CN115542099B - Online GIS partial discharge detection method and device - Google Patents
Online GIS partial discharge detection method and device Download PDFInfo
- Publication number
- CN115542099B CN115542099B CN202211497961.9A CN202211497961A CN115542099B CN 115542099 B CN115542099 B CN 115542099B CN 202211497961 A CN202211497961 A CN 202211497961A CN 115542099 B CN115542099 B CN 115542099B
- Authority
- CN
- China
- Prior art keywords
- partial discharge
- signal
- fault
- ultrahigh frequency
- gis
- 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
- G01R31/12—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
- G01R31/1209—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 using acoustic measurements
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/10—Machine learning using kernel methods, e.g. support vector machines [SVM]
-
- 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
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Mathematical Physics (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Health & Medical Sciences (AREA)
- Acoustics & Sound (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Testing Relating To Insulation (AREA)
Abstract
The invention discloses an online GIS partial discharge detection method and device, belonging to the technical field of live detection, wherein an ultrahigh frequency acquisition signal is obtained through an ultrahigh frequency test module; obtaining an ultrasonic acquisition signal through an ultrasonic testing module; extracting characteristic signals of the collected ultrahigh frequency collected signals to obtain waveform characteristic signals and amplitude characteristic signals; the partial discharge analysis module sends the ultrasonic acquisition signals and the waveform characteristic signals obtained by analysis to a trained PSO-SVM-based combined prediction model, and outputs corresponding fault levels; when the fault level is in a serious state or an emergency state, the fault level, the ultrasonic acquisition signal, the amplitude characteristic signal and the waveform characteristic signal are sent into a classification model based on an SVM algorithm to obtain a corresponding partial discharge fault type, and a corresponding fault processing strategy is output according to the fault level and the partial discharge fault type, so that the accuracy of GIS detection is greatly improved.
Description
Technical Field
The invention relates to the technical field of GIS detection, in particular to an online GIS partial discharge detection method and device.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
Petrochemical plants belong to high-risk industries, and normal, safe and uninterrupted power supply is very important for the petrochemical plants. In order to ensure normal power supply and reliable power supply of a petrochemical plant, 110kVGIS electrical equipment is core equipment in all equipment in specific power supply equipment of a transformer substation, and only when the electrical equipment is operated safely and reliably, a complete and reliable power supply system can be provided to provide power support for production after a petrochemical plant is built. In order to realize online detection of the GIS, GIS partial discharge detection is a very common detection measure. Because GIS's coaxial arrangement, the electromagnetic wave not only can propagate in GIS inside, can see through nonmetal parts such as basin formula insulation moreover and leak outside GIS, use the electromagnetic wave that the UHF antenna can detect GIS partial discharge and produce. The method has the main advantages of high sensitivity and strong anti-interference capability, can position the discharge source according to the time difference of electromagnetic waves from the discharge source to different sensors, and is successfully applied to GIS production and operation detection at present. When the partial discharge signal is generated inside the GIS, impact vibration and sound can be generated, so that the partial discharge signal can be measured by an ultrasonic sensor arranged on the outer wall of the cavity. The ultrasonic method is a well-established partial discharge monitoring method except the UHF method.
The existing GIS partial discharge testing device mainly comprises an ultrasonic testing module and an ultrahigh frequency testing module, wherein the ultrasonic testing module is simple to use but low in testing efficiency and easy to miss judgment; the ultrahigh frequency testing module is high in sensitivity, but has the problems that field electromagnetic interference is complex, and operation can be performed by an experienced expert, and the like, and meanwhile, as the acquired signals obtained by the ultrahigh frequency testing module have obvious time sequence characteristics, an SVM algorithm which is used for processing time sequence characteristic data and has great advantages is lacked to combine the data of the ultrasonic testing module and the data of the ultrahigh frequency testing module to analyze partial discharge signals of the GIS, the efficiency and the precision of partial discharge detection are improved, and meanwhile, the characteristic signals obtained by ultrasonic waves are lacked to be comprehensively calculated to obtain waveform characteristic signals and amplitude characteristic signals, so that the input data volume is large, and the calculation efficiency is low.
Therefore, it is necessary to establish an online GIS partial discharge detection method and device in consideration of the above factors.
Disclosure of Invention
In order to solve the technical problems in the background art, the invention provides an online GIS partial discharge detection method, which greatly improves the accuracy of GIS detection and reduces the working intensity of detection personnel.
And measuring the surrounding environment of the running GIS through an ultrahigh frequency test module to obtain a background signal of the GIS and obtain an ultrahigh frequency acquisition signal.
And measuring the surrounding environment of the running GIS through an ultrasonic testing module, and acquiring a background signal of the GIS to obtain an ultrasonic acquisition signal.
And inputting the collected ultrahigh frequency collected signals and the collected ultrasonic collected signals into a partial discharge analysis module, and extracting characteristic signals of the collected ultrahigh frequency collected signals to obtain waveform characteristic signals and amplitude characteristic signals.
The waveform characteristic signal is obtained based on the rising time of the ultrahigh frequency acquisition signal, the falling edge time of the ultrahigh frequency acquisition signal and the main peak time of the ultrahigh frequency acquisition signal, and the calculation formula of the waveform characteristic signal is as follows:
whereint 1 、t 2 、t 3 Respectively the rising time of the ultrahigh frequency acquisition signal, the falling edge time of the ultrahigh frequency acquisition signal and the main peak time of the ultrahigh frequency acquisition signal,K 1 、K 2 、K 3 、K 4 is constant and has a value ranging from 0 to 1.
The amplitude characteristic signal is obtained by construction based on the mean value of the amplitude of the ultrahigh frequency acquisition signal, the standard deviation of the amplitude and the entropy value of the amplitude, and the calculation formula of the amplitude characteristic signal is as follows:
whereinK 5 、K 6 、K 7 、K 8 、K 9 Is constant and has a value ranging from 0 to 1,J 1 、J 2 、Srespectively is the mean value of the amplitude of the ultrahigh frequency acquisition signal, the standard deviation of the amplitude and the entropy value of the amplitude.
And the partial discharge analysis module forms a preliminary input set by the ultrasonic acquisition signals and the waveform characteristic signals obtained by analysis, sends the preliminary input set into a trained PSO-SVM-based combined prediction model, and outputs corresponding fault levels.
And when the fault level is in a serious state or an emergency state, taking the fault level, the ultrasonic acquisition signal, the amplitude characteristic signal and the waveform characteristic signal as an input set, sending the input set into a classification model based on an SVM algorithm to obtain a corresponding partial discharge fault type, and outputting a corresponding fault processing strategy according to the fault level and the partial discharge fault type.
Preferably, the PSO-SVM algorithm training specifically comprises the steps of:
s1, a waveform characteristic signal, an amplitude characteristic signal and an ultrasonic acquisition signal are used as input data and are divided into a training set and a testing set, and partial discharge fault or normal data with a fault level corresponding to the signals are used as output data.
S2, establishing an SVM model, determining parameters needing optimization, the number L1 of neurons, the learning rate epsilon and the training iteration times k, and determining respective optimization ranges.
And S3, initializing PSO parameters including initial speed and position of the particles, learning weight, training times and scale.
And S4, determining a fitness function of the particles, taking the improved error function of the prediction model as the fitness function of the particles, and searching for the optimal model parameters.
S5, comparing the fitness values of the particles to obtain an individual optimal position and a global optimal position, and updating the optimal fitness value.
S6, judging whether the maximum iteration times is reached, if the maximum iteration times is reached, transmitting the obtained optimal parameters to the SVM model, training and predicting, and if the maximum iteration times is not reached, returning to the step S5.
Preferably, the failure levels include: mild, severe, emergency; in a slight state, the fault is in a slight state and needs to be observed intensively; in a severe state, the fault is serious, but the normal operation of the GIS is not influenced, and the GIS needs to be retested regularly; in an emergency state, the fault is serious, and the shutdown needs to be immediately contacted.
Preferably, the specific formula of the improved error function is as follows:
wherein, the first and the second end of the pipe are connected with each other,
is the output value of the t-th neural network,
is the average of the outputs of all N neurons,
is the desired output value of the neuron,K 10 is a constant.
Preferably, the partial discharge fault types specifically include internal impurities, surface burrs, poor contacts, floating potential, and solid insulation surface contamination.
An online GIS partial discharge detection device specifically includes: the system comprises an ultrasonic testing module, an ultrahigh frequency testing module and a partial discharge analysis module;
the ultrasonic testing module is used for measuring the running GIS to obtain an ultrahigh frequency acquisition signal;
the ultrasonic testing module is used for measuring the running GIS to obtain an ultrasonic acquisition signal;
and the partial discharge analysis module is used for analyzing the ultrahigh frequency acquisition signals and the ultrasonic acquisition signals obtained by analysis and outputting corresponding partial discharge fault types with fault levels.
The invention has the advantages that: by adopting the ultrasonic testing module and the ultrahigh frequency testing module, the acquisition of GIS partial discharge signals can be realized, the analysis of the partial discharge signals is realized by arranging the partial discharge analysis module, and the identification of partial discharge types is quicker and more accurate by adopting a combined prediction model of PSO-SVM, so that the problems that the original ultrasonic testing module is low in testing efficiency and easy to miss judgment, the ultrahigh frequency testing module needs to be positioned by using a high-grade oscilloscope and an experienced expert to operate and the like are solved, and the working intensity of detection personnel is reduced.
The fault grade is obtained through the combined prediction model prediction based on the PSO-SVM, and then the fault type is judged according to the grade of the fault grade, so that the judgment accuracy is further improved, unnecessary calculation is avoided, and the efficiency is improved.
The partial discharge fault type with the fault level is output, so that the partial discharge type and the fault level are output to a detector, the experience requirement and the use requirement of the detector are further reduced, the operation of the module is simpler, and the judgment on the module is more accurate.
And outputting a corresponding fault processing strategy, so that a detector can process the module according to the suggestion and the regulation requirement, the module becomes more humanized, and the detection result becomes more instructive.
Drawings
FIG. 1 is a flowchart of example 1 of the present invention.
Fig. 2 is a flowchart of the PSO-SVM algorithm in embodiment 1 of the present invention.
Fig. 3 is a schematic block diagram of embodiment 2 of the present invention.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the application, its application, or uses. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application, but the present application may be practiced in other ways than those described herein, and it will be apparent to those of ordinary skill in the art that the present application is not limited to the specific embodiments disclosed below.
Example 1:
the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the application, its application, or uses. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application, but the present application may be practiced in other ways than those described herein, and it will be appreciated by those skilled in the art that the present application may be practiced without departing from the spirit and scope of the present application, and that the present application is not limited to the specific embodiments disclosed below.
The existing GIS partial discharge testing device mainly comprises an ultrasonic testing module and an ultrahigh frequency testing module, wherein the ultrasonic testing module is simple to use but has low testing efficiency and is easy to miss judgment; the ultrahigh frequency testing module has high sensitivity, but has the problems of complicated field electromagnetic interference, positioning operation needing a high-grade oscilloscope and an experienced expert and the like.
Example 1:
as shown in fig. 1, the invention discloses an online GIS partial discharge detection method:
in order to solve the technical problems in the background technology, the invention provides an online GIS partial discharge detection method.
And measuring the running GIS by an Ultra High Frequency (UHF) test module to obtain an Ultra High Frequency acquisition signal.
And measuring the running GIS through an ultrasonic testing module to obtain an ultrasonic acquisition signal.
The method comprises the steps of inputting collected ultrahigh frequency collected signals and ultrasonic collected signals into a partial discharge analysis module, extracting characteristic signals of the collected ultrahigh frequency collected signals, and obtaining waveform characteristic signals and amplitude characteristic signals according to the characteristic signals, wherein the waveform characteristic signals are obtained based on the rising time of the ultrahigh frequency collected signals, the falling edge time of the ultrahigh frequency collected signals and the main peak time of the ultrahigh frequency collected signals, and the amplitude characteristic signals are obtained based on the mean value of the amplitude of the ultrahigh frequency collected signals, the standard deviation of the amplitude and the entropy of the amplitude.
And the partial discharge analysis module forms a preliminary input set by the ultrasonic acquisition signals and the waveform characteristic signals obtained by analysis, sends the preliminary input set into a trained PSO-SVM-based combined prediction model, and outputs corresponding fault levels.
And when the fault level is in a serious state or an emergency state, taking the fault level, the ultrasonic acquisition signal, the amplitude characteristic signal and the waveform characteristic signal as an input set, sending the input set into a classification model based on an SVM algorithm to obtain a corresponding partial discharge fault type, and outputting a corresponding fault processing strategy according to the fault level and the partial discharge fault type.
Specifically, for example, when the fault level obtained by the combined prediction model based on the PSO-SVM algorithm is emergency, an emergency level signal, a 50mV ultrasonic acquisition signal, an amplitude characteristic signal and a waveform characteristic signal are transmitted to the classification model based on the SVM algorithm, and the obtained partial discharge fault type is internal impurities.
By adopting the ultrasonic testing module and the ultrahigh frequency testing module, the acquisition of GIS partial discharge signals can be realized, the analysis of the partial discharge signals is realized by arranging the partial discharge analysis module, and the identification of partial discharge types is quicker and more accurate by adopting a combined prediction model of PSO-SVM, so that the problems that the original ultrasonic testing module is low in testing efficiency and easy to miss judgment, the ultrahigh frequency testing module needs to be positioned by using a high-grade oscilloscope and an experienced expert to operate and the like are solved, and the working intensity of detection personnel is reduced.
Through the construction of the waveform characteristic signal and the amplitude characteristic signal, the quantity of input data is further reduced, and the efficiency of model processing is further improved.
The fault grade is obtained through the combined prediction model prediction based on the PSO-SVM, and then the fault type is judged according to the grade of the fault grade, so that the judgment accuracy is further improved, unnecessary calculation is avoided, and the efficiency is improved.
The partial discharge fault type with the fault level is output, so that the partial discharge type and the fault level are output to a detector, the experience requirement and the use requirement of the detector are further reduced, the operation of the module is simpler, and the judgment on the module is more accurate.
And outputting a corresponding fault processing strategy, so that a detector can process the module according to the suggestion and the regulation requirement, the module becomes more humanized, and the detection result becomes more instructive.
In another possible embodiment, an Ultra High Frequency-UHF (Ultra High Frequency-UHF) test module measures the surrounding environment of the GIS to obtain the background signal.
By acquiring the background signal, the detection precision can be further improved, and peripheral interference factors are eliminated.
In a possible other embodiment, the ultrasonic testing module measures the surrounding environment of the GIS to obtain the background signal.
By acquiring the background signal, the detection precision can be further improved, and peripheral interference factors are eliminated.
In a possible further embodiment, the waveform characteristic signal is calculated by the formula:
whereint 1 、t 2 、t 3 Respectively the rising time of the ultrahigh frequency acquisition signal, the falling edge time of the ultrahigh frequency acquisition signal and the main peak time of the ultrahigh frequency acquisition signal,K 1 、K 2 、K 3 、K 4 the value is constant and ranges from 0 to 1, and is determined specifically according to the influence degree of the time based on the rising time, the falling edge time and the main peak time on the prediction result.
In a possible further embodiment, the amplitude characteristic signal is calculated by the formula:
whereinK 5 、K 6 、K 7 、K 8 、K 9 The value range is between 0 and 1, and is determined according to the average value of the amplitude values, the standard deviation of the amplitude values and the influence degree of the entropy values of the amplitude values on the prediction result,J 1 、J 2 、Srespectively is the mean value of the amplitude of the ultrahigh frequency acquisition signal, the standard deviation of the amplitude and the entropy value of the amplitude.
In another possible embodiment, the PSO-SVM algorithm is trained by the following specific steps:
s1, a waveform characteristic signal, an amplitude characteristic signal and an ultrasonic acquisition signal are used as input data, partial discharge fault or normal data with a fault level corresponding to the signal data are used as output data, a data set is constructed based on the input data and the output data, and the data set is divided into a training set and a testing set.
S2, establishing an SVM model, determining parameters needing optimization, the number L1 of neurons, the learning rate epsilon and the training iteration times k, and determining respective optimization ranges of the parameters.
S3 initializes PSO parameters. Including the initial velocity and position of the particle, the learning weight, the number of training passes, and the scale.
And S4, determining a fitness function of the particles. The improved error function of the prediction model is used as the fitness function of the particles, and the optimal model parameters are searched.
S5, the fitness values of the particles are compared. And searching the individual optimal position and the global optimal position, and updating the optimal fitness value.
And S6, judging whether the maximum iteration number is reached. And if the maximum iteration times are reached, transmitting the obtained optimal parameters to the SVM model, and carrying out training and prediction. If the request is not met, the procedure returns to step S5.
The PSO-SVM algorithm is adopted, the fault detection speed can be further improved, results can be obtained within several seconds of fault detection on site, each fault detection point only has a few minutes of time, if the speed is low, real-time output cannot be achieved, the final detection result is disordered, the detection device cannot normally reflect the state of the GIS, after training is completed, the related algorithm can quickly identify faults, and the risk that the fault problem cannot be detected due to the fact that the speed is too low is avoided.
In a possible further embodiment, the failure level comprises: mild status, severe status, emergency status.
In another possible embodiment, when the fault is in the slight state, only the observation needs to be strengthened; in the severe state, the fault is severe at the moment, but the normal operation of the GIS is not influenced, and the GIS needs to be retested regularly; in the emergency state, the fault is serious, and the shutdown needs to be immediately contacted.
In a possible further embodiment, the partial discharge fault type: including internal impurities, surface burrs, poor contact, suspension potential, and solid insulation surface contamination.
In another possible embodiment, the modified error function is specifically formulated as:
wherein the content of the first and second substances,
for the output value of the t-th neural network,
is the average of the outputs of all N neurons,
is the desired output value of the neuron,K 10 is a constant.
And an improved error function is adopted, so that the final prediction result becomes more accurate, and the fault state can be better determined.
In another possible embodiment, the fault handling policy specifically includes: strengthening detection; periodic retesting; and (5) stopping the machine.
And outputting a corresponding fault processing strategy, so that a detector can process the module according to the suggestion and the regulation requirement, the module becomes more humanized, and the detection result becomes more instructive.
Example 2
As shown in fig. 3, the present invention discloses an online GIS partial discharge detection device:
the online GIS partial discharge detection device adopts the partial discharge detection method and specifically comprises an ultrasonic testing module, an ultrahigh frequency testing module and a partial discharge analysis module.
And the ultrasonic testing module is used for measuring the running GIS to obtain an ultrahigh frequency acquisition signal.
And the ultrasonic testing module is used for measuring the running GIS to obtain an ultrasonic acquisition signal.
And the partial discharge analysis module is used for analyzing the ultrahigh frequency acquisition signals and the ultrasonic acquisition signals obtained by analysis and outputting corresponding partial discharge fault types with fault levels.
In embodiments of the present invention, the term "plurality" means two or more unless explicitly defined otherwise. The terms "mounted," "connected," "fixed," and the like are to be construed broadly and include, for example, "connected" that may be fixedly connected, detachably connected, or integrally connected. Specific meanings of the above terms in the embodiments of the present invention can be understood by those of ordinary skill in the art according to specific situations.
In the description of the embodiments of the present invention, it should be understood that the terms "upper", "lower", and the like indicate orientations or positional relationships based on those shown in the drawings, and are only for convenience in describing the embodiments of the present invention and simplifying the description, but do not indicate or imply that the referred devices or units must have a specific direction, be configured in a specific orientation, and operate, and thus, should not be construed as limiting the embodiments of the present invention.
In the description herein, the appearances of the phrase "one embodiment," "a preferred embodiment," or the like, are intended to mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the embodiments of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
The above description is only a preferred embodiment of the present invention, and is not intended to limit the present invention, and various modifications and changes may be made to the present embodiment by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the embodiments of the present invention should be included in the protection scope of the embodiments of the present invention.
Claims (6)
1. An online GIS partial discharge detection method is characterized by specifically comprising the following steps:
measuring the surrounding environment of the running GIS through an ultrahigh frequency test module to obtain a background signal of the GIS and obtain an ultrahigh frequency acquisition signal;
measuring the surrounding environment of the running GIS through an ultrasonic testing module, and acquiring a background signal of the GIS to obtain an ultrasonic acquisition signal;
inputting the collected ultrahigh frequency collected signal and the collected ultrasonic collected signal into a partial discharge analysis module, and extracting a characteristic signal of the collected ultrahigh frequency collected signal to obtain a waveform characteristic signal and an amplitude characteristic signal;
the waveform characteristic signal is obtained based on the rising time of the ultrahigh frequency acquisition signal, the falling edge time of the ultrahigh frequency acquisition signal and the main peak time of the ultrahigh frequency acquisition signal, and the calculation formula of the waveform characteristic signal is as follows:
whereint 1 、t 2 、t 3 Respectively the rising time of the ultrahigh frequency acquisition signal, the falling edge time of the ultrahigh frequency acquisition signal and the main peak time of the ultrahigh frequency acquisition signal,K 1 、K 2 、K 3 、K 4 is constant and has a value ranging from 0 to 1;
the amplitude characteristic signal is obtained by construction based on the mean value of the amplitude of the ultrahigh frequency acquisition signal, the standard deviation of the amplitude and the entropy value of the amplitude, and the calculation formula of the amplitude characteristic signal is as follows:
whereinK 5 、K 6 、K 7 、K 8 、K 9 Is constant and has a value ranging from 0 to 1,J 1 、J 2 、Srespectively is the mean value of the amplitude of the ultrahigh frequency acquisition signal, the standard deviation of the amplitude and the entropy value of the amplitude;
the partial discharge analysis module constructs ultrasonic acquisition signals and waveform characteristic signals into a primary input set, inputs the primary input set into a trained PSO-SVM-based combined prediction model, and outputs fault levels through the prediction model;
and when the fault level is in a serious state or an emergency state, inputting the fault level, the ultrasonic acquisition signal, the amplitude characteristic signal and the waveform characteristic signal as an input set into a classification model based on an SVM algorithm to obtain a partial discharge fault type, and outputting a fault processing strategy according to the fault level and the partial discharge fault type.
2. The online GIS partial discharge detection method of claim 1, wherein the PSO-SVM algorithm training comprises the following steps:
s1, taking waveform characteristic signals, amplitude characteristic signals and ultrasonic acquisition signals as input data, dividing the input data into a training set and a testing set, and taking partial discharge fault or normal data with a fault level corresponding to the signals as output data;
s2, establishing an SVM model, determining parameters needing optimization, the number L1 of neurons, the learning rate epsilon and the training iteration times k, and determining respective optimization ranges;
s3, initializing PSO parameters including initial speed and position of particles, learning weight, training times and scale;
s4, determining a fitness function of the particles, taking an improved error function of the prediction model as the fitness function of the particles, and searching for optimal model parameters;
s5, comparing the fitness values of the particles to obtain an individual optimal position and a global optimal position, and updating the optimal fitness value;
s6, judging whether the maximum iteration times is reached, if the maximum iteration times is reached, transmitting the obtained optimal parameters to the SVM model, training and predicting, and if the maximum iteration times is not reached, returning to the step S5.
3. The online GIS partial discharge detection method of claim 1, wherein the fault level comprises: mild, severe, emergency; in a slight state, the fault is in a slight state and needs to be observed intensively; in a severe state, the fault is serious, but the normal operation of the GIS is not influenced, and the GIS needs to be retested regularly; in an emergency state, the fault is serious, and the shutdown needs to be immediately contacted.
4. The online GIS partial discharge detection method of claim 2, wherein the improved error function has a specific formula:
wherein, the first and the second end of the pipe are connected with each other,
for the output value of the t-th neural network,
is the average of the outputs of all N neurons,
is the desired output value of the neuron,K 10 is a constant.
5. The online GIS partial discharge detection method of claim 1, wherein the partial discharge fault types include internal impurities, surface burrs, poor contacts, floating potential, and solid insulation surface contamination.
6. An online GIS partial discharge detection device, which adopts the online GIS partial discharge detection method of any one of claims 1-5, and is characterized by specifically comprising: the system comprises an ultrasonic testing module, an ultrahigh frequency testing module and a partial discharge analysis module;
the ultrasonic testing module is used for measuring the running GIS to obtain an ultrahigh frequency acquisition signal;
the ultrasonic testing module is used for measuring the running GIS to obtain an ultrasonic acquisition signal;
and the partial discharge analysis module is used for analyzing the ultrahigh frequency acquisition signal and the ultrasonic acquisition signal obtained by analysis and outputting a corresponding partial discharge fault type with a fault level.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211497961.9A CN115542099B (en) | 2022-11-28 | 2022-11-28 | Online GIS partial discharge detection method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211497961.9A CN115542099B (en) | 2022-11-28 | 2022-11-28 | Online GIS partial discharge detection method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115542099A CN115542099A (en) | 2022-12-30 |
| CN115542099B true CN115542099B (en) | 2023-02-07 |
Family
ID=84722086
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211497961.9A Active CN115542099B (en) | 2022-11-28 | 2022-11-28 | Online GIS partial discharge detection method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115542099B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115840794B (en) * | 2023-02-14 | 2023-05-02 | 国网山东省电力公司东营供电公司 | Photovoltaic system planning method based on GIS and RL models |
| CN116359683B (en) * | 2023-02-28 | 2023-12-26 | 四川大唐国际甘孜水电开发有限公司 | Partial discharge mode identification method and system based on information interaction |
| CN116046077B (en) * | 2023-03-14 | 2023-07-14 | 浙江省邮电工程建设有限公司 | GIS state detection method based on temperature and humidity monitoring |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104751209A (en) * | 2015-04-02 | 2015-07-01 | 中国航天科工集团第二研究院七〇六所 | Optional-forward-link encoding method of ultra-high frequency reader |
| CN105548832A (en) * | 2015-12-10 | 2016-05-04 | 国网四川省电力公司电力科学研究院 | High-voltage power cable fault recognition method |
| CN107907807A (en) * | 2017-12-25 | 2018-04-13 | 国网湖北省电力公司信息通信公司 | A kind of local discharge of gas-insulator switchgear mode identification method |
| CN108896878A (en) * | 2018-05-10 | 2018-11-27 | 国家电网公司 | A kind of detection method for local discharge based on ultrasound |
| CN112801127A (en) * | 2020-12-09 | 2021-05-14 | 西安华谱电力设备制造有限公司 | Cable partial discharge defect identification method based on oscillation waves |
| CN114414955A (en) * | 2021-12-24 | 2022-04-29 | 贵州电网有限责任公司 | Method and system for monitoring and evaluating state of switch cabinet |
-
2022
- 2022-11-28 CN CN202211497961.9A patent/CN115542099B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104751209A (en) * | 2015-04-02 | 2015-07-01 | 中国航天科工集团第二研究院七〇六所 | Optional-forward-link encoding method of ultra-high frequency reader |
| CN105548832A (en) * | 2015-12-10 | 2016-05-04 | 国网四川省电力公司电力科学研究院 | High-voltage power cable fault recognition method |
| CN107907807A (en) * | 2017-12-25 | 2018-04-13 | 国网湖北省电力公司信息通信公司 | A kind of local discharge of gas-insulator switchgear mode identification method |
| CN108896878A (en) * | 2018-05-10 | 2018-11-27 | 国家电网公司 | A kind of detection method for local discharge based on ultrasound |
| CN112801127A (en) * | 2020-12-09 | 2021-05-14 | 西安华谱电力设备制造有限公司 | Cable partial discharge defect identification method based on oscillation waves |
| CN114414955A (en) * | 2021-12-24 | 2022-04-29 | 贵州电网有限责任公司 | Method and system for monitoring and evaluating state of switch cabinet |
Non-Patent Citations (1)
| Title |
|---|
| GIS局部放电多特征量传感研究进展;张兴辉,等;《高压电器》;20220416;第58卷(第4期);第8-16页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115542099A (en) | 2022-12-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115542099B (en) | Online GIS partial discharge detection method and device | |
| CN102053016B (en) | System for monitoring vibration of rotating machinery rolling bearing in wireless mode | |
| CN113124929A (en) | Transformer substation multi-parameter signal acquisition comprehensive analysis system and method | |
| CN101576593B (en) | Array type inferior insulator local electric field detector and inverse diagnosis method | |
| CN110647924B (en) | GIS equipment state evaluation method based on support vector description and K-nearest neighbor algorithm | |
| WO2023284127A1 (en) | Vibration-signal-based online monitoring system for gil defects, and method | |
| CN107543581A (en) | Multi-functional substation framework health monitoring and damnification recognition method | |
| CN106918389A (en) | It is a kind of based on the vibration modal analysis method of doppler optical displacement method and its application | |
| CN111999614A (en) | Local discharge multi-source fusion detection early warning method for high-voltage switch cabinet | |
| CN113702778B (en) | GIL arc discharge fault positioning method and system | |
| CN207992717U (en) | A kind of gate of hydropower station on-line condition monitoring system | |
| CN112198472B (en) | Online remote verification method and system for partial discharge intelligent sensor | |
| Song et al. | Multi-objective-based Acoustic Sensor Configuration for Structural Health Monitoring of Compressor Blade | |
| CN115754008A (en) | Combined monitoring method and system for structural damage, computer equipment and storage medium | |
| Riahi et al. | Differentiation of leakage and corrosion signals in acoustic emission testing of aboveground storage tank floors with artificial neural networks | |
| CN112505501B (en) | Partial discharge severity grade judging and early warning method based on electroacoustic combination | |
| Bi et al. | A fault diagnosis algorithm for wind turbine blades based on bp neural network | |
| CN113156278A (en) | CNN network-based GIS partial discharge positioning method | |
| Zhao et al. | Vibration feature extraction and fault detection method for transmission towers | |
| Heidari | Combined diagnosis of pd based on the multidimensional parameters | |
| Zhu et al. | Blocking detection of underground drainage pipe based on CEEMD and GG clustering | |
| CN115078540B (en) | Wall nondestructive testing system and wall nondestructive testing method for wall climbing detection robot | |
| Xu et al. | Research on identification of occasional ultra-high frequency partial discharge in ultra-high voltage GIS based on K-Means | |
| CN118130984A (en) | Cable partial discharge fault real-time monitoring method based on data driving | |
| Huang et al. | On-line Condition Monitoring of Vibration Sensor Based on Wavelet and Particle Swarm Optimization-back Propagation Neural Network |
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 |