CN109164382B - Method for diagnosing electrical erosion fault of high-voltage circuit breaker contact - Google Patents
Method for diagnosing electrical erosion fault of high-voltage circuit breaker contact Download PDFInfo
- Publication number
- CN109164382B CN109164382B CN201811053083.5A CN201811053083A CN109164382B CN 109164382 B CN109164382 B CN 109164382B CN 201811053083 A CN201811053083 A CN 201811053083A CN 109164382 B CN109164382 B CN 109164382B
- Authority
- CN
- China
- Prior art keywords
- parameters
- contact
- support vector
- circuit breaker
- vector machine
- 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.)
- Active
Links
Images
Classifications
-
- 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
- G01R31/327—Testing of circuit interrupters, switches or circuit-breakers
- G01R31/3271—Testing of circuit interrupters, switches or circuit-breakers of high voltage or medium voltage devices
- G01R31/3275—Fault detection or status indication
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M13/00—Testing of machine parts
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Testing Electric Properties And Detecting Electric Faults (AREA)
- Testing Relating To Insulation (AREA)
Abstract
The invention relates to the field of circuit breaker diagnosis, and particularly discloses a method for diagnosing electrical erosion faults of a contact of a high-voltage circuit breaker, which comprises the following steps: acquiring contact ablation evaluation parameters of the circuit breaker: resistance-travel curve and static resistance value signals; acquiring a contact ablation state parameter value of the circuit breaker according to the contact ablation evaluation parameter; optimizing parameters of a support vector machine by adopting a bat algorithm to obtain optimal parameters, and establishing an optimal nonlinear support vector machine by adopting the optimal parameters; constructing sample data; training an optimal nonlinear support vector machine by using sample data, inputting contact ablation evaluation parameters, and outputting corresponding contact ablation state parameter values to obtain the nonlinear support vector machine capable of being evaluated; and predicting contact ablation evaluation parameters of the circuit breaker to be evaluated by adopting the trained nonlinear support vector machine. The method can accurately evaluate the electric shock erosion fault of the high-voltage circuit breaker.
Description
Technical Field
The invention belongs to the field of high-voltage circuit breaker diagnosis, and particularly relates to a method for diagnosing electrical erosion faults of a contact of a high-voltage circuit breaker.
Background
The contact resistance generated in the motion process of the contact of the electric switch or the contact resistance generated in the vibration environment during working are both embodied as dynamic contact resistance, and the characteristics of the dynamic contact resistance are the real reflection and the embodiment of the surface electric contact state of the electric switch contact, so the dynamic contact resistance can be used as the main basis for the evaluation of the electric switch contact.
The contact pair of the high-voltage circuit breaker is mainly formed by connecting a main contact and an arc contact in parallel, wherein the main contact bears rated working current, and the arc contact bears arc burning loss. The arcing contact status is the most important factor affecting the electrical life of a high voltage circuit breaker. The overhaul of power equipment in China is in a transition period from planned overhaul to state overhaul, and the state detection of an arc contact is an important part of the state overhaul of an arc extinguishing chamber. The electrical erosion of the arc contact can cause the reduction of the on-off short circuit current capability and the insulation capability of the arc extinguish chamber, and the explosion caused by the failure of the on-off current of the arc extinguish chamber can be caused under extreme conditions, thereby seriously threatening the reliability of a power system. Therefore, the research based on arc contact state detection and electric service life prediction has important theoretical significance and engineering practical value for improving the stability of the power system.
The contact resistance between the contacts of a circuit breaker during closing or opening can also be understood as a function of the contact slip travel. Normally, the contact resistance of a contact of the high-voltage circuit breaker in a closed state is in the magnitude of 10-20 u omega, in order to prevent a measured contact voltage drop signal from being interfered, the high-voltage circuit breaker has good robustness, a constant current source applied by the industrial standard is direct current, and the amplitude is not less than 1000A. Through the test to contact resistance, can assess the contact degree of ablation, realize the condition maintenance of circuit breaker.
For example, patent document No. 201710953217.8 discloses a method for evaluating ablation state of arc contact of circuit breaker based on neural network, which discloses that the ablation state of arc contact is evaluated and judged by using neural network algorithm, but the learning time of neural network algorithm is too long, the efficiency is not high, and the problem that the accuracy is not high due to the fact that the neural network algorithm may fall into local minimum value is solved.
Disclosure of Invention
The invention aims to provide a method for diagnosing electrical erosion faults of a contact of a high-voltage circuit breaker, which is used for accurately evaluating the electrical erosion faults of the contact of the high-voltage circuit breaker by combining an optimized support vector machine.
In order to achieve the above object, the present invention provides a method for diagnosing electrical erosion faults of a contact of a high-voltage circuit breaker, comprising:
s101, after a plurality of circuit breakers operate for a period of time under different voltage currents respectively, acquiring dynamic contact resistance signals and static resistance value signals of moving arc contacts of the circuit breakers to obtain dynamic contact resistance-time curves; acquiring a dynamic stroke when the dynamic contact resistance signal of the moving arc contact occurs to obtain a stroke-time curve;
s102, obtaining a resistance-travel curve according to the dynamic contact resistance-time curve and the travel-time curve;
s103, taking the static resistance value signal and the resistance-stroke curve as contact ablation evaluation parameters, and acquiring contact ablation state parameter values of the circuit breaker according to the contact ablation evaluation parameters;
s104, optimizing parameters of a support vector machine by adopting a bat algorithm to obtain optimal parameters, and establishing an optimal nonlinear support vector machine by adopting the optimal parameters;
s105, taking the contact ablation evaluation parameter and the corresponding contact ablation state parameter value of each breaker as a group of sample data;
s106, training the optimal nonlinear support vector machine by using multiple groups of sample data, inputting the contact ablation evaluation parameters, and outputting corresponding contact ablation state parameter values by the nonlinear support vector machine to obtain the nonlinear support vector machine capable of evaluating the contact ablation fault of the high-voltage circuit breaker;
s107, predicting the contact ablation evaluation parameters of the breaker to be diagnosed by adopting the trained nonlinear support vector machine, and directly evaluating the ablation state according to the output contact ablation state parameter values.
Preferably, in the above technical solution, the step 102 specifically includes:
s201, setting parameters of the support vector machine: penalty parameter C, RBF kernel parameter, parameter range of loss function; initializing bat group related parameters: setting initial population number n and pulse loudness A0Pulse emissivity r0A bat pulse emission rate increasing coefficient gamma, a pulse loudness attenuation coefficient alpha, and bat search pulse frequency upper and lower limits fmin,fmaxMaximum number of iterations tmaxAnd the search precision;
s202, initializing the bat position xiAnd velocity vi；
S203, determine the fitness evaluation function f (x), where x is (x)1,…xd)TAccording to whatEvaluating the fitness value of each bat by the fitness evaluation function to find a current optimal solution x;
s204, adjusting the bat search pulse frequency, and updating the speed and the position of the bat according to the formulas (1), (2) and (3):
fi＝fmin+(fmax-fmin)β (1)
in the formula: beta is [0,1 ]]A randomly generated uniform random number; f. ofiRepresents a frequency of the acoustic wave; x represents the current global optimal solution;
indicating the position of the ith bat at time t,
representing the speed at that moment;
s205, generating uniformly distributed random number rand, if rand > riS206 is entered, otherwise S207 is entered, wherein riThe pulse emissivity of the ith bat;
s206, randomly disturbing the current optimal solution to generate a new solution, and carrying out border-crossing processing on the new solution, namely searching a local solution near the currently selected optimal solution and recording the current optimal solution;
s207, generating a new solution through random flight if rand is less than AiAnd f (x)i) F (x), then go to S208, otherwise go to S209, where aiThe pulse loudness of the ith bat;
s208, recording the new solution, and updating r by using the formulas (4) and (5)iAnd Ai；
ri t+1＝ri 0[1-exp(-γ*t)] (4)
In the formula, ri t+1Represents the pulse emissivity of the ith bat in the t +1 generation, ri 0Represents the maximum pulse emissivity of the ith bat, gamma is a pulse emissivity increasing coefficient, wherein gamma is more than 0,
respectively represents the pulse loudness of the ith bat in t +1 and t generations, and a is equal to 0,1]Is the pulse loudness attenuation coefficient;
s209, sorting the fitness values of all bats in the bat group, and finding out the current optimal solution and the optimal fitness value;
s210, if the preset search precision is met or the maximum search times are reached, turning to the step S211, otherwise, returning to the step S204;
and S211, outputting the current global optimal solution, and based on the current optimal nonlinear support vector machine model and the parameters thereof.
Preferably, in the above technical solution, step S103 specifically includes:
s301, dividing sample data into a training sample set and a test sample set;
s302, normalizing the data of the test sample set and the training sample set;
s303, setting training parameters of a support vector machine according to the optimal parameters selected in the step S211, training and learning a training sample set, and training a test sample by using the support vector machine;
s304, obtaining the prediction result of the test sample set.
Preferably, in the above technical solution, the step S211 based on the currently selected optimal nonlinear support vector machine model and the parameters thereof includes: training parameters, type of model, kernel function type, loss function and its parameters.
Preferably, in the above technical scheme, a dynamic contact resistance tester of the circuit breaker is used to collect dynamic contact resistance signals of the circuit breaker.
Preferably, in the above technical scheme, a stroke sensor is used for measuring the movement track of the moving arc contact.
Compared with the prior art, the method for diagnosing the electrical erosion fault of the contact of the high-voltage circuit breaker obtains the contact erosion data of the high-voltage circuit breaker, establishes the nonlinear compensation model by applying the bat algorithm BA-SVM support vector regression machine, and accurately predicts the contact erosion state parameter values obtained by measuring the high-voltage circuit breaker under different voltage currents.
Drawings
Fig. 1 is a flow chart of a method for diagnosing electrical erosion faults of contacts of a high-voltage circuit breaker according to the invention.
Fig. 2 is a flow chart of the BA-SVM algorithm according to the present invention.
Fig. 3 is a graph of the dynamic resistance versus time waveform of phase a of the test circuit breaker according to the present invention.
Fig. 4 is a waveform diagram of the dynamic resistance versus the movable contact travel of phase a of the test circuit breaker according to the present invention.
Fig. 5 is a graph of the dynamic resistance versus time waveform of phase C of the test circuit breaker according to the present invention.
Fig. 6 is a waveform diagram of dynamic resistance versus movable contact travel for phase C of a test circuit breaker according to the present invention.
Detailed Description
The following detailed description of the present invention is provided in conjunction with the accompanying drawings, but it should be understood that the scope of the present invention is not limited to the specific embodiments.
As shown in fig. 1, the main idea of the method for diagnosing the electrical erosion fault of the contact of the high-voltage circuit breaker in this embodiment is as follows: setting collected contact ablation evaluation parameters (a resistance-stroke curve and a static resistance value) of the circuit breaker as x, taking contact ablation state parameter values as target parameters y, obviously, taking y as f (x) as a nonlinear relation, taking the contact ablation evaluation parameters x as input samples of a BA-SVM (support vector machine) model, and outputting contact ablation state parameter values after the processing of the BA-SVM support vector machine model, wherein the output contact ablation state parameter values are the target parameters y expected to eliminate the influence of voltage and current, specifically, the method comprises the following steps:
step S101, after a plurality of circuit breakers operate for a period of time under different voltage currents respectively, acquiring dynamic contact resistance signals and static resistance value signals of moving arc contacts of the circuit breakers to obtain dynamic contact resistance-time curves; acquiring a dynamic stroke when the dynamic contact resistance signal of the moving arc contact occurs to obtain a stroke-time curve;
specifically, when a high-voltage circuit breaker is used for an experiment for acquiring dynamic resistance, an adjustable voltage current source is adopted, a super capacitor is utilized to generate impact current, the impact current output by the super capacitor can reach 2500A, a switching-on and switching-off test mode is adopted for measurement, namely, a switching-on experiment with a set time length of 250mS is carried out, then, a switching-off experiment with a time length of 250mS is carried out, and resistance can be obtained through voltage current, and the DB-8016 circuit breaker dynamic contact resistance tester calculates a dynamic contact resistance signal and a static resistance signal of the contact resistance through voltage data and current data on two sides of the contact resistance, and then, a curve of the dynamic contact resistance and time is drawn according to the relation between the resistance and the time; the instrument can draw a voltage versus time curve, a current versus time curve, and the like.
And S102, obtaining a resistance-travel curve according to the dynamic contact resistance-time curve and the travel-time curve.
In the step, the motion trail of the moving arc contact can be measured simultaneously by the stroke sensor tester and transmitted to the DB-8016 circuit breaker dynamic contact resistance tester, and a curve of stroke and time can be drawn on the liquid crystal display. The relation between the dynamic contact resistance and the stroke is deduced according to the curve between the dynamic contact resistance and the time and the curve between the stroke and the time, and the curve between the contact resistance and the stroke of the moving arc contact can be drawn on a liquid crystal display.
And step S103, taking the static resistance value signal and the resistance-stroke curve as contact ablation evaluation parameters, and acquiring contact ablation state parameter values of the circuit breaker according to the contact ablation evaluation parameters.
In the step, the relation between the contact ablation evaluation parameters (resistance-stroke curve and static resistance value) and the contact ablation state parameter values can be obtained according to an expert investigation method, and if the contact ablation state parameter values are numerical values between 0 and 1, the relation is divided into three sections, namely 0 to 0.4, 0.4 to 0.6 and 0.6 to 1.0; the first and second intervals are normal, the third interval is undetermined, and the fourth and fifth intervals are serious in ablation and need to be replaced, for example:
1. the resistance-stroke curve is normal, and the static resistance value is normal; the resistance-stroke curve is slightly abnormal, and the static resistance value is normal; the resistance-travel curve is slightly abnormal, and the static resistance value is slightly abnormal; the resistance-stroke curve is normal, and the static resistance value is slightly abnormal; all the above cases are in the interval 0-0.2.
2. The resistance-travel curve is normal, and the static resistance value is moderately abnormal; the resistance-travel curve is moderately abnormal, and the static resistance value is normal; moderate anomalies in the resistance-travel curve, and moderate anomalies in the static resistance values; the above cases are all in the interval of 0.4-0.6.
3. The resistance-travel curve is within moderate abnormity, and the static resistance value is severely abnormal; the resistance-travel curve is severely abnormal, and the static resistance value is moderately abnormal; the resistance-travel curve is severely abnormal, and the static resistance value is severely abnormal; the above cases are all in the interval of 0.6-1.0.
Step S104, optimizing parameters of a support vector machine by adopting a bat algorithm to obtain optimal parameters, and establishing an optimal nonlinear support vector machine by adopting the optimal parameters;
step S105, taking the contact ablation evaluation parameter and the corresponding contact ablation state parameter value of each breaker as a group of sample data;
and S106, training the optimal nonlinear support vector machine by using multiple groups of sample data, inputting the contact ablation evaluation parameters, and outputting corresponding contact ablation state parameter values by the nonlinear support vector machine to obtain the nonlinear support vector machine capable of evaluating the contact ablation fault of the high-voltage circuit breaker.
In step S106, the method specifically includes:
step S301, dividing sample data into a training sample set and a test sample set, randomly extracting the first 90% of the sample data as the training set, and the last 10% as the test set;
step S302, normalizing the data of the test sample set and the training sample set;
step S303, training and learning a training sample set according to a BA-SVM support vector machine set by the selected optimal parameters, and training a test sample by using the support vector machine;
and S304, acquiring a prediction result of the test sample set, and evaluating and analyzing the prediction effect of the SVM nonlinear correction model through the test data.
And S107, predicting the contact ablation evaluation parameters of the circuit breaker to be diagnosed by adopting the trained nonlinear support vector machine, and directly evaluating the ablation state according to the output contact ablation state parameter values.
With continuing reference to fig. 2, in this embodiment, step 102 specifically includes:
step S201, setting parameters of an SVM (support vector machine): the penalty parameter C has a parameter range of [1, 100 ]]RBF nuclear parameters in the range of 0.1, 100]The parameter range of the loss function is [0.001, 1 ]](ii) a Initializing bat group related parameters: setting initial population number n and pulse loudness A0Pulse emissivity r0A bat pulse emission rate increasing coefficient gamma, a pulse loudness attenuation coefficient alpha, and bat search pulse frequency upper and lower limits fmin,fmaxMaximum number of iterations tmaxAnd the search precision;
step S202, initializing the bat position xiAnd velocity vi；
In step S203, a fitness evaluation function f (x) is determined, where x is (x)1,…xd)TEvaluating the fitness value of each bat according to the fitness evaluation function to find a current optimal solution x;
step S204, adjusting the bat search pulse frequency, and updating the speed and the position of the bat according to the formulas (1), (2) and (3):
fi＝fmin+(fmax-fmin)β (1)
in the formula: beta is [0,1 ]]A randomly generated uniform random number; f. ofiRepresents a frequency of the acoustic wave; x represents the current global optimal solution;
indicating the position of the ith bat at time t,
representing the speed at that moment;
step S205, generating uniformly distributed random number rand, if rand > riS206 is entered, otherwise S207 is entered, wherein riThe pulse emissivity of the ith bat;
step S206, randomly disturbing the current optimal solution to generate a new solution, and carrying out border-crossing processing on the new solution, namely searching a local solution near the currently selected optimal solution and recording the current optimal solution;
step S207, generating a new solution by random flight if rand < AiAnd f (x)i) F (x), then go to S208, otherwise go to S209, where aiThe pulse loudness of the ith bat;
step S208, record the new solution, and update r by the formulas (4) and (5)iAnd AiI.e. (increase r)iDecrease Ai)；
ri t+1＝ri 0[1-exp(-γ*t)] (4)
In the formula, ri t+1Express the ith batPulse emissivity at the t +1 generation, ri 0Represents the maximum pulse emissivity of the ith bat, gamma is a pulse emissivity increasing coefficient, wherein gamma is more than 0,
respectively represents the pulse loudness of the ith bat in t +1 and t generations, and a is equal to 0,1]Is the pulse loudness attenuation coefficient;
step S209, sorting the fitness values of all bats in the bat group, and finding out the current optimal solution and the optimal fitness value;
step S210, if the preset search accuracy is met or the maximum search frequency is reached, go to step S211, otherwise return to step S204;
step S211, outputting a current global optimal solution, based on the currently selected optimal nonlinear support vector machine model and its parameters, including: training parameters (including penalty factor C, radial basis kernel function parameters, etc.), type of model, kernel function type, loss function and its parameters.
Further, in step S102, a resistance-travel curve is obtained according to the dynamic contact resistance-time curve and the travel-time curve, which is specifically as follows:
the dynamic resistance test is carried out on an SF6 high-voltage circuit breaker for a certain transformer substation, and the waveforms of the dynamic resistance of the A phase of the circuit breaker, the time and the stroke are shown in fig. 3 and 4:
after the waveform is measured for many times and is relatively stable, the dynamic resistance test is carried out on the phase B of the circuit breaker, and the dynamic resistance of the phase B is the same as the waveform of the phase A.
When the circuit breaker C-phase dynamic resistance test is carried out, the measured dynamic resistance and time and stroke waveform diagrams are shown in FIGS. 5 and 6:
the closing time point of the circuit breaker is 100mS, and the stability time of a moving contact of the circuit breaker is about 150 mS. And after a plurality of times of C-phase dynamic resistance tests of the circuit breaker, waveforms are shown in fig. 5 and 6.
Therefore, the dynamic contact resistance of the circuit breaker fluctuates greatly in the closing overtravel stage. It is presumed that there should be severe ablation of the breaker C phase contact, as opposed to the smooth waveform of the contact resistance of the a phase in fig. 3 and 4.
And (3) data analysis, namely extracting the data of the travel and the contact resistance of the moving contact of the A phase and the C phase through PC analysis software of the dynamic resistance tester to perform data analysis.
The resistance-stroke data of the A phase and the C phase are made into data curves, and the comparison of A, C phase data curves shows that the dynamic resistance of the A phase is monotonically decreased along with the increase of the stroke of the moving contact, so that the good contact between the moving contact and the fixed contact of the A phase can be inferred, and the obvious ablation condition does not exist. And finally, combining a static resistance value signal to form a PSO-SVM support vector regression machine for training after inputting the ablation evaluation parameters of the contact and optimizing.
The method for diagnosing the electrical erosion fault of the contact of the high-voltage circuit breaker obtains the contact erosion data of the high-voltage circuit breaker, establishes a nonlinear compensation model by applying a bat algorithm BA-SVM (building-support vector machine) support vector regression machine, and accurately predicts the contact erosion state parameter values obtained by measuring the high-voltage circuit breaker under different voltage currents.
The foregoing descriptions of specific exemplary embodiments of the present invention have been presented for purposes of illustration and description. It is not intended to limit the invention to the precise form disclosed, and obviously many modifications and variations are possible in light of the above teaching. The exemplary embodiments were chosen and described in order to explain certain principles of the invention and its practical application to enable one skilled in the art to make and use various exemplary embodiments of the invention and various alternatives and modifications as are suited to the particular use contemplated. It is intended that the scope of the invention be defined by the claims and their equivalents.
Claims (1)
1. A method for diagnosing electrical erosion faults of contacts of a high-voltage circuit breaker is characterized by comprising the following steps:
s101, after a plurality of circuit breakers operate for a period of time under different voltage currents respectively, acquiring dynamic contact resistance signals and static resistance value signals of moving arc contacts of the circuit breakers to obtain dynamic contact resistance-time curves; acquiring a dynamic stroke when the dynamic contact resistance signal of the moving arc contact occurs to obtain a stroke-time curve;
s102, obtaining a resistance-travel curve according to the dynamic contact resistance-time curve and the travel-time curve;
s103, taking the static resistance value signal and the resistance-stroke curve as contact ablation evaluation parameters, and acquiring contact ablation state parameter values of the circuit breaker according to the contact ablation evaluation parameters;
s104, optimizing parameters of a support vector machine by adopting a bat algorithm to obtain optimal parameters, and establishing an optimal nonlinear support vector machine by adopting the optimal parameters;
s105, taking the contact ablation evaluation parameter and the corresponding contact ablation state parameter value of each breaker as a group of sample data;
s106, training the optimal nonlinear support vector machine by using multiple groups of sample data, inputting the contact ablation evaluation parameters, and outputting corresponding contact ablation state parameter values by the nonlinear support vector machine to obtain the nonlinear support vector machine capable of evaluating the contact ablation fault of the high-voltage circuit breaker;
s107, predicting the contact ablation evaluation parameters of the breaker to be diagnosed by adopting the trained nonlinear support vector machine, and directly evaluating the ablation state according to the output contact ablation state parameter values;
step 102 specifically includes:
s201, setting parameters of the support vector machine: penalty parameter C, RBF kernel parameter, parameter range of loss function; initializing bat group related parameters: setting initial population number n and pulse loudness A0Pulse emissivity r0A bat pulse emission rate increasing coefficient gamma, a pulse loudness attenuation coefficient alpha, and bat search pulse frequency upper and lower limits fmin,fmaxMaximum number of iterations tmaxAnd the search precision;
s202, initializing the bat position xiAnd velocity vi；
S203, determine the fitness evaluation function f (x), where x is (x)1,…xd)TEvaluating the fitness value of each bat according to the fitness evaluation function to find the current optimal solution x*；
S204, adjusting the bat search pulse frequency, and updating the speed and the position of the bat according to the formulas (1), (2) and (3):
fi＝fmin+(fmax-fmin)β (1)
in the formula: beta is [0,1 ]]A randomly generated uniform random number; f. ofiRepresents a frequency of the acoustic wave; x is the number of*Representing a current global optimal solution;
indicating the position of the ith bat at time t,
representing the speed at that moment;
s205, generating uniformly distributed random number rand, if rand > riS206 is entered, otherwise S207 is entered, wherein riThe pulse emissivity of the ith bat;
s206, randomly disturbing the current optimal solution to generate a new solution, and carrying out border-crossing processing on the new solution, namely searching a local solution near the currently selected optimal solution and recording the current optimal solution;
s207, generating a new solution through random flight if rand is less than AiAnd f (x)i) F (x), then go to S208, otherwise go to S209, where aiThe pulse loudness of the ith bat;
s208, recording the new solution, and updating r by using the formulas (4) and (5)iAnd Ai；
ri t+1＝ri 0[1-exp(-γ*t)] (4)
In the formula, ri t+1Represents the pulse emissivity of the ith bat in the t +1 generation, ri 0Represents the maximum pulse emissivity of the ith bat, and gamma is a pulse emissivity increasing coefficient, wherein gamma is＞0，
Respectively represents the pulse loudness of the ith bat in t +1 and t generations, and a is equal to 0,1]Is the pulse loudness attenuation coefficient;
s209, sorting the fitness values of all bats in the bat group, and finding out the current optimal solution and the optimal fitness value;
s210, if the preset search precision is met or the maximum search times are reached, turning to the step S211, otherwise, returning to the step S204;
s211, outputting a current global optimal solution, and based on a currently selected optimal nonlinear support vector machine model and parameters thereof;
step S103 specifically includes:
s301, dividing sample data into a training sample set and a test sample set;
s302, normalizing the data of the test sample set and the training sample set;
s303, setting training parameters of a support vector machine according to the optimal parameters selected in the step S211, training and learning a training sample set, and training a test sample by using the support vector machine;
s304, obtaining a prediction result of the test sample set;
the step S211 includes, based on the currently selected optimal nonlinear support vector machine model and its parameters: training parameters, model types, kernel function types, loss functions and parameters thereof;
a dynamic contact resistance tester of the circuit breaker is adopted to collect dynamic contact resistance signals of the circuit breaker, and a travel sensor is adopted to measure the motion trail of the moving arc contact.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811053083.5A CN109164382B (en) | 2018-09-10 | 2018-09-10 | Method for diagnosing electrical erosion fault of high-voltage circuit breaker contact |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811053083.5A CN109164382B (en) | 2018-09-10 | 2018-09-10 | Method for diagnosing electrical erosion fault of high-voltage circuit breaker contact |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109164382A CN109164382A (en) | 2019-01-08 |
| CN109164382B true CN109164382B (en) | 2021-01-05 |
Family
ID=64894661
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811053083.5A Active CN109164382B (en) | 2018-09-10 | 2018-09-10 | Method for diagnosing electrical erosion fault of high-voltage circuit breaker contact |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109164382B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110412373B (en) * | 2019-07-23 | 2021-08-27 | 安徽升隆电气有限公司 | Switch cabinet fault early warning system and replacement method thereof |
| CN111505490A (en) * | 2020-03-23 | 2020-08-07 | 温州大学乐清工业研究院 | AC contactor ablation condition evaluation method based on convolutional neural network regression |
| CN112084662A (en) * | 2020-09-11 | 2020-12-15 | 西安高压电器研究院有限责任公司 | Method and device for detecting electrical service life of circuit breaker |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103336243A (en) * | 2013-07-01 | 2013-10-02 | 东南大学 | Breaker fault diagnosis method based on separating/closing coil current signals |
| CN103439650A (en) * | 2013-08-07 | 2013-12-11 | 王岩 | Method and device used for state monitoring and fault diagnosis of relay |
| CN103575525A (en) * | 2013-11-18 | 2014-02-12 | 东南大学 | Intelligent diagnosis method for mechanical fault of circuit breaker |
| CN103616635A (en) * | 2013-12-05 | 2014-03-05 | 国家电网公司 | Method and device for diagnosing mechanical characteristic failures of high-voltage circuit-breaker |
| CN104764993A (en) * | 2014-01-07 | 2015-07-08 | 国家电网公司 | Detection method and device for high-voltage circuit breaker |
| CN104793134A (en) * | 2015-04-29 | 2015-07-22 | 中国电力科学研究院 | Breaker operating mechanism fault diagnosis method based on least square support vector machine |
| CN106482937A (en) * | 2016-09-30 | 2017-03-08 | 南方电网科学研究院有限责任公司 | A kind of monitoring method of mechanical state of high-voltage circuit breaker |
| CN108121999A (en) * | 2017-12-10 | 2018-06-05 | 北京工业大学 | Support vector machines parameter selection method based on mixing bat algorithm |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9287713B2 (en) * | 2011-08-04 | 2016-03-15 | Siemens Aktiengesellschaft | Topology identification in distribution network with limited measurements |
| CN105548867A (en) * | 2015-12-01 | 2016-05-04 | 天津市电力科技发展公司 | Diagnostic system and diagnostic method of contact state of high-voltage circuit breaker |
| CN105467309A (en) * | 2015-12-01 | 2016-04-06 | 天津市电力科技发展公司 | State evaluation method and maintenance strategy for contact of high-voltage circuit breaker |
| CN106556796A (en) * | 2016-10-21 | 2017-04-05 | 中国电力科学研究院 | A kind of SF6 choppers arcing contact abatement detecting method |
| CN107680835B (en) * | 2017-10-13 | 2019-07-26 | 中国电力科学研究院 | A kind of breaker arcing contact ablation state evaluating method neural network based |
-
2018
- 2018-09-10 CN CN201811053083.5A patent/CN109164382B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103336243A (en) * | 2013-07-01 | 2013-10-02 | 东南大学 | Breaker fault diagnosis method based on separating/closing coil current signals |
| CN103439650A (en) * | 2013-08-07 | 2013-12-11 | 王岩 | Method and device used for state monitoring and fault diagnosis of relay |
| CN103575525A (en) * | 2013-11-18 | 2014-02-12 | 东南大学 | Intelligent diagnosis method for mechanical fault of circuit breaker |
| CN103616635A (en) * | 2013-12-05 | 2014-03-05 | 国家电网公司 | Method and device for diagnosing mechanical characteristic failures of high-voltage circuit-breaker |
| CN104764993A (en) * | 2014-01-07 | 2015-07-08 | 国家电网公司 | Detection method and device for high-voltage circuit breaker |
| CN104793134A (en) * | 2015-04-29 | 2015-07-22 | 中国电力科学研究院 | Breaker operating mechanism fault diagnosis method based on least square support vector machine |
| CN106482937A (en) * | 2016-09-30 | 2017-03-08 | 南方电网科学研究院有限责任公司 | A kind of monitoring method of mechanical state of high-voltage circuit breaker |
| CN108121999A (en) * | 2017-12-10 | 2018-06-05 | 北京工业大学 | Support vector machines parameter selection method based on mixing bat algorithm |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109164382A (en) | 2019-01-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109164382B (en) | Method for diagnosing electrical erosion fault of high-voltage circuit breaker contact | |
| Schichler et al. | Risk assessment on defects in GIS based on PD diagnostics | |
| JP4629113B2 (en) | Method and apparatus for determining the closing time of an electrical switchgear | |
| CN109061462A (en) | A kind of High Voltage Circuit Breaker Contacts ablation assessment of failure method | |
| RU2551645C2 (en) | Method and device for determination of wear of contact elements | |
| CN110275096A (en) | Insulator surface defect local discharge detection device and detection method | |
| JP4511162B2 (en) | Fuel cell evaluation system | |
| CN114252731A (en) | Relay action characteristic evaluation method and device based on multiple parameters | |
| Lewin et al. | Locating partial discharge sources in high voltage transformer windings | |
| CN114114001B (en) | GIS equipment isolating switch mechanical state monitoring method and system | |
| CN111722060B (en) | Distribution line early fault severity evaluation method based on waveform characteristics | |
| Strobl et al. | Resonant electric arcs in DC microgrids with low system impedance in the VLF-band | |
| Pochanke et al. | Experimental studies of circuit breaker drives and mechanisms diagnostics | |
| KR101420729B1 (en) | Apparatus and method for diagnosing a ground network | |
| CN114200258A (en) | Electric arc arcing detection method based on electric signals | |
| CN109492339B (en) | Arc model construction method and system | |
| Patel et al. | Simulation and mathematical analysis of partial discharge measurement in transformer | |
| Glotic et al. | Determining a gas-discharge arrester model's parameters by measurements and optimization | |
| CN111142017A (en) | Breaker state diagnosis system and method | |
| CN116840641A (en) | Switch cabinet insulation micro-component defect evaluation method, device and test system | |
| Shen et al. | Localization of partial discharges using UHF sensors in power transformers | |
| KR102535587B1 (en) | Method and apparatus for predicting partial discharge based on power facility temperature measurement using artificial neural network | |
| CN117436288B (en) | Aviation direct current fault arc model simulation method and storage medium | |
| RU2807606C1 (en) | Method for assessing electrical erosion wear resistance of material of electrical contact parts | |
| Schavemaker et al. | Digital testing of high-voltage circuit breakers |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |