CN113740674B - Dielectric response test method, system and storage medium - Google Patents
Dielectric response test method, system and storage medium Download PDFInfo
- Publication number
- CN113740674B CN113740674B CN202110814843.5A CN202110814843A CN113740674B CN 113740674 B CN113740674 B CN 113740674B CN 202110814843 A CN202110814843 A CN 202110814843A CN 113740674 B CN113740674 B CN 113740674B
- Authority
- CN
- China
- Prior art keywords
- curve
- insulating medium
- dielectric
- test
- frequency domain
- 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
- 230000004044 response Effects 0.000 title claims abstract description 154
- 238000010998 test method Methods 0.000 title claims abstract description 62
- 238000012360 testing method Methods 0.000 claims abstract description 207
- 238000001453 impedance spectrum Methods 0.000 claims abstract description 68
- 230000032683 aging Effects 0.000 claims abstract description 46
- 238000007781 pre-processing Methods 0.000 claims abstract description 44
- 238000012549 training Methods 0.000 claims abstract description 37
- 238000003062 neural network model Methods 0.000 claims abstract description 35
- 239000000463 material Substances 0.000 claims abstract description 27
- 230000010287 polarization Effects 0.000 claims description 40
- 238000001228 spectrum Methods 0.000 claims description 38
- 238000000034 method Methods 0.000 claims description 25
- 230000028161 membrane depolarization Effects 0.000 claims description 24
- 238000002203 pretreatment Methods 0.000 claims description 22
- 238000004364 calculation method Methods 0.000 claims description 19
- 238000004590 computer program Methods 0.000 claims description 11
- 238000001514 detection method Methods 0.000 claims description 8
- 230000006870 function Effects 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 5
- 230000002336 repolarization Effects 0.000 claims description 4
- 239000003990 capacitor Substances 0.000 claims description 3
- 238000001566 impedance spectroscopy Methods 0.000 claims description 3
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 239000003921 oil Substances 0.000 description 22
- 238000009413 insulation Methods 0.000 description 9
- 239000002253 acid Substances 0.000 description 6
- 238000002474 experimental method Methods 0.000 description 6
- 239000011810 insulating material Substances 0.000 description 6
- 239000000243 solution Substances 0.000 description 4
- 239000002131 composite material Substances 0.000 description 3
- 238000001035 drying Methods 0.000 description 3
- 238000007796 conventional method Methods 0.000 description 2
- 238000003745 diagnosis Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- YLLIGHVCTUPGEH-UHFFFAOYSA-M potassium;ethanol;hydroxide Chemical compound [OH-].[K+].CCO YLLIGHVCTUPGEH-UHFFFAOYSA-M 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 238000003878 thermal aging Methods 0.000 description 2
- 229920000784 Nomex Polymers 0.000 description 1
- 230000002999 depolarising effect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 239000002480 mineral oil Substances 0.000 description 1
- 235000010446 mineral oil Nutrition 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000004763 nomex Substances 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 229920002545 silicone oil Polymers 0.000 description 1
- 239000012086 standard solution Substances 0.000 description 1
- 238000004154 testing of material Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
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
- G01R31/12—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing
- G01R31/1227—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing of components, parts or materials
- G01R31/1263—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing of components, parts or materials of solid or fluid materials, e.g. insulation films, bulk material; of semiconductors or LV electronic components or parts; of cable, line or wire insulation
- G01R31/1272—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing of components, parts or materials of solid or fluid materials, e.g. insulation films, bulk material; of semiconductors or LV electronic components or parts; of cable, line or wire insulation of cable, line or wire insulation, e.g. using partial discharge measurements
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R27/00—Arrangements for measuring resistance, reactance, impedance, or electric characteristics derived therefrom
- G01R27/02—Measuring real or complex resistance, reactance, impedance, or other two-pole characteristics derived therefrom, e.g. time constant
- G01R27/26—Measuring inductance or capacitance; Measuring quality factor, e.g. by using the resonance method; Measuring loss factor; Measuring dielectric constants ; Measuring impedance or related variables
- G01R27/2688—Measuring quality factor or dielectric loss, e.g. loss angle, or power factor
- G01R27/2694—Measuring dielectric loss, e.g. loss angle, loss factor or power factor
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Power Engineering (AREA)
- Measurement Of Resistance Or Impedance (AREA)
- Investigating Or Analyzing Materials By The Use Of Electric Means (AREA)
Abstract
The application relates to a dielectric response test method, comprising the following steps: obtaining a dielectric response prediction model, wherein the dielectric response prediction model is a model obtained by training a preprocessing curve through a preset neural network model, and the preprocessing curve is a frequency domain dielectric spectrum curve of a preprocessing insulating medium; acquiring a characteristic curve of the insulating medium to be tested according to the dielectric response prediction model and the characteristic parameters of the insulating medium to be tested, wherein the characteristic curve is a frequency domain dielectric spectrum curve of the insulating medium to be tested, and the characteristic parameters comprise at least one of the material type, the temperature and the aging degree of the insulating medium to be tested; and acquiring dielectric response information of the insulating medium to be tested according to the characteristic curve. The application also relates to a dielectric response test system and a storage medium applying the dielectric response test method. The invention improves the dielectric response test efficiency of the insulating medium and is beneficial to saving resources.
Description
Technical Field
The present disclosure relates to the field of insulating material testing technologies, and in particular, to a dielectric response testing method, a dielectric response testing system, and a storage medium.
Background
In diagnosis and evaluation of insulation conditions of insulating materials such as insulating paper, insulating oil, etc., dielectric response experiments have been increasingly employed in recent years as a nondestructive diagnosis method of insulation conditions. Dielectric characteristic parameters of the insulating material in the frequency domain can be obtained through dielectric response experiments.
Although the current dielectric response experimental conditions are mature, in order to know the dielectric characteristic parameters of a specific type of insulating material under specific conditions (wherein the conditions include temperature conditions, aging degree conditions and the like), a dielectric response experiment is still required to be specially performed on the specific type of insulating material under the specific conditions, and the dielectric characteristic parameters of the insulating material can be obtained according to the experimental result.
Disclosure of Invention
Accordingly, it is necessary to provide a dielectric response testing method, system and storage medium for solving the above-mentioned technical problems of low dielectric response testing efficiency and resource waste.
The invention provides a dielectric response test method, which comprises the following steps:
Acquiring a dielectric response prediction model; the dielectric response prediction model is a model obtained by training a preprocessing curve through a preset neural network model, wherein the preprocessing curve is a frequency domain dielectric spectrum curve of a preprocessing insulating medium;
acquiring a characteristic curve of the insulating medium to be tested according to the dielectric response prediction model and the characteristic parameters of the insulating medium to be tested; the characteristic curve is a frequency domain dielectric spectrum curve of the insulating medium to be tested, and the characteristic parameter comprises at least one of the material type, the temperature and the aging degree of the insulating medium to be tested;
and acquiring dielectric response information of the insulating medium to be tested according to the characteristic curve.
In one embodiment, prior to the step of obtaining the dielectric response prediction model, the method comprises:
testing the pretreatment insulating medium according to a preset test mode to obtain the pretreatment curve; the test mode comprises at least one of a PDC test method and an FDS test method;
training the preprocessing curve by using the neural network model to generate the dielectric response prediction model.
In one embodiment, the pretreatment curve includes at least one of a first curve and a second curve;
Testing the pretreatment insulating medium according to a preset test mode to obtain the pretreatment curve, wherein the step of testing the pretreatment insulating medium comprises the following steps:
carrying out current test on the pretreated insulating medium by using the PDC test method, and acquiring the first curve according to a current test result; and/or the number of the groups of groups,
and carrying out frequency domain dielectric test on the pretreated insulating medium by using the FDS test method, and obtaining the second curve according to a frequency domain dielectric test result.
In one embodiment, the pretreatment curve includes a first curve, a second curve, and a third curve;
in the step of obtaining the first curve according to the current test result, the first curve is located in a first sub-frequency domain; the frequency value corresponding to the first sub-frequency domain is smaller than a preset frequency;
in the step of obtaining the second curve according to the frequency domain dielectric test result, the second curve is located in a second sub-frequency domain; the frequency value corresponding to the second sub-frequency domain is larger than a preset frequency;
testing the pretreatment insulating medium according to a preset test mode, and obtaining the pretreatment curve, wherein the method further comprises the following steps:
performing curve splicing processing on the first curve and the second curve at the preset frequency to obtain the third curve;
Training the preprocessing curve by using the neural network model, wherein the training comprises the following steps:
training the third curve by using the neural network model.
In one embodiment, the step of performing a current test on the pre-treated insulating medium by using the PDC test method and obtaining the first curve according to a current test result includes:
carrying out current test on the pretreated insulating medium according to the PDC test method to obtain a current curve; wherein the current test comprises at least one of a polarized current test and a depolarized current test, the current profile comprising at least one of a polarized current profile and a depolarized current profile of the preconditioned insulating medium;
and acquiring the first curve according to the current curve.
In one embodiment, the step of obtaining the first curve according to the current curve includes:
acquiring the conductivity of the pretreatment insulating medium according to the polarization current curve and the depolarization current curve;
acquiring the repolarization rate of the pretreatment insulating medium according to the depolarization current curve and the conductivity;
calculating the complex polarization rate according to a preset relation function to obtain a first complex dielectric constant of the pretreatment insulating medium; wherein the relationship function is used to characterize a functional relationship between the complex polarizability of the pre-treated insulating medium and a first complex permittivity that includes at least one of a real part and an imaginary part of the complex permittivity of the pre-treated insulating medium;
Generating a first complex permittivity spectrum curve of the pretreatment insulating medium according to the first complex permittivity;
and acquiring the first curve according to the first complex dielectric constant spectrum curve.
In one embodiment, the step of performing a frequency domain dielectric test on the pre-treated insulating medium by using the FDS test method and obtaining the second curve according to a frequency domain dielectric test result includes:
calculating according to the conductivity, the characteristic angular frequency and the vacuum capacitance of the pretreatment insulating medium to obtain the real part of the second complex dielectric constant of the pretreatment insulating medium; the characteristic angular frequency is an angular frequency value of the pretreatment insulating medium under the current voltage, and the vacuum capacitor is a capacitance value of the pretreatment insulating medium under the vacuum environment;
acquiring an imaginary part of a second complex dielectric constant of the pre-treatment insulating medium according to the absolute dielectric constant, the relative dielectric constant and the thickness of the pre-treatment insulating medium and the area of the electrode plate; the area of the electrode plate is an area value of the electrode plate used in the frequency domain dielectric test of the pretreatment insulating medium;
calculating the real part of the second complex dielectric constant and the imaginary part of the second complex dielectric constant to obtain the second complex dielectric constant of the pretreatment insulating medium;
Determining a second complex permittivity spectrum curve of the pre-treated insulating medium by the second complex permittivity;
and obtaining the second curve according to the second complex dielectric constant spectrum curve.
In one embodiment, after the step of generating the dielectric response prediction model, the method further includes:
acquiring a predicted frequency domain dielectric spectrum curve of the test insulating medium according to the dielectric response prediction model and the test parameters of the test insulating medium; wherein the characteristic parameter comprises at least one of a material type, a temperature and an aging degree of the test insulating medium;
testing the test insulating medium according to a preset test mode to obtain an actual frequency domain dielectric spectrum curve of the test insulating medium;
matching the fitting degree of the predicted frequency domain dielectric spectrum curve with the actual frequency domain dielectric spectrum curve, and obtaining the prediction accuracy of the dielectric response prediction model according to the comparison result;
when the prediction accuracy is greater than a first preset threshold, judging that the dielectric response prediction model is trained;
a step of obtaining a dielectric response prediction model, comprising:
and obtaining a trained dielectric response prediction model.
A dielectric response test system, the system comprising a detection module and a test computation module, wherein:
the detection module is used for detecting characteristic parameters of the insulating medium to be detected; wherein the characteristic parameters comprise at least one of the material type, the temperature and the aging degree of the insulating medium to be tested;
the test calculation module is used for obtaining a dielectric response prediction model, wherein the dielectric response prediction model is a model obtained by training a preprocessing curve through a preset neural network model, and the preprocessing curve is a frequency domain dielectric spectrum curve of a preprocessing insulating medium; acquiring a characteristic curve of the insulating medium to be tested according to the dielectric response prediction model and characteristic parameters of the insulating medium to be tested, wherein the characteristic curve is a frequency domain dielectric spectrum curve of the insulating medium to be tested; and acquiring dielectric response information of the insulating medium to be tested according to the characteristic curve.
A dielectric response test system comprising a memory storing a computer program and a processor implementing the steps of the dielectric response test method described above when the processor executes the computer program.
A computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of the dielectric response test method described above.
According to the dielectric response test method, the dielectric response test system and the storage medium, the characteristic parameters of the insulating medium to be tested are used as the input parameters of the dielectric response prediction model, the dielectric response prediction model is calculated according to the characteristic parameters to obtain the characteristic curve of the insulating medium to be tested, then the dielectric response information of the insulating medium to be tested can be directly obtained according to the characteristic curve, the dielectric characteristics of the insulating medium to be tested can be known according to the dielectric response information, the dielectric response test is omitted, the dielectric response test efficiency is improved, the dielectric characteristics of the insulating medium of various material types under various conditions (including temperature conditions and aging degree conditions) are facilitated to be quickly known, multiple experiments on the insulating medium to be tested with the same material types and the same conditions are effectively avoided, and manpower resources and materials required by the experiments are saved.
Drawings
In order to more clearly illustrate the technical solutions of embodiments or conventional techniques of the present application, the drawings required for the descriptions of the embodiments or conventional techniques will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present application, and other drawings may be obtained according to these drawings without inventive effort for a person of ordinary skill in the art. Wherein:
FIG. 1 is a flow chart of a method of testing dielectric response in one embodiment;
FIG. 2 is a flow chart of one embodiment of step 104 of FIG. 1;
FIG. 3 is a flow chart illustrating an embodiment of step 204 in FIG. 2;
FIG. 4 is a flow chart of a method for obtaining a dielectric response prediction model based on the PDC test method and the FDS test method in one embodiment;
FIG. 5 is a flow chart of a first curve obtained based on the PDC test method in one embodiment;
FIG. 6 is a flow chart illustrating one embodiment of step 504 of FIG. 5;
FIG. 7 is a flow chart of a second curve obtained based on the FDS test method in one embodiment;
FIG. 8 is a flowchart illustrating steps for verifying the prediction accuracy of a dielectric response prediction model in one embodiment;
FIG. 9 is a schematic block diagram of a medium response test system according to one embodiment.
DETAILED DESCRIPTION OF EMBODIMENT (S) OF INVENTION
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be further described in detail with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present application.
In one embodiment, as shown in FIG. 1, the present invention provides a dielectric response test method that is applied to a dielectric response test system as shown in FIG. 9.
It should be noted that the dielectric response test method is used to test dielectric characteristics of an insulating medium, for example, in some embodiments, the insulating medium includes, but is not limited to, insulating oil and insulating paper applied to a transformer, wherein the insulating oil includes, but is not limited to, at least one type of mineral oil and silicone oil, and the insulating paper includes, but is not limited to, at least one type of tufQUIN composite insulating paper and Nomex insulating paper.
The insulating medium is insulating oil paper applied to the transformer.
The method comprises the following steps:
step 102, a dielectric response prediction model is obtained.
In step 102, the dielectric response prediction model is a model obtained by training a preprocessing curve through a preset neural network model, the preprocessing curve is a frequency domain dielectric spectrum curve of a preprocessing insulating medium, and the preprocessing curve is used as a training sample of the neural network model; it should be noted that, the execution subject of the process of training the preprocessing curve by the neural network model is not limited, for example, in order to improve the integration of the dielectric response test system of the present invention, in some embodiments, the neural network model is instructed by the dielectric response test system of the present invention to train the preprocessing curve to obtain a dielectric response prediction model, and the dielectric response prediction model is stored; of course, the neural network model may also be instructed by an external device connected to the dielectric response test system of the present invention to train the preprocessing curve to obtain a dielectric response prediction model, and the external device transmits the dielectric response prediction model to the dielectric response test system of the present invention.
And 104, acquiring a characteristic curve of the insulating medium to be tested according to the dielectric response prediction model and the characteristic parameters of the insulating medium to be tested.
In step 104, the characteristic curve is a frequency domain dielectric spectrum curve of the insulating medium to be tested, the characteristic parameter includes at least one of a material type, a temperature and an aging degree of the insulating medium to be tested, the characteristic parameter may be specifically selected according to practical requirements, and in some embodiments, in order to improve accuracy of a prediction result of the dielectric response prediction model, the characteristic parameter includes the material type, the temperature and the aging degree of the insulating medium to be tested; in essence, the characteristic parameters are used as input parameters of a dielectric response prediction model, the dielectric response prediction model is calculated according to the characteristic parameters, and finally, a characteristic curve corresponding to the insulating medium to be tested is output.
Specifically, three parameters of the material type, the temperature and the aging degree of the insulating medium to be tested are input into a dielectric response prediction model, the dielectric response prediction model predicts according to the parameters, and finally a frequency domain dielectric spectrum prediction curve of the insulating medium to be tested is output as a characteristic curve. For example, in one embodiment, assume that parameters are input into a dielectric response prediction model: the dielectric response prediction model outputs a frequency domain dielectric spectrum curve of the tufQUIN composite insulating paper with the aging degree of 3 grades at the experimental temperature of 13 ℃ according to the parameters of the tufQUIN composite insulating paper with the 13 ℃ and the 3 grades.
And 106, acquiring dielectric response information of the insulating medium to be tested according to the characteristic curve.
The dielectric response information is used for representing the dielectric characteristics of the insulating medium to be tested.
According to the dielectric response test method, the characteristic parameters of the insulating medium to be tested are used as the input parameters of the dielectric response prediction model, the dielectric response prediction model is calculated according to the characteristic parameters to obtain the characteristic curve of the insulating medium to be tested, then the dielectric response information of the insulating medium to be tested can be directly obtained according to the characteristic curve, and the dielectric characteristics of the insulating medium to be tested can be known according to the dielectric response information. For example, a frequency domain dielectric spectrum curve of a certain insulating medium with the aging degree of 100 degrees at 200 ℃ needs to be obtained, dielectric response characteristics are observed through the frequency domain dielectric spectrum curve, the dielectric response curve is not easy to achieve and dangerous under the condition that the dielectric response curve is obtained through experiments, at the moment, the dielectric response prediction model provided by the application can be used for predicting the dielectric characteristics of the insulating medium to be detected, the accuracy of a prediction result can be ensured, the time of an experimenter can be greatly saved, and the efficiency of obtaining the frequency domain dielectric spectrum curve is improved.
It should be noted that, before step 102, the preprocessing curve needs to be trained by using the neural network model to obtain the dielectric response prediction model, while before training using the neural network model, the insulating medium needs to be preprocessed to obtain the preprocessed insulating medium, then the frequency domain dielectric spectrum curve corresponding to the preprocessed insulating medium is obtained as the preprocessing curve, and the obtained preprocessing curve is input into the neural network model as the training sample, so in some embodiments, before step 102, the method includes:
and 202, performing heat aging treatment on the insulating mediums of various material types for different durations to obtain the insulating mediums with different aging degrees.
In step 202, the accelerated heat aging process of the insulating medium is substantially performed, specifically, the insulating medium of various material types is first subjected to pre-drying treatment in a vacuum environment, if the insulating medium is insulating paper, the insulating paper of different types after the pre-drying treatment is soaked in insulating oil, and the accelerated heat aging process is performed on the insulating paper of different material types at a preset temperature.
In the accelerated heat aging process, insulating paper is sampled based on a preset time interval, and an insulating paper sample is obtained. If the insulating medium is insulating oil, performing accelerated thermal aging treatment on different types of insulating oil after the pre-drying treatment at a preset temperature, and sampling the insulating oil based on a preset time interval in the accelerated thermal aging treatment process to obtain an insulating oil sample. The insulating paper sample and the insulating oil sample are collectively referred to as a pre-treated insulating medium hereinafter.
Further, according to the formulaDetermining the total acid value X of the insulating oil in the pretreated insulating medium;
wherein V is 1 The volume of the potassium hydroxide ethanol solution with standard concentration consumed for titrating the insulating oil; v (V) 0 To empty the titrationWhite consumed standard concentration potassium hydroxide ethanol solution volume; c is the concentration of the standard solution; m is the mass of the insulating oil to be tested. The extent of aging of the pretreated insulating medium is characterized by the total acid number.
In one embodiment, the different types of new insulating papers are pre-dried in vacuum at 105 ℃ for 48 hours, after 48 hours, all insulating papers are soaked in insulating oil, and accelerated heat aging treatment is performed in an environment at 140 ℃. And taking out a batch of insulating paper samples of different types every 250 hours to obtain insulating paper samples with different ageing times. The total acid value of the residual insulating oil in the insulating paper samples with different aging times is calculated to characterize the aging degree of the insulating paper samples.
In one embodiment, different types of new insulating oils are dried, degassed. The different types of insulating oil were then subjected to an accelerated heat aging treatment in an environment of 140 degrees celsius. And taking out a batch of insulating oil samples of different types every 250 hours to obtain insulating oil samples with different ageing degrees. The total acid numbers of the insulating oil samples at different aging times were calculated to characterize the aging degree of the insulating oil samples.
In another embodiment, the aging degree of the oiled paper insulating paper sample can be classified according to the total acid value of the insulating oil in the oiled paper insulating sample, for example, the total acid value is 1-grade aging degree between 0 and 0.1, and the aging degree is 2-grade aging degree between 0.1 and 0.2, so that the aging degree of the oiled paper insulating sample can be represented according to the classification.
Of course, the step 202 is determined according to the test requirement, and in other embodiments, the step 202 may not be provided in any way, which does not require the accelerated heat aging treatment of the insulating medium.
And 204, testing the pretreatment insulating medium according to a preset test mode to obtain a pretreatment curve.
In step 204, the test mode includes at least one of PDC (full scale polarizat ion and depolar izat ion current, that is, polarization depolarization current) test method and FDS (full scale frequency domain dielectric spectroscopy, that is, frequency domain dielectric spectrum) test method, and the test mode may be specifically selected according to the needs of the actual test; it should be noted that, in some embodiments, a test time domain with a certain duration may be preset according to a test requirement, and the pre-processing insulating medium is tested according to the test time domain by using a preset test mode, so as to obtain a pre-processing curve corresponding to the test time domain.
In addition, in some embodiments, before executing step 204, a preset frequency may be selectively set according to the test requirement, while when executing step 204, a portion of the frequency domain dielectric spectrum curve of the pre-processing insulating medium located in the preset frequency domain may be selected as a pre-processing curve, and a specific range of the preset frequency domain may be specifically set according to the actual test requirement; of course, the preset frequency domain may be divided into a plurality of sub-frequency domains, and the preprocessing curve corresponding to each sub-frequency domain may be obtained from the frequency domain dielectric spectrum curve of the preprocessing insulating medium according to the range of each sub-frequency domain.
It should be noted that step 204 is substantially a step of collecting training samples of the neural network model, specifically, the pre-treatment insulating medium of various material types may be tested according to a preset test mode under different aging degrees or different temperatures, so as to obtain pre-treatment curves corresponding to the pre-treatment insulating medium of various material types under different aging degrees and different temperatures.
And 206, training the preprocessing curve by using the neural network model to generate a dielectric response prediction model.
Wherein, in step 206, the preprocessing curve is used as a training sample of the neural network model; specifically, a neural network model is first preset, and the pretreatment curves corresponding to the pretreatment insulating mediums of various material types in the step 204 under the conditions of different aging degrees and different temperatures are input into the dielectric response prediction model for training, so as to obtain the dielectric response prediction model, that is, the characteristics of the frequency domain dielectric spectrum curves (i.e., the pretreatment curves) of the pretreatment insulating mediums of different material types, different aging degrees and different temperatures are analyzed and learned by using the neural network model, so that the dielectric response prediction model can predict the frequency domain dielectric spectrum curves (i.e., the characteristic curves) of the insulating mediums to be tested according to the three parameters of the material types, the temperatures, the aging degrees and the like of the insulating mediums.
It should be noted that, because of the number of training samples of the neural network model, the prediction accuracy of the dielectric response prediction model on the characteristic curve of the insulating medium to be detected is also directly affected, and through the cooperation setting of the steps 202-206, a large number of preprocessing curves can be obtained as training samples to generate a dielectric response prediction model with high prediction accuracy, and the accuracy of predicting the characteristic curve of the insulating medium to be detected by using the dielectric response prediction model is high.
As shown in fig. 3, in one embodiment, the pretreatment curve includes at least one of a first curve and a second curve; testing the pretreatment insulating medium according to a preset test mode to obtain a pretreatment curve, wherein the step comprises at least one of the following steps:
step 302, performing current test on the pretreated insulating medium by using a PDC test method, and acquiring a first curve according to a current test result.
In step 302, the first curve may be obtained in a preset frequency domain, or may be obtained in any sub-frequency domain.
And step 304, performing frequency domain dielectric test on the pretreated insulating medium by using an FDS test method, and acquiring a second curve according to the frequency domain dielectric test result.
In step 304, the second curve may be obtained in a preset frequency domain, or may be obtained in any sub-frequency domain.
It should be noted that the setting of the steps 302 and 304 is not limited, and may be set according to the actual requirement, for example, in some embodiments, only one of the steps 302 and 304 may be set, where the neural network model uses the first curve as the training sample when only the step 302 is set, and uses the second curve as the training sample when only the step 304 is set. Of course, in some embodiments, step 302 and step 304 are set simultaneously, that is, the PDC test method is adopted to obtain the first curve, the FDS test method is also adopted to obtain the second curve, the neural network model uses the first curve and the second curve as training samples at the same time, or the first curve and the second curve are processed to obtain the processed curve, and the neural network model uses the processed curve as the training samples.
Further, the pretreatment insulating medium is tested according to a preset test mode, and in the step of obtaining a pretreatment curve, the pretreatment curve comprises a first curve, a second curve and a third curve, wherein the third curve is a curve obtained after the first curve and the second curve are processed, and the method specifically comprises the following steps of:
step 402, performing a current test on the pre-treated insulating medium by using a PDC test method, and obtaining a first curve in a first sub-frequency domain according to a current test result.
And step 404, performing frequency domain dielectric test on the pretreated insulating medium by using an FDS test method, and acquiring a second curve in a second sub-frequency domain according to the frequency domain dielectric test result.
And 406, performing curve splicing processing on the first curve and the second curve at a preset frequency to obtain a third curve.
The first curve is located in the first sub-frequency domain, and the second curve is located in the second sub-frequency domain; the preset frequency is any frequency value in a preset frequency domain, the preset frequency can be specifically set according to the actual test requirement, the preset frequency domain is divided into a first word frequency domain and a second sub-frequency domain according to the position corresponding to the preset frequency, the frequency value corresponding to the first sub-frequency domain is smaller than the preset frequency, and the frequency value corresponding to the second sub-frequency domain is larger than the preset frequency.
For example, in one embodiment, assuming that the preset frequency is 0.1Hz, in a frequency domain range with a frequency less than 0.1Hz, calculating a first curve of each pre-treated insulating medium from data obtained by a current test (i.e., a test performed by a PDC test method), calculating a second curve of each pre-treated insulating medium from data obtained by a frequency domain dielectric test (i.e., a test performed by a FDS test method), first unifying coordinate systems of the first curve and the second curve, and then connecting a point closest to 0.1Hz on the first curve with a point closest to 0.1Hz on the second curve, thereby obtaining a frequency domain dielectric spectrum curve of each insulating medium.
Training the preprocessing curve by using a neural network model, wherein the training step comprises the following steps:
step 408, training the third curve by using the neural network model to generate a dielectric response prediction model.
In the above steps 402-408, it should be noted that, since the time required for testing in the low frequency band by using the PDC test method is far lower than the time required for testing in the low frequency band by using the FDS test method, the PDC test method and the FDS test method are mutually matched, that is, the frequency domain dielectric spectrum curve (i.e., the first curve) of the pre-treated insulating medium is calculated in the low frequency band by using the PDC test method, the frequency domain dielectric spectrum curve (i.e., the second curve) of the pre-treated insulating medium is calculated in other frequency bands by using the FDS test method, and finally, the first curve and the second curve are spliced to obtain the third curve covering the whole preset frequency domain, so that the problem that the time for obtaining the third curve of the pre-treated insulating medium in the whole preset frequency domain by using only one test method is long is avoided, the time for obtaining the frequency domain dielectric spectrum curve of the pre-treated insulating medium can be greatly saved, and the efficiency for generating the dielectric response prediction model is beneficial to be improved.
As shown in fig. 5, in one embodiment, the step of performing a current test on the pre-treated insulating medium by using a PDC test method and obtaining a first curve according to a result of the current test includes:
step 502, performing current test on the pretreated insulating medium according to a PDC test method to obtain a current curve.
Wherein, in step 502, the current test comprises at least one of a polarization current test and a depolarization current test, the current profile comprising at least one of a polarization current profile and a depolarization current profile of the preconditioning dielectric. Specifically, in some embodiments, polarization current tests and/or depolarization current tests are performed on pre-treated insulating mediums of different material types and different aging degrees at different temperatures in a test time domain, a polarization current curve of the insulating medium in the test time domain if the insulating medium is subjected to intervention is obtained according to the polarization current test result, and/or a depolarization current curve of the insulating medium in the test time domain if the insulating medium is subjected to intervention is obtained according to the depolarization current test result.
Step 504, obtaining a first curve according to the current curve.
Wherein in step 502, a first curve is obtained according to the polarization current curve and/or the depolarization current curve, and in some embodiments, a frequency domain dielectric spectrum curve of the pre-treated insulating medium is obtained as the first curve in the first sub-frequency domain according to the polarization current curve and the depolarization current curve.
Further, as shown in fig. 6, in one embodiment, the step of obtaining a first curve according to the current curve includes:
step 602, obtaining the conductivity of the pretreated insulating medium according to the polarization current curve and the depolarization current curve.
Wherein, in step 602, as a possible implementation, according to the formula:
the conductivity G of the pre-treated insulating medium can be calculated.
Wherein ε 0 Is the dielectric constant of the insulating medium; c (C) 0 A geometric capacitance that is an insulating medium; u (U) C Is the polarization voltage used in the current test (i.e., when testing with PDC); t is t C Polarization time for current testing; t is t 1 The time corresponding to half peak value of the polarized current curve; i.e 1 Is a depolarizing current; i.e 2 Is a polarized current.
Step 604, obtaining the repolarization rate of the pretreated insulating medium according to the depolarization current curve and the conductivity.
Wherein, in step 604, the complex polarization rate includes at least one of a real part of the complex polarization rate and a real part of the complex polarization rate.
Specifically, the polarization current i is obtained from the polarization current curve 1 And based on polarization current i 1 Relationship with complex polarization χ (ω), the real part of the complex polarization is obtained asThe imaginary part of the complex polarization is +. >Where ω is the angular frequency of the voltage used in the current test.
Step 606, calculating the complex polarization rate according to the preset relation function, and obtaining the first complex dielectric constant of the pre-treated insulating medium.
Wherein in step 606, the relationship function is used to characterize a functional relationship between the complex polarizability of the pre-processed dielectric medium and a first complex permittivity that includes at least one of a real part and an imaginary part of the complex permittivity of the pre-processed dielectric medium. The method specifically comprises the following steps:
step one, calculating and obtaining the real part of the first complex dielectric constant according to the real part of the complex polarization rate.
Specifically, the real part of the first complex dielectric constant of the pretreatment insulating medium is obtained as epsilon' =epsilon according to the real part of the complex polarization rate ∞ +χ' (ω), where ε ∞ The dielectric constant in vacuum for several pre-treated insulating mediums was 8.85×10 -12 F/m。
And step two, calculating and obtaining the imaginary part of the first complex dielectric constant according to the imaginary part of the complex polarization rate.
Specifically, according to the imaginary part of the complex polarization rate, the imaginary part of the first complex dielectric constant of the pretreatment insulating medium is obtained as
And thirdly, calculating to obtain the first complex dielectric constant according to the real part and the imaginary part of the first complex dielectric constant.
Specifically, the first complex dielectric constant epsilon of the pre-treated dielectric medium is obtained according to the real part epsilon 'and the imaginary part epsilon' of the first complex dielectric constant 1 =ε' +jε). In the first sub-frequency domain, a first complex dielectric constant epsilon corresponding to each frequency is calculated 1 A first complex permittivity spectrum curve is generated.
Step 608 generates a first complex permittivity spectrum curve of the preconditioning dielectric medium based on the first complex permittivity.
Step 610, obtaining a first curve according to the first complex permittivity spectrum curve.
The step 610 specifically includes the following steps:
step one, obtaining a first dielectric loss tangent value of the pretreated insulating medium according to a first complex dielectric constant.
Specifically, according to the real part epsilon' and the imaginary part epsilon″ of the first complex dielectric constant, the first dielectric loss tangent of the pretreated insulating medium is obtained as follows:and in the first sub-frequency domain, calculating a first dielectric loss tangent corresponding to each frequency, and generating a first dielectric loss tangent spectrum curve.
And step two, acquiring a first dielectric loss tangent spectrum curve of the pretreated insulating medium according to the first dielectric loss tangent.
And thirdly, generating a first curve according to the first complex dielectric constant spectrum curve and the first dielectric loss tangent spectrum curve.
As shown in fig. 7, in the step of performing a frequency domain dielectric test on the pre-treated insulating medium by using the FDS test method and obtaining the second curve according to the frequency domain dielectric test result, specifically, performing frequency domain dielectric spectrum tests on the plurality of pre-treated insulating mediums by using the FDS test method, so as to obtain the second frequency domain dielectric spectrum curve of the pre-treated insulating medium as the second curve in the second sub-frequency domain.
More specifically, the steps include:
step 702, calculating according to the conductivity, the characteristic angular frequency and the vacuum capacitance of the pre-treated insulating medium, and obtaining the real part of the second complex dielectric constant of the pre-treated insulating medium.
Wherein, in step 702, a calculation is performed according to a formulaCalculating the real part epsilon' of the second complex dielectric constant, wherein G is the conductivity of the pre-treatment insulating medium, omega is the characteristic angular frequency (i.e. the angular frequency value of the pre-treatment insulating medium at the current voltage), and C 0 Is vacuum capacitance (i.e. the capacitance value of the pre-treatment insulating medium in a vacuum environment).
Step 704, obtaining an imaginary part of a second complex dielectric constant of the pre-processed insulating medium according to the absolute dielectric constant, the relative dielectric constant and the thickness of the pre-processed insulating medium and the electrode plate area.
Wherein, in step 704, the formula is followedCalculating the imaginary part epsilon, epsilon of the second complex dielectric constant r For the relative permittivity of the pre-treated insulating medium, S is the electrode plate area (i.e., the area value of the electrode plate used in performing the frequency domain dielectric test on the pre-treated insulating medium), and d is the thickness of the pre-treated insulating medium.
Step 706, calculating the real part of the second complex permittivity and the imaginary part of the second complex permittivity to obtain the second complex permittivity of the pre-processed insulating medium.
Wherein in step 706, the real part ε', the imaginary part ε ", and the equation ε of the second complex permittivity are calculated * Calculation of second complex permittivity ε * 。
Step 708, determining a second complex permittivity spectrum curve of the pre-treated insulating medium by the second complex permittivity.
In step 708, in the second sub-frequency domain, a second complex permittivity corresponding to each frequency value is calculated, and a second complex permittivity spectrum curve is generated.
Step 710, obtaining a second curve according to the second complex permittivity spectrum curve.
In step 710, the method specifically includes the following steps:
step one, obtaining a characteristic phase difference.
The characteristic phase difference is a phase difference value between a voltage phasor and a current phasor used when the frequency domain dielectric test is carried out on the pretreated insulating medium.
And step two, obtaining a second dielectric loss tangent value of the pretreated insulating medium under different frequency conditions according to the characteristic phase difference.
Wherein, according to the formulaCalculating a second dielectric loss tangent tan delta of the pre-treated dielectric medium, wherein->For the characteristic phase difference (i.e., the phase difference between the voltage and current phasors used in performing the frequency domain dielectric test), δ is the second dielectric loss angle (i.e., the residual angle of the phase difference between the voltage and current phasors used in the frequency domain dielectric test).
And thirdly, acquiring a second dielectric loss tangent spectrum curve of the pretreated insulating medium according to the second dielectric loss tangent.
And in the second sub-frequency domain, calculating a second dielectric loss tangent corresponding to each frequency to generate a second dielectric loss tangent spectrum curve.
And step four, generating a second curve according to the second complex dielectric constant spectrum curve and the second dielectric loss tangent spectrum curve.
It should be noted that in the step 406, performing curve splicing processing on the first curve and the second curve at the preset frequency to obtain a third curve specifically includes:
the preset frequency is set as a splicing point, the first curve comprises a first complex dielectric constant spectrum curve and a first dielectric loss tangent spectrum curve, and the second curve comprises a second complex dielectric constant spectrum curve and a second dielectric loss tangent spectrum curve.
Firstly, correspondingly splicing a first complex dielectric constant spectrum curve and a second complex dielectric constant spectrum curve of a pre-treatment insulating medium to obtain a complex dielectric constant spectrum curve of the pre-treatment insulating medium in the whole preset frequency domain (namely the sum of the first frequency domain and the second frequency domain), then correspondingly splicing the first dielectric loss tangent spectrum curve and the second dielectric loss tangent spectrum curve of the pre-treatment insulating medium to obtain a dielectric loss tangent spectrum curve of the pre-treatment insulating medium in the whole preset frequency domain, and finally, generating a frequency domain dielectric spectrum curve of the pre-treatment insulating medium to be used as a third curve according to the complex dielectric constant spectrum curve and the dielectric loss tangent spectrum curve.
As shown in fig. 8, in one embodiment, after the step of generating the dielectric response prediction model, the method further includes:
step 802, obtaining a predicted frequency domain dielectric spectrum curve of the test insulating medium according to the dielectric response prediction model and the test parameters of the test insulating medium.
In step 802, a plurality of groups of insulating mediums are selected as test insulating mediums, and the characteristic parameters comprise at least one of material type, temperature and aging degree of the test insulating mediums; specifically, the test parameters are input into a dielectric response prediction model, and the dielectric response prediction model is operated according to the information such as the material type, the temperature, the aging degree and the like so as to predict and generate a predicted frequency domain dielectric spectrum curve matched with the test insulating medium.
Step 804, testing the test insulation medium according to a preset test mode, and obtaining an actual frequency domain dielectric spectrum curve of the test insulation medium.
In step 804, the test insulation mediums with different material types and different aging degrees are tested at different temperatures by using PDC test method and/or FDS test to obtain actual frequency domain dielectric spectrum curves of the test insulation mediums.
And step 806, matching the predicted frequency domain dielectric spectrum curve with the actual frequency domain dielectric spectrum curve, and obtaining the prediction accuracy of the dielectric response prediction model according to the comparison result.
In step 806, all the predicted frequency-domain dielectric spectrum curves and the actual frequency-domain dielectric spectrum curves corresponding to the predicted frequency-domain dielectric spectrum curves are compared respectively, when the fitting degree between any one of the predicted frequency-domain dielectric spectrum curves and the actual frequency-domain dielectric spectrum curve corresponding to the predicted frequency-domain dielectric spectrum curve exceeds a second preset threshold, the predicted frequency-domain dielectric spectrum curve is determined to be a correct predicted result, and finally the number of correct predicted results is obtained, and the prediction accuracy of the dielectric response prediction model is obtained according to the ratio between the number of correct predicted results and the total amount of the frequency-domain dielectric spectrum predicted curves. It should be noted that, the specific value of the second preset threshold may be set according to the actual requirement, for example, in some embodiments, the second preset threshold is 95%, and when the fitting degree between the predicted frequency domain dielectric spectrum curve and the actual frequency domain dielectric spectrum curve corresponding to the predicted frequency domain dielectric spectrum curve exceeds 95%, the predicted frequency domain dielectric spectrum curve is judged to be a correct predicted result.
And 808, when the prediction accuracy is greater than the first preset threshold, determining that the dielectric response prediction model is trained.
Wherein, in step 808, in the case that the prediction accuracy is greater than the first preset threshold, the dielectric response prediction model training is completed. If the prediction accuracy is smaller than or equal to the first preset threshold value, the number of training samples is increased, and training is continued on the dielectric response prediction model until the prediction accuracy of the dielectric response prediction model is larger than the first preset threshold value.
In one embodiment, if the number of frequency domain dielectric spectrum prediction curves is 100, where the number of correct prediction results is 90, the prediction accuracy of the dielectric response prediction model is 90%, and if the first preset threshold is 95%, the prediction accuracy of the dielectric response prediction model does not meet the requirement, and the model needs to be trained continuously.
Further, under the condition that a predictable insulating medium type needs to be newly added in the dielectric response prediction model, a plurality of frequency domain dielectric spectrum curves of the insulating medium type to be tested at different temperatures and different ageing degrees are obtained, and the dielectric response prediction model is trained again through the frequency domain dielectric spectrum curves, so that the dielectric response model learns the frequency domain dielectric spectrum curve characteristics of the insulating medium type to be tested. And finally, verifying the prediction accuracy of the dielectric response prediction model for predicting the type of the insulating medium to be detected.
A step of obtaining a dielectric response prediction model, comprising:
step 810, obtaining a trained dielectric response prediction model.
It should be understood that, although the steps in the flowcharts of fig. 1-8 are shown in order as indicated by the arrows, these steps are not necessarily performed in order as indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in FIGS. 1-8 may include multiple steps or stages that are not necessarily performed at the same time, but may be performed at different times, nor do the order in which the steps or stages are performed necessarily performed in sequence, but may be performed alternately or alternately with at least a portion of the steps or stages in other steps or other steps.
Referring to fig. 9, in one embodiment, the present invention provides a dielectric response testing system 900 to which the above dielectric response testing method can be applied, where the dielectric response testing system 900 includes a detection module 910 and a test calculation module 920, and the detection module 910 is connected to the test calculation module 920.
Wherein:
the detection module 910 is configured to detect a characteristic parameter of the insulating medium.
Wherein the characteristic parameter comprises at least one of a material type, a temperature and an aging degree of the insulating medium, and the insulating medium comprises at least one of the pretreated insulating medium, the test insulating medium and the insulating medium to be tested.
The test calculation module 920 is configured to obtain a dielectric response prediction model, where the dielectric response prediction model is a model obtained by training a preprocessing curve through a preset neural network model, and the preprocessing curve is a frequency domain dielectric spectrum curve of the preprocessing insulating medium; acquiring a characteristic curve of the insulating medium to be tested according to the dielectric response prediction model and the characteristic parameters of the insulating medium to be tested, wherein the characteristic curve is a frequency domain dielectric spectrum curve of the insulating medium to be tested; and acquiring dielectric response information of the insulating medium to be tested according to the characteristic curve.
Of course, the dielectric response test system 900 may select and set a storage module 930 according to the requirement of use, where the storage module 930 is connected to the detection module 910 and the test calculation module 920, and the storage module 930 is used to store a preset neural network model, and the storage module 930 sends the neural network model to the test calculation module 920; the storage module 930 is further configured to store the dielectric response prediction model generated by training the test calculation module 920, and when the test calculation module 920 needs to use the dielectric response prediction model, the storage module 930 sends the dielectric response prediction model to the test calculation module 920; of course, the storage module 930 may also store the characteristic parameters of the insulating medium.
In one embodiment, the test calculation module 920 is further configured to test the pre-treated insulating medium according to a preset test mode, to obtain a pre-treatment curve, where the test mode includes at least one of a PDC test method and an FDS test method; training the preprocessing curve by using a neural network model to generate a dielectric response prediction model.
In one embodiment, the pretreatment curve includes at least one of a first curve and a second curve;
the test calculation module 920 is further configured to perform a current test on the pretreated insulating medium by using a PDC test method, and obtain a first curve according to a current test result; and/or performing frequency domain dielectric test on the pretreated insulating medium by using an FDS test method, and acquiring a second curve according to the frequency domain dielectric test result.
In one embodiment, the pretreatment curve includes a first curve, a second curve, and a third curve; the test calculation module 920 is further configured to obtain a first curve according to a current test result, where the first curve is located in a first sub-frequency domain, and a frequency value corresponding to the first sub-frequency domain is smaller than a preset frequency; in the step of acquiring a second curve according to the frequency domain dielectric test result, the second curve is located in a second sub-frequency domain, wherein the frequency value corresponding to the second sub-frequency domain is larger than the preset frequency; performing curve splicing processing on the first curve and the second curve at a preset frequency to obtain a third curve; the third curve is trained using the neural network model.
In one embodiment, the test calculation module 920 is further configured to perform a current test on the pre-treated insulating medium according to a PDC test method, and obtain a current curve, where the current test includes at least one of a polarization current test and a depolarization current test, and the current curve includes at least one of a polarization current curve and a depolarization current curve of the pre-treated insulating medium; and acquiring a first curve according to the current curve.
In one embodiment, the test calculation module 920 is further configured to obtain the conductivity of the pre-treated insulating medium according to the polarization current curve and the depolarization current curve; acquiring the repolarization rate of the pretreated insulating medium according to the depolarization current curve and the conductivity; calculating the complex polarization rate according to a preset relation function to obtain a first complex dielectric constant of the pre-treatment insulating medium, wherein the relation function is used for representing a functional relation between the complex polarization rate of the pre-treatment insulating medium and the first complex dielectric constant, and the first complex dielectric constant comprises at least one of a real part and an imaginary part in the complex dielectric constant of the pre-treatment insulating medium; generating a first complex dielectric constant spectrum curve of the pretreatment insulating medium according to the first complex dielectric constant; and acquiring a first curve according to the first complex dielectric constant spectrum curve.
In one embodiment, the test calculation module 920 is further configured to calculate according to the conductivity, the characteristic angular frequency, and the vacuum capacitance of the pre-processed insulating medium, and obtain a real part of the second complex dielectric constant of the pre-processed insulating medium; the characteristic angular frequency is the angular frequency value of the pretreatment insulating medium under the current voltage, and the vacuum capacitor is the capacitance value of the pretreatment insulating medium under the vacuum environment; acquiring an imaginary part of a second complex dielectric constant of the pre-treatment insulating medium according to the absolute dielectric constant, the relative dielectric constant and the thickness of the pre-treatment insulating medium and the area of the electrode plate; the area of the electrode plate is the area value of the electrode plate used in the frequency domain dielectric test of the pretreatment insulating medium; calculating the real part of the second complex dielectric constant and the imaginary part of the second complex dielectric constant to obtain the second complex dielectric constant of the pretreatment insulating medium; determining a second complex permittivity spectrum curve of the pre-treated insulating medium by the second complex permittivity; and obtaining a second curve according to the second complex dielectric constant spectrum curve.
In one embodiment, the test calculation module 920 is further configured to obtain a predicted frequency domain dielectric spectrum curve of the test insulation medium according to the dielectric response prediction model and a test parameter of the test insulation medium, where the feature parameter includes at least one of a material type, a temperature, and an aging degree of the test insulation medium; testing the test insulating medium according to a preset test mode to obtain an actual frequency domain dielectric spectrum curve of the test insulating medium; matching the fitting degree of the predicted frequency domain dielectric spectrum curve with that of the actual frequency domain dielectric spectrum curve, and obtaining the prediction accuracy of the dielectric response prediction model according to the comparison result; when the prediction accuracy is greater than a first preset threshold, judging that the dielectric response prediction model is trained; and obtaining a trained dielectric response prediction model.
Those skilled in the art will appreciate that the structure shown in fig. 9 is merely a block diagram of a portion of the structure associated with the present application and is not limiting of the dielectric response test system to which the present application is applied, and that a particular dielectric response test system may include more or fewer components than shown, or may combine certain components, or have a different arrangement of components.
A dielectric response test system comprising a memory storing a computer program and a processor implementing the steps of the dielectric response test method described above when the processor executes the computer program.
A computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of the dielectric response test method described above.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, or the like. Volatile memory can include random access memory (Random AccEWs Memory, RAM) or external cache memory. By way of illustration, and not limitation, RAM can be in the form of a variety of forms, such as static random access memory (Static Random AccEWs Memory, SRAM) or dynamic random access memory (Dynamic Random AccEWs Memory, DRAM), and the like.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The above-described embodiments represent only a few embodiments of the present application, which are described in more detail and are not to be construed as limiting the scope of the invention. It should be noted that it would be apparent to those skilled in the art that various modifications and improvements could be made without departing from the spirit of the present application, which would be within the scope of the present application. Accordingly, the scope of protection of the present application is to be determined by the claims appended hereto.
Claims (11)
1. A method of dielectric response testing, the method comprising:
testing the pretreatment insulating medium according to a preset test mode to obtain a pretreatment curve; the test mode comprises a polarization depolarization current test method;
acquiring a dielectric response prediction model; the dielectric response prediction model is a model obtained by training the preprocessing curve through a preset neural network model, and the preprocessing curve is a frequency domain dielectric spectrum curve of the preprocessing insulating medium;
Acquiring a characteristic curve of the insulating medium to be tested according to the dielectric response prediction model and the characteristic parameters of the insulating medium to be tested; the characteristic curve is a frequency domain dielectric spectrum curve of the insulating medium to be tested, and the characteristic parameter comprises at least one of the material type, the temperature and the aging degree of the insulating medium to be tested;
and acquiring dielectric response information of the insulating medium to be tested according to the characteristic curve.
2. The method of claim 1, wherein the test mode further comprises a frequency domain dielectric spectroscopy test mode; before the step of obtaining the dielectric response prediction model, the method comprises the following steps:
training the preprocessing curve by using the neural network model to generate the dielectric response prediction model.
3. The method of claim 2, wherein the pre-processing curve comprises a first curve and a second curve;
testing the pretreatment insulating medium according to a preset test mode to obtain the pretreatment curve, wherein the step of testing the pretreatment insulating medium comprises the following steps:
carrying out current test on the pretreated insulating medium by using the polarization depolarization current test method, and obtaining the first curve according to a current test result;
And carrying out frequency domain dielectric test on the pretreated insulating medium by using the frequency domain dielectric spectrum test method, and obtaining the second curve according to the frequency domain dielectric test result.
4. A dielectric response test method according to claim 3, wherein the pre-treatment profile comprises a first profile, a second profile and a third profile;
in the step of obtaining the first curve according to the current test result, the first curve is located in a first sub-frequency domain; the frequency value corresponding to the first sub-frequency domain is smaller than a preset frequency;
in the step of obtaining the second curve according to the frequency domain dielectric test result, the second curve is located in a second sub-frequency domain; the frequency value corresponding to the second sub-frequency domain is larger than a preset frequency;
testing the pretreatment insulating medium according to a preset test mode, and obtaining the pretreatment curve, wherein the method further comprises the following steps:
performing curve splicing processing on the first curve and the second curve at the preset frequency to obtain the third curve;
training the preprocessing curve by using the neural network model, wherein the training comprises the following steps:
training the third curve by using the neural network model.
5. The method of claim 3 or 4, wherein the step of performing a current test on the pre-treated insulating medium using the polarization depolarization current test method and obtaining the first curve based on the current test result comprises:
carrying out current test on the pretreatment insulating medium according to the polarization depolarization current test method to obtain a current curve; wherein the current test comprises at least one of a polarized current test and a depolarized current test, the current profile comprising at least one of a polarized current profile and a depolarized current profile of the preconditioned insulating medium;
and acquiring the first curve according to the current curve.
6. The method of claim 5, wherein the step of obtaining the first curve from the current curve comprises:
acquiring the conductivity of the pretreatment insulating medium according to the polarization current curve and the depolarization current curve;
acquiring the repolarization rate of the pretreatment insulating medium according to the depolarization current curve and the conductivity;
calculating the complex polarization rate according to a preset relation function to obtain a first complex dielectric constant of the pretreatment insulating medium; wherein the relationship function is used to characterize a functional relationship between the complex polarizability of the pre-treated insulating medium and a first complex permittivity that includes at least one of a real part and an imaginary part of the complex permittivity of the pre-treated insulating medium;
Generating a first complex permittivity spectrum curve of the pretreatment insulating medium according to the first complex permittivity;
and acquiring the first curve according to the first complex dielectric constant spectrum curve.
7. The method of claim 3 or 4, wherein the step of performing a frequency-domain dielectric test on the pre-treated dielectric medium using the frequency-domain dielectric spectroscopy test method and obtaining the second curve based on the frequency-domain dielectric test result comprises:
calculating according to the conductivity, the characteristic angular frequency and the vacuum capacitance of the pretreatment insulating medium to obtain the real part of the second complex dielectric constant of the pretreatment insulating medium; the characteristic angular frequency is an angular frequency value of the pretreatment insulating medium under the current voltage, and the vacuum capacitor is a capacitance value of the pretreatment insulating medium under the vacuum environment;
acquiring an imaginary part of a second complex dielectric constant of the pre-treatment insulating medium according to the absolute dielectric constant, the relative dielectric constant and the thickness of the pre-treatment insulating medium and the area of the electrode plate; the area of the electrode plate is an area value of the electrode plate used in the frequency domain dielectric test of the pretreatment insulating medium;
Calculating the real part of the second complex dielectric constant and the imaginary part of the second complex dielectric constant to obtain the second complex dielectric constant of the pretreatment insulating medium;
determining a second complex permittivity spectrum curve of the pre-treated insulating medium by the second complex permittivity;
and obtaining the second curve according to the second complex dielectric constant spectrum curve.
8. The method of claim 2, further comprising, after the step of generating the dielectric response prediction model:
acquiring a predicted frequency domain dielectric spectrum curve of the test insulating medium according to the dielectric response prediction model and the test parameters of the test insulating medium; wherein the characteristic parameter comprises at least one of a material type, a temperature and an aging degree of the test insulating medium;
testing the test insulating medium according to a preset test mode to obtain an actual frequency domain dielectric spectrum curve of the test insulating medium;
matching the fitting degree of the predicted frequency domain dielectric spectrum curve with the actual frequency domain dielectric spectrum curve, and obtaining the prediction accuracy of the dielectric response prediction model according to the comparison result;
When the prediction accuracy is greater than a first preset threshold, judging that the dielectric response prediction model is trained;
a step of obtaining a dielectric response prediction model, comprising:
and obtaining a trained dielectric response prediction model.
9. A dielectric response test system, the system comprising a detection module and a test computation module, wherein:
the detection module is used for detecting characteristic parameters of the insulating medium to be detected; wherein the characteristic parameters comprise at least one of the material type, the temperature and the aging degree of the insulating medium to be tested;
the test calculation module is used for testing the pretreatment insulating medium according to a preset test mode to obtain a pretreatment curve; the test mode comprises a polarization depolarization current test method; obtaining a dielectric response prediction model, wherein the dielectric response prediction model is a model obtained by training the preprocessing curve through a preset neural network model, and the preprocessing curve is a frequency domain dielectric spectrum curve of the preprocessing insulating medium; acquiring a characteristic curve of the insulating medium to be tested according to the dielectric response prediction model and characteristic parameters of the insulating medium to be tested, wherein the characteristic curve is a frequency domain dielectric spectrum curve of the insulating medium to be tested; and acquiring dielectric response information of the insulating medium to be tested according to the characteristic curve.
10. A dielectric response test system comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the dielectric response test method of any one of claims 1 to 8.
11. A computer readable storage medium having stored thereon a computer program, characterized in that the computer program when executed by a processor implements the steps of the dielectric response test method of any of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110814843.5A CN113740674B (en) | 2021-07-19 | 2021-07-19 | Dielectric response test method, system and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110814843.5A CN113740674B (en) | 2021-07-19 | 2021-07-19 | Dielectric response test method, system and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113740674A CN113740674A (en) | 2021-12-03 |
| CN113740674B true CN113740674B (en) | 2024-01-02 |
Family
ID=78728850
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110814843.5A Active CN113740674B (en) | 2021-07-19 | 2021-07-19 | Dielectric response test method, system and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113740674B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115015684B (en) * | 2022-08-10 | 2022-11-29 | 山东和兑智能科技有限公司 | High-voltage bushing digital evaluation method and system based on time-frequency domain feature fusion |
| CN115480137A (en) * | 2022-10-14 | 2022-12-16 | 重庆大学 | Method and system for measuring insulation broadband high-voltage dielectric response of electrical equipment |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110045245A (en) * | 2019-04-27 | 2019-07-23 | 西南交通大学 | A kind of appraisal procedure of oil-immersed transformer casing X wax content |
| CN112505494A (en) * | 2020-10-30 | 2021-03-16 | 西安交通大学 | Method and device for evaluating insulation water content of oiled paper |
| CN112666231A (en) * | 2020-11-17 | 2021-04-16 | 国网上海市电力公司 | Method for testing water content of solid insulation of converter transformer |
-
2021
- 2021-07-19 CN CN202110814843.5A patent/CN113740674B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110045245A (en) * | 2019-04-27 | 2019-07-23 | 西南交通大学 | A kind of appraisal procedure of oil-immersed transformer casing X wax content |
| CN112505494A (en) * | 2020-10-30 | 2021-03-16 | 西安交通大学 | Method and device for evaluating insulation water content of oiled paper |
| CN112666231A (en) * | 2020-11-17 | 2021-04-16 | 国网上海市电力公司 | Method for testing water content of solid insulation of converter transformer |
Non-Patent Citations (1)
| Title |
|---|
| 基于Havriliak-Negami介电模型的油纸绝缘老化参数提取;云浩等;《电力工程技术》;20210331;第40卷(第2期);文章摘要、引言 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113740674A (en) | 2021-12-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113740674B (en) | Dielectric response test method, system and storage medium | |
| Bundy et al. | An electrochemical impedance spectroscopy method for prediction of the state of charge of a nickel-metal hydride battery at open circuit and during discharge | |
| CN115032540B (en) | Lithium ion battery health state estimation method, device, equipment and medium | |
| Nagpal et al. | Artificial neural network approaches for fault classification: comparison and performance | |
| CN109507554A (en) | A kind of insulation of electrical installation state evaluating method | |
| Zhu et al. | Predicting battery life with early cyclic data by machine learning | |
| CN113083739B (en) | Battery cell sorting method and device and computer equipment | |
| CN117171696B (en) | Sensor production monitoring method and system based on Internet of things | |
| CN116572747B (en) | Battery fault detection method, device, computer equipment and storage medium | |
| Heins et al. | Electrode‐Resolved Monitoring of the Ageing of Large‐Scale Lithium‐Ion Cells by using Electrochemical Impedance Spectroscopy | |
| CN117471227A (en) | Automobile wire harness parameter performance test method and test system | |
| CN113340353B (en) | Monitoring method, equipment and medium for power transmission line | |
| CN115856529A (en) | CVT insulation aging state evaluation method based on PDC multi-feature-quantity weight | |
| Yang et al. | A novel multiple feature-based engine knock detection system using sparse Bayesian extreme learning machine | |
| Fan et al. | Polarization loss analysis and ageing characterisation of transformer oil‐immersed insulation by using decoupled frequency domain spectroscopy | |
| Tagayi et al. | Employment of relaxation times distribution with improved elastic net regularization for advanced impedance data analysis of a lithium-ion battery | |
| CN113125965A (en) | Method, device and equipment for detecting lithium separation of battery and storage medium | |
| CN113670987B (en) | Method, device, equipment and storage medium for identifying oil paper insulation aging state | |
| CN115508678A (en) | Oil paper insulation aging state evaluation method and system based on interface charge decay time | |
| CN113466749A (en) | Method, device, equipment and storage medium for evaluating aging state of main insulation of transformer | |
| CN114970749A (en) | Method, device and readable storage medium for acquiring characteristic parameters of capacitor | |
| Sin et al. | Data-driven prediction of battery degradation using EIS-based robust features | |
| Feng et al. | Convolutional Neural Network‐Long Short‐Term Memory‐Based State of Health Estimation for Li‐Ion Batteries under Multiple Working Conditions | |
| dos Santos et al. | Study of machine learning algorithms to state of health estimation of iron phosphate lithium-ion battery used in fully electric vehicles | |
| Mahdjoubi et al. | A new approach using Least Squares Support Vector Machines (LS-SVM) to predict Furan in power transformers |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |