CN117725397A - Partial discharge characteristic extraction method for switch cabinet - Google Patents
Partial discharge characteristic extraction method for switch cabinet Download PDFInfo
- Publication number
- CN117725397A CN117725397A CN202311686411.6A CN202311686411A CN117725397A CN 117725397 A CN117725397 A CN 117725397A CN 202311686411 A CN202311686411 A CN 202311686411A CN 117725397 A CN117725397 A CN 117725397A
- Authority
- CN
- China
- Prior art keywords
- partial discharge
- characteristic
- cat
- switch cabinet
- parameter
- 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
- 238000000605 extraction Methods 0.000 title claims abstract description 28
- 241000879809 Felis margarita Species 0.000 claims abstract description 52
- 238000000034 method Methods 0.000 claims abstract description 38
- 238000004458 analytical method Methods 0.000 claims abstract description 28
- 230000008569 process Effects 0.000 claims abstract description 23
- 238000010168 coupling process Methods 0.000 claims abstract description 21
- 238000012545 processing Methods 0.000 claims abstract description 21
- 238000001514 detection method Methods 0.000 claims abstract description 20
- 230000008878 coupling Effects 0.000 claims abstract description 16
- 238000005859 coupling reaction Methods 0.000 claims abstract description 16
- ZINJLDJMHCUBIP-UHFFFAOYSA-N ethametsulfuron-methyl Chemical compound CCOC1=NC(NC)=NC(NC(=O)NS(=O)(=O)C=2C(=CC=CC=2)C(=O)OC)=N1 ZINJLDJMHCUBIP-UHFFFAOYSA-N 0.000 claims abstract description 15
- 239000000284 extract Substances 0.000 claims abstract description 8
- 238000001914 filtration Methods 0.000 claims abstract description 4
- 230000035945 sensitivity Effects 0.000 claims description 21
- 241000282326 Felis catus Species 0.000 claims description 20
- 230000006870 function Effects 0.000 claims description 19
- 238000013507 mapping Methods 0.000 claims description 11
- 238000013459 approach Methods 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 7
- 238000012544 monitoring process Methods 0.000 claims description 7
- 239000003795 chemical substances by application Substances 0.000 claims description 6
- 230000000739 chaotic effect Effects 0.000 claims description 5
- 238000011161 development Methods 0.000 claims description 5
- 238000012549 training Methods 0.000 claims description 5
- 230000007704 transition Effects 0.000 claims description 4
- 102100029469 WD repeat and HMG-box DNA-binding protein 1 Human genes 0.000 claims description 3
- 101710097421 WD repeat and HMG-box DNA-binding protein 1 Proteins 0.000 claims description 3
- 230000005540 biological transmission Effects 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 3
- 230000003247 decreasing effect Effects 0.000 claims description 3
- 238000012804 iterative process Methods 0.000 claims description 3
- 238000004519 manufacturing process Methods 0.000 claims description 3
- 230000007246 mechanism Effects 0.000 claims description 3
- 238000012546 transfer Methods 0.000 claims description 3
- 230000003044 adaptive effect Effects 0.000 claims description 2
- 241001465754 Metazoa Species 0.000 abstract description 3
- 238000005516 engineering process Methods 0.000 description 4
- 238000003745 diagnosis Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- NAWXUBYGYWOOIX-SFHVURJKSA-N (2s)-2-[[4-[2-(2,4-diaminoquinazolin-6-yl)ethyl]benzoyl]amino]-4-methylidenepentanedioic acid Chemical compound C1=CC2=NC(N)=NC(N)=C2C=C1CCC1=CC=C(C(=O)N[C@@H](CC(=C)C(O)=O)C(O)=O)C=C1 NAWXUBYGYWOOIX-SFHVURJKSA-N 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000009827 uniform distribution Methods 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a partial discharge characteristic extraction method of a switch cabinet, which comprises the steps of collecting electric signals generated by the switch cabinet, performing filtering treatment, performing parameter analysis on a partial discharge coupling process of the switch cabinet, constructing a parameter identification model according to parameter constraint conditions in the operation process of the switch cabinet, taking characteristic parameters and characteristic constants as keys for partial discharge signal extraction, constructing a fuzzy information analysis model by acquiring the characteristic constants and identifying various characteristic parameters, and introducing a salsa algorithm to further extract key information and characteristics of the processed information; and the extracted partial discharge signal characteristic signals are represented by a partial discharge characteristic function, and are subjected to coupling processing, so that the discharge characteristic detection of the switch cabinet is realized. The invention regards the partial discharge signal as a walking path of a 'sand cat' (an animal), extracts the characteristics of the partial discharge signal by simulating the walking path of the sand cat, and evaluates the running state of the switch cabinet.
Description
Technical Field
The invention relates to the technical field of signal feature extraction, in particular to a partial discharge feature extraction method of a switch cabinet.
Background
The feature extraction of the partial discharge signals of the switch cabinet is one of important technologies in the state monitoring of the power equipment. With the complexity of power equipment and the diversification of operating environments, conventional monitoring methods have been difficult to meet the requirements of modern power systems. Therefore, the development of a novel partial discharge monitoring technology and a signal characteristic extraction method has important significance.
The feature extraction of the partial discharge signal of the switch cabinet refers to extracting feature information capable of reflecting the state of equipment, such as discharge intensity, discharge frequency, discharge duration and the like, from the partial discharge signal. Such feature information may provide powerful support for state assessment and fault diagnosis of the device. There are various methods and techniques for extracting the partial discharge signal characteristics of a switchgear. Among them, the time domain analysis method is one of the most commonly used. The method extracts characteristic values such as peak value, mean value, variance and the like of the discharge signal by carrying out time domain waveform analysis on the partial discharge signal. In addition, the frequency domain analysis method is also a common technique, and extracts characteristics such as frequency components and energy distribution in the partial discharge signal by converting the partial discharge signal into a frequency domain signal. In addition to time and frequency domain analysis methods, there are many other partial discharge signal feature extraction methods, such as wavelet transform, short-time fourier transform, empirical mode decomposition, etc. The methods have advantages and disadvantages and are suitable for different application scenes and equipment types. In summary, the extraction of the partial discharge signal characteristics of the switch cabinet is one of key technologies for monitoring the state of the power equipment. By selecting a proper feature extraction method and a proper processing technology, feature information reflecting the state of the equipment can be effectively extracted, and powerful support is provided for state evaluation and fault diagnosis of the power equipment.
The existing method for extracting the partial discharge signal characteristics of the switch cabinet is easy to receive obstacle interference, has insufficient convergence speed and cannot well adapt to different path planning problems.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides a partial discharge characteristic extraction method of a switch cabinet.
The present invention is so achieved. A partial discharge characteristic extraction method of a switch cabinet comprises the following steps:
step 1, collecting an electric signal generated by a switch cabinet, performing filtering treatment, performing parameter analysis on a partial discharge coupling process of the switch cabinet, constructing a parameter identification model according to parameter constraint conditions in the operation process of the switch cabinet, taking characteristic parameters and characteristic constants as keys for extracting partial discharge signals, effectively acquiring the parameter signals through acquiring the characteristic constants, processing the partial discharge signals after the extraction of the partial discharge signals is completed, and constraining the extracted characteristic information during the processing to obtain partial discharge characteristic information;
step 2: in order to obtain a more stable signal characteristic constant, the construction of a fuzzy information analysis model is completed by identifying various characteristic parameters, a salsa algorithm is introduced to further extract key information and characteristics of the processed information, and the extraction process comprises four processes of initializing a population, searching hunting, searching for a second time and attacking hunting;
step 3: the extracted partial discharge signal characteristic signals are represented by partial discharge characteristic functions and are subjected to coupling processing, so that a relatively complete parameter component which can be used for describing the characteristics of the discharge signals is generated, and the parameter component is adjusted to realize the detection of the discharge characteristics of the switch cabinet.
Further preferably, the parameter constraint is expressed as:
in the formula (1): a represents a constraint condition of operation parameters of the switch cabinet; e represents a control parameter, and the value of e is a constant value in normal cases; i represents the harmonic wave generated by the partial discharge signal of the switch cabinet in operation in transmission; delta is expressed as the device electrical signal capacity; t is denoted as the frequency at which the electrical signal is locally generated.
Further preferably, the feature parameters and the feature constants are used as keys for extracting the partial discharge signals, and the feature constants are obtained to effectively obtain the parameter signals; the acquisition process is as follows:
in the formula (2): g represents the characteristics of the electrical signal; f represents electric signal disturbance compensation; s represents the fluctuation amplitude of the signal, and the value of S is controlled between +1 and-1.
Further preferably, the constraint is performed on the extracted characteristic information to obtain partial discharge characteristic information, and the discharge of the control terminal is expressed as:
in the formula (3): h represents a control end discharge signal; epsilon represents the reliability of the parameter; ρ represents a signal blur characteristic constant. And fusing the parameter characteristics with the same performance, constructing a brand new signal, outputting the signal, and finishing the signal processing.
Further preferably, the specific process of step 2 is:
step 2.1 initializing a sandcat population,
X i ＝LB+S i ×(UB-LB) (5)
wherein: x is X i E (0.9,1.08), taking initial value S for S sequence 1 ＝rand[0,1]，S i For the mapping coefficient corresponding to the ith sand cat,to generate three coefficient values with certain randomness through chaotic mapping, S i+1 Mapping coefficient corresponding to ith+1st sand cat, X i For the initial position of the ith cat, UB and LB are the upper and lower boundaries of the variable respectively;
step 2.2, the hunting mechanism of the sand cat depends on low-frequency noise emission; solution expression of each sand cat is X i ＝[x i1 ,x i2 ,…,x id ]；x id SCSO algorithm for the d dimension of the ith cat simulates the hearing ability of the cat in terms of low frequency detection, the cat perceives low frequencies below 2kHz, assuming the general sensitivity range of the catFrom 0 to 2kHz, to improve the searching speed in the initial stage of iteration and the searching precision in the later stage of iteration, the general sensitivity range is +.>Non-linearly decreasing from 2 to 0 as the iterative process proceeds to gradually approach the prey without losing or skipping;
step 2.3, in order to search for prey, assume that the sensitivity range of the sand cat is 2kHz to 0; s is(s) M Simulating acoustic characteristic parameters of the sand cat;
in formula (6): t is the current iteration number, and T is the maximum iteration number;
step 2.4. The final and main parameters of the control exploration and development phase transition are R, due to this adaptive strategy:
in the formula (7): rand (0, 1) is a random number between (0, 1), R is the intervalA random value of (a) is determined; the search space is randomly initialized between defined boundaries and, in a search step,the location update of each current search agent is based on a random location; the search agent explores the new space in the search space;
step 2.5 to avoid trapping in local optima, the sensitivity range is different for each sand cat, defined as:
in formula (8):sensitivity ranges for each cat; furthermore, the->For exploring or utilizing the operation of the phase, but +.>For steering the parameter R to achieve transfer control between these phases;
step 2.6, when the R is more than 1, the salsa executes a search task, and the position Pos is determined according to the optimal solution bc (t) and the current position Pos c (t) sensitivity RangeUpdating the position of the user; enabling the sandcat to find other best possible locations:
in the formula (9): pos (t+1) is the updated position of the salsa;
step 2.7, when the sand cat searches for the prey, after updating the position information, performing secondary search on the individual with the worst fitness value:
x＝r×cos(θ) y＝r×sin(θ) z＝r×θ r＝u×e θv (10)
Pos(t+1)＝Pos bc (t)×x×y×z+Pos c (t) (11)
wherein: x, y, z are three search directions of three-dimensional space coordinates, pos (t+1) is the updated position of the sand cat, r is the radius of the spiral, and θ is a random angle in the range of [0,2 pi ]; u and v are constants related to the spiral shape for controlling the spiral radius;
step 2.8: when |R| is less than or equal to 1, the sand cat carries out attack hunting:
Pos rnd (t)＝|rand(0,1)·Pos bc (t)-Pos c (t)| (12)
in the formula (12): pos rnd (t) is to use the optimal solution position Pos bc (t) and the current position Pos c (t) random positions of production.
Further preferably, assuming that the sensitivity range of the sand cat is a circle, each sand cat is randomly selected by an angle θ using roulette,
the random positions can ensure that the sand cat approaches to the prey, and the random angles are helpful for the algorithm to jump out of local optimum.
Further preferably, the specific process of step 3 is:
step 3.1: constructing an information analysis model aiming at the partial discharge of the switch cabinet, and training and optimizing the model by using a training data set; finally, the optimized information analysis model is applied to information analysis of actual partial discharge data, so that the functions of real-time monitoring and prediction of the information analysis model are realized; the discharge layer in the information analysis model is expressed as y, the value of the corresponding y is 1-0, a partial discharge characteristic function is output, the function is expressed as Z, and the description of Z is expressed by the following calculation formula:
in formula (14): z represents a detection feature output function; omega y Representing the output quantity;
step 3.2: and (3) coupling the partial discharge characteristic function output result, wherein the processing process is as follows (15):
in formula (15):representing the coupling processing result of the output signal; q represents a blurring component; q represents a detection axis; and outputting the detection result, outputting the coupling result, and generating a relatively complete parameter component which can be used for describing the characteristics of the discharge signal after the coupling result is fused, and adjusting the parameter component to realize the detection of the discharge characteristics of the switch cabinet.
The salsa algorithm is one method that may be used to detect and analyze the partial discharge signal of a switchgear. The core idea is to consider the partial discharge signal as a walking path of a 'sand cat' (an animal), and extract the characteristics of the discharge signal by simulating the walking path of the sand cat. Specifically, the algorithm treats the partial discharge signal as a random process and simulates such a random process using the path of travel of the sand cat. By analyzing the walking path of the sand cat, various characteristics of the discharge signal, such as discharge intensity, discharge frequency, discharge duration, and the like, can be extracted. These features can be used to identify and classify different types of partial discharge signals and to evaluate the operating state of the switchgear.
The method has the advantages of strong robustness, high convergence speed, strong adaptability and the like, and when the problem of path planning is solved, the sand cat algorithm has stronger robustness by simulating animal behaviors. The algorithm can find a better path even in the case of a large environmental change or the presence of obstacles. The salsa algorithm gradually optimizes the value of the objective function by continually updating the candidate solutions of the paths. This iterative approach may allow the algorithm to converge to the optimal solution in a shorter time. The sand cat algorithm can be adjusted and optimized according to different environments and task requirements. By adjusting parameters and limiting conditions in the algorithm, the algorithm can be better adapted to different path planning problems.
Drawings
FIG. 1 is a flow chart of a partial discharge feature extraction method of a switchgear of the present invention;
fig. 2 is a flow chart of feature extraction of the salsa algorithm.
Detailed Description
The invention is further elucidated in detail below in connection with the accompanying drawings.
Referring to fig. 1, a partial discharge feature extraction method of a switchgear includes the steps of:
the method comprises the steps of 1, collecting electric signals generated by a switch cabinet, mainly comprising electric signals such as ultrasonic waves and ground electric waves, performing simple filtering and the like on the electric signals, performing parameter analysis on a partial discharge coupling process of the switch cabinet, constructing a parameter identification model according to parameter constraint conditions in the operation process of the switch cabinet, taking characteristic parameters and characteristic constants as keys for extracting partial discharge signals, effectively acquiring the parameter signals through acquiring the characteristic constants, processing the partial discharge signals after the extraction of the partial discharge signals is completed, and constraining the extracted characteristic information during the processing to obtain partial discharge characteristic information;
step 2: in order to obtain a more stable signal characteristic constant, the construction of a fuzzy information analysis model is completed by identifying various characteristic parameters, a salsa algorithm is introduced to further extract key information and characteristics of the processed information, and the extraction process is divided into four processes of initializing population, searching hunting (exploration), secondary searching and attacking hunting (development);
step 3: the extracted partial discharge signal characteristic signals are represented by partial discharge characteristic functions and are subjected to coupling processing, so that a relatively complete parameter component which can be used for describing the characteristics of the discharge signals is generated, and the parameter component is adjusted to realize the detection of the discharge characteristics of the switch cabinet.
The specific process of the step 1 of the invention is as follows:
step 1.1: according to load fluctuation, carrying out parameter analysis on the partial discharge coupling process of the switch cabinet, and constructing a parameter identification model according to parameter constraint conditions in the operation process of the switch cabinet, wherein the parameter constraint conditions are expressed as the following calculation formula (1):
in the formula (1): a represents a constraint condition of operation parameters of the switch cabinet; e represents a control parameter, and the value of e is a constant value in normal cases; i represents the harmonic wave generated by the partial discharge signal of the switch cabinet in operation in transmission; delta is expressed as the device electrical signal capacity; t is denoted as the frequency at which the electrical signal is locally generated.
Step 1.2: under the constraint condition, the characteristic parameters and the characteristic constants are used as keys for extracting the partial discharge signals, and the characteristic constants are obtained, so that the parameter signals are effectively obtained. The acquisition process is shown in the following calculation formula (2):
in the formula (2): g represents the characteristics of the electrical signal; f represents electric signal disturbance compensation; s represents the fluctuation amplitude of the signal, and the value of S is controlled between +1 and-1.
Step 1.3: the effective acquisition of the characteristic parameters is completed, the characteristic parameters are processed, the extracted characteristic information is restrained during the processing, at the moment, partial discharge characteristic information can be obtained, and the discharge of the control end is expressed as the following calculation formula (3):
in the formula (3): h represents a control end discharge signal; epsilon represents the reliability of the parameter; ρ represents a signal blur characteristic constant. And fusing the parameter characteristics with the same performance, constructing a brand new signal, outputting the signal, and finishing the signal processing.
The specific process of the step 2 of the invention is as follows:
step 2.1 in the SCSO algorithm, the population is called a group of sandcats, each sandcat displaying a value of the problem variable. In the detection of the partial discharge of the switch cabinet, the population is the brand new signal which is output after being processed in the step 1, and each cat is acquired to obtain each signal. The algorithm is a population-based method, initial population positions are uniformly distributed in a search space, the global search capability of the algorithm is enhanced, and the search efficiency is improved. The standard sand cat swarm optimization algorithm randomly initializes the population, the risk of reducing the diversity of the population exists, and the chaotic sequence generated by the chaotic mapping has the characteristics of ergodic property, nonlinearity, unpredictability and the like and is commonly used for initializing the population instead of the random sequence. The mapping has the advantages of simple parameters, uniform distribution and the like, initializes the sand cat population,
X i ＝LB+S i ×(UB-LB) (5)
wherein: x is X i E (0.9,1.08), taking initial value S for S sequence 1 ＝rand[0,1]，S i For the mapping coefficient corresponding to the ith sand cat,in order to generate three coefficient values with certain randomness through chaotic mapping, the method is used for initializing the positions of the satay cat groups in the search space, so that the satay cat groups have randomness and are more beneficial to global search. S is S i+1 Mapping coefficient corresponding to ith+1st sand cat, X i For the initial position of the ith cat, UB and LB are the upper and lower boundaries of the variables, respectively.
Step 2.2. A search mechanism for a cat's prey relies on low frequency noise emissions. Solution expression of each sand cat is X i ＝[x i1 ,x i2 ,…,x id ]。x id SCSO algorithm for the d dimension of the ith cat simulates the hearing ability of the cat in terms of low frequency detection, the cat perceives low frequencies below 2kHz, assuming the general sensitivity range of the catFrom 0 to 2kHz, to improve the searching speed in the initial stage of iteration and the searching precision in the later stage of iteration, the general sensitivity range is +.>Non-linearly decreasing from 2 to 0 as the iterative process proceeds to gradually approach the prey without losing or skipping.
Step 2.3 to search for prey, assume that the sensitivity range of the sandcat is 2kHz to 0.s is(s) M To simulate the acoustic feature parameters of a sand cat, its value of inspiration comes from the acoustic feature of the sand cat, assuming a value of 2.
In formula (6): t is the current iteration number, T is the maximum iteration number, s M ＝2。
Step 2.4 the final and main parameters of the control exploration and development phase transition are R, due to which the two phases of transition and possibilities will be more balanced.
In the formula (7): rand (0, 1) is a random number between (0, 1), R is the intervalIs a random value of (a) in the memory. The search space is randomly initialized between defined boundaries and in the search step the location update of each current search agent is based on a random location. In this way, the search agent is able to explore new spaces in the search space.
Step 2.5 to avoid trapping in local optima, the sensitivity range is different for each sand cat, defined as:
in formula (8):sensitivity range for each cat. Furthermore, the->For exploring or utilizing the operation of the phase, but +.>For steering the parameter R to achieve transfer control between these phases.
Step 2.6, when the I R I is more than 1, the salsa executes a search task according to the optimal solution position Pos bc (t) and the current position Pos c (t) sensitivity RangeUpdating its own location. So that the sand cat can find the best possible position.
In the formula (9): pos (t+1) is the updated position of the salsa. The formula provides another opportunity for the algorithm to find new local optima in the search area. Thus, the obtained location is located between the current location and the prey location. Furthermore, this is achieved by randomness, not by an exact method. Thus, the search agent in the algorithm is advantageous for increasing randomness. This results in a low cost and efficient algorithm operation.
Step 2.7, the salsa may not be able to converge to the optimal solution effectively when searching the prey, so that after updating the position information, the individual with the worst fitness value is searched for a second time, and the searching performance of the salsa group optimization algorithm is improved.
x＝r×cos(θ) y＝r×sin(θ) z＝r×θ r＝u×e θv (10)
Pos(t+1)＝Pos bc (t)×x×y×z+Pos c (t) (11)
In the formula (10) and the formula (11): x, y, z are three search directions of three-dimensional space coordinates, pos (t+1) is the updated position of the sand cat, r is the radius of the spiral, and θ is the random angle in the range of [0,2 pi ]. u and v are constants associated with the spiral shape, used to control the spiral radius, here commonly taken as 1; e is the base of the natural logarithm.
Step 2.8: when the R is less than or equal to 1, the sand cat attacks hunting.
Pos rnd (t)＝|rand(0,1)·Pos bc (t)-Pos c (t)| (12)
In the formula (12): pos rnd (t) is to use the optimal solution position Pos bc (t) and the current position Pos c (t) random positions of production.
Further preferably, assuming that the sensitivity range of the sand cat is a circle, each sand cat is randomly selected by an angle θ using roulette,
the random positions can ensure that the sand cat approaches to the prey, and the random angles are helpful for the algorithm to jump out of local optimum.
Further preferably, the specific process of step 3 is
Step 3.1: an information analysis model for the partial discharge of the switchgear is constructed and the model is trained and optimized using a training dataset. And finally, applying the optimized information analysis model to information analysis of actual partial discharge data, and realizing functions of real-time monitoring, prediction and the like of the information analysis model. In the information analysis model, a discharge layer is expressed as y, the value of the corresponding y is 1-0, a partial discharge characteristic function is output, the function is expressed as Z, and the description of Z can be expressed by the following calculation formula:
in formula (14): z represents a detection feature output function; omega y Representing the output.
Step 3.2: and (3) coupling the partial discharge characteristic function output result, wherein the processing process is as follows (15):
in formula (15):representing the coupling processing result of the output signal; q represents a blurring component; q represents the detection axis. According to the calculation formula, the detection result is output, and the coupling result is output, so that a relatively complete parameter component which can be used for describing the characteristics of the discharge signal is generated after the coupling result is fused, and the parameter component is adjusted, so that the discharge characteristic detection of the switch cabinet is realized.
It will be appreciated by persons skilled in the art that the foregoing description is a preferred embodiment of the invention, and is not intended to limit the invention, but rather to limit the invention to the specific embodiments described, and that modifications may be made to the technical solutions described in the foregoing embodiments, or equivalents may be substituted for elements thereof, for the purposes of those skilled in the art. Modifications, equivalents, and alternatives falling within the spirit and principles of the invention are intended to be included within the scope of the invention.
Claims (7)
1. The partial discharge characteristic extraction method of the switch cabinet is characterized by comprising the following steps of:
step 1, collecting an electric signal generated by a switch cabinet, performing filtering treatment, performing parameter analysis on a partial discharge coupling process of the switch cabinet, constructing a parameter identification model according to parameter constraint conditions in the operation process of the switch cabinet, taking characteristic parameters and characteristic constants as keys for extracting partial discharge signals, effectively acquiring the parameter signals through acquiring the characteristic constants, processing the partial discharge signals after the extraction of the partial discharge signals is completed, and constraining the extracted characteristic information during the processing to obtain partial discharge characteristic information;
step 2: in order to obtain a more stable signal characteristic constant, the construction of a fuzzy information analysis model is completed by identifying various characteristic parameters, a salsa algorithm is introduced to further extract key information and characteristics of the processed information, and the extraction process comprises four processes of initializing a population, searching hunting, searching for a second time and attacking hunting;
step 3: the extracted partial discharge signal characteristic signals are represented by partial discharge characteristic functions and are subjected to coupling processing, so that a relatively complete parameter component which can be used for describing the characteristics of the discharge signals is generated, and the parameter component is adjusted to realize the detection of the discharge characteristics of the switch cabinet.
2. The partial discharge feature extraction method of a switchgear cabinet according to claim 1, characterized in that the parameter constraints are expressed as:
in the formula (1): a represents a constraint condition of operation parameters of the switch cabinet; e represents a control parameter, and the value of e is a constant value in normal cases; i represents the harmonic wave generated by the partial discharge signal of the switch cabinet in operation in transmission; delta is expressed as the device electrical signal capacity; t is denoted as the frequency at which the electrical signal is locally generated.
3. The partial discharge characteristic extraction method of the switch cabinet according to claim 2, wherein the characteristic parameters and the characteristic constants are used as keys for extracting the partial discharge signals, and the parameter signals are effectively obtained through obtaining the characteristic constants; the acquisition process is as follows:
in the formula (2): g represents the characteristics of the electrical signal; f represents electric signal disturbance compensation; s represents the fluctuation amplitude of the signal, and the value of S is controlled between +1 and-1.
4. The method for extracting partial discharge characteristics of a switchgear according to claim 3, wherein the constraint on the extracted characteristic information obtains partial discharge characteristic information, and the control end discharge is expressed as:
in the formula (3): h represents a control end discharge signal; epsilon represents the reliability of the parameter; ρ represents a signal blur feature constant; and fusing the parameter characteristics with the same performance, constructing a brand new signal, outputting the signal, and finishing the signal processing.
5. The partial discharge characteristic extraction method of a switchgear according to claim 1, wherein the specific process of step 2 is:
step 2.1 initializing a sandcat population,
X i ＝LB+S i ×(UB-LB) (5)
wherein: x is X i E (0.9,1.08), taking initial value S for S sequence 1 ＝rand[0,1]，S i For the mapping coefficient corresponding to the ith sand cat,s is three coefficient values generated by chaotic mapping i+1 Mapping coefficient corresponding to ith+1st sand cat, X i For the initial position of the ith cat, UB and LB are the upper and lower boundaries of the variable respectively;
step 2.2 search for hunting of a Sai catThe cable mechanism relies on low frequency noise emissions; solution expression of each sand cat is X i ＝[x i1 ,x i2 ,…,x id ]；x id SCSO algorithm for the d dimension of the ith cat simulates the hearing ability of the cat in terms of low frequency detection, the cat perceives low frequencies below 2kHz, assuming the general sensitivity range of the catFrom 0 to 2kHz, to improve the searching speed in the initial stage of iteration and the searching precision in the later stage of iteration, the general sensitivity range is +.>Non-linearly decreasing from 2 to 0 as the iterative process proceeds to gradually approach the prey without losing or skipping;
step 2.3, in order to search for prey, assume that the sensitivity range of the sand cat is 2kHz to 0; s is(s) M Simulating acoustic characteristic parameters of the sand cat;
in formula (6): t is the current iteration number, and T is the maximum iteration number;
step 2.4. The final and main parameters of the control exploration and development phase transition are R, due to this adaptive strategy:
in the formula (7): rand (0, 1) is a random number between (0, 1), R is the intervalA random value of (a) is determined; randomly initializing search spaces between defined boundaries, wherein in the step of searching, the location update of each current search agent is based on a random location; search agentExploring new space in the search space;
step 2.5 to avoid trapping in local optima, the sensitivity range is different for each sand cat, defined as:
in formula (8):sensitivity ranges for each cat; furthermore, the->For exploring or utilizing the operation of the phase, but +.>For steering the parameter R to achieve transfer control between these phases;
step 2.6, when the I R I is more than 1, the salsa executes a search task according to the optimal solution position Pos bc (t) and the current position Pos c (t) sensitivity RangeUpdating the position of the user; enabling the sandcat to find other best possible locations:
in the formula (9): pos (t+1) is the updated position of the salsa;
step 2.7, when the sand cat searches for the prey, after updating the position information, performing secondary search on the individual with the worst fitness value:
x＝r×cos(θ) y＝r×sin(θ) z＝r×θ r＝u×e θv (10)
Pos(t+1)＝Pos bc (t)×x×y×z+Pos c (t) (11)
wherein: x, y, z are three search directions of three-dimensional space coordinates, pos (t+1) is the updated position of the sand cat, r is the radius of the spiral, and θ is a random angle in the range of [0,2 pi ]; u and v are constants related to the spiral shape for controlling the spiral radius;
step 2.8: when |R| is less than or equal to 1, the sand cat carries out attack hunting:
Pos rnd (t)＝|rand(0,1)·Pos bc (t)-Pos c (t)| (12)
in the formula (12): pos rnd (t) is to use the optimal solution position Pos bc (t) and the current position Pos c (t) random positions of production.
6. The method of claim 5, wherein assuming that the sensitivity range of the cat is a circle, randomly selecting an angle θ for each cat using roulette,
the random positions can ensure that the sand cat approaches to the prey, and the random angles are helpful for the algorithm to jump out of local optimum.
7. The partial discharge characteristic extraction method of a switchgear according to claim 1, wherein the specific process of step 3 is:
step 3.1: constructing an information analysis model aiming at the partial discharge of the switch cabinet, and training and optimizing the model by using a training data set; finally, the optimized information analysis model is applied to information analysis of actual partial discharge data, so that the functions of real-time monitoring and prediction of the information analysis model are realized; the discharge layer in the information analysis model is expressed as y, the value of the corresponding y is 1-0, a partial discharge characteristic function is output, the function is expressed as Z, and the description of Z is expressed by the following calculation formula:
in formula (14): z represents a detection feature output function; omega y Representing the output quantity;
step 3.2: and (3) coupling the partial discharge characteristic function output result, wherein the processing process is as follows (15):
in formula (15):representing the coupling processing result of the output signal; q represents a blurring component; q represents a detection axis; and outputting the detection result, outputting the coupling result, and generating a relatively complete parameter component which can be used for describing the characteristics of the discharge signal after the coupling result is fused, and adjusting the parameter component to realize the detection of the discharge characteristics of the switch cabinet.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311686411.6A CN117725397A (en) | 2023-12-11 | 2023-12-11 | Partial discharge characteristic extraction method for switch cabinet |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311686411.6A CN117725397A (en) | 2023-12-11 | 2023-12-11 | Partial discharge characteristic extraction method for switch cabinet |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117725397A true CN117725397A (en) | 2024-03-19 |
Family
ID=90206359
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311686411.6A Pending CN117725397A (en) | 2023-12-11 | 2023-12-11 | Partial discharge characteristic extraction method for switch cabinet |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117725397A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118094823A (en) * | 2024-04-23 | 2024-05-28 | 北京邮电大学 | Mechanical element parameter optimization method and device and computer product |
-
2023
- 2023-12-11 CN CN202311686411.6A patent/CN117725397A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118094823A (en) * | 2024-04-23 | 2024-05-28 | 北京邮电大学 | Mechanical element parameter optimization method and device and computer product |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117725397A (en) | Partial discharge characteristic extraction method for switch cabinet | |
| US5781698A (en) | Method of autonomous machine learning | |
| CN113406579B (en) | Camouflage interference waveform generation method based on deep reinforcement learning | |
| CN114970378B (en) | Sea clutter sample library construction method based on GAN network | |
| Ajibade et al. | A hybrid chaotic particle swarm optimization with differential evolution for feature selection | |
| Xi et al. | Harmonic estimation in power systems using an optimised adaptive Kalman filter based on PSO‐GA | |
| CN110958263B (en) | Network attack detection method, device, equipment and storage medium | |
| Zhang et al. | MANomaly: Mutual adversarial networks for semi-supervised anomaly detection | |
| Lei et al. | High voltage shunt reactor acoustic signal denoising based on the combination of VMD parameters optimized by coati optimization algorithm and wavelet threshold | |
| Li et al. | Adaptive denoising model for ship-radiated noise based on dynamic weighted filtering | |
| Asaduzzaman et al. | An adversarial approach for intrusion detection using hybrid deep learning model | |
| CN117031194B (en) | Ultrasonic hidden danger detection method and system for power distribution network | |
| CN112487933B (en) | Radar waveform identification method and system based on automatic deep learning | |
| CN117636871A (en) | Optical fiber distributed acoustic sensing signal identification method, system, equipment and medium | |
| Ni et al. | Power quality disturbances identification based on adaptive symplectic geometric mode decomposition and improved marine predators algorithm | |
| CN115356599B (en) | Multi-mode urban power grid fault diagnosis method and system | |
| CN115601653A (en) | High-resolution satellite image matching method | |
| Liu et al. | Swarm intelligence for classification of remote sensing data | |
| CN112564834B (en) | Intelligent cognition and interference method and system for wireless communication system | |
| CN109901030B (en) | Reactor turn-to-turn insulation state monitoring method, system and application | |
| Xie et al. | Denoising of partial discharge signal using rapid sparse decomposition | |
| Jiang et al. | Modulation recognition of underwater acoustic communication signals based on neural architecture search | |
| Serrano et al. | Rta3: A real time adversarial attack on recurrent neural networks | |
| CN118444086B (en) | Fault positioning method and system based on line carrier | |
| Jie et al. | Radiation emitter signal recognition based on VMD and IALO-SVM |
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 |