CN112147494A - Mechanical fault detection method for high-voltage vacuum circuit breaker - Google Patents
Mechanical fault detection method for high-voltage vacuum circuit breaker Download PDFInfo
- Publication number
- CN112147494A CN112147494A CN202010667756.7A CN202010667756A CN112147494A CN 112147494 A CN112147494 A CN 112147494A CN 202010667756 A CN202010667756 A CN 202010667756A CN 112147494 A CN112147494 A CN 112147494A
- Authority
- CN
- China
- Prior art keywords
- coil
- circuit breaker
- current
- switching
- support vector
- 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
- 238000001514 detection method Methods 0.000 title claims description 7
- 239000002245 particle Substances 0.000 claims abstract description 50
- 238000003745 diagnosis Methods 0.000 claims abstract description 45
- 238000012843 least square support vector machine Methods 0.000 claims abstract description 45
- 238000000034 method Methods 0.000 claims abstract description 36
- 238000004458 analytical method Methods 0.000 claims abstract description 16
- 230000008569 process Effects 0.000 claims abstract description 15
- 238000007405 data analysis Methods 0.000 claims abstract description 8
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical group [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 claims description 37
- 230000006870 function Effects 0.000 claims description 18
- 230000007246 mechanism Effects 0.000 claims description 14
- 238000012706 support-vector machine Methods 0.000 claims description 12
- 238000012549 training Methods 0.000 claims description 11
- 238000004364 calculation method Methods 0.000 claims description 9
- 102100029469 WD repeat and HMG-box DNA-binding protein 1 Human genes 0.000 claims description 6
- 101710097421 WD repeat and HMG-box DNA-binding protein 1 Proteins 0.000 claims description 6
- 150000001875 compounds Chemical class 0.000 claims description 3
- 230000004907 flux Effects 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 3
- 230000003068 static effect Effects 0.000 claims description 3
- 239000000284 extract Substances 0.000 claims description 2
- 238000010586 diagram Methods 0.000 description 8
- 230000002068 genetic effect Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
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
- G01R31/3272—Apparatus, systems or circuits therefor
- G01R31/3274—Details related to measuring, e.g. sensing, displaying or computing; Measuring of variables related to the contact pieces, e.g. wear, position or resistance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/11—Complex mathematical operations for solving equations, e.g. nonlinear equations, general mathematical optimization problems
- G06F17/13—Differential equations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/15—Correlation function computation including computation of convolution operations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Mathematical Physics (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Databases & Information Systems (AREA)
- Algebra (AREA)
- Computing Systems (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Operations Research (AREA)
- Arc-Extinguishing Devices That Are Switches (AREA)
Abstract
The invention discloses a method for detecting mechanical faults of a high-voltage vacuum circuit breaker, which comprises the following steps: constructing a current characteristic analysis model of a switching-on and switching-off coil of the high-voltage vacuum circuit breaker, and extracting a coil current signal characteristic value in a switching-on and switching-off process of circuit breaking; constructing a fault diagnosis model of a least square support vector machine, and realizing data analysis based on the least square support vector machine and multi-classification of the least square support vector machine; and solving the fault diagnosis model by adopting a particle swarm algorithm. By constructing a current characteristic analysis model of a switching-on and switching-off coil of the high-voltage vacuum circuit breaker and a fault diagnosis model of a least square support vector machine, the diagnosis of mechanical faults of the high-voltage vacuum circuit breaker is realized, the problem that no proper fault diagnosis method exists for small sample data of current parameters is solved, the operation state of the high-voltage vacuum circuit breaker can be monitored in real time, the fault type is determined, the mechanical faults are prevented and rapidly eliminated, the operation safety of a power system is improved, and the power supply reliability of a power distribution network is improved.
Description
Technical Field
The invention relates to the technical field of power equipment detection, in particular to a mechanical fault detection method for a high-voltage vacuum circuit breaker.
Background
The high-voltage vacuum circuit breaker is an important node in a power system, and when a line fails, the circuit breaker can rapidly cut off a fault part, so that the safety of a power grid is guaranteed. The high-voltage vacuum circuit breaker can cause performance reduction when being in a complex working environment for a long time, for example, the opening and closing speed is reduced or the power requirement of an operating mechanism cannot be provided, and large-area power failure can be caused in a corresponding area under severe conditions. Therefore, the operating state of the high voltage vacuum circuit breaker is related to the operational safety of the whole power system. The high-voltage vacuum circuit breaker takes the operating mechanism as a power source, switching-on and switching-off are realized through conversion of electric energy and mechanical energy, and the poor mechanical performance of the operating mechanism can cause misoperation or closing rejection of the circuit breaker, so that the mechanical performance of the operating mechanism is very important for the high-voltage vacuum circuit breaker. According to the statistical data of the power system, the mechanical failure of the high-voltage vacuum circuit breaker accounts for about three quarters of all the failures. Therefore, monitoring the operating state of the operating mechanism and diagnosing mechanical faults are of great significance.
At present, the diagnosis method for the mechanical fault of the high-voltage vacuum circuit breaker at home and abroad has a certain foundation, but the following problems still exist: firstly, the method for optimizing the support vector machine based on the genetic algorithm carries out fault diagnosis on the high-voltage circuit breaker, but the genetic algorithm needs complicated selection, variation and cross operation, and the complexity of the genetic operation of the genetic algorithm is exponentially increased along with the complication of the problem; secondly, the diagnosis of the mechanical fault of the high-voltage vacuum circuit breaker is carried out based on an expert system, although a better diagnosis effect is achieved theoretically, in practical application, the reasoning process is complex, and the online monitoring requirement in the practical process is difficult to meet: and thirdly, the mechanical fault diagnosis of the high-voltage vacuum circuit breaker is carried out based on the neural network, but the diagnosis capability of the neural network is greatly limited by the number of samples, and the diagnosis speed is low and the convergence is difficult under the condition of small samples.
Disclosure of Invention
The invention aims to provide a high-voltage vacuum circuit breaker mechanical fault detection method, which realizes the diagnosis of the high-voltage vacuum circuit breaker mechanical fault by constructing a least square support vector machine fault diagnosis model of a high-voltage vacuum circuit breaker opening and closing coil current characteristic analysis model, solves the problem that no proper fault diagnosis method exists for small sample data of current parameters, can monitor the running state of a high-voltage vacuum circuit breaker in a power system in real time, senses the fault type of the high-voltage vacuum circuit breaker, can prevent the high-voltage vacuum circuit breaker mechanical fault, can eliminate the fault after the high-voltage vacuum circuit breaker has the mechanical fault, improves the running safety of the power system, and improves the power supply reliability of a power distribution network.
In order to solve the technical problem, an embodiment of the present invention provides a method for detecting a mechanical fault of a high voltage vacuum circuit breaker, including the following steps:
constructing a current characteristic analysis model of a switching-on and switching-off coil of the high-voltage vacuum circuit breaker, and extracting a coil current signal characteristic value in a switching-on and switching-off process of circuit breaking;
constructing a fault diagnosis model of a least square support vector machine, and realizing data analysis based on the least square support vector machine and multi-classification of the least square support vector machine;
and solving the fault diagnosis model by adopting a particle swarm algorithm.
Further, the method for constructing the analysis model of the current characteristics of the switching-on and switching-off coil of the high-voltage vacuum circuit breaker comprises the following steps:
the switching-on electromagnet of the high-voltage vacuum circuit breaker operating mechanism consists of an iron core and a coil, and when the switching-on electromagnet receives a switching-on signal, a differential equation of voltage U and current i of the coil is as follows:
wherein R is the resistance of the iron core, psi is the flux linkage of the coil, and t is time;
at the initial moment when the closing signal is sent, the iron core is in an unsaturated state, the current i of the coil does not affect the inductance L of the coil but is affected by the air gap of the iron core, and the differential equation of the coil voltage U and the current i is as follows:
wherein v is the moving speed of the iron core;
the initial moment of the switching-on signal is constant, the iron core is static, and the differential equation of the voltage U and the current i is specially solved as follows:
in the formula, L1The inductance of the coil is the initial moment of the switching-on signal;
after the switching-on signal is sent out, the current in the coil increases exponentially at t1At the moment, the current in the coil is increased to move the iron core, and the current value in the coil is I1(ii) a After the core moves, a back electromotive force is generated, and then the current starts to decrease until t2At the moment, the iron core is attracted, and the current value is I2The differential equation of the voltage U and the current i of the coil becomes:
in the formula, L2The inductor of the coil is attracted to the iron core;
after the iron core is attracted, the current in the coil increases exponentially at t3The current in the coil is I when the steady state is reached3;
At t4At that time, the auxiliary contact is cut off;
at t5At the moment, the current is reduced to the minimum value, and the closing process is finished;
the high-voltage vacuum circuit breaker opening and closing coil current characteristic analysis model extracts the I1The said I2The said I3The t is1The t is2The t is3The t is4The t is5And the characteristic value is used as the coil current signal characteristic value in the switching-on and switching-off process.
Further, the implementation of the data analysis based on the least squares support vector machine includes:
the training set sample in the fault diagnosis model of the least square support vector machine is { (a)1,b1),(a2,b2),…,(an,bn) In which a isiTo input samples, biIs an output;
the objective function f (omega) of the fault diagnosis model of the least square support vector machine is as follows:
in the formula, omega is a normal vector, and C is a planning factor; e.g. of the typeiIs an error variable;
the constraint conditions of the fault diagnosis model of the least square support vector machine are as follows:
in the formula (I), the compound is shown in the specification,is aiMapping in a high-dimensional feature space, wherein beta is an offset;
according to the statistical theory, the Lagrangian function for constructing the model is as follows:
in the formula, alphaiLagrange operator;
and (3) deriving variables of the Lagrangian function, and enabling the derivative to be zero:
erasure parameter eiAnd ω, changing the above equation to a linear equation:
wherein Ω is bibjK(ai,aj),K(ai,aj) Is a kernel function, B ═ B1,b2,...,bn]TI is an identity matrix, a ═ α1,α2,...,αn]T;
The decision function of the least square support vector machine is obtained as follows:
in the formula, sgn is a sign function, and returns the positive and negative of the parameter.
Further, the operating state of the circuit breaker operating mechanism comprises: normal (R)1) And jamming of iron core (R)2) Coil supply voltage failure (R)3) And jamming of operating mechanism (R)4) Auxiliary contact failure (R)5) Over-large idle stroke of iron core (R)6) The fault set Y ═ R1,R2,R3,R4,R5,R6]TAnd constructing six times of two classifications of the least square support vector machine, wherein the two classifications take 1 and-1 as record values, so that the fault set Y is obtained:
six rows of the fault set Y respectively correspond to the R1The R is2The R is3The R is4Station, stationR is5The R is6The operating state of (c).
Further, the implementing the least squares support vector machine multi-classification includes:
when classification is carried out, firstly training of input samples is carried out, the training target is the serial number of the class to which the input samples belong, then the least square support vector machine is used for classifying the newly input samples to obtain corresponding classification distances, the classifier with the largest classification distance is used for determining the classification result, the obtained codes consisting of 1 and-1 are used for determining the class to which the input samples belong, and finally, the predicted classification result is restored through decoding and the classification accuracy is output.
Further, the solving the established least squares support vector machine fault diagnosis model by adopting the particle swarm algorithm includes:
initializing the particle swarm algorithm, and setting parameters of the particle swarm algorithm;
predicting a learning sample by using the least square support vector machine corresponding to each particle vector respectively, and taking a prediction error as a fitness value of each particle;
comparing the fitness value of each particle with the optimal fitness value of each particle, and selecting a better fitness value as the optimal fitness value;
comparing the particle fitness of the individual particles with the particle fitness of the group particles, and selecting better group fitness;
updating the position and velocity of the particle;
judging whether a calculation termination requirement is met;
if the calculation termination requirement is met, obtaining a result meeting the requirement;
if the calculation termination requirement is not met, the least square support vector machine corresponding to each particle vector is reused for predicting the learning sample.
The technical scheme of the embodiment of the invention has the following beneficial technical effects:
the method has the advantages that the fault diagnosis model of the least square support vector machine is established through the current characteristic analysis model of the opening and closing coils of the high-voltage vacuum circuit breaker, the diagnosis of the mechanical fault of the high-voltage vacuum circuit breaker is realized, the problem that no proper fault diagnosis method exists for the small sample data of the current parameter is solved, the operation state of the high-voltage vacuum circuit breaker in the power system can be monitored in real time, the fault type of the high-voltage vacuum circuit breaker is sensed, the mechanical fault of the high-voltage vacuum circuit breaker can be prevented, the fault can be eliminated more quickly after the mechanical fault of the high-voltage vacuum circuit breaker occurs, the operation safety of the power.
Drawings
Fig. 1 is a flow chart of a method for detecting a mechanical fault of a high-voltage vacuum circuit breaker according to an embodiment of the present invention;
fig. 2 is a waveform diagram of a closing current provided in an embodiment of the present invention;
FIG. 3 is an equivalent circuit diagram of an electromagnet according to an embodiment of the present invention;
FIG. 4 is a diagram of coil current waveforms for various fault conditions provided by an embodiment of the present invention;
fig. 5 is a diagram of a solution result of the particle swarm algorithm provided by the embodiment of the invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be described in further detail with reference to the accompanying drawings in conjunction with the following detailed description. It should be understood that the description is intended to be exemplary only, and is not intended to limit the scope of the present invention. Moreover, in the following description, descriptions of well-known structures and techniques are omitted so as to not unnecessarily obscure the concepts of the present invention.
Fig. 1 is a flowchart of a method for detecting a mechanical fault of a high-voltage vacuum circuit breaker according to an embodiment of the present invention.
Referring to fig. 1, an embodiment of the present invention provides a method for detecting a mechanical fault of a high voltage vacuum circuit breaker, including the following steps:
s100, constructing a high-voltage vacuum circuit breaker opening and closing coil current characteristic analysis model, and extracting a coil current signal characteristic value in the opening and closing process.
S200, constructing a fault diagnosis model of the least square support vector machine, and realizing data analysis based on the least square support vector machine and multi-classification of the least square support vector machine.
And S300, solving the fault diagnosis model by adopting a particle swarm algorithm.
Specifically, establish high-pressure vacuum circuit breaker divide-shut brake coil current characteristic analysis model, include:
the switching-on electromagnet of the high-voltage vacuum circuit breaker operating mechanism consists of an iron core and a coil, and when the switching-on electromagnet receives a switching-on signal, a differential equation of voltage U and current i of the coil is as follows:
in the formula, R is the resistance of an iron core, psi is the flux linkage of a coil, and t is time;
at the initial moment when the closing signal is sent out, the iron core is in an unsaturated state, the current i of the coil does not affect the inductance L of the coil but is affected by the air gap of the iron core, and the differential equation of the voltage U of the coil and the current i is as follows:
wherein v is the moving speed of the iron core;
the initial moment that closing signal sent, for the constant, the iron core is static, and the differential equation special solution of voltage U and electric current i does:
in the formula, L1The inductance of the coil at the initial moment is sent out for a closing signal;
after the closing signal is sent out, the current in the coil increases exponentially at t1At the moment, the current in the coil is increased to move the iron core, and the current value in the coil is I1(ii) a After the iron core moves, a reverse electromotive force is generated, and thenThe current starts to decrease until t2The iron core is attracted at the moment, and the current value is I2The differential equation of the voltage U and the current i of the coil becomes:
in the formula, L2The iron core is attracted by the inductor of the coil.
Further, after the iron core is attracted, the current in the coil increases exponentially at t3The current in the coil is I when the steady state is reached3(ii) a At t4At that time, the auxiliary contact is cut off; at t5At the moment, the current is reduced to the minimum value, and the closing process is finished; high-voltage vacuum circuit breaker opening and closing coil current characteristic analysis model extraction I1、I2、I3、t1、t2、t3、t4、t5And the characteristic value is used as the coil current signal characteristic value in the switching-on and switching-off process.
Specifically, the data analysis based on the least square support vector machine is realized, and comprises the following steps:
the training set sample in the fault diagnosis model of the least square support vector machine is { (a)1,b1),(a2,b2),…,(an,bn) In which a isiTo input samples, biIs an output;
the objective function f (omega) of the fault diagnosis model of the least square support vector machine is as follows:
in the formula, omega is a normal vector, and C is a planning factor; e.g. of the typeiIs an error variable.
The constraint conditions of the fault diagnosis model of the least square support vector machine are as follows:
in the formula (I), the compound is shown in the specification,is aiThe mapping in the high-dimensional feature space, β, is the offset.
According to the statistical theory, the Lagrangian function for constructing the model is as follows:
in the formula, alphaiLagrange operator;
and (3) deriving variables of the Lagrangian function, and enabling the derivative to be zero:
erasure parameter eiAnd ω, changing the above equation to a linear equation:
wherein Ω is bibjK(ai,aj),K(ai,aj) Is a kernel function, B ═ B1,b2,...,bn]TI is an identity matrix, a ═ α1,α2,...,αn]T;
The decision function of the least square support vector machine is obtained as follows:
in the formula, sgn is a sign function, and returns the positive and negative of the parameter.
Specifically, the operating state of the circuit breaker operating mechanism includes: normal (R)1) And jamming of iron core (R)2) Coil supply voltage failure (R)3) And jamming of operating mechanism (R)4) Auxiliary contact failure (R)5) Over-large idle stroke of iron core (R)6) The fault set Y ═ R1,R2,R3,R4,R5,R6]TAnd constructing two classifications of a six-time least square support vector machine, wherein the two classifications take 1 and-1 as record values, so that a fault set Y is obtained:
six rows of the fault set Y respectively correspond to R1、R2、R3、R4、R5、R6The operating state of (c).
Further, implementing a least squares support vector machine multi-classification, comprising:
when classification is carried out, training of input samples is carried out firstly, the training target is the affiliated class number of the input samples, then a least square support vector machine is used for classifying the newly input samples to obtain corresponding classification distances, a classifier with the largest classification distance is used for determining classification results, the affiliated class is determined by the obtained 1 and-1 codes, and finally the predicted classification results are restored through decoding and the classification accuracy is output.
In an implementation manner of the embodiment of the present invention, solving the established least squares support vector machine fault diagnosis model by using a particle swarm algorithm includes:
s310, initializing a particle swarm algorithm and setting parameters of the particle swarm algorithm.
And S320, predicting the learning sample by using a least square support vector machine corresponding to each particle vector, and taking the prediction error as the fitness value of each particle.
S330, comparing the fitness value of each particle with the optimal fitness value of each particle, and selecting a better fitness value as the optimal fitness value.
S340, comparing the particle fitness of the individual particles and the population particles, and selecting the better population fitness.
And S350, updating the position and the speed of the particles.
And S360, judging whether the calculation termination requirement is met.
And S370, if the calculation termination requirement is met, obtaining a result meeting the requirement.
And S380, if the calculation termination requirement is not met, predicting the learning sample by using the least square support vector machine corresponding to each particle vector again.
Fig. 2 is a waveform diagram of a closing current according to an embodiment of the present invention.
Fig. 3 is an equivalent circuit diagram of an electromagnet according to an embodiment of the present invention.
Fig. 4 is a waveform diagram of coil current under different fault conditions provided by an embodiment of the present invention.
Fig. 5 is a diagram of a solution result of the particle swarm algorithm provided by the embodiment of the invention.
Referring to fig. 2, 3, 4 and 5, since the reliability of the high voltage vacuum circuit breaker is high and it is difficult to collect mechanical fault data of the high voltage vacuum circuit breaker, a fault coil current waveform is collected by a method of simulating a fault. By combining the fault category of the high-voltage vacuum circuit breaker and the feasibility of simulating the fault, the following fault types are selected: insufficient coil supply voltage (R2), closing spring jamming (R3), closing coil failure (R4), auxiliary contact failure (R5) and normal operating conditions (R1).
The coil current waveforms for the different fault conditions are shown in fig. 4. As shown in fig. 4(a), the fault of the coil supply voltage affects the closing time of the circuit breaker, and it can be seen that the whole current waveform curve moves to the lower right; as shown in fig. 4(b), the total closing time after the failure mode failure of the closing coil is changed from 30ms to 34ms, t2And t3Regular changes are also produced; as can be seen from fig. 4(c), the maximum current, the duration of excitation, and the duration of latching under the condition of a closing spring jam fault vary; as shown in fig. 4(d), the contact time characteristics are affected by the non-ideal contact of the contacts because the auxiliary contacts are failed when the circuit breaker completes the closingIn the state, the coil power supply is not cut off in time, so the time used in the whole closing process is prolonged.
Through analog analysis of fault types, the coil current waveforms are changed by the main amounts of I1, I2, I3, t1, t2, t3 and t 4. I1, I3, t1 and t2 show whether coil supply voltage and resistance are normal or not, t1, t2, t3 and t4 show whether a closing spring fails or not, and coil current waveforms in stages t 3-t 4 show the working state of the auxiliary contact. The correspondence relationship between the types of failures and the associated characteristic quantities is shown in table 1.
TABLE 1 Change feature quantities for different fault types
The method adopts a particle swarm algorithm to solve the established least square support vector machine fault diagnosis model, and specific data of a training sample is shown in a table 2.
TABLE 2 least squares support vector machine fault diagnosis model training sample
In the particle swarm algorithm solving process, specific parameters are set as follows: controlling group optimal particle motion velocity parameter c11.5; controlling the optimal motion speed c of a single particle21.7; the population number is 20; the number of iterations was 200. The solving result of the particle swarm algorithm is shown in fig. 4, and the fault classification parameters of the obtained least squares support vector machine fault diagnosis model are shown in table 3.
TABLE 3 Fault classification parameters of the least squares support vector machine fault diagnosis model
The embodiment of the invention aims to protect a mechanical fault detection method of a high-voltage vacuum circuit breaker, which comprises the following steps: constructing a current characteristic analysis model of a switching-on and switching-off coil of the high-voltage vacuum circuit breaker, and extracting a coil current signal characteristic value in a switching-on and switching-off process of circuit breaking; constructing a fault diagnosis model of a least square support vector machine, and realizing data analysis based on the least square support vector machine and multi-classification of the least square support vector machine; and solving the fault diagnosis model by adopting a particle swarm algorithm. The technical scheme has the following effects:
by constructing a current characteristic analysis model of a switching-on and switching-off coil of the high-voltage vacuum circuit breaker and a fault diagnosis model of a least square support vector machine, the diagnosis of mechanical faults of the high-voltage vacuum circuit breaker is realized, the problem that no proper fault diagnosis method exists for small sample data of current parameters is solved, the running state of the high-voltage vacuum circuit breaker in a power system can be monitored in real time, the fault type of the high-voltage vacuum circuit breaker is sensed, the mechanical faults of the high-voltage vacuum circuit breaker can be prevented, the faults can be eliminated more quickly after the mechanical faults of the high-voltage vacuum circuit breaker occur, the running safety of the power system is improved, and the.
It is to be understood that the above-described embodiments of the present invention are merely illustrative of or explaining the principles of the invention and are not to be construed as limiting the invention. Therefore, any modification, equivalent replacement, improvement and the like made without departing from the spirit and scope of the present invention should be included in the protection scope of the present invention. Further, it is intended that the appended claims cover all such variations and modifications as fall within the scope and boundaries of the appended claims or the equivalents of such scope and boundaries.
Claims (6)
1. A method for detecting mechanical faults of a high-voltage vacuum circuit breaker is characterized by comprising the following steps:
constructing a current characteristic analysis model of a switching-on and switching-off coil of the high-voltage vacuum circuit breaker, and extracting a coil current signal characteristic value in a switching-on and switching-off process of circuit breaking;
constructing a fault diagnosis model of a least square support vector machine, and realizing data analysis based on the least square support vector machine and multi-classification of the least square support vector machine;
and solving the fault diagnosis model by adopting a particle swarm algorithm.
2. The method for detecting the mechanical fault of the high-voltage vacuum circuit breaker according to claim 1, wherein the constructing of the analysis model of the current characteristic of the opening and closing coil of the high-voltage vacuum circuit breaker comprises the following steps:
the switching-on electromagnet of the high-voltage vacuum circuit breaker operating mechanism consists of an iron core and a coil, and when the switching-on electromagnet receives a switching-on signal, a differential equation of voltage U and current i of the coil is as follows:
wherein R is the resistance of the iron core, psi is the flux linkage of the coil, and t is time;
at the initial moment when the closing signal is sent, the iron core is in an unsaturated state, the current i of the coil does not affect the inductance L of the coil but is affected by the air gap of the iron core, and the differential equation of the coil voltage U and the current i is as follows:
wherein v is the moving speed of the iron core;
the initial moment of the switching-on signal is constant, the iron core is static, and the differential equation of the voltage U and the current i is specially solved as follows:
in the formula, L1The inductance of the coil is the initial moment of the switching-on signal;
after the switching-on signal is sent out, the current in the coil increases exponentially at t1Time of day, the lineThe iron core is moved by the increase of current in the coil, and the current value in the coil is I1(ii) a After the core moves, a back electromotive force is generated, and then the current starts to decrease until t2At the moment, the iron core is attracted, and the current value is I2The differential equation of the voltage U and the current i of the coil becomes:
in the formula, L2The inductor of the coil is attracted to the iron core;
after the iron core is attracted, the current in the coil increases exponentially at t3The current in the coil is I when the steady state is reached3;
At t4At that time, the auxiliary contact is cut off;
at t5At the moment, the current is reduced to the minimum value, and the closing process is finished;
the high-voltage vacuum circuit breaker opening and closing coil current characteristic analysis model extracts the I1The said I2The said I3The t is1The t is2The t is3The t is4The t is5And the characteristic value is used as the coil current signal characteristic value in the switching-on and switching-off process.
3. The method for detecting mechanical faults of a high-voltage vacuum circuit breaker according to claim 1, wherein the realizing of the data analysis based on a least squares support vector machine comprises:
the training set sample in the fault diagnosis model of the least square support vector machine is { (a)1,b1),(a2,b2),…,(an,bn) In which a isiTo input samples, biIs an output;
the objective function f (omega) of the fault diagnosis model of the least square support vector machine is as follows:
in the formula, omega is a normal vector, and C is a planning factor; e.g. of the typeiIs an error variable;
the constraint conditions of the fault diagnosis model of the least square support vector machine are as follows:
in the formula (I), the compound is shown in the specification,is aiMapping in a high-dimensional feature space, wherein beta is an offset;
according to the statistical theory, the Lagrangian function for constructing the model is as follows:
in the formula, alphaiLagrange operator;
and (3) deriving variables of the Lagrangian function, and enabling the derivative to be zero:
erasure parameter eiAnd ω, changing the above equation to a linear equation:
wherein Ω is bibjK(ai,aj),K(ai,aj) Is a kernel function, B ═ B1,b2,...,bn]TI is an identity matrix, a ═ α1,α2,...,αn]T;
The decision function of the least square support vector machine is obtained as follows:
in the formula, sgn is a sign function, and returns the positive and negative of the parameter.
4. The mechanical fault detection method for high-voltage vacuum circuit breaker according to claim 1,
the operating state of the circuit breaker operating mechanism comprises: normal (R)1) And jamming of iron core (R)2) Coil supply voltage failure (R)3) And jamming of operating mechanism (R)4) Auxiliary contact failure (R)5) Over-large idle stroke of iron core (R)6) The fault set Y ═ R1,R2,R3,R4,R5,R6]TAnd constructing six times of two classifications of the least square support vector machine, wherein the two classifications take 1 and-1 as record values, so that the fault set Y is obtained:
six rows of the fault set Y respectively correspond to the R1The R is2The R is3The R is4The R is5The R is6The operating state of (c).
5. The method for detecting mechanical faults of a high-voltage vacuum circuit breaker according to claim 4, wherein the realizing of the multi-classification of the least squares support vector machine comprises:
when classification is carried out, firstly training of input samples is carried out, the training target is the serial number of the class to which the input samples belong, then the least square support vector machine is used for classifying the newly input samples to obtain corresponding classification distances, the classifier with the largest classification distance is used for determining the classification result, the obtained codes consisting of 1 and-1 are used for determining the class to which the input samples belong, and finally, the predicted classification result is restored through decoding and the classification accuracy is output.
6. The method for detecting mechanical faults of a high-voltage vacuum circuit breaker according to claim 1, wherein the solving of the established least squares support vector machine fault diagnosis model by adopting a particle swarm algorithm comprises the following steps:
initializing the particle swarm algorithm, and setting parameters of the particle swarm algorithm;
predicting a learning sample by using the least square support vector machine corresponding to each particle vector respectively, and taking a prediction error as a fitness value of each particle;
comparing the fitness value of each particle with the optimal fitness value of each particle, and selecting a better fitness value as the optimal fitness value;
comparing the particle fitness of the individual particles with the particle fitness of the group particles, and selecting better group fitness;
updating the position and velocity of the particle;
judging whether a calculation termination requirement is met;
if the calculation termination requirement is met, obtaining a result meeting the requirement;
if the calculation termination requirement is not met, the least square support vector machine corresponding to each particle vector is reused for predicting the learning sample.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010667756.7A CN112147494A (en) | 2020-07-10 | 2020-07-10 | Mechanical fault detection method for high-voltage vacuum circuit breaker |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010667756.7A CN112147494A (en) | 2020-07-10 | 2020-07-10 | Mechanical fault detection method for high-voltage vacuum circuit breaker |
Publications (1)
Publication Number | Publication Date |
---|---|
CN112147494A true CN112147494A (en) | 2020-12-29 |
Family
ID=73887742
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010667756.7A Pending CN112147494A (en) | 2020-07-10 | 2020-07-10 | Mechanical fault detection method for high-voltage vacuum circuit breaker |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112147494A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113552479A (en) * | 2021-08-12 | 2021-10-26 | 北京慧智神光科技有限公司 | Fault analysis capability evaluation method and system of circuit breaker monitoring device |
CN113656977A (en) * | 2021-08-25 | 2021-11-16 | 绵阳市维博电子有限责任公司 | Coil fault intelligent diagnosis method and device based on multi-mode feature learning |
CN117438241A (en) * | 2023-12-15 | 2024-01-23 | 霍立克电气有限公司 | Switching-on control method of vacuum circuit breaker |
JP7467781B2 (en) | 2021-12-16 | 2024-04-15 | 株式会社東芝 | System for monitoring condition of circuit breaker spring operating mechanism, and method for monitoring circuit breaker spring operating mechanism |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103323770A (en) * | 2013-05-23 | 2013-09-25 | 国家电网公司 | Device for detection of mechanical characteristics and diagnosis of faults of high-voltage circuit breaker |
CN103336243A (en) * | 2013-07-01 | 2013-10-02 | 东南大学 | Breaker fault diagnosis method based on separating/closing coil current signals |
WO2016176913A1 (en) * | 2015-05-06 | 2016-11-10 | 杨启蓓 | Smart multi-dimensional big data analyzing expert system for high-voltage circuit breaker in power grid |
CN109061463A (en) * | 2018-09-29 | 2018-12-21 | 华南理工大学 | A kind of monitoring of mechanical state of high-voltage circuit breaker and method for diagnosing faults |
CN110530566A (en) * | 2019-10-10 | 2019-12-03 | 国网河南省电力公司电力科学研究院 | A kind of breaker switching on and off spring pressure monitoring system and detection method |
CN111060813A (en) * | 2019-12-09 | 2020-04-24 | 国网北京市电力公司 | Fault diagnosis method and device for high-voltage circuit breaker operating mechanism and electronic equipment |
-
2020
- 2020-07-10 CN CN202010667756.7A patent/CN112147494A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103323770A (en) * | 2013-05-23 | 2013-09-25 | 国家电网公司 | Device for detection of mechanical characteristics and diagnosis of faults of high-voltage circuit breaker |
CN103336243A (en) * | 2013-07-01 | 2013-10-02 | 东南大学 | Breaker fault diagnosis method based on separating/closing coil current signals |
WO2016176913A1 (en) * | 2015-05-06 | 2016-11-10 | 杨启蓓 | Smart multi-dimensional big data analyzing expert system for high-voltage circuit breaker in power grid |
CN109061463A (en) * | 2018-09-29 | 2018-12-21 | 华南理工大学 | A kind of monitoring of mechanical state of high-voltage circuit breaker and method for diagnosing faults |
CN110530566A (en) * | 2019-10-10 | 2019-12-03 | 国网河南省电力公司电力科学研究院 | A kind of breaker switching on and off spring pressure monitoring system and detection method |
CN111060813A (en) * | 2019-12-09 | 2020-04-24 | 国网北京市电力公司 | Fault diagnosis method and device for high-voltage circuit breaker operating mechanism and electronic equipment |
Non-Patent Citations (4)
Title |
---|
姜海龙等: "基于PSO⁃LSSVM的高压真空断路器弹簧机构", 《高压电器》 * |
姜海龙等: "基于PSO-LSSVM的高压真空断路器弹簧机构机械故障诊断", 《高压电器》 * |
张卫正等: "基于最小二乘支持向量机的高压断路器故障诊断", 《高压电器》 * |
张永奎: "基于分合闸线圈电流信号的高压断路器机械故障诊断", 《高压电器》 * |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113552479A (en) * | 2021-08-12 | 2021-10-26 | 北京慧智神光科技有限公司 | Fault analysis capability evaluation method and system of circuit breaker monitoring device |
CN113552479B (en) * | 2021-08-12 | 2023-10-31 | 北京慧智神光科技有限公司 | Fault analysis capability evaluation method and system of breaker monitoring device |
CN113656977A (en) * | 2021-08-25 | 2021-11-16 | 绵阳市维博电子有限责任公司 | Coil fault intelligent diagnosis method and device based on multi-mode feature learning |
JP7467781B2 (en) | 2021-12-16 | 2024-04-15 | 株式会社東芝 | System for monitoring condition of circuit breaker spring operating mechanism, and method for monitoring circuit breaker spring operating mechanism |
CN117438241A (en) * | 2023-12-15 | 2024-01-23 | 霍立克电气有限公司 | Switching-on control method of vacuum circuit breaker |
CN117438241B (en) * | 2023-12-15 | 2024-03-15 | 霍立克电气有限公司 | Switching-on control method of vacuum circuit breaker |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112147494A (en) | Mechanical fault detection method for high-voltage vacuum circuit breaker | |
CN110929763B (en) | Multi-source data fusion-based mechanical fault diagnosis method for medium-voltage vacuum circuit breaker | |
Li et al. | A hierarchical data-driven method for event-based load shedding against fault-induced delayed voltage recovery in power systems | |
CN109270442B (en) | DBN-GA neural network-based high-voltage circuit breaker fault detection method | |
CN103336243B (en) | Based on the circuit breaker failure diagnostic method of divide-shut brake coil current signal | |
CN106291351B (en) | High-voltage circuitbreaker fault detection method based on convolutional neural networks algorithm | |
CN104595170B (en) | A kind of air compressor machine monitoring and diagnosis system and method for self-adaptive kernel gauss hybrid models | |
CN106019131B (en) | A kind of primary cut-out operating mechanism state comprehensive estimation method based on divide-shut brake coil current | |
CN109462521B (en) | Network flow abnormity detection method suitable for source network load interaction industrial control system | |
Li et al. | Adaptive online monitoring of voltage stability margin via local regression | |
CN111060815B (en) | GA-Bi-RNN-based high-voltage circuit breaker fault diagnosis method | |
CN111060813B (en) | Fault diagnosis method and device for operating mechanism of high-voltage circuit breaker and electronic equipment | |
CN110703077B (en) | HPSO-TSVM-based high-voltage circuit breaker fault diagnosis method | |
CN102707256A (en) | Fault diagnosis method based on BP-Ada Boost nerve network for electric energy meter | |
CN106019138A (en) | Online diagnosis method for mechanical fault of high-voltage circuit breaker | |
CN109814529A (en) | A kind of train door prognostic and health management system | |
CN109298330A (en) | Fault Diagnosis for HV Circuit Breakers method based on GHPSO-BP | |
CN111325315A (en) | Distribution transformer power failure and power loss prediction method based on deep learning | |
CN105891708A (en) | ANFIS-based condition assessment for high-voltage circuit breaker operating mechanism | |
CN108020781A (en) | A kind of circuit breaker failure diagnostic method | |
CN110147827A (en) | A kind of failure prediction method based on IAALO-SVM and similarity measurement | |
CN116244279A (en) | High-voltage switch cabinet defect prediction method and system based on causal graph attention mechanism | |
CN108846489A (en) | A kind of large scale equipment health Evaluation method based on SVM | |
Oh et al. | Storm-induced power grid damage forecasting method for solving low probability event data | |
CN113326663A (en) | Railway relay running state evaluation method based on extreme learning machine |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20201229 |