- zero
- current
- value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000010187 selection method Methods 0.000 title claims abstract description 10
- 230000001052 transient effect Effects 0.000 claims abstract description 36
- 238000012549 training Methods 0.000 claims abstract description 25
- 238000000034 method Methods 0.000 claims abstract description 19
- 238000003062 neural network model Methods 0.000 claims abstract description 13
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 11
- 238000012545 processing Methods 0.000 claims abstract description 11
- 238000010606 normalization Methods 0.000 claims abstract description 7
- 238000013528 artificial neural network Methods 0.000 claims description 22
- 241000283153 Cetacea Species 0.000 claims description 20
- 230000000739 chaotic effect Effects 0.000 claims description 14
- 238000013507 mapping Methods 0.000 claims description 7
- 238000005070 sampling Methods 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 4
- 230000010354 integration Effects 0.000 claims description 3
- 238000006243 chemical reaction Methods 0.000 claims 1
- 230000001537 neural effect Effects 0.000 claims 1
- 238000012360 testing method Methods 0.000 abstract description 9
- 238000009826 distribution Methods 0.000 abstract description 7
- 238000004364 calculation method Methods 0.000 abstract description 3
- 238000011156 evaluation Methods 0.000 abstract 1
- 238000005457 optimization Methods 0.000 description 5
- 238000000605 extraction Methods 0.000 description 3
- 230000007547 defect Effects 0.000 description 2
- 238000010891 electric arc Methods 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/08—Locating faults in cables, transmission lines, or networks
- G01R31/088—Aspects of digital computing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2113/00—Details relating to the application field
- G06F2113/04—Power grid distribution networks
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
- Y04S10/52—Outage or fault management, e.g. fault detection or location
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Data Mining & Analysis (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Geometry (AREA)
- Computer Hardware Design (AREA)
- Locating Faults (AREA)
Abstract
The invention discloses a rapid line selection method for small-current ground faults, belongs to the technical field of power fault processing, and solves the problems of low line selection accuracy and low line selection speed in single-phase ground fault line selection in the prior art. The invention comprises the following steps: step 1: collecting electric information quantity; step 2: signal processing is carried out, and steady-state and transient fault characteristics are extracted; step 3: acquiring harmonic amplitude values; solving transient zero sequence active power integral components; acquiring line energy based on variation modal decomposition; carrying out normalization processing on the data to obtain a sample data set; step 4: training by using the sample data set obtained in the step 3 to obtain a trained optimal neural network model, and inputting a test sample into the trained model to judge whether low-current grounding occurs or not. According to the method, the calculation efficiency of small-current ground fault evaluation analysis of the power distribution network is improved, the rapid line selection efficiency of the small-current ground fault is improved, and the operation supporting capacity of the power distribution network is improved.
Description
Technical Field
The invention relates to the technical field of power fault processing, in particular to a rapid line selection method for a small-current ground fault.
Background
The power distribution network is used as a network connected with users to directly distribute electric energy, and the safety and reliability of the power distribution network play a vital role in the production and development of national economy. At present, a low-current grounding system is a main grounding mode of a power distribution network, and fault line selection of the low-current grounding system is always a very difficult problem in a power system. And a small-current grounding system fault line selection device is additionally arranged at the station end, line selection is performed by utilizing signal characteristics in a transient period, and the reliability of power supply of the power distribution network is ensured. However, the transient process has short duration and difficult extraction of fault signals. The method for carrying out fusion line selection on multiple fault characteristics can overcome the defect of a single line selection mode. After the data characteristics of the system during faults are obtained, the conventional data optimization algorithm such as a BP neural network, a support vector machine and the like have the problems of low iteration speed, easiness in sinking into local optimum and the like, and low line selection accuracy and low line selection speed are caused. The low accuracy of line selection and the slow line selection speed can lead to the system to be maintained in a fault state for a long time, and the grounding current generated during single-phase grounding can form an electric arc at a grounding point, and if the electric arc is not cleared timely, equipment can be burnt out, so that economic loss and even safety accidents are caused, and therefore the accuracy and the rapidity of line selection are required to be improved urgently.
Disclosure of Invention
Aiming at the problems existing in the prior art, the invention provides a rapid line selection method for small-current ground faults, which aims at: the method solves the problems of low line selection accuracy and low line selection speed of single-phase grounding fault line selection in the prior art.
The technical scheme adopted by the invention is as follows:
a rapid line selection method for a small-current ground fault comprises the following steps:
step 1: when the small-current grounding system fails, the failure acquisition device is used for recording the electrical information quantity after various failures under different conditions. The step of extracting the zero sequence current and bus zero sequence voltage signals of each circuit is as follows: when the low-current grounding system fails, the fault acquisition device is used for recording the electrical information quantity after various faults under different conditions, wherein the electrical information quantity comprises zero-sequence current signals of each branch feeder line in a steady state, zero-sequence current signals of each branch feeder line in a transient state and zero-sequence voltage signals of a bus.
Step 2: after the zero sequence current and zero sequence voltage signals are obtained, signal processing is carried out to extract the steady-state and transient fault characteristics of the circuit.
Step 3: performing fast Fourier transform on the extracted steady-state component of the zero sequence current to obtain the amplitude I of the fifth harmonic component of the jth line in steady state j250HZ . Taking the ratio of the amplitude of the fifth harmonic component of each line to the sum of the amplitudes of the fifth harmonic components of all lines as a fault characteristic quantity F 1 。
In the formula (1), I j,250HZ For the fifth harmonic component amplitude of line j,
is the sum of the magnitudes of the fifth harmonic components of all lines.
The zero sequence voltage and the zero sequence current component are multiplied and then integrated in transient state, the characteristic quantity is amplified, the characteristic value difference between a fault line and a normal line is favorably distinguished, and the calculation formula is as follows:
in the formula (2), V (T) is a zero-sequence voltage at the time T, I (T) is a zero-sequence active current at the time T, and T represents an integration period and is taken as a cycle after a fault.
Further, the fault characteristic data normalization processing method comprises the following steps: assuming the transient zero sequence active power integral component of the jth line to be P j The ratio of the value to the sum of the integral components of all lines is taken as a fault characteristic quantity F 2 The calculation formula is:
in the formula (3), P j For the transient zero sequence active power integral value of line j,
and the sum of transient zero sequence active power integral values of all lines.
And acquiring line energy based on VMD, performing VMD decomposition on the extracted line transient zero sequence current signal under the condition that the decomposition number K is 2, and obtaining the line energy after VMD decomposition. And taking the ratio of the line energy value to the sum of all line energy values as a fault characteristic value F 3 。
Further, line energy is obtained based on VMD, VMD decomposition is carried out on the extracted transient zero sequence current signal, and under the condition that the number of decomposition K is 2, zero sequence current is decomposed into two IMF functions and then recorded as d 1(t) ，d 2(t) The band energy of the jth IMF component is:
in the formula (4), k is the number of sampling points, n is the sampling data length, and j=1, 2; and adding the energy of the frequency bands of the two IMF components to obtain the energy value corresponding to the line (outgoing line l).
Adding the energy of the frequency bands of the two IMF components to obtain the energy value corresponding to the line (outgoing line l):
E l ＝E 1 +E 2 (5)
further, the ratio of the energy value of the line l to the sum of all the line energy values is used as an energy proportion value as a third fault characteristic F in line fault 3 ：
In the formula (6), E l As the energy value of the line l,
is the sum of the energy values of all the lines.
Step 4: according to the fault characteristic quantity F of the optimized BP neural network 1 ，F 2 ，F 3 And processing and selecting lines.
Further, before the BP neural network is optimized by using the whale algorithm, the whale algorithm is optimized, global search diversity of the whale algorithm is enhanced, and the optimization steps of the method comprise chaotic mapping and inertial weight optimization are as follows:
the initial population mode of the system is improved by adopting the cubic map chaotic operation as chaotic map, and the expression of the cubic map chaotic map is as follows:
in the formula (7), ρ is a control parameter, which is set to 1; y is k Secondly, introducing an inertial weight formula for balancing global searching capability and later local searching capability of a whale algorithm for the whale population sequence after the kth iteration, wherein the inertial weight formula is as follows:
ω(t)＝ω min +(ω max -ω min )×m×exp(-t/M) (8)
in the formula (8), ω (t) is an inertia weight value after t times of iteration, m is an adjustment coefficient, ω min And omega max The initial minimum and maximum weights are respectively, and M is the maximum iteration number.
Further, the BP neural network is optimized by using an optimized whale algorithm (C-I-WOA), the initial weight and the threshold (C-I-WOA-BP) of the BP neural network are optimized by optimizing the whale algorithm, so that the BP neural network has the optimal initial weight and the threshold, the problem that the BP neural network is easy to fall into local optimum can be solved, and then the part of the obtained fault characteristic quantity is used as training sample data to be input into the optimized BP neural network for training, and the optimal C-I-WOA-BP neural network model is obtained through training. And sending the line characteristic state to be tested into an optimized C-I-WOA-BP neural network model, and judging whether small current grounding occurs or not.
In summary, the beneficial effects of the invention are as follows:
the C-I-WOA-BP neural network model can overcome the defects and limitations of the traditional single BP neural network, and the optimized algorithm can solve the problems that the BP neural network is easy to fall into local optimum and has low iteration speed; after the small-current grounding system breaks down, the method can be used for improving the accuracy rate and the line selection speed of fault line selection.
Drawings
The invention will now be described by way of example and with reference to the accompanying drawings in which:
FIG. 1 is a schematic flow chart of the method of the present invention;
the diagram is: the method comprises the steps of 1-line ground fault, 2-extraction of zero sequence current and bus zero sequence voltage signals of each line, 3-steady state transient fault feature extraction, 4-FFT acquisition of line fifth harmonic amplitude, 5-acquisition of transient zero sequence active power integral quantity, 6-acquisition of line energy based on VMD, 7-data normalization processing acquisition of a sample set, 8-training set, 9-chaotic mapping equation, 10-nonlinear self-adaptive weight, 11-optimization of initial population and inertia weight (C-I-WOA) of WOA algorithm, 12-C-I-WOA optimization of BP neural network, 13-training acquisition of optimal C-I-WOA-BP neural network model, 14-test set, 15-test of trained model and 16-judgment of whether low current ground connection occurs.
Detailed Description
For the purposes of making the objects, technical solutions and advantages of the embodiments of the present application more clear, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is apparent that the described embodiments are only some embodiments of the present application, but not all embodiments. The components of the embodiments of the present application, which are generally described and illustrated in the figures herein, may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present application, as provided in the accompanying drawings, is not intended to limit the scope of the application, as claimed, but is merely representative of selected embodiments of the application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present application without making any inventive effort, are intended to be within the scope of the present application.
In the description of the embodiments of the present application, it should be noted that, directions or positional relationships indicated by terms such as "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., are directions or positional relationships based on those shown in the drawings, or those that are conventionally put in use of the inventive product, are merely for convenience of description and simplicity of description, and are not indicative or implying that the apparatus or element to be referred to must have a specific direction, be configured and operated in a specific direction, and thus should not be construed as limiting the present application. Furthermore, the terms "first," "second," "third," and the like are used merely to distinguish between descriptions and should not be construed as indicating or implying relative importance.
The present invention is described in detail below with reference to fig. 1.
Examples:
the invention discloses a rapid line selection method for a small-current ground fault, which is realized by the following steps:
as shown in fig. 1, in the figure, 1 is that a ground fault occurs on a line, and 2 is that zero sequence current and bus zero sequence voltage signals of each line are extracted: when the ground fault of the power distribution network is monitored, the zero sequence current signals of all lines and the zero sequence voltage signals of the bus are acquired by the acquisition device at the first time.
And 3, extracting steady-state transient fault characteristics: after the zero sequence current and zero sequence voltage signals are acquired, the signals are processed to obtain three characteristic value inputs of the line, wherein the three characteristic inputs are respectively a fifth harmonic amplitude ratio of the line, a transient zero sequence active power integral ratio and a VMD-based line energy ratio, and meanwhile, the characteristic value output of the line is set to 0 or 1, wherein the output 0 indicates that the line is a normal line, and the 1 indicates that the line is a fault line so as to facilitate subsequent line selection.
4, obtaining line fifth harmonic amplitude for FFT: collecting zero-sequence current steady-state signals of each line after faults by using a collecting device, carrying out FFT (fast Fourier transform) on each signal to obtain fifth harmonic wave amplitude values, and taking the ratio of the fifth harmonic wave component amplitude value of each line to the sum of the fifth harmonic wave component amplitude values of all lines as a first fault characteristic quantity F 1 ：
In the formula (1), I j,250HZ For the fifth harmonic component amplitude of line j,
is the sum of the magnitudes of the fifth harmonic components of all lines.
And 5, acquiring transient zero sequence active power integral quantity: when the acquisition device is used for extracting the transient zero-sequence current of each line and the zero-sequence voltage of the bus, multiplying the zero-sequence current of each line by the zero-sequence voltage of the bus, and then integrating, namely:
in the formula (2), V (T) is a zero-sequence voltage at the time T, I (T) is a zero-sequence active current at the time T, and T represents an integration period and is taken as a cycle after a fault.
Thereby obtaining the transient zero sequence active power integral value of each line after the fault and combining the transient zero sequence active power of each lineThe ratio of the power integral value to the sum of the transient zero-sequence active power integral values of all lines is taken as a second fault characteristic quantity F 2 ：
In the formula (3), P j For the transient zero sequence active power integral value of line j,
and the sum of transient zero sequence active power integral values of all lines.
And 6, obtaining line energy based on VMD: after decomposing the zero sequence current into two IMF functions, the IMF functions are marked as d 1(t) ，d 2(t) The band energy of the jth IMF component is:
in the formula (4), k is the number of sampling points, n is the sampling data length, and j=1, 2. The energy of the frequency bands of the two IMF components is added to obtain the corresponding energy value of the line (outgoing line l):
E l ＝E 1 +E 2 (5)
the ratio of the energy value of line/and the sum of all line energy values is used as the energy proportion value to be used as the third fault characteristic quantity F in line fault 3 ：
In the formula (6), E l As the energy value of the line l,
is the sum of the energy values of all the lines.
And 7, acquiring a sample set for data normalization processing: after the data features are obtained by the method, normalization processing is needed for the data due to differentiation among the data features, and the normalization standard is the sum of the corresponding feature values of the lines.
8 is training set: after normalizing the data set, the normalized data set should be divided into a training set, wherein the data volume of the training set accounts for 80% of all data, and the training set is used for training a subsequent neural network, so that an optimal model is obtained by training for line selection.
9 is a chaotic mapping equation: according to the invention, an improved whale algorithm is adopted to optimize the initial weight and the threshold value of the BP neural network, a large number of chaotic initial populations with diversity are generated by using chaotic variables, then a population with a good fitness value is selected as the whale algorithm initial population, so that the searching efficiency is improved, and a cubic map chaotic operation formula is selected on a chaotic mapping formula to improve the whale algorithm initial population. The expression of the cubic map chaotic map is:
in the formula (7), ρ is a control parameter, which is set to 1; y is k For the sequence of whale population after the kth iteration,
10 is nonlinear inertial adaptive weight: for whale algorithm, the inertial weight has great influence on the convergence speed and the global optimizing capability. Secondly, introducing an inertial weight formula to balance the global searching capability and the later local searching capability of a whale algorithm, wherein the inertial weight formula is as follows:
ω(t)＝ω min +(ω max -ω min )×m×exp(-t/M) (8)
in the formula (8), ω (t) is an inertia weight value after t times of iteration, m is an adjustment coefficient, ω min And omega max The initial minimum weight and the maximum weight are respectively, M is the maximum iteration number, and as can be known from the formula (8), the early stage of iteration is kept at a smaller value along with the iteration number t, the self-adaptive weight omega is kept at a higher value, and the global searching capability of the algorithm can be increased; with the increase of the later period t, omega gradually decreases, and the local part of the algorithm can be improvedOptimizing capability.
11 is the initial population and inertial weights (C-I-WOA) to optimize the WOA algorithm: and optimizing the initial population and the inertia weight of the WOA algorithm by using the chaotic mapping equation and the inertia weight to obtain an optimized whale algorithm (C-I-WOA).
And 12, optimizing the BP neural network by using an optimized whale algorithm (C-I-WOA), namely, optimizing an initial weight and a threshold value of the BP neural network to form a C-I-WOA-BP model, and accelerating the training precision and training speed of the neural network.
13 to obtain the optimal C-I-WOA-BP neural network model: and inputting the training set in the sample into a C-I-WOA-BP neural network for training, and obtaining an optimal network model through training convergence of the neural network.
14 is the test set: the normalized data set is divided into a training set and a test set, wherein the training set is used for training the front neural network, and the rest part of data features are used as the test set.
15 is to test the trained model: and inputting the test set into the optimal C-I-WOA-BP neural network model obtained by the previous training for testing, and evaluating the reliability of the network model optimized before.
16 is a determination of whether a low current ground occurs: and sending the line characteristic state to be tested into an optimized C-I-WOA-BP neural network model, and judging whether small current grounding occurs or not.
The foregoing examples merely represent specific embodiments of the present application, which are described in more detail and are not to be construed as limiting the scope of the present application. It should be noted that, for those skilled in the art, several variations and modifications can be made without departing from the technical solution of the present application, which fall within the protection scope of the present application.
Claims (10)
1. The rapid line selection method for the low-current ground fault is characterized by comprising the following steps of:
step 1: when the small-current grounding system fails, the fault acquisition device is utilized to record the electrical information quantity of each line after the fault under various conditions;
step 2: after acquiring zero sequence current and zero sequence voltage signals of each line based on the step 1, performing signal processing, and extracting steady-state and transient fault characteristics;
step 3: extracting the amplitude of the fifth harmonic component of each line by FFT conversion, and taking the ratio of the amplitude of the fifth harmonic component of each line to the sum of the amplitudes of the fifth harmonic components of all lines as a fault characteristic quantity F 1 ；
The integral component of the transient zero-sequence active power is obtained, namely, the zero-sequence voltage and the zero-sequence current component in transient state are multiplied and then integrated; the ratio of the value to the sum of the integral components of all lines is taken as a fault characteristic quantity F 2 ；
VMD decomposition is carried out on the extracted transient zero sequence current signals, under the condition that the decomposition number K is 2, the transient zero sequence current signals of the ith line are subjected to VMD decomposition, the energy ratio of the energy of the ith line to the energy sum of all lines is calculated, and the fault characteristic quantity F is obtained 3 ；
Will fail characteristic quantity F 1 、F 2 、F 3 Carrying out normalization processing to obtain a sample data set;
step 4: optimizing parameters of the C-I-WOA-BP neural network model, training by using the sample data set obtained in the step 3 to obtain an optimized C-I-WOA-BP neural network model, sending a line characteristic state to be tested into the optimized C-I-WOA-BP neural network model, and judging whether low-current grounding occurs or not.
2. The method for fast selecting a small current ground fault according to claim 1, wherein the step 1 is specifically: the acquisition device records the electrical information quantity after faults under various conditions, wherein the electrical information quantity comprises zero sequence current signals of each branch feeder line in a steady state, and zero sequence current signals and bus zero sequence voltage signals in a transient state.
3. The method for rapid line selection of small current ground fault according to claim 1, wherein the fault characteristic quantity F in the step 3 1 The specific acquisition process of (1) is as follows: by means of a pick-up deviceCollecting zero sequence current steady-state signals of each line after faults, performing FFT (fast Fourier transform) on each signal to obtain fifth harmonic amplitude values, and taking the ratio of the fifth harmonic component amplitude value of each line to the sum of the fifth harmonic component amplitude values of all lines as a fault characteristic quantity F 1 。
4. A method for rapid line selection for a low current ground fault according to claim 1 or 3, characterized in that the fault characteristic quantity F 1 The specific mathematical expression of (2) is:
in the formula (1), I j,250HZ For the fifth harmonic component amplitude of line j,
is the sum of the magnitudes of the fifth harmonic components of all lines.
5. The method for rapid line selection of small current ground fault according to claim 1, wherein the fault characteristic quantity F in the step 3 2 The specific acquisition process of (1) is as follows: when the acquisition device is used for extracting the transient zero-sequence current of each line and the zero-sequence voltage of the bus, multiplying the zero-sequence current of each line by the zero-sequence voltage and integrating to obtain the transient zero-sequence active power integral value of the line; taking the ratio of the transient zero-sequence active power integral value of each line to the sum of the transient zero-sequence active power integral values of all lines as a fault characteristic quantity F 2 。
6. The rapid line selection method for small current grounding faults according to claim 1 or 5, wherein the specific mathematical expression of multiplying and re-integrating the zero sequence active current and the zero sequence voltage of each line is as follows:
in the formula (2), V (T) is a zero-sequence voltage at the time T, I (T) is a zero-sequence active current at the time T, and T represents an integration period and is taken as a cycle after a fault.
7. A method for rapid line selection for a low current ground fault according to claim 1 or 5, characterized in that the fault characteristic quantity F 2 The specific mathematical expression of (2) is:
in the formula (3), P j For the transient zero sequence active power integral value of line j,
and the sum of transient zero sequence active power integral values of all lines.
8. The rapid line selection method for small current ground faults according to claim 1, characterized in that in the step 3, the characteristic quantity F 3 The specific acquisition process of (1) is as follows: VMD decomposition is carried out on the extracted transient zero sequence current signal, and under the condition that the decomposition number K is 2, the zero sequence current is decomposed into two IMF functions and then recorded as d 1 (t)，d 2 (t) the band energy of the jth IMF component is:
in the formula (4), k is the number of sampling points, n is the sampling data length, and j=1, 2; adding the energy of the frequency bands of the two IMF components to obtain an energy value corresponding to the line;
E l ＝E 1 +E 2 (5)
the ratio of the energy value of each line to the sum of all line energy valuesAs the energy proportion value, the energy proportion value is used as the fault characteristic quantity F in the line fault 3 。
9. A method for rapid line selection for small current ground fault according to claim 1 or 8, characterized in that the fault characteristic quantity F 3 The specific mathematical expression of (2) is:
in the formula (6), E l As the energy value of the line l,
is the sum of the energy values of all the lines.
10. The method for fast selecting a small current ground fault according to claim 1, wherein the step 6 is specifically: firstly, adopting a cubic map chaotic operation as chaotic mapping to improve an initial population mode of a whale algorithm, wherein the expression of the cubic map chaotic mapping is as follows:
in the formula (7), ρ is a control parameter, which is set to 1; y is k Secondly, introducing an inertial weight formula for balancing global searching capability and later local searching capability of a whale algorithm for the whale population sequence after the kth iteration, wherein the inertial weight formula is as follows:
ω(t)＝ω min +(ω max -ω min )×m×exp(-t/M) (8)
in the formula (8), ω (t) is an inertia weight value after t times of iteration, m is an adjustment coefficient, ω min And omega max Respectively an initial minimum weight and a maximum weight, wherein M is the maximum iteration number;
as can be seen from the formula (8), in the initial stage of iteration, the weight value omega (t) is large, so that the convergence rate in the early stage is accelerated; in the later iteration stage, the weight value omega (t) is small, and the later search precision is increased;
optimizing the BP neural network by using an optimized whale algorithm (C-I-WOA), continuously optimizing the initial weight and the threshold value of the BP neural network by optimizing the whale algorithm, so that the BP neural has the optimal initial weight and the threshold value (C-I-WOA-BP), and further inputting part of the obtained fault characteristic value as training sample data into the optimized BP neural network for training, and obtaining an optimal C-I-WOA-BP neural network model by training; and sending the line characteristic state to be tested into an optimized C-I-WOA-BP neural network model, and judging whether small current grounding occurs or not.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310386910.7A CN116413553A (en) | 2023-04-12 | 2023-04-12 | Rapid line selection method for small-current ground fault |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310386910.7A CN116413553A (en) | 2023-04-12 | 2023-04-12 | Rapid line selection method for small-current ground fault |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116413553A true CN116413553A (en) | 2023-07-11 |
Family
ID=87059298
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310386910.7A Pending CN116413553A (en) | 2023-04-12 | 2023-04-12 | Rapid line selection method for small-current ground fault |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116413553A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117892250A (en) * | 2024-03-18 | 2024-04-16 | 青岛鼎信通讯股份有限公司 | Single-phase earth fault positioning method based on fault characteristics and BP neural network |
-
2023
- 2023-04-12 CN CN202310386910.7A patent/CN116413553A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117892250A (en) * | 2024-03-18 | 2024-04-16 | 青岛鼎信通讯股份有限公司 | Single-phase earth fault positioning method based on fault characteristics and BP neural network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107803350B (en) | A kind of method of lithium battery automatic sorting, storage medium and battery sorting device | |
| CN107609569B (en) | Power distribution network ground fault positioning method based on multi-dimensional feature vectors | |
| CN109307824B (en) | Clustering-based power distribution network single-phase earth fault section positioning method | |
| CN109359271B (en) | Transformer winding deformation degree online detection method based on logistic regression | |
| Wu et al. | Induction machine fault detection using SOM-based RBF neural networks | |
| CN103115789B (en) | Second generation small-wave support vector machine assessment method for damage and remaining life of metal structure | |
| CN110994604B (en) | Power system transient stability assessment method based on LSTM-DNN model | |
| CN109283432B (en) | Method and device for analyzing fault section positioning based on spectral sequence kurtosis | |
| CN113391164A (en) | Intelligent identification method and device for single-phase earth fault of power distribution network | |
| CN109188210A (en) | A kind of urban electric power cable Two-terminal Fault Location method based on VMD-Hilbert transformation | |
| CN116413553A (en) | Rapid line selection method for small-current ground fault | |
| CN109657720B (en) | On-line diagnosis method for turn-to-turn short circuit fault of power transformer | |
| CN110703078A (en) | GIS fault diagnosis method based on spectral energy analysis and self-organizing competition algorithm | |
| CN111157843B (en) | Power distribution network line selection method based on time-frequency domain traveling wave information | |
| CN112485590A (en) | Power distribution network single-phase line-breaking fault identification method | |
| CN108004565A (en) | Full distributed the phonetic warning method and its system of a kind of aluminium cell | |
| CN113237619B (en) | Fault early warning method, device, equipment and storage medium for variable-speed rotating machinery vibration | |
| CN110824299B (en) | Fault line selection method based on two-dimensional plane judgment of zero-sequence current curve cluster | |
| CN109872511B (en) | Self-adaptive two-stage alarm method for monitoring axial displacement sudden change | |
| CN113883014B (en) | Method, device and equipment for detecting unbalance of wind turbine generator impeller and storage medium | |
| CN111695543A (en) | Method for identifying hidden danger discharge type of power transmission line based on traveling wave time-frequency characteristics | |
| CN114002550B (en) | Direct-current power distribution network ground fault line selection method and system | |
| CN114113882B (en) | Power transmission line fault positioning method and system based on fuzzy calculation | |
| Wu et al. | Research on short-circuit fault-diagnosis strategy of lithium-ion battery in an energy-storage system based on voltage cosine similarity | |
| CN114062845A (en) | Line fault detection method and device and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |