CN108508298B - Service life assessment method and device for switching-on/off coil - Google Patents
Service life assessment method and device for switching-on/off coil Download PDFInfo
- Publication number
- CN108508298B CN108508298B CN201810330732.5A CN201810330732A CN108508298B CN 108508298 B CN108508298 B CN 108508298B CN 201810330732 A CN201810330732 A CN 201810330732A CN 108508298 B CN108508298 B CN 108508298B
- Authority
- CN
- China
- Prior art keywords
- opening
- closing coil
- coil
- preset
- test
- 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
- G01R31/003—Environmental or reliability tests
-
- 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
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Environmental & Geological Engineering (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
- Testing Resistance To Weather, Investigating Materials By Mechanical Methods (AREA)
Abstract
The invention discloses a service life assessment method of an opening and closing coil, which comprises the following steps: acquiring original current data of at least two opening and closing coils; according to preset data, alternating salt spray test operation is carried out on each opening and closing coil; acquiring test current data after the alternating salt spray test operation is executed; the switching-on/off coil is arranged on a circuit breaker, and whether the switching-on/off coil cannot touch a switching-on/off mechanism of the circuit breaker is judged; if yes, stopping the alternating salt spray test operation; if not, continuing to execute the alternating salt spray test operation on the opening and closing coil until the opening and closing coil cannot touch the opening and closing mechanism; and training operation based on a support vector machine regression algorithm is carried out according to the original current data and the test current data, so as to obtain a training model. The invention also discloses a service life assessment device of the switching-on/off coil, which can simulate climate conditions, effectively obtain service life assessment information of the switching-on/off coil and timely replace the switching-on/off coil.
Description
Technical Field
The invention relates to the technical field of power electronics, in particular to a service life assessment method and device of an opening and closing coil.
Background
The switching-on/off coil is a core component of the circuit breaker and is also a fault multiple component, and due to the reasons of the design, manufacture, material quality, operation and the like of the circuit breaker, the fault of the switching-on/off coil occurs, and the circuit breaker gradually deteriorates under the action of environmental factors such as temperature, humidity, salt mist and the like when in actual operation, and the conditions of moving iron core movement jam, poor coil loop contact and the like seriously influence the safe and stable operation of a power grid. According to southern power grid data statistics, about 45% of circuit breaker failure is due to faults of opening and closing coils. At present, an accurate and effective method is not available for an aging test of an opening and closing coil of a circuit breaker, so that the phenomenon that the circuit breaker is refused to operate due to the fact that the opening and closing coil cannot be replaced in time is easy, and serious loss is caused to a power grid.
Disclosure of Invention
The embodiment of the invention aims to provide a service life assessment method and device for an opening and closing coil, which can simulate weather conditions and effectively obtain service life assessment information of the opening and closing coil.
To achieve the above object, an embodiment of the present invention provides a lifetime assessment method for an opening/closing coil, including:
acquiring original current data of at least two opening and closing coils;
according to preset data, alternating salt spray test operation is carried out on each opening and closing coil;
acquiring test current data of each opening and closing coil after the alternating salt spray test operation is executed;
the switching-on/off coil is arranged on the circuit breaker, and whether the switching-on/off coil can trigger a switching-on/off mechanism of the circuit breaker is judged;
if yes, continuing to execute the alternating salt spray test operation on the opening and closing coil until the opening and closing coil cannot touch the opening and closing mechanism;
and if not, training operation based on a support vector machine regression algorithm is carried out according to the original current data and the test current data, so as to obtain a training model.
Compared with the prior art, the service life assessment method of the switching-on/off coils disclosed by the invention is characterized in that alternating salt spray test operation is carried out on each switching-on/off coil according to the preset parameters, test current data are obtained, and then whether the switching-on/off coil can trigger a switching-on/off mechanism of a circuit breaker is judged; if yes, continuing to execute the alternating salt spray test operation on the opening and closing coil; and otherwise, training operation based on a support vector machine regression algorithm is performed according to the pre-selected and acquired original current data of the opening and closing coil and the test current data, so that a training model is obtained. The aging test of the opening and closing coil of the circuit breaker in the prior art is solved, no accurate and effective method is available, so that the phenomenon that the circuit breaker is refused due to the fact that the opening and closing coil cannot be replaced in time is easy, the climate condition can be simulated, the service life evaluation information of the opening and closing coil can be effectively obtained, and the opening and closing coil can be replaced in time.
As an improvement of the above scheme, the training operation based on the support vector machine regression algorithm specifically includes:
constructing a training set input matrix and a regression value corresponding to the training set input matrix;
and carrying out regression modeling by using a support vector machine according to the training set input matrix and the regression value.
As an improvement of the above scheme, the training operation based on the support vector machine regression algorithm is performed according to the raw current data and the test current data, and after a training model is obtained, the method further comprises:
and according to the training model, taking historical current data of the opening and closing coil of the circuit breaker as input data, and carrying out service life assessment on the opening and closing coil.
As an improvement of the above scheme, the preset data includes a preset spraying time, a preset spraying amount, a preset temperature, a preset humidity and a preset test time.
As a modification of the above, the alternating salt spray test operation comprises four test cycles, each comprising a two hour spray cycle and a seven hour storage cycle.
The embodiment of the invention also provides a service life evaluation device of the opening and closing coil, which comprises the following steps:
the test box is used for executing alternating salt spray test operation on the opening and closing coil;
the upper computer module is used for setting preset data and simultaneously carrying out training operation based on a support vector machine regression algorithm according to the original current data and the test current data to obtain a training model;
the main control circuit module is used for adjusting the control parameters of the regulation and control module according to the preset data;
the data acquisition module is used for acquiring original current data and test current data;
and the regulation and control module is used for carrying out parameter regulation and control on the test box according to the control parameters.
Compared with the prior art, the service life assessment device for the opening and closing coil disclosed by the invention has the advantages that the main control circuit module adjusts the control parameters of the regulation and control module according to the preset parameters set by the upper computer module, so that the regulation and control module carries out parameter regulation and control on the test box according to the control parameters, further the test box carries out alternating salt spray test operation on the opening and closing coil, the data acquisition module acquires original current data and test current data, and the upper computer module carries out training operation based on a regression algorithm of a support vector machine according to the original current data and the test current data, so that a training model is obtained. The aging test of the opening and closing coil of the circuit breaker in the prior art is solved, no accurate and effective method is available, so that the phenomenon that the circuit breaker is refused due to the fact that the opening and closing coil cannot be replaced in time is easy, the climate condition can be simulated, the service life evaluation information of the opening and closing coil can be effectively obtained, and the opening and closing coil can be replaced in time.
As an improvement of the above scheme, the training operation based on the support vector machine regression algorithm specifically includes:
constructing a training set input matrix and a regression value corresponding to the training set input matrix;
and carrying out regression modeling by using a support vector machine according to the training set input matrix and the regression value.
As an improvement of the above scheme, the upper computer module is further configured to:
and according to the training model, taking historical current data of the opening and closing coil of the circuit breaker as input data, and carrying out service life assessment on the opening and closing coil.
As an improvement of the above scheme, the preset data includes a preset spraying time, a preset spraying amount, a preset temperature, a preset humidity and a preset test time.
As a modification of the above, the alternating salt spray test operation comprises four test cycles, each comprising a two hour spray cycle and a seven hour storage cycle.
Drawings
Fig. 1 is a flowchart of a service life evaluation method of an opening and closing coil provided by an embodiment of the present invention;
fig. 2 is a flowchart of an alternating salt spray test operation in a service life evaluation method of an opening and closing coil provided by an embodiment of the invention;
fig. 3 is a flowchart of constructing a training model and using the training model in a lifetime assessment method of an opening and closing coil provided by an embodiment of the present invention;
fig. 4 is a block diagram of a lifetime assessment device for an opening/closing coil according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Example 1
Referring to fig. 1, fig. 1 is a flowchart of a service life evaluation method of an opening and closing coil provided by an embodiment of the present invention; comprising the following steps:
s1, acquiring original current data of at least two opening and closing coils;
s2, performing alternating salt spray test operation on each opening and closing coil according to preset data;
s3, acquiring test current data of each opening and closing coil after the alternating salt spray test operation is executed;
s4, installing the opening and closing coil on a circuit breaker, and judging whether the opening and closing coil can trigger an opening and closing mechanism of the circuit breaker;
s5, if yes, continuing to execute the alternating salt spray test operation on the opening and closing coil until the opening and closing coil cannot touch the opening and closing mechanism;
and S6, if not, training operation based on a support vector machine regression algorithm is carried out according to the original current data and the test current data, and a training model is obtained.
Specifically, in step S1, before the alternating salt spray test operation is performed on the opening and closing coils, the original current data of the opening and closing coils are obtained, and preferably, the opening and closing coils include at least two opening and closing coils, and each opening and closing coil has a corresponding number. In the embodiment of the present invention, the most preferred solution is that the number of the switching coils is 10.
Specifically, in step S2, an alternating salt spray test operation may be performed on each of the switching coils according to preset data in a test box, preferably, the test box performs an alternating salt spray test operation on each of the switching coils by using an alternating salt spray test grade of GB/T2423.18, and a specific flow may refer to fig. 2. Preferably, the preset data includes a preset spraying time, a preset spraying amount, a preset temperature, a preset humidity and a preset test time. Preferably, the alternating salt spray test operation comprises four test cycles, each comprising a two hour spray cycle and a seven hour storage cycle. Preferably, the spray is a 5% sodium chloride solution, the preset temperature is 40 ℃, the preset humidity is 95%, and the preset time is the total duration of four test periods of the alternating salt spray test operation.
Specifically, in step S3, test current data of each of the switching-on/off coils is obtained after the alternating salt spray test operation is performed, and the test time and the corresponding coil number are recorded in detail.
Specifically, in step S4, the opening and closing coil after the execution of the alternating salt spray test operation is installed on a circuit breaker, and whether the opening and closing coil can trigger an opening and closing mechanism of the circuit breaker is judged; if the opening and closing coil can also touch the opening and closing mechanism of the circuit breaker, the opening and closing coil can also work normally; if the opening and closing coil cannot touch the opening and closing mechanism of the circuit breaker, the performance of the opening and closing coil is deteriorated, and the normal operation of the opening and closing coil cannot be performed any more.
Specifically, in step S5, if the opening/closing coil can also touch the opening/closing mechanism of the circuit breaker, the alternating salt spray test operation is continuously performed on the opening/closing coil until the opening/closing coil cannot touch the opening/closing mechanism.
Specifically, in step S6, if the opening/closing coil cannot trigger the opening/closing mechanism of the circuit breaker, training operation based on a support vector machine regression algorithm is performed according to the original current data and the test current data, so as to obtain a training model. Preferably, step S6 specifically includes:
s61, constructing a training set input matrix and a regression value corresponding to the training set input matrix;
s62, carrying out regression modeling by using a support vector machine according to the training set input matrix and the regression value.
Specifically, in step S61, test current data of the opening and closing coil after the alternating salt spray test operation is collated, preferably, the test current data may be a current curve of the opening and closing coil, and a current average value of the opening and closing coil after N times of alternating salt spray test operation is calculated to obtain the current average value array, where N is an integer, and N is greater than or equal to 0; wherein, the current average value array is:
iN＝[xN1，xN2，…，xNn]；
wherein iN represents an array of current average values of the opening and closing coils after the alternating salt spray test operation is performed for N times; n represents the number of test points of the opening and closing coil; xNn represents the average value of the current at the nth test point on each of the opening and closing coils in the nth alternating salt spray test operation. And when N is equal to 0, the average value of the current curve of the opening and closing coil when the alternating salt spray test operation is not performed is represented.
For example, a current curve, which is composed of a plurality of test points, if the sampling rate of the current curve i is 15kHZ and the sampling time is 100ms, then a curve has 15kHZ and 0.1s=1500 test points, i.e., n=1500. At this time, each of the switching coils adopts a current curve of 1500 test points. If the number of the switching-on/off coils is 10, current average values are calculated corresponding to each test point, and finally a current average value array of 1500 test points is obtained. The current average value array is the result after each alternating salt spray test operation.
Specifically, if N is greater than or equal to 5, i0 to i4 are taken to form a training set, and i5 to iN are taken as a test set. The training set can be used for modeling so as to obtain the training model, the testing set is used for testing the accuracy of the training model, and the service life of the opening and closing coil can be evaluated by using the training model on the basis of ensuring that the training model is accurate and reliable. If tested, the accuracy is low, and then re-modeling is required.
Specifically, the current average value arrays i0 to i4 are connected into a current curve one-dimensional array, and then the current curve one-dimensional array is expressed as: i= [ i0, i1, i2, i3, i4], i being renumbered for convenience of description, let i= [ x1, x2, …, x5n ], the window function T [ m ] = { tk, tk+1, …, tk+m-1}, k=1, 2, …,5n-m; m=3n, t is time. The window function Tm is used to slide the value on i, each time the data with length m is intercepted, tm= { tk, tk+1, …, tk+m-1} is taken as the input value, and tk+m is taken as the regression value of the sequence. Constructing the training set input matrix X and a regression value Y corresponding to the training set input matrix.
Y＝[y1,y2,…,yq] T ＝[x3n+1,x3n+2,…,x5n] T ；
Wherein the training set input matrix X is a matrix of p×q, p=3n, q=2n.
Specifically, in step S62, the training set input matrix X is represented in the form of a row vector, where x= [ X1, X2, …, xq] T The method comprises the steps of carrying out a first treatment on the surface of the Wherein Xq represents x2n, …, x5n-1; the regression values are simultaneously expressed as y= [ Y1, Y2, …, yq] T The method comprises the steps of carrying out a first treatment on the surface of the Wherein y1=x3n+1, yq=x5n.
Specifically, regression modeling is performed by using a support vector machine according to the training set input matrix and the regression value, so as to obtain a regression equation (i.e. training model) of current changing along with time, wherein the expression of the regression equation is as follows:
y＝xw+b；
wherein; w= [ w1, w2, …, wp] T ，b＝[b1，b2，…，bq] T . Specifically, when the regression equation is obtained, x=i2 (the average value of the current of the opening and closing coil after the 2 nd alternating salt spray test operation) is input, that is, the future change condition of the current waveform of the opening and closing coil, that is, the predicted value y, can be predicted, and at this time, the predicted value can be compared with the test set, so that the reliability of the regression equation can be determined.
Specifically, the process of solving w, b refers to the following formula (translated to minimize a quadratic convex programming problem with linear inequality constraints):
y i -((x i ·w)+b)≤ε+ξ i formula (3);
wherein, the liquid crystal display device comprises a liquid crystal display device,
c is penalty factor, and->
For relaxation variables, min R (w) refers to the minimum of the solving function R (w), s.t. (x) i W) +b) means at (x) i Solving for the minimum value of R (w) under the conditions of w) +b.
The penalty factor C is a positive constant, is a compromise between the complexity of the function regression model and the fitting precision of the sample, and is higher as the value is larger; epsilon is the maximum error allowed by regression, and controls the number of support vectors and generalization capability, and the larger the value is, the fewer the support vectors are. By utilizing the dual principle, simultaneously introducing a Lagrangian multiplier a and a kernel function K, and converting the formulas (1) to (3) into the following formulas:
wherein, the liquid crystal display device comprises a liquid crystal display device,
a i and->
Represents that two times of accumulation are carried out in one equation, j is the number of times of the control accumulation process, and the kernel function K selects a radial basis kernel function K (x i ,x j )＝exp(-γ||x i -x j I) 2), where γ is a parameter of the radial basis function itself, implicitly determining the distribution of the data mapped to the new feature space, the larger γ, the fewer support vectors, the smaller γ values, and the more support vectors. The number of support vectors affects the speed of training and prediction. In the formula (4), a is a Lagrange multiplier introduced when a solving problem is applied to solving by a Lagrange multiplier method, and the key of solving the Lagrange multiplier is that a solving process of a can be realized by a lot of software, such as a sklearn library in libsvm, python, so that w can be obtained. The Lagrangian multiplier method is a common method of solving the optimization problem. For the training set, y, x is known and w is found. Since y=xw+b, substituting this formula can solve b.
Further, after the training model is obtained in step S6, step S7 is further included:
and S7, taking historical current data of the opening and closing coil of the circuit breaker as input data according to the training model, and carrying out service life assessment on the opening and closing coil. Specifically, referring to fig. 3, a historical current waveform of an opening and closing coil of a circuit breaker in use is taken as an input value X, and a future change condition of the opening and closing coil current waveform is predicted. If the current waveform has a tendency of gradually becoming smaller or larger, the performance of the switching-on/switching-off coil is gradually deteriorated, and the switching-on/switching-off coil needs to be replaced.
Compared with the prior art, the service life assessment method of the switching-on/off coils disclosed by the invention is characterized in that alternating salt spray test operation is carried out on each switching-on/off coil according to the preset parameters, test current data are obtained, and then whether the switching-on/off coil can trigger a switching-on/off mechanism of a circuit breaker is judged; if yes, continuing to execute the alternating salt spray test operation on the opening and closing coil; and otherwise, training operation based on a support vector machine regression algorithm is performed according to the pre-selected and acquired original current data of the opening and closing coil and the test current data, so that a training model is obtained. The aging test of the opening and closing coil of the circuit breaker in the prior art is solved, no accurate and effective method is available, so that the phenomenon that the circuit breaker is refused due to the fact that the opening and closing coil cannot be replaced in time is easy, the climate condition can be simulated, the service life evaluation information of the opening and closing coil can be effectively obtained, and the opening and closing coil can be replaced in time.
Example two
Referring to fig. 4, fig. 4 is a block diagram of a lifetime assessment device of an opening/closing coil according to an embodiment of the present invention; comprising the following steps:
the test box 1 is used for executing alternating salt spray test operation on the opening and closing coil;
the upper computer module 2 is used for setting preset data and simultaneously carrying out training operation based on a support vector machine regression algorithm according to the original current data and the test current data to obtain a training model;
the main control circuit module 3 is used for adjusting the control parameters of the regulation and control module 5 according to the preset data;
the data acquisition module 4 is used for acquiring original current data and test current data;
and the regulation and control module 5 is used for carrying out parameter regulation and control on the test box 1 according to the control parameters.
Specifically, the data acquisition module 4 includes a temperature sensor 41, a humidity sensor 42, and a hall current sensor 43. Before the alternating salt spray test operation is performed on the opening and closing coils, the initial current data of the opening and closing coils are obtained through the hall current sensor 43, and preferably, the opening and closing coils comprise at least two opening and closing coils, and each opening and closing coil has a corresponding number. In the embodiment of the present invention, the most preferred solution is that the number of the switching coils is 10.
Specifically, the test box 1 may perform an alternating salt spray test operation on each of the switching-on/off coils according to preset data, and preferably, the test box 1 performs an alternating salt spray test operation on each of the switching-on/off coils by adopting an alternating salt spray test grade of GB/T2423.18, and a specific flow may refer to fig. 2.
Specifically, the upper computer module 2 mainly comprises a computer, a serial port communication program and a control program, the upper computer module 2 is used for controlling the starting and stopping of the service life evaluation device of the whole opening and closing coil, setting the preset parameters and simultaneously receiving the data of the data acquisition module 4, so that training operation based on a support vector machine regression algorithm is performed according to the original current data and the test current data, and a training model is obtained. Preferably, the preset data includes a preset spraying time, a preset spraying amount, a preset temperature, a preset humidity and a preset test time. Preferably, the alternating salt spray test operation comprises four test cycles, each comprising a two hour spray cycle and a seven hour storage cycle. Preferably, the spray is a 5% sodium chloride solution, the preset temperature is 40 ℃, the preset humidity is 95%, and the preset time is the total duration of four test periods of the alternating salt spray test operation.
Specifically, the main control circuit module 3 includes a signal conditioning circuit 31, a singlechip 32, a serial communication circuit 33 and a driving circuit 34; wherein; the singlechip 32 is respectively connected with the signal conditioning circuit 31, the serial communication circuit 33 and the driving circuit 34, the signal conditioning circuit 31 is connected with the data acquisition module 4, the serial communication circuit 33 is connected with the upper computer module 2, the singlechip 32 plays a control role, and the driving circuit 34 is used for driving the operation of the regulation and control module 5. The main control circuit module 3 receives preset data of the upper computer module 1 through the serial communication circuit 33, adjusts control parameters of the regulation and control module 5 according to the preset data, and meanwhile, the main control circuit module 3 receives signals acquired by the data acquisition module 4 through the signal conditioning circuit 31 and transmits the signals back to the upper computer module 1 for analysis.
Preferably, the test chamber 1 is connected with the control module 5; the regulation and control module 5 comprises a spraying device 51, a temperature regulating device 52 and a humidity regulating device 53, wherein the spraying device 51 comprises an automatic nozzle and a beaker, the temperature regulating device 52 comprises a resistance wire heater and a compressor, and the humidity regulating device 53 comprises a humidifier and an air drying agent. And the parameter control module is used for carrying out parameter control on the test box 1 according to the control parameters sent by the main control circuit module 3. So that the spraying time in the test box 1 meets the preset spraying time, the spraying quantity meets the preset spraying quantity, the temperature meets the preset temperature, the humidity meets the preset humidity, and the test time meets the preset test time.
Specifically, after the alternating salt spray test operation is performed, test current data of each opening and closing coil is obtained through the hall current sensor 43, and the test time and the corresponding coil number are recorded in detail.
Specifically, the switching-on/off coil after the alternating salt spray test operation is executed is installed on a circuit breaker, and whether the switching-on/off coil can trigger a switching-on/off mechanism of the circuit breaker is judged; if the opening and closing coil can also touch the opening and closing mechanism of the circuit breaker, the opening and closing coil can also work normally; if the opening and closing coil cannot touch the opening and closing mechanism of the circuit breaker, the performance of the opening and closing coil is deteriorated, and the normal operation of the opening and closing coil cannot be performed any more.
Specifically, if the opening and closing coil can also touch the opening and closing mechanism of the circuit breaker, the alternating salt spray test operation is continuously performed on the opening and closing coil until the opening and closing coil cannot touch the opening and closing mechanism.
Specifically, if the opening and closing coil cannot touch the opening and closing mechanism of the circuit breaker, the upper computer module 1 is further configured to construct a training set input matrix and a regression value corresponding to the training set input matrix; and carrying out regression modeling by using a support vector machine according to the training set input matrix and the regression value.
Specifically, test current data of the switching-on/off coil after the alternating salt spray test operation is arranged, preferably, the test current data can be a current curve of the switching-on/off coil, and a current average value of the switching-on/off coil after N times of alternating salt spray test operation is calculated to obtain the current average value array, wherein N is an integer, and N is greater than or equal to 0; wherein, the current average value array is:
iN＝[xN1，xN2，…，xNn]；
wherein iN represents an array of current average values of the opening and closing coils after the alternating salt spray test operation is performed for N times; n represents the number of test points of the opening and closing coil; xNn represents the average value of the current at the nth test point on each of the opening and closing coils in the nth alternating salt spray test operation. And when N is equal to 0, the average value of the current curve of the opening and closing coil when the alternating salt spray test operation is not performed is represented.
For example, a current curve, which is composed of a plurality of test points, if the sampling rate of the current curve i is 15kHZ and the sampling time is 100ms, then a curve has 15kHZ and 0.1s=1500 test points, i.e., n=1500. At this time, each of the switching coils adopts a current curve of 1500 test points. If the number of the switching-on/off coils is 10, current average values are calculated corresponding to each test point, and finally a current average value array of 1500 test points is obtained. The current average value array is the result after each alternating salt spray test operation.
Specifically, if N is greater than or equal to 5, i0 to i4 are taken to form a training set, and i5 to iN are taken as a test set. The training set can be used for modeling so as to obtain the training model, the testing set is used for testing the accuracy of the training model, and the service life of the opening and closing coil can be evaluated by using the training model on the basis of ensuring that the training model is accurate and reliable. If tested, the accuracy is low, and then re-modeling is required.
Specifically, the current average value arrays i0 to i4 are connected into a current curve one-dimensional array, and then the current curve one-dimensional array is expressed as: i= [ i0, i1, i2, i3, i4], i being renumbered for convenience of description, let i= [ x1, x2, …, x5n ], the window function T [ m ] = { tk, tk+1, …, tk+m-1}, k=1, 2, …,5n-m; m=3n, t is time. The window function Tm is used to slide the value on i, each time the data with length m is intercepted, tm= { tk, tk+1, …, tk+m-1} is taken as the input value, and tk+m is taken as the regression value of the sequence. Constructing the training set input matrix X and a regression value Y corresponding to the training set input matrix.
Y＝[y1,y2,…,yq] T ＝[x3n+1,x3n+2,…,x5n] T ；
Wherein the training set input matrix X is a matrix of p×q, p=3n, q=2n.
Specifically, the training set input matrix X is expressed in the form of a row vector, where x= [ X1, X2, …, xq] T The method comprises the steps of carrying out a first treatment on the surface of the Wherein Xq represents x2n, …, x5n-1; the regression values are simultaneously expressed as y= [ Y1, Y2, …, yq] T The method comprises the steps of carrying out a first treatment on the surface of the Wherein y1=x3n+1, yq=x5n.
Specifically, regression modeling is performed by using a support vector machine according to the training set input matrix and the regression value, so as to obtain a regression equation (i.e. training model) of current changing along with time, wherein the expression of the regression equation is as follows:
y＝xw+b；
wherein; w= [ w1, w2, …, wp] T ，b＝[b1，b2，…，bq] T . Specifically, when the regression equation is obtained, x=i2 (the average value of the current of the opening and closing coil after the 2 nd alternating salt spray test operation) is input, that is, the future change condition of the current waveform of the opening and closing coil, that is, the predicted value y, can be predicted, and at this time, the predicted value can be compared with the test set, so that the reliability of the regression equation can be determined.
Specifically, the process of solving w, b refers to the following formula (translated to minimize a quadratic convex programming problem with linear inequality constraints):
y i -((x i ·w)+b)≤ε+ξ i formula (3);
wherein, the liquid crystal display device comprises a liquid crystal display device,
c is penalty factor, and->
For relaxation variables, min R (w) refers to the minimum of the solving function R (w), s.t. (x) i W) +b) means at (x) i Solving for the minimum value of R (w) under the conditions of w) +b.
The penalty factor C is a positive constant, is a compromise between the complexity of the function regression model and the fitting precision of the sample, and is higher as the value is larger; epsilon is the maximum error allowed by regression, and controls the number of support vectors and generalization capability, and the larger the value is, the fewer the support vectors are. By utilizing the dual principle, simultaneously introducing a Lagrangian multiplier a and a kernel function K, and converting the formulas (1) to (3) into the following formulas:
wherein, the liquid crystal display device comprises a liquid crystal display device,
a i and->
Represents that two times of accumulation are carried out in one equation, j is the number of times of the control accumulation process, and the kernel function K selects a radial basis kernel function K (x i ,x j )＝exp(-γ||x i -x j I) 2), where γ is a parameter of the radial basis function itself, implicitly determining the distribution of the data mapped to the new feature space, the larger γ, the fewer support vectors, the smaller γ values, and the more support vectors. The number of support vectors affects the speed of training and prediction. In the formula (4), a is a Lagrange multiplier introduced when a solving problem is applied to solving by a Lagrange multiplier method, and the key of solving the Lagrange multiplier is that a solving process of a can be realized by a lot of software, such as a sklearn library in libsvm, python, so that w can be obtained. The Lagrangian multiplier method is a common method of solving the optimization problem. For the training set, y, x is known and w is found. Since y=xw+b, substituting this formula can solve b.
Further, after the training model is obtained, the upper computer module 1 is further configured to perform life assessment on the opening and closing coil according to the training model by using historical current data of the opening and closing coil of the circuit breaker as input data.
Specifically, according to the training model, historical current data of an opening and closing coil of the circuit breaker is used as input data, and service life of the opening and closing coil is evaluated. Specifically, referring to fig. 3, a historical current waveform of an opening and closing coil of a circuit breaker in use is taken as an input value X, and a future change condition of the opening and closing coil current waveform is predicted. If the current waveform has a tendency of gradually becoming smaller or larger, the performance of the switching-on/switching-off coil is gradually deteriorated, and the switching-on/switching-off coil needs to be replaced.
Compared with the prior art, the service life assessment device for the opening and closing coil disclosed by the invention has the advantages that the control parameters of the regulation and control module 5 are adjusted through the main control circuit module 3 according to the preset parameters set by the upper computer module 2, so that the regulation and control module 5 carries out parameter regulation and control on the test box 1 according to the control parameters, further the test box 1 carries out alternating salt spray test operation on the opening and closing coil, the data acquisition module 4 acquires original current data and test current data, and the upper computer module 2 carries out training operation based on a regression algorithm of a support vector machine according to the original current data and the test current data, so that a training model is obtained. The aging test of the opening and closing coil of the circuit breaker in the prior art is solved, no accurate and effective method is available, so that the phenomenon that the circuit breaker is refused due to the fact that the opening and closing coil cannot be replaced in time is easy, the climate condition can be simulated, the service life evaluation information of the opening and closing coil can be effectively obtained, and the opening and closing coil can be replaced in time.
While the foregoing is directed to the preferred embodiments of the present invention, it will be appreciated by those skilled in the art that changes and modifications may be made without departing from the principles of the invention, such changes and modifications are also intended to be within the scope of the invention.
Claims (8)
1. The service life assessment method of the opening and closing coil is characterized by comprising the following steps of:
acquiring original current data of at least two opening and closing coils;
according to preset data, alternating salt spray test operation is carried out on each opening and closing coil;
acquiring test current data of each opening and closing coil after the alternating salt spray test operation is executed;
the switching-on/off coil is arranged on the circuit breaker, and whether the switching-on/off coil can trigger a switching-on/off mechanism of the circuit breaker is judged;
if yes, continuing to execute the alternating salt spray test operation on the opening and closing coil until the opening and closing coil cannot touch the opening and closing mechanism;
if not, training operation based on a support vector machine regression algorithm is carried out according to the original current data and the test current data, and a training model is obtained;
and according to the training model, taking historical current data of the opening and closing coil of the circuit breaker as input data, and carrying out service life assessment on the opening and closing coil.
2. The method for evaluating the service life of an opening and closing coil according to claim 1, wherein the training operation based on the support vector machine regression algorithm specifically comprises:
constructing a training set input matrix and a regression value corresponding to the training set input matrix;
and carrying out regression modeling by using a support vector machine according to the training set input matrix and the regression value.
3. The lifetime assessment method of an opening/closing coil according to claim 1, wherein the preset data includes a preset spraying time, a preset spraying amount, a preset temperature, a preset humidity, and a preset test time.
4. The method of claim 1, wherein the alternating salt spray test operation comprises four test cycles, each comprising a two hour spray cycle and a seven hour storage cycle.
5. A lifetime assessment device for an opening/closing coil, comprising:
the test box is used for executing alternating salt spray test operation on the opening and closing coil;
the upper computer module is used for setting preset data and simultaneously carrying out training operation based on a support vector machine regression algorithm according to the original current data and the test current data to obtain a training model;
the upper computer module is also used for:
according to the training model, historical current data of an opening and closing coil of a circuit breaker is used as input data, and service life evaluation is carried out on the opening and closing coil;
the main control circuit module is used for adjusting control parameters of the regulation and control module according to the preset data;
the data acquisition module is used for acquiring original current data and test current data;
and the regulation and control module is used for carrying out parameter regulation and control on the test box according to the control parameters.
6. The lifetime assessment device of an opening/closing coil according to claim 5, wherein the training operation based on the support vector machine regression algorithm specifically comprises:
constructing a training set input matrix and a regression value corresponding to the training set input matrix;
and carrying out regression modeling by using a support vector machine according to the training set input matrix and the regression value.
7. The lifetime assessment device of an opening/closing coil according to claim 5, wherein the preset data includes a preset spraying time, a preset spraying amount, a preset temperature, a preset humidity, and a preset test time.
8. The switching-on/off coil life assessment device according to claim 5, wherein the alternating salt spray test operation includes four test periods, each of which includes a two-hour spray period and a seven-hour storage period.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810330732.5A CN108508298B (en) | 2018-04-13 | 2018-04-13 | Service life assessment method and device for switching-on/off coil |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810330732.5A CN108508298B (en) | 2018-04-13 | 2018-04-13 | Service life assessment method and device for switching-on/off coil |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108508298A CN108508298A (en) | 2018-09-07 |
| CN108508298B true CN108508298B (en) | 2023-05-23 |
Family
ID=63381720
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810330732.5A Active CN108508298B (en) | 2018-04-13 | 2018-04-13 | Service life assessment method and device for switching-on/off coil |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108508298B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109752609B (en) * | 2018-12-06 | 2021-02-26 | 国网江苏省电力有限公司南京供电分公司 | Electric power equipment operation reliability testing device, testing center and testing judgment method |
| CN110426586A (en) * | 2019-08-16 | 2019-11-08 | 南方电网科学研究院有限责任公司 | A kind of pilot system for assessing the secondary cabinet service life |
| CN111947906A (en) * | 2020-08-10 | 2020-11-17 | 广西电网有限责任公司电力科学研究院 | GIS opening and closing coil's action life test system |
| CN112611924B (en) * | 2020-11-20 | 2022-07-22 | 南方电网科学研究院有限责任公司 | Method for evaluating service life of circuit breaker mechanism |
| CN112731122A (en) * | 2020-11-25 | 2021-04-30 | 广西电网有限责任公司电力科学研究院 | Method and device for evaluating service life of circuit breaker mechanism |
| CN112798948B (en) * | 2020-12-30 | 2022-12-06 | 广西电网有限责任公司电力科学研究院 | State evaluation method and device for GIS opening and closing coil |
| CN113654977B (en) * | 2021-08-17 | 2024-04-30 | 苏州布鲁比科技有限公司 | Method and device for predicting residual life of opening and closing coil |
| CN113435795B (en) * | 2021-08-26 | 2021-12-14 | 南方电网科学研究院有限责任公司 | State evaluation method, device, equipment and medium for power equipment |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090063115A1 (en) * | 2007-08-31 | 2009-03-05 | Zhao Lu | Linear programming support vector regression with wavelet kernel |
| CN102331543A (en) * | 2011-06-23 | 2012-01-25 | 上海市安全生产科学研究所 | Support vector machine based fault electric arc detection method |
| CN102721920A (en) * | 2012-06-29 | 2012-10-10 | 沈阳工业大学 | Prediction device and method for remaining life of operating mechanism of 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 |
| CN103698698A (en) * | 2014-01-10 | 2014-04-02 | 中国南方电网有限责任公司超高压输电公司检修试验中心 | Diagnostic method of electrical life of high-voltage circuit breaker based on fuzzy theory |
| CN104777092A (en) * | 2015-04-14 | 2015-07-15 | 电子科技大学 | Printed circuit board service life assessment method based on salt-spray environment test |
| CN105242199A (en) * | 2015-09-28 | 2016-01-13 | 国家电网公司 | Real-time diagnosis method and apparatus of opening-closing coil state of high-voltage breaker |
| US20160084919A1 (en) * | 2013-05-06 | 2016-03-24 | University Of Saskatchewan | Methods and apparatus for detection of transient instability and out-of-step conditions by state deviation |
| CN105868770A (en) * | 2016-03-23 | 2016-08-17 | 国网山东省电力公司电力科学研究院 | High-voltage circuit breaker fault diagnosis method based on unsupervised learning model |
| CN106019131A (en) * | 2016-05-18 | 2016-10-12 | 四川大学 | State comprehensive evaluation method for high-voltage circuit breaker operating mechanism based on switching-on and switching-off coil currents |
| 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 |
-
2018
- 2018-04-13 CN CN201810330732.5A patent/CN108508298B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090063115A1 (en) * | 2007-08-31 | 2009-03-05 | Zhao Lu | Linear programming support vector regression with wavelet kernel |
| CN102331543A (en) * | 2011-06-23 | 2012-01-25 | 上海市安全生产科学研究所 | Support vector machine based fault electric arc detection method |
| CN102721920A (en) * | 2012-06-29 | 2012-10-10 | 沈阳工业大学 | Prediction device and method for remaining life of operating mechanism of circuit breaker |
| US20160084919A1 (en) * | 2013-05-06 | 2016-03-24 | University Of Saskatchewan | Methods and apparatus for detection of transient instability and out-of-step conditions by state deviation |
| CN103336243A (en) * | 2013-07-01 | 2013-10-02 | 东南大学 | Breaker fault diagnosis method based on separating/closing coil current signals |
| CN103698698A (en) * | 2014-01-10 | 2014-04-02 | 中国南方电网有限责任公司超高压输电公司检修试验中心 | Diagnostic method of electrical life of high-voltage circuit breaker based on fuzzy theory |
| CN104777092A (en) * | 2015-04-14 | 2015-07-15 | 电子科技大学 | Printed circuit board service life assessment method based on salt-spray environment test |
| CN105242199A (en) * | 2015-09-28 | 2016-01-13 | 国家电网公司 | Real-time diagnosis method and apparatus of opening-closing coil state of high-voltage breaker |
| CN105868770A (en) * | 2016-03-23 | 2016-08-17 | 国网山东省电力公司电力科学研究院 | High-voltage circuit breaker fault diagnosis method based on unsupervised learning model |
| CN106019131A (en) * | 2016-05-18 | 2016-10-12 | 四川大学 | State comprehensive evaluation method for high-voltage circuit breaker operating mechanism based on switching-on and switching-off coil currents |
| 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 |
Non-Patent Citations (2)
| Title |
|---|
| 李恒真等.基于K-S检验法和SVM的高压断路器分合闸线圈回路故障诊断.《电气应用》.2016,第35卷(第9期),全文. * |
| 陈亚林等.基于拟合优度检验法和支持向量机的断路器分合闸回路故障辨识研究.《自动化与仪器仪表》.2017,第2017卷(第12期),全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108508298A (en) | 2018-09-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108508298B (en) | Service life assessment method and device for switching-on/off coil | |
| Mathieu et al. | State estimation and control of electric loads to manage real-time energy imbalance | |
| JP6298214B2 (en) | System and method for maximizing the expected utility of signal injection test patterns within a utility grid | |
| CN109387779B (en) | Method for predicting residual life of operation accessory of universal circuit breaker based on statistical data drive | |
| Mathieu et al. | State estimation and control of heterogeneous thermostatically controlled loads for load following | |
| US10248084B2 (en) | Controller, control method, and control program | |
| CN103218675B (en) | A kind of based on the short-term load forecasting method of cluster and moving window | |
| Ziekow et al. | The potential of smart home sensors in forecasting household electricity demand | |
| WO2017153991A1 (en) | A method and system for optimizing and predicting demand response | |
| KR102516910B1 (en) | Method and device for diagnosing wear of an electrical switching unit, and electrical unit comprising such a device | |
| Ledva et al. | Real-time energy disaggregation of a distribution feeder's demand using online learning | |
| CN108062608A (en) | The distribution switchgear reliability estimation method of meter and environmental factor | |
| CN115408864A (en) | Electronic transformer error state self-adaptive prediction method, system and equipment | |
| Wang et al. | A DRL-aided multi-layer stability model calibration platform considering multiple events | |
| Godina et al. | Model predictive control technique for energy optimization in residential sector | |
| CN106093130B (en) | Method and system for carrying out thermoelectric combined test on test article coated with electric composite grease | |
| CN114252731A (en) | Relay action characteristic evaluation method and device based on multiple parameters | |
| CN111121971A (en) | Method for judging fault type through surface temperature change of high-voltage switch cabinet | |
| WO2020137260A1 (en) | Relay state prediction device, relay state prediction system, relay state prediction method, and program | |
| CN110174588A (en) | The estimation of electrical network parameter | |
| CN113515723B (en) | Comprehensive evaluation method for minimum action voltage of circuit breaker | |
| KR20170014358A (en) | Temperature-rise test automation system and method for electric power equipment using prediction algorithm | |
| Chen et al. | Condition assessment of distribution automation remote terminal units based on double-layer improved cloud model | |
| Galiev et al. | Development of an Integrated Expert System for Distribution Network Diagnostics Based on Artificial Intelligence Technology | |
| Nakutis et al. | A short-term power fluctuations averaging in low voltage AC grid |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20200602 Address after: 510663 Building 3, 4, 5 and 3 of Building J1, 11 Kexiang Road, Science City, Luogang District, Guangzhou City, Guangdong Province Applicant after: ELECTRIC POWER Research Institute CHINA SOUTHERN POWER GRID Applicant after: CHINA SOUTHERN POWER GRID Co.,Ltd. Address before: 510670 No. 11 Kexiang Road, Science City, Huangpu District, Guangzhou City, Guangdong Province Applicant before: ELECTRIC POWER Research Institute CHINA SOUTHERN POWER GRID |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |