CN111239263B - Method and system for detecting foreign matter defects in GIS equipment - Google Patents
Method and system for detecting foreign matter defects in GIS equipment Download PDFInfo
- Publication number
- CN111239263B CN111239263B CN202010062368.6A CN202010062368A CN111239263B CN 111239263 B CN111239263 B CN 111239263B CN 202010062368 A CN202010062368 A CN 202010062368A CN 111239263 B CN111239263 B CN 111239263B
- Authority
- CN
- China
- Prior art keywords
- foreign matter
- training
- training sample
- classification model
- matter defect
- 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
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N29/00—Investigating or analysing materials by the use of ultrasonic, sonic or infrasonic waves; Visualisation of the interior of objects by transmitting ultrasonic or sonic waves through the object
- G01N29/04—Analysing solids
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01H—MEASUREMENT OF MECHANICAL VIBRATIONS OR ULTRASONIC, SONIC OR INFRASONIC WAVES
- G01H17/00—Measuring mechanical vibrations or ultrasonic, sonic or infrasonic waves, not provided for in the preceding groups
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N29/00—Investigating or analysing materials by the use of ultrasonic, sonic or infrasonic waves; Visualisation of the interior of objects by transmitting ultrasonic or sonic waves through the object
- G01N29/44—Processing the detected response signal, e.g. electronic circuits specially adapted therefor
- G01N29/4409—Processing the detected response signal, e.g. electronic circuits specially adapted therefor by comparison
- G01N29/4418—Processing the detected response signal, e.g. electronic circuits specially adapted therefor by comparison with a model, e.g. best-fit, regression analysis
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N29/00—Investigating or analysing materials by the use of ultrasonic, sonic or infrasonic waves; Visualisation of the interior of objects by transmitting ultrasonic or sonic waves through the object
- G01N29/44—Processing the detected response signal, e.g. electronic circuits specially adapted therefor
- G01N29/4445—Classification of defects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Theoretical Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Pathology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- Immunology (AREA)
- Chemical & Material Sciences (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Biophysics (AREA)
- Signal Processing (AREA)
- Computational Linguistics (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Acoustics & Sound (AREA)
- Investigating Or Analyzing Materials By The Use Of Ultrasonic Waves (AREA)
Abstract
The embodiment of the invention discloses a method and a system for detecting foreign matter defects in GIS equipment. The detection method comprises the following steps: acquiring ultrasonic signals and vibration signals in the GIS equipment at preset time intervals; extracting the characteristic component of each section of ultrasonic signal and the characteristic component of the vibration signal to obtain each characteristic component matrix; and inputting each characteristic component matrix into a foreign matter defect classification model, and outputting a foreign matter defect category corresponding to each characteristic component matrix through the foreign matter defect classification model. According to the embodiment of the invention, the foreign matter defect signals in the GIS equipment are acquired by the ultrasonic sensor and the vibration sensor, and the foreign matter defect classification model outputs the category of the foreign matter defect, so that the detection is efficient and the detection result is accurate.
Description
Technical Field
The invention relates to the technical field of GIS equipment, in particular to a method and a system for detecting foreign matter defects in GIS equipment.
Background
The generation of foreign objects inside a Gas Insulated metal enclosed Switchgear (GIS) may occur during a device manufacturing process, an in-plant assembly process, a device transportation process, an in-situ installation process, and a GIS device operation process. By 2018, the failure condition of a combined electrical appliance of a certain power company in nearly 5 five years is statistically analyzed, the failure frequency of 363kV and above combined electrical appliance equipment in less than 5 years is 44, the failure frequency accounts for 80% of all failures, the failure is mainly caused by foreign matters and comprises 32 insulation breakdown discharges and 72.7% of the failure frequency, the internal cleaning of the combined electrical appliance is not in place in the split charging, general assembly and field connection processes before delivery in the production process of the combined electrical appliance, and after the combined electrical appliance is put into initial operation, the foreign matters are gradually exposed under the action of mechanical vibration and an electric field to cause discharge failures. When the voltage reaches a certain value, the foreign matters jump between the grounding shell and the high-voltage conductor due to resultant force of electric field force, gravity, viscous resistance and the like applied to the foreign matters, and partial discharge occurs when the foreign matters do not contact the high-voltage conductor in the jumping process. The motion characteristic of foreign matter depends on factors such as material, the shape of foreign matter, and because the inside foreign matter position of GIS equipment is different, the atress condition not only receives the effect of GIS equipment radial electric field power, can receive axial electric field power simultaneously for the foreign matter progressively is close to insulating parts such as basin insulator, causes electric field distortion, causes the inside insulating part surface flashover trouble of equipment.
In present electric power system, the foreign matter detection to GIS equipment inside is most to be carried out based on ultrasonic signal, but utilize ultrasonic signal only can assist electric power system's staff to judge whether there is the foreign matter defect in the GIS equipment, and the shape of judgement foreign matter defect that can not be fine and size and information and the risk of aassessment defect, this just makes the staff often just can go to disassemble GIS equipment and maintain after the accident that influences electric power operation, greatly increased cost of maintenance and reduced maintenance efficiency, simultaneously with the target that improves electric wire netting operating stability not conform.
Disclosure of Invention
The embodiment of the invention provides a method and a system for detecting a foreign matter defect inside GIS equipment, which aim to solve the problem that the prior art cannot effectively detect the shape, size and other information of the foreign matter defect inside the GIS equipment.
In a first aspect, a method for detecting a foreign object defect inside a GIS device is provided, including:
acquiring ultrasonic signals and vibration signals in the GIS equipment at preset time intervals;
extracting the characteristic component of each section of ultrasonic signal and the characteristic component of the vibration signal to obtain each characteristic component matrix;
inputting each characteristic component matrix into a foreign matter defect classification model, and outputting a foreign matter defect category corresponding to each characteristic component matrix through the foreign matter defect classification model;
wherein the foreign matter defect classification model is o (x) i )＝h(x i )w，o(x i ) Representing the corresponding foreign matter defect category, x, of each characteristic component matrix of the output i Representing the characteristic component matrix, h (x) i ) And w represents a weight value.
In a second aspect, a system for detecting a foreign object defect inside a GIS device is provided, including:
the acquisition module is used for acquiring ultrasonic signals and vibration signals in the GIS equipment at preset time intervals;
the extraction module is used for extracting the characteristic component of each section of ultrasonic signal and the characteristic component of the vibration signal to obtain each characteristic component matrix;
the classification module is used for inputting each characteristic component matrix into a foreign matter defect classification model and outputting a foreign matter defect category corresponding to each characteristic component matrix through the foreign matter defect classification model;
wherein the foreign matter defect classification model is o: (x i )＝h(x i )w，o(x i ) Representing the output foreign matter defect type x corresponding to each characteristic component matrix i Representing said characteristic component matrix, h (x) i ) And a matrix representing a non-linear mapping of the feature component matrix from the input space to the feature space, wherein w represents a weight.
Therefore, according to the embodiment of the invention, the foreign matter defect classification model outputs the category of the foreign matter defect by analyzing and processing the foreign matter defect signals collected by the ultrasonic sensor and the vibration sensor in the GIS equipment, so that the detection is efficient and the detection result is accurate.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments of the present invention will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to these drawings without inventive labor.
Fig. 1 is a flowchart of a method for detecting a foreign object defect inside a GIS device according to an embodiment of the present invention;
fig. 2 is a block diagram of a system for detecting a foreign object defect inside a GIS device according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, not all, embodiments of the present invention. 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 invention.
The embodiment of the invention discloses a method for detecting foreign matter defects in GIS equipment. As shown in fig. 1, the method comprises the following steps:
step S101: and acquiring ultrasonic signals and vibration signals in the GIS equipment at preset time intervals.
Specifically, an ultrasonic signal and a vibration signal are detected by an ultrasonic sensor and a vibration sensor, respectively. The ultrasonic sensor and the vibration sensor can be installed on the GIS equipment and used for converting physical information inside the GIS equipment into electric signals.
The preset time interval and the preset duration can be set according to experience and actual conditions. It should be understood that the ultrasonic signal of each preset time length corresponds to the vibration signal of each preset time length, that is, not only the time lengths are equal, but also the starting time and the stopping time are the same.
Step S102: and extracting the characteristic component of each section of ultrasonic signal and the characteristic component of the vibration signal to obtain each characteristic component matrix.
Specifically, the step may include the following processes:
(1) And intercepting a section of characteristic ultrasonic signal from each section of ultrasonic signal according to the waveform diagram of each section of ultrasonic signal.
Wherein, the abscissa of the oscillogram of the ultrasonic signal is time, and the ordinate is amplitude. The amplitudes in the truncated characteristic ultrasound signal include amplitudes greater than a preset ultrasound threshold.
(2) And intercepting a section of characteristic vibration signal from each section of vibration signal according to the waveform diagram of each section of vibration signal.
The abscissa of the waveform diagram of the vibration signal is time, and the ordinate is amplitude. The amplitude values in the intercepted characteristic vibration signal include amplitude values greater than a preset vibration threshold value.
In addition, it should be understood that each section of the characteristic ultrasonic signal and each section of the characteristic vibration signal are intercepted and correspond to each other, namely, not only the time length is equal, but also the starting time and the cut-off time are also the same.
(3) And performing discrete Fourier transform on each section of the characteristic ultrasonic signal and the characteristic vibration signal to obtain a frequency domain graph of each section of the characteristic ultrasonic signal and a frequency domain graph of the characteristic vibration signal.
(4) And extracting the characteristic component of each section of characteristic ultrasonic signal and the characteristic component of the characteristic vibration signal according to the frequency domain graph of each section of characteristic ultrasonic signal and the frequency domain graph of the characteristic vibration signal to obtain each characteristic component matrix.
Wherein the characteristic component matrix can be x i ＝[x i ′；x i ″]And (4) showing. And i is the serial number of the acquired ultrasonic signal and the acquired vibration signal. For example, the serial number of the first acquired segment of the ultrasonic signal and the vibration signal is 1, the serial number of the second acquired segment of the ultrasonic signal and the vibration signal is 2, and so on.
Specifically, the characteristic component of the ultrasonic signal is the frequency and amplitude of the ultrasonic signal, and can be used
And (4) showing. Omega mi ' denotes the frequency of the ultrasonic signal, A mi ' denotes the amplitude of the ultrasonic signal. m represents the number of fundamental waves and harmonics selected for each feature component. For example, in a preferred embodiment of the present invention, the fundamental, second harmonic and third harmonic are selected such that m =3. (omega) 1i ′,A 1i ') corresponding to the fundamental wave, (ω) 2i ′,A 2i ') corresponding to the second harmonic, (ω) 3i ′,A 3i ') corresponds to the third harmonic.
Specifically, the characteristic components of the vibration signal are the frequency and amplitude of the vibration signal, which can be used
And (4) showing. Omega mi "indicates the frequency of the vibration signal, A mi "denotes the amplitude of the vibration signal. Similarly, m represents the number of fundamental waves and harmonics selected for each feature component. For example, in a preferred embodiment of the present invention, the fundamental, second harmonic and third harmonic are selected such that m =3. (omega) 1i ″,A 1i Corresponding to fundamental wave, (omega) 2i ″,A 2i Corresponding to the second harmonic, (omega) 3i ″,A 3i ") corresponds to the third harmonic.
Step S103: and inputting each characteristic component matrix into a foreign matter defect classification model, and outputting a foreign matter defect category corresponding to each characteristic component matrix through the foreign matter defect classification model.
The foreign matter defect classification model of the embodiment of the invention is based on a single hidden layer feedforward neural network. Specifically, the foreign matter defect classification model is as follows:
o(x i )＝h(x i )w。
wherein, o (x) i ) And representing the corresponding foreign matter defect type of each output characteristic component matrix. x is the number of i A characteristic component matrix is represented. h (x) i ) Matrices representing a non-linear mapping of the feature component matrix from the input space to the feature space, i.e.
Where u represents the center point and σ represents the width (which can be randomly generated). w represents a weight value. The weight value can be determined through a pre-training process.
In a preferred embodiment of the present invention, the foreign object defect categories are divided by information such as the shape, size, and number of the foreign object defects. For example, the foreign object defect categories can be classified into four categories, respectively: (1) a single equally divided hemisphere with the radius of 1 mm; (2) a sphere with a radius of 1 mm; (3) a sphere with a radius of 0.5 mm; and (4) two equally-divided hemispheroids with the radius of 1 mm. It should be understood that the type of the foreign object defect of the embodiment of the present invention is not limited to the above examples, and may include other types.
Therefore, through the process, the foreign matter defect type corresponding to each section of ultrasonic signal and vibration signal can be obtained, so that the number, the shape, the size and the like of the foreign matter defects can be determined, the danger level caused by the foreign matter defects can be determined according to the number, the shape, the size and the like of the foreign matter defects, the three-dimensional dynamic motion process of the foreign matter defects in the GIS equipment can be reconstructed, and the GIS equipment can be conveniently guided to be overhauled. In addition, the ultrasonic signal and the vibration signal are adopted for analysis, so that noise interference can be eliminated, and the detection result is more accurate.
Preferably, the foreign object defect classification model can be trained by using samples in advance so as to determine a more accurate weight value, and thus, the detection result is more accurate.
Specifically, before step S101, the detection method further includes:
and training a foreign body defect classification model by adopting a training sample, and determining a weight.
Wherein, training the sample includes: the characteristic component of the ultrasonic signal, the characteristic component of the vibration signal and the standard foreign matter defect category corresponding to the characteristic component of the ultrasonic signal and the characteristic component of the vibration signal.
The training process specifically comprises the following steps:
(1) And setting a random weight value of the foreign matter defect classification model.
(2) And inputting the characteristic components of the ultrasonic signals and the vibration signals of the training samples into a foreign matter defect classification model, and outputting the foreign matter defect category corresponding to each training sample through the foreign matter defect classification model.
(3) And calculating the difference value between the standard foreign body defect type of the training sample and the foreign body defect type corresponding to the training sample output by the foreign body defect classification model to obtain the training error of each training sample.
It should be understood that each foreign object defect category may be represented by a matrix of binary numbers. That is, the training sample belongs to the foreign object defect category, the value of the foreign object defect category is 1, and the values of the other foreign object defect categories are 0. Therefore, the difference value calculation of the standard foreign matter defect type of the training sample and the foreign matter defect type corresponding to the training sample output by the foreign matter defect classification model can be carried out.
(4) An energy function is calculated based on the training error for each training sample.
Specifically, the energy function is represented by the following formula:
wherein J (w) represents an energy function. e.g. of a cylinder k And e n Both represent training errors of the training samples. n and k each represent the rank of the training error for the training sample, k =1,2, … …, L, n =1,2, … …, L.
The embodiment of the invention introduces a Minimum Error Entropy (MEE) criterion on the basis of an ELM (Extreme Learning Machines) algorithm, and obtains an energy function through the following process:
the error entropy is used to evaluate the training error, namely:
wherein H α (e) The error entropy is represented. Alpha represents the order of the error entropy, alpha ≠ 1, alpha>0。V α (e) Representing the potential energy of the information. In particular, V α (e) Represented by the following formula:
V α (e)＝∫p α (e)de＝E[p α-1 (e)]。
where p (e) represents the probability density function of the training error. E [ p ] α-1 (e)]Representing the desired operator. In general, p (e) is estimated by the Parzen window method, as shown in the following equation:
wherein, the Gaussian nucleus
Binding V α (e)＝∫p α (e)de＝E[p α-1 (e)]And
two calculations can obtain an estimate of the second-order information potential, as shown in the following equation:
since-log is monotonically increasing, the error entropy H α (e) Minimization of, i.e. potential energy V of, information α (e) Is maximized.
Based on the above formula, the energy function can be defined as
By maximizing the energy function, an optimal solution of the weight can be obtained, namely:
(5) And determining the training error of each training sample corresponding to the maximum energy function as a standard training error.
(6) And calculating to obtain a weight by adopting a weight formula according to the standard training error of each training sample, the characteristic component of the ultrasonic signal of each training sample and the characteristic component of the vibration signal.
Specifically, the weight formula is: w = (H) T ΛH) -1 H T ΛT。
Wherein, H = [ (H (x)) 1 )) T ,(h(x 2 )) T ,…,(h(x L )) T ] T . H is the output matrix of the hidden layer node.
G σ Representing a gaussian kernel of width sigma. e.g. of a cylinder j Represents the standard training error of the training sample, j represents the index of the standard training error, j =1,2, … …, L. T represents the standard foreign object defect category matrix of the training sample. T can be defined by T = [ T ] 1 ,t 2 ,…t j ,…,t L ] T And (4) showing. t is t j The standard foreign object defect type matrix corresponding to the jth training sample is represented by the following formula j ＝[t j,1 ,t j,2 ,…t j,M ] T And M represents the number of categories.
Therefore, through the training process, the minimum error entropy criterion is introduced aiming at the non-Gaussian noise of the training sample, and through the autonomous updating process of the weight, the extreme learning machine algorithm suitable for the multi-mode distributed noise is obtained to determine the optimal weight, so that the foreign matter defect classification model adopting the weight outputs more accurate foreign matter defect types.
In a specific embodiment of the present invention, the accuracy of the detected foreign object defect type is 98%, and the accuracy of the prior art is 90%, so that the accuracy of the detection method of the embodiment of the present invention is higher than that of the prior art.
The foreign matter defect type obtained by the detection method of the embodiment of the invention can be applied to a three-dimensional dynamic process for reconstructing foreign matter defects in GIS equipment and risk degree evaluation of the GIS equipment.
Specifically, after the detection method of the embodiment of the invention determines the type of the foreign object defect, the flight trajectory of the foreign object in the GIS device can be calculated by using a computer simulation technology according to the information such as the shape, the size, the quantity and the like in the type of the foreign object defect, the three-dimensional dynamic process of the foreign object defect in the GIS device is simulated and reconstructed, and the three-dimensional dynamic process is displayed, so that related workers can intuitively know the motion state of the foreign object defect in the GIS device.
Specifically, a set of risk assessment criteria can be established, information such as the shape, size and number of the foreign matters is divided into different risk levels according to expert experience, the detected defect types of the foreign matters are matched with the risk levels, and finally the risk assessment results of the GIS equipment to be tested are given. The evaluation results are divided into three cases of danger, mild and safe. When the evaluation result is dangerous, in order to prevent the result from being caused by the influence of the environment, a secondary detection method can be adopted, namely, the detection is carried out again, the data is collected again and predicted again, if the result is still dangerous, the evaluation conclusion of the danger is given, the maintainers are informed in time, the defect development trend is diagnosed, and the maintenance suggestion is provided; if the display result is slight, the record is needed to be recorded on the case so as to be convenient for paying attention to in time; when the display result is safe, attention is not needed.
To sum up, the method for detecting the foreign matter defect inside the GIS equipment in the embodiment of the invention utilizes the extreme learning machine network training to obtain the foreign matter defect classification model inside the GIS equipment, analyzes and processes the foreign matter defect signals inside the GIS equipment collected by the ultrasonic sensor and the vibration sensor, outputs the foreign matter defect classification by the foreign matter defect classification model, has high detection efficiency and accurate detection result, so as to assist a worker to judge whether the GIS equipment needs to be disassembled to eliminate hidden dangers, and provides suggestions for later maintenance, thereby greatly improving the maintenance efficiency, and the foreign matter defect can be found in time, thereby reducing the economic loss caused by the internal defect of the GIS equipment, ensuring the operation stability of a power grid, and promoting the realization of the target of an intelligent power grid.
The embodiment of the invention also discloses a system for detecting the foreign matter defect in the GIS equipment. As shown in fig. 2, the detection system includes the following modules:
the acquisition module 201 is configured to acquire an ultrasonic signal and a vibration signal inside the GIS device for a preset time period at preset time intervals.
The extracting module 202 is configured to extract a characteristic component of each ultrasonic signal and a characteristic component of the vibration signal, so as to obtain each characteristic component matrix.
The classification module 203 is configured to input each feature component matrix into a foreign object defect classification model, and output a foreign object defect category corresponding to each feature component matrix through the foreign object defect classification model.
Wherein the foreign matter defect classification model is o (x) i )＝h(x i )w，o(x i ) Representing the corresponding foreign matter defect type x of each characteristic component matrix of the output i Representing a characteristic component matrix, h (x) i ) And a matrix representing a non-linear mapping of the feature component matrix from the input space to the feature space, wherein w represents the weight.
Preferably, the detection system further comprises:
and the training module is used for training the foreign matter defect classification model by adopting a training sample before the step of acquiring the ultrasonic signal and the vibration signal in the GIS equipment with a preset time length, and determining the weight.
Preferably, the training module comprises:
and the setting submodule is used for setting the random weight of the foreign matter defect classification model.
And the classification submodule is used for inputting the characteristic components of the ultrasonic signals and the vibration signals of the training samples into the foreign matter defect classification model and outputting the foreign matter defect categories corresponding to the training samples through the foreign matter defect classification model.
And the first calculation submodule is used for calculating the difference value between the standard foreign body defect type of the training sample and the foreign body defect type corresponding to the training sample output by the foreign body defect classification model, and obtaining the training error of each training sample.
And the second calculation submodule is used for calculating an energy function according to the training error of each training sample.
And the determining submodule is used for determining the training error of each training sample corresponding to the maximum energy function as a standard training error.
And the third calculation submodule is used for calculating to obtain a weight by adopting a weight formula according to the standard training error of each training sample, the characteristic component of the ultrasonic signal of each training sample and the characteristic component of the vibration signal.
The weight formula is: w = (H) T ΛH) -1 H T ΛT。
Wherein, H = [ (H (x)) 1 )) T ,(h(x 2 )) T ,…,(h(x L )) T ] T ，
G σ Denotes a Gaussian kernel of width σ, e j The standard training error of the training sample is represented, j represents the serial number of the standard training error, j =1,2, … …, L and T represent the standard foreign matter defect type matrix of the training sample.
Preferably, the energy function is:
wherein J (w) represents an energy function, e k And e n Each representing a training error of the training sample, n and k each representing a sequence number of the training error of the training sample, k =1,2, ……，L，n＝1，2，……，L。
Preferably, the characteristic component of the ultrasonic signal is a frequency and an amplitude of the ultrasonic signal, and the characteristic component of the vibration signal is a frequency and an amplitude of the vibration signal.
For the device embodiment, since it is basically similar to the method embodiment, the description is simple, and for the relevant points, refer to the partial description of the method embodiment.
To sum up, the system for detecting the foreign matter defects inside the GIS equipment in the embodiment of the invention utilizes the extreme learning machine network training to obtain the foreign matter defect classification model inside the GIS equipment, analyzes and processes the foreign matter defect signals inside the GIS equipment collected by the ultrasonic sensor and the vibration sensor, outputs the foreign matter defect types by the foreign matter defect classification model, has high detection efficiency and accurate detection result, so as to assist workers to judge whether the GIS equipment needs to be disassembled to eliminate hidden dangers, and provides suggestions for later maintenance, thereby greatly improving the maintenance efficiency, finding the foreign matter defects in time, reducing the economic loss caused by the internal defects of the GIS equipment, ensuring the operation stability of a power grid, and promoting the realization of the target of an intelligent power grid.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: various media capable of storing program codes, such as a U disk, a removable hard disk, a ROM, a RAM, a magnetic disk, or an optical disk.
The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present invention, and all the changes or substitutions should be covered within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (6)
1. A method for detecting foreign matter defects inside GIS equipment is characterized by comprising the following steps:
acquiring ultrasonic signals and vibration signals in the GIS equipment at preset time intervals;
extracting the characteristic component of each section of the ultrasonic signal and the characteristic component of the vibration signal to obtain each characteristic component matrix;
inputting each characteristic component matrix into a foreign matter defect classification model, and outputting a foreign matter defect category corresponding to each characteristic component matrix through the foreign matter defect classification model;
wherein the foreign matter defect classification model is o (x) i )＝h(x i )w，o(x i ) Representing the output foreign matter defect type x corresponding to each characteristic component matrix i Representing said characteristic component matrix, h (x) i ) A matrix representing a non-linear mapping of the feature component matrix from an input space to a feature space, w representing a weight;
before the step of acquiring the ultrasonic signal and the vibration signal inside the GIS equipment with a preset time length, the detection method further comprises the following steps:
training the foreign body defect classification model by adopting a training sample, and determining the weight;
the step of training the foreign matter defect classification model by adopting a training sample and determining the weight comprises the following steps:
setting a random weight value of the foreign matter defect classification model;
inputting the characteristic components of the ultrasonic signals and the characteristic components of the vibration signals of the training samples into the foreign matter defect classification model, and outputting the foreign matter defect category corresponding to each training sample through the foreign matter defect classification model;
calculating the difference value between the standard foreign body defect type of the training sample and the foreign body defect type corresponding to the training sample output by the foreign body defect classification model to obtain the training error of each training sample;
calculating an energy function according to the training error of each training sample;
determining the training error of each training sample corresponding to the maximum energy function as a standard training error;
calculating to obtain the weight by adopting a weight formula according to the standard training error of each training sample, the characteristic component of the ultrasonic signal and the characteristic component of the vibration signal of each training sample;
wherein, the weight formula is: w = (H) T ΛH) -1 H T ΛT；H＝[(h(x 1 )) T ,(h(x 2 )) T ,…,(h(x L )) T ] T ，
G σ Denotes a Gaussian kernel of width σ, e j And j =1,2, … …, L and T represent a standard foreign matter defect type matrix of the training sample.
2. The detection method according to claim 1, wherein the energy function is:
wherein J (w) represents an energy function, e k And e n Each representing a training error of the training sample, n and k each representing a rank of the training error of the training sample, k =1,2, … …, L, n =1,2, … …, L.
3. The detection method according to claim 1, characterized in that: the characteristic components of the ultrasonic signal are the frequency and the amplitude of the ultrasonic signal, and the characteristic components of the vibration signal are the frequency and the amplitude of the vibration signal.
4. A detection system for foreign matter defects inside GIS equipment is characterized by comprising:
the acquisition module is used for acquiring ultrasonic signals and vibration signals in the GIS equipment at preset time intervals;
the extraction module is used for extracting the characteristic component of each section of ultrasonic signal and the characteristic component of the vibration signal to obtain each characteristic component matrix;
the classification module is used for inputting each characteristic component matrix into a foreign matter defect classification model and outputting a foreign matter defect category corresponding to each characteristic component matrix through the foreign matter defect classification model;
wherein the foreign matter defect classification model is o (x) i )＝h(x i )w，o(x i ) Representing the output foreign matter defect type x corresponding to each characteristic component matrix i Representing said characteristic component matrix, h (x) i ) A matrix representing a non-linear mapping of the feature component matrix from an input space to a feature space, w represents a weight
The detection system further comprises:
the training module is used for training the foreign matter defect classification model by adopting a training sample before the step of acquiring the ultrasonic signal and the vibration signal inside the GIS equipment with a preset time length, and determining the weight;
the training module comprises:
the setting submodule is used for setting a random weight of the foreign matter defect classification model;
the classification submodule is used for inputting the characteristic components of the ultrasonic signals and the vibration signals of the training samples into the foreign matter defect classification model and outputting the foreign matter defect category corresponding to each training sample through the foreign matter defect classification model;
the first calculation submodule is used for calculating the difference value between the standard foreign body defect type of the training sample and the foreign body defect type corresponding to the training sample output by the foreign body defect classification model to obtain the training error of each training sample;
the second calculation submodule is used for calculating an energy function according to the training error of each training sample;
the determining submodule is used for determining the training error of each training sample corresponding to the maximum energy function as a standard training error;
the third calculation submodule is used for calculating to obtain the weight by adopting a weight formula according to the standard training error of each training sample, the characteristic component of the ultrasonic signal of each training sample and the characteristic component of the vibration signal;
wherein, the weight formula is: w = (H) T ΛH) -1 H T ΛT；H＝[(h(x 1 )) T ,(h(x 2 )) T ,…,(h(x L )) T ] T ，
G σ Denotes a Gaussian kernel of width σ, e j And j =1,2, … …, L and T represent a standard foreign matter defect type matrix of the training sample.
5. The detection system of claim 4, wherein the energy function is:
wherein J (w) represents an energy function, e k And e n Each representing a training error of the training sample, n and k each representing a rank of the training error of the training sample, k =1,2, … …, L, n =1,2, … …, L.
6. The detection system of claim 4, wherein: the characteristic components of the ultrasonic signal are the frequency and the amplitude of the ultrasonic signal, and the characteristic components of the vibration signal are the frequency and the amplitude of the vibration signal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010062368.6A CN111239263B (en) | 2020-01-19 | 2020-01-19 | Method and system for detecting foreign matter defects in GIS equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010062368.6A CN111239263B (en) | 2020-01-19 | 2020-01-19 | Method and system for detecting foreign matter defects in GIS equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111239263A CN111239263A (en) | 2020-06-05 |
| CN111239263B true CN111239263B (en) | 2022-10-28 |
Family
ID=70871296
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010062368.6A Active CN111239263B (en) | 2020-01-19 | 2020-01-19 | Method and system for detecting foreign matter defects in GIS equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111239263B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111768372B (en) * | 2020-06-12 | 2024-03-12 | 国网智能科技股份有限公司 | Method and system for detecting foreign matters in cavity of GIS (gas insulated switchgear) |
| CN112711023A (en) * | 2020-12-16 | 2021-04-27 | 国网青海省电力公司电力科学研究院 | Method for detecting residues in GIS |
| CN112946078B (en) * | 2021-02-03 | 2022-08-05 | 山东大学 | Composite material bonding quality evaluation and early damage identification method and system |
| CN113465664A (en) * | 2021-06-17 | 2021-10-01 | 国网宁夏电力有限公司电力科学研究院 | Method for detecting and identifying foreign matters in tank type power equipment |
| CN113654646A (en) * | 2021-06-29 | 2021-11-16 | 国网江苏省电力有限公司电力科学研究院 | Method, device and system for positioning and diagnosing internal mechanical defects of GIL (general information language) |
| CN114264945B (en) * | 2021-12-13 | 2023-08-04 | 广东电网有限责任公司 | Method and device for detecting equipment state |
Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5214595A (en) * | 1988-05-16 | 1993-05-25 | Hitachi, Ltd. | Abnormality diagnosing system and method for a high voltage power apparatus |
| JPH11271382A (en) * | 1998-01-26 | 1999-10-08 | Mitsubishi Electric Corp | Device for detecting partial discharge generation position of conducting device and device for detecting partial discharge strength |
| KR20100048529A (en) * | 2008-10-31 | 2010-05-11 | 엘에스전선 주식회사 | Power apparatus defect detection method and system |
| CN103558519A (en) * | 2013-11-02 | 2014-02-05 | 国家电网公司 | GIS partial discharge ultrasonic signal identification method |
| KR101515231B1 (en) * | 2013-11-22 | 2015-04-24 | 주식회사 효성 | A method of partial discharge diagnosis for gas insulated switchgear |
| CN106846316A (en) * | 2017-02-10 | 2017-06-13 | 云南电网有限责任公司电力科学研究院 | A kind of GIS inside typical defect automatic distinguishing method for image |
| CN106932493A (en) * | 2017-03-05 | 2017-07-07 | 北京工业大学 | A kind of analysis method of the ultrasonic sensor array parameter based on genetic algorithm |
| CN107728014A (en) * | 2017-08-23 | 2018-02-23 | 国网山东省电力公司检修公司 | Method and system based on multiple sensor signals characteristic identificating equipment insulation defect |
| CN108535616A (en) * | 2018-06-13 | 2018-09-14 | 云南电网有限责任公司昆明供电局 | Free metal particle insulation defect analogue experiment installation and method in a kind of GIS |
| CN109444682A (en) * | 2018-11-02 | 2019-03-08 | 国网四川省电力公司广安供电公司 | The construction method of partial discharge of switchgear diagnostic system based on multi-information fusion |
| CN109839263A (en) * | 2019-02-26 | 2019-06-04 | 北京航空航天大学 | A kind of GIS feature extraction and mechanical defect diagnostic method based on vibration information |
| CN109974986A (en) * | 2019-03-05 | 2019-07-05 | 国网江苏省电力有限公司电力科学研究院 | A kind of GIS device mechanical defect diagnostic system and method based on vibration detection |
| CN110632484A (en) * | 2019-11-06 | 2019-12-31 | 广东电网有限责任公司 | ELM-based GIS partial discharge defect diagnosis and classification system and method |
| CN110646517A (en) * | 2019-10-25 | 2020-01-03 | 华南理工大学 | Probe coupling device for ultrasonically detecting internal defects of GIS insulating part |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011158711A1 (en) * | 2010-06-14 | 2011-12-22 | 住友金属工業株式会社 | Defect determination device |
| US11157014B2 (en) * | 2016-12-29 | 2021-10-26 | Tesla, Inc. | Multi-channel sensor simulation for autonomous control systems |
-
2020
- 2020-01-19 CN CN202010062368.6A patent/CN111239263B/en active Active
Patent Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5214595A (en) * | 1988-05-16 | 1993-05-25 | Hitachi, Ltd. | Abnormality diagnosing system and method for a high voltage power apparatus |
| JPH11271382A (en) * | 1998-01-26 | 1999-10-08 | Mitsubishi Electric Corp | Device for detecting partial discharge generation position of conducting device and device for detecting partial discharge strength |
| KR20100048529A (en) * | 2008-10-31 | 2010-05-11 | 엘에스전선 주식회사 | Power apparatus defect detection method and system |
| CN103558519A (en) * | 2013-11-02 | 2014-02-05 | 国家电网公司 | GIS partial discharge ultrasonic signal identification method |
| KR101515231B1 (en) * | 2013-11-22 | 2015-04-24 | 주식회사 효성 | A method of partial discharge diagnosis for gas insulated switchgear |
| CN106846316A (en) * | 2017-02-10 | 2017-06-13 | 云南电网有限责任公司电力科学研究院 | A kind of GIS inside typical defect automatic distinguishing method for image |
| CN106932493A (en) * | 2017-03-05 | 2017-07-07 | 北京工业大学 | A kind of analysis method of the ultrasonic sensor array parameter based on genetic algorithm |
| CN107728014A (en) * | 2017-08-23 | 2018-02-23 | 国网山东省电力公司检修公司 | Method and system based on multiple sensor signals characteristic identificating equipment insulation defect |
| CN108535616A (en) * | 2018-06-13 | 2018-09-14 | 云南电网有限责任公司昆明供电局 | Free metal particle insulation defect analogue experiment installation and method in a kind of GIS |
| CN109444682A (en) * | 2018-11-02 | 2019-03-08 | 国网四川省电力公司广安供电公司 | The construction method of partial discharge of switchgear diagnostic system based on multi-information fusion |
| CN109839263A (en) * | 2019-02-26 | 2019-06-04 | 北京航空航天大学 | A kind of GIS feature extraction and mechanical defect diagnostic method based on vibration information |
| CN109974986A (en) * | 2019-03-05 | 2019-07-05 | 国网江苏省电力有限公司电力科学研究院 | A kind of GIS device mechanical defect diagnostic system and method based on vibration detection |
| CN110646517A (en) * | 2019-10-25 | 2020-01-03 | 华南理工大学 | Probe coupling device for ultrasonically detecting internal defects of GIS insulating part |
| CN110632484A (en) * | 2019-11-06 | 2019-12-31 | 广东电网有限责任公司 | ELM-based GIS partial discharge defect diagnosis and classification system and method |
Non-Patent Citations (2)
| Title |
|---|
| Field Applications of Ultra High Frequency Techniques for Defect Detection in GIS;Jun Xiong;《SENSOR》;20180626;第18卷;全文 * |
| GIS内部缺陷检测及诊断分析技术研究;何宁辉 等;《电测与仪表》;20191225;第56卷(第24期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111239263A (en) | 2020-06-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111239263B (en) | Method and system for detecting foreign matter defects in GIS equipment | |
| CN109116193B (en) | Electrical equipment dangerous discharge distinguishing method based on partial discharge signal comprehensive entropy | |
| CN102016607B (en) | Method and apparatus for analyzing waveform signals of a power system | |
| EP3699614A1 (en) | Method and system of partial discharge recognition for diagnosing electrical networks | |
| CN107907320B (en) | Method and device for diagnosing mechanical state characteristics of on-load tap-changer of transformer | |
| US10539602B2 (en) | Active shaft grounding system with diagnostic waveform analysis | |
| CN108375718B (en) | Evaluation of phase-resolved partial discharges | |
| CN112816838B (en) | GIS equipment defect diagnosis device and method based on FFT, VMD and LS-SVM | |
| Hussain et al. | Review on partial discharge diagnostic techniques for high voltage equipment in power systems | |
| JP2019191142A (en) | Winding insulation deterioration diagnostic device of rotary machine | |
| CN117590172A (en) | Partial discharge acoustic-electric combined positioning method, device and equipment applied to transformer | |
| Martinovič et al. | Fast algorithm for contactless partial discharge detection on remote gateway device | |
| CN115951185A (en) | Insulator flashover detection method and device, electronic equipment and storage medium | |
| Zhao et al. | Rolling Bearing Fault Feature Extraction Based on Adaptive Tunable Q‐Factor Wavelet Transform and Spectral Kurtosis | |
| CN116910470A (en) | GIS combined electrical apparatus partial discharge fault mode identification method | |
| WO2017176401A1 (en) | Shaft ground monitoring with diagnostic waveform analysis | |
| CN114264945B (en) | Method and device for detecting equipment state | |
| CN115455684A (en) | Oil-immersed transformer non-contact state evaluation method and system based on noise analysis | |
| CN114184911B (en) | Method and device for detecting defect type of equipment and electronic equipment | |
| CN112729531B (en) | Fault studying and judging method and system for distribution transformer equipment | |
| CN110703080B (en) | GIS spike discharge diagnosis method, discharge degree identification method and device | |
| CN109357751B (en) | Power transformer winding looseness defect detection system | |
| CN109917245B (en) | Ultrasonic detection partial discharge signal mode identification method considering phase difference | |
| Rossetti et al. | Classification of bearing faults through time-frequency analysis and image processing | |
| CN117849560B (en) | Valve side sleeve insulation monitoring method and system combining end screen voltage and partial discharge |
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 |