CN113191437A - Transformer mechanical fault detection method based on vibration signal composite eigenvector - Google Patents
Transformer mechanical fault detection method based on vibration signal composite eigenvector Download PDFInfo
- Publication number
- CN113191437A CN113191437A CN202110496964.XA CN202110496964A CN113191437A CN 113191437 A CN113191437 A CN 113191437A CN 202110496964 A CN202110496964 A CN 202110496964A CN 113191437 A CN113191437 A CN 113191437A
- Authority
- CN
- China
- Prior art keywords
- transformer
- vibration signal
- vibration
- principal component
- feature vector
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01H—MEASUREMENT OF MECHANICAL VIBRATIONS OR ULTRASONIC, SONIC OR INFRASONIC WAVES
- G01H11/00—Measuring mechanical vibrations or ultrasonic, sonic or infrasonic waves by detecting changes in electric or magnetic properties
- G01H11/06—Measuring mechanical vibrations or ultrasonic, sonic or infrasonic waves by detecting changes in electric or magnetic properties by electric means
- G01H11/08—Measuring mechanical vibrations or ultrasonic, sonic or infrasonic waves by detecting changes in electric or magnetic properties by electric means using piezoelectric devices
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01P—MEASURING LINEAR OR ANGULAR SPEED, ACCELERATION, DECELERATION, OR SHOCK; INDICATING PRESENCE, ABSENCE, OR DIRECTION, OF MOVEMENT
- G01P15/00—Measuring acceleration; Measuring deceleration; Measuring shock, i.e. sudden change of acceleration
- G01P15/02—Measuring acceleration; Measuring deceleration; Measuring shock, i.e. sudden change of acceleration by making use of inertia forces using solid seismic masses
- G01P15/08—Measuring acceleration; Measuring deceleration; Measuring shock, i.e. sudden change of acceleration by making use of inertia forces using solid seismic masses with conversion into electric or magnetic values
- G01P15/09—Measuring acceleration; Measuring deceleration; Measuring shock, i.e. sudden change of acceleration by making use of inertia forces using solid seismic masses with conversion into electric or magnetic values by piezoelectric pick-up
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Testing Of Short-Circuits, Discontinuities, Leakage, Or Incorrect Line Connections (AREA)
Abstract
本发明属于电力设备检测技术领域，具体涉及一种基于振动信号的变压器机械故障在线检测方法；采用的技术方案为：布置传感器采集变压器振动信号传输至数据采集仪；从变压器振动信号中提取出七类振动信号波形特征量；对变压器振动信号进行分解，取前七组IMF分量作为振动信号能量特征量；组成十四维的复合特征向量，利用主成分分析法将其降维成两维主成分特征向量；提取已知状态下主成分特征向量作为样本，过DBN网络两阶段的训练得到特征向量与变压器机械状态之间的对应关系；将目标变压器的2维主成分特征向量输入训练好的DBN网络得到该变压器的机械状态；本发明主要用于变压器机械故障在线检测。The invention belongs to the technical field of power equipment detection, and in particular relates to an online detection method for transformer mechanical faults based on vibration signals; the adopted technical scheme is: arranging sensors to collect transformer vibration signals and transmitting them to a data acquisition instrument; The characteristic quantity of vibration-like signal waveform; the transformer vibration signal is decomposed, and the first seven groups of IMF components are taken as the energy characteristic quantity of the vibration signal; 14-dimensional composite eigenvectors are formed, and the principal component analysis method is used to reduce the dimension into two-dimensional principal components Feature vector; extract the principal component feature vector in the known state as a sample, and obtain the corresponding relationship between the feature vector and the mechanical state of the transformer through two-stage training of the DBN network; input the 2-dimensional principal component feature vector of the target transformer into the trained DBN The network obtains the mechanical state of the transformer; the invention is mainly used for on-line detection of transformer mechanical faults.
Description
Technical Field
The invention belongs to the technical field of power equipment detection, and particularly relates to a transformer mechanical fault online detection method based on vibration signals.
Background
The transformer is an important device in the power system, and bears a series of important tasks such as voltage conversion, electric energy distribution and transmission, and the operation state of the transformer directly influences the safety and stability of the whole power system. The collision, extrusion and external short circuit faults of the transformer during use and transportation can cause the loosening and deformation of the internal winding and the iron core, and the conditions can threaten the safe operation of the power system.
At present, the traditional detection methods for the mechanical fault of the transformer mainly comprise a low-voltage pulse method, a frequency response method, a short-circuit impedance method and the like, the method needs to stop the transformer to be detected and then can detect the transformer to be detected, the influence on the normal operation of a power system is large, the detection process is complicated, and the detection efficiency is low. Therefore, the research on the online detection method which is simple to operate and does not need to stop running the transformer to be detected has important practical significance.
Disclosure of Invention
The invention aims to provide a transformer online detection method based on a vibration signal composite eigenvector, so that the purpose of online detection of mechanical faults of a transformer is achieved by fully utilizing state information contained in a vibration signal of the transformer under the condition of ensuring normal operation of the transformer.
In order to achieve the purpose, the invention discloses the following technical scheme:
a transformer mechanical fault detection method based on vibration signal composite eigenvector is realized by the following steps:
step 1: a vibration signal acquisition system is formed by a PCB356A16 piezoelectric acceleration sensor and an NI-9234 data acquisition instrument, the acceleration sensor is attached to a transformer box body, and a vibration signal of the transformer is acquired and transmitted to the data acquisition instrument;
step 2: extracting the waveform characteristic quantities of the average value, skewness, peak-to-peak value, kurtosis, fundamental frequency ratio, 50Hz odd-order frequency multiplication ratio and seven types of vibration signals from the vibration signals of the transformer;
and step 3: EEMD decomposition is carried out on the transformer vibration signal to obtain IMF components of the vibration signal, and the first seven groups of IMF components are extracted to serve as vibration signal energy characteristic quantities;
and 4, step 4: combining the seven types of waveform characteristic quantities and the seven types of energy characteristic quantities of the transformer into a fourteen-dimensional composite characteristic vector;
and 5: reducing the dimension of the fourteen-dimensional composite feature vector into a two-dimensional principal component feature vector by using a principal component analysis method;
step 6: extracting a plurality of groups of vibration signal principal component characteristic vectors under a known state as training samples of the DBN network, and obtaining the corresponding relation between the characteristic vectors and the mechanical state of the transformer by the training samples through two stages of training of the DBN network;
and 7: inputting the 2-dimensional principal component feature vector of the target transformer into the trained DBN network to obtain the mechanical state of the transformer; the working flow of the detection method is shown in the attached figure 2.
The transformer vibration signal acquisition points are provided with six, which are respectively as follows: a1, B2, C3, a4, B5 and C6.
The acquisition points A1, B2 and C3 of the transformer vibration signal are uniformly distributed at the top of the transformer box body and respectively correspond to the positions of A, B, C three phases.
The acquisition points A4, B5 and C6 of the transformer vibration signal are uniformly distributed on the side surface of the transformer box body and respectively correspond to the positions of A, B, C three phases.
Advantages and advantageous effects of the invention
(1) The transformer mechanical fault detection method based on the vibration signal does not need to stop operating the transformer to be detected, is not electrically connected with the power system, and can ensure the normal operation of the power system; the interference of environmental factors is small, and the accuracy of state identification is high; the operation is simple, and the detection efficiency is high.
(2) According to the invention, through the combination of the waveform characteristic vector and the energy characteristic vector of the vibration signal, the information complementarity among various characteristic quantities in the vibration signal is fully utilized, and the accuracy of the mechanical fault detection of the transformer is improved.
(3) The invention optimizes the composite characteristic vector by using a principal component analysis method, not only retains the original information of the composite characteristic vector, but also reduces the dimension of the characteristic vector, improves the working efficiency of the classifier, and effectively improves the precision and speed of the mechanical fault detection of the transformer.
Drawings
The invention is further described below with reference to the accompanying drawings.
FIG. 1 is a block diagram of a vibration signal acquisition system of the present invention;
FIG. 2 is a flow chart of a fault detection method of the present invention;
FIG. 3 is a distribution diagram of vibration signal acquisition points according to the present invention;
FIG. 4 is a frequency spectrum diagram of vibration signals before and after the loosening of the transformer core according to the present invention;
fig. 5 is a diagram of DBN recognition results before and after optimization of a vibration signal according to the present invention.
Detailed Description
Example (b):
as shown in fig. 1, loosening fault setting is performed on a 10kV transformer, and a transformer vibration signal acquisition system is set up to acquire vibration signals of the transformer so as to verify the validity of the detection method; a transformer vibration signal acquisition system is formed by utilizing a PCB356A16 piezoelectric acceleration sensor and an NI-9234 data acquisition instrument, the vibration sensor is adsorbed on the surface of a box body of a transformer to be detected, then a voltage regulator is opened, the adjusted voltage is applied to the transformer, and a transformer vibration signal acquired by the sensor is transmitted to the data acquisition instrument through a signal transmission line.
As shown in fig. 3, in order to reduce the influence of the transformer structure on the vibration signal, vibration signal acquisition points are arranged at the top and the side of the transformer box body; a1, B2 and C3 respectively correspond to the top position of A, B, C three phases, and A4, B5 and C6 respectively correspond to the middle position of A, B, C three-phase oil tanks.
In a transformer vibration signal acquisition system built in a laboratory, power supply voltage is applied to a transformer after being regulated by a voltage regulator, and three-phase vibration signals can be respectively acquired by a sensor of the vibration signals on the top of a box body.
Under the condition of a laboratory, vibration signals of an iron core and a winding of an experimental transformer are collected through a no-load experiment and a short-circuit experiment, when the transformer is in no-load, the power voltage is rated voltage, the winding current is no-load current, the no-load current is very small, the winding vibration can be ignored, and the vibration of the iron core is a main vibration source; when the transformer is subjected to a short-circuit test, the applied power voltage is low, the vibration of an iron core can be ignored, the winding current is rated current, and the winding vibration is a main vibration source; therefore, the vibration signals of the winding and the iron core are obtained through the no-load and short-circuit experiments of the transformer.
The spectrum distribution of the measuring points before and after the occurrence of the loosening fault of the iron core is shown in fig. 4, wherein the solid line is the vibration spectrum distribution of a certain collecting point before the loosening fault, and the dotted line is the vibration spectrum distribution of the collecting point after the fault.
It can be known from fig. 4 that the vibration signal change laws of the corresponding top and front middle collecting points of the three phases of the transformer are different, the frequency spectrum characteristics of the vibration signal of the collecting point at the top are kept consistent before and after the iron core is loosened, and the frequency spectrum distribution law of the vibration signal of the collecting point at the middle part of the side surface of the transformer box is greatly changed, because the transmission path of the vibration signal of the iron core collected at the middle part of the side surface of the transformer box is single relative to the transmission medium at the top of the box, the path is shorter, and the transmission and the characteristic retention of the signal are facilitated.
FIG. 4(a) is a graph showing the variation of the vibration spectrum at point A1;
FIG. 4(b) is a graph showing the variation of the vibration spectrum at point A4;
FIG. 4(c) is a graph showing the variation of the vibration spectrum at point B2;
FIG. 4(d) is a graph showing the variation of the vibration spectrum at point B5;
FIG. 4(e) is a graph showing the variation of the vibration spectrum at point C3;
fig. 4(f) is a graph of the variation of the vibration spectrum at point C6.
From FIG. 4(b), (d), (f)The distribution rule of the dynamic signal frequency spectrum is known, and the iron core vibration signals are mainly distributed at the fundamental frequency of 100Hz and the frequency multiplication part thereof under the normal working state; after the iron core loosens, the amplitude of each frequency point is increased, but the increase at the 500Hz position is particularly obvious; wherein the vibration acceleration amplitude at 500Hz of FIG. 4(b) is 4.89X 10 from before the failure-3g sudden increase to 7.96 × 10 after failure-3g, vibration acceleration amplitude at 500Hz of FIG. 4(d) from 3.78X 10 before failure-3g sudden increase to 10.21 × 10 after failure-3g, FIG. 4(f) vibration acceleration amplitude at 500Hz from 7.26X 10 before failure-3g burst to 14.35 × 10 after failure-3g, when the iron core has a loosening fault, the pressing force of the silicon steel sheets of the iron core is reduced, so that the inherent vibration frequency of the iron core is changed; the second-order natural vibration frequency in the normal state is reduced to be near 500Hz along with the loosening fault, and the second-order natural vibration frequency and the vibration frequency of the iron core at 500Hz are in resonance when the transformer runs, so that the amplitude of a signal component at the frequency point is increased suddenly.
The test data is taken as an example to verify the proposed detection method:
calculating waveform characteristic quantities of the transformer vibration signals in a normal state and an iron core loosening fault state to obtain an attached table 1 (the waveform characteristic quantities of the transformer vibration signals before and after the fault);
as can be seen from table 1, the waveform characteristic quantities of the vibration signal after the fault all change to different degrees, which indicates that the above characteristic quantities can be used to detect the mechanical state of the transformer, and the above characteristic quantities are combined into the waveform characteristic vector vb。
Decomposing the transformer vibration signals in a normal state and an iron core loosening fault state by using the EEMD to obtain energy characteristic quantities (the transformer vibration signal energy characteristic quantities before and after the fault) as shown in an attached table 2;
as can be seen from the attached Table 2, the energy characteristic quantities of each IMF component of the vibration signal after the fault are changed in different degrees, which shows that the above characteristic quantities can be used for detecting the mechanical fault of the transformer, and the above characteristic quantities form an energy characteristic vector vn。
The identification of the transformer vibration signal by using the single eigenvector has limitation; extracting characteristic quantities of the acquired 4 types of transformer vibration signals, and performing normalization processing as shown in an attached table 3 (characteristic quantities of four types of transformer vibration signals);
as shown in the attached table 3, the peak-to-peak value, the principal component frequency, the fundamental frequency ratio, and the 50Hz odd harmonic ratio are taken as examples for explanation, each characteristic quantity changes with the state change of the transformer, and comparing the value of each characteristic quantity in each state, it can be known that different characteristic quantities do not have a distinct distinction degree for each state, for example: for a normal winding vibration signal, a fault winding vibration signal and a normal iron core vibration signal of the transformer, the main component frequency is 100Hz, and the main component frequency of the iron core fault vibration signal is 200Hz, which shows that the main component frequency can identify the iron core fault vibration signal, and the identification effect for other three types of vibration signals is poor; the difference of the fundamental frequency ratio values of the normal winding vibration signal, the fault winding vibration signal and the normal iron core vibration signal is obvious, but the fundamental frequency ratio of the fault iron core vibration signal is closer to the fundamental frequency ratio of the normal iron core vibration signal, which shows that the characteristic quantity has good recognition effect on the first three types of signals, but false judgment is easily caused when the normal iron core vibration signal and the fault iron core vibration signal are recognized.
Vibration signal waveform feature vector vbAnd IMF component energy feature vector vnThe mechanical state of the transformer is reflected from the integral angle and the local angle respectively, and the integral angle and the local angle form a composite characteristic vector VF。
VF＝[vb,vn]
Aiming at a composite eigenvector comprehensively reflecting waveform characteristics and energy characteristics of the vibration signals of the transformer, a principal component analysis method is used for solving a covariance matrix of the composite eigenvector and extracting principal components of the composite eigenvector, and an obtained covariance matrix characteristic root and a contribution rate of the covariance matrix characteristic root are shown in an attached table 4 (the covariance matrix characteristic root and the contribution rate of the covariance matrix characteristic root);
8 characteristic roots and related parameters thereof are listed in an attached table 4, and the other characteristic roots are ignored because the contribution rate is too small and the expressed characteristic information is limited.
As can be seen from table 4, the number of the characteristic roots corresponding to the first two characteristic factors is large, the cumulative contribution rate of the two characteristic factors is greater than 90%, the most characteristic information of the original characteristic vector can be expressed, and the contribution rates of the other characteristic factors are small and can be ignored; therefore, the two characteristic values are used as principal component characteristic vectors to replace the original characteristic vectors for analysis, and the dimensionality of the characteristic vectors is reduced.
Obtaining principal component characteristic vectors of vibration signals of the transformer winding and the iron core under normal state and fault state according to the method, such as table 5 (principal components of vibration signals of different types of transformers);
as can be seen from table 5, the 14-dimensional composite feature vector optimized by the principal component analysis method is processed into a 2-dimensional principal component feature vector, which is beneficial to improving the working efficiency of the classifier.
The method comprises the steps of collecting 4 types of vibration signals of a normal-state iron core vibration signal, a fault-state iron core vibration signal, a normal-state winding vibration signal and a fault-state winding vibration signal of a transformer, wherein each type of vibration signal comprises 350 groups of data, and 1400 groups of data are obtained, each group of data is a composite feature vector comprising 14 feature quantities, optimizing the composite feature vector by a principal component analysis method to obtain 2-dimensional principal component feature vectors, respectively selecting 1200 groups of data before and after optimization of the principal component analysis method as training samples, and inputting 200 groups of data as test samples into a DBN network for classification and recognition, wherein the result is shown in figure 4.
As shown in fig. 5(a), before feature quantity optimization, 3 of 50 normal winding vibration signals are identified as faulty winding vibration signals, 1 of the 50 normal winding vibration signals is identified as normal core vibration signals, 1 of the 50 normal winding vibration signals is identified as faulty core vibration signals, and certain identification errors occur in the remaining three vibration signals, so that the identification accuracy of the DBN network on the basis of non-optimized feature quantity data is 91.0% as a whole; using a principal component analysis method to classify the DBN after feature quantity optimization, as shown in fig. 5(b), after feature quantity optimization, 1 group of the 50 groups of normal winding vibration signals is identified as fault winding vibration, 1 group is identified as normal core vibration, 1 group of the 50 groups of normal core vibration signals is identified as fault winding vibration signal, and the identification precision of the DBN network is 98.5%; the principal component characteristic vector can well retain the information of the original characteristic vector, the workload of the classifier is reduced, the identification accuracy of the classifier is improved, and therefore the working efficiency of detecting the mechanical fault of the transformer is improved.
The above embodiments are merely illustrative of the principles of the present invention and its effects, and do not limit the present invention. It will be apparent to those skilled in the art that modifications and improvements can be made to the above-described embodiments without departing from the spirit and scope of the invention. Accordingly, it is intended that all equivalent modifications or changes be made by those skilled in the art without departing from the spirit and technical spirit of the present invention, and be covered by the claims of the present invention.
Claims (4)
1. A transformer mechanical fault detection method based on vibration signal composite eigenvectors is characterized by comprising the following steps:
step 1: a vibration signal acquisition system is formed by a PCB356A16 piezoelectric acceleration sensor and an NI-9234 data acquisition instrument, the acceleration sensor is attached to a transformer box body, and a vibration signal of the transformer is acquired and transmitted to the data acquisition instrument;
step 2: extracting the waveform characteristic quantities of the average value, skewness, peak-to-peak value, kurtosis, fundamental frequency ratio, 50Hz odd-order frequency multiplication ratio and seven types of vibration signals from the vibration signals of the transformer;
and step 3: EEMD decomposition is carried out on the transformer vibration signal to obtain IMF components of the vibration signal, and the first seven groups of IMF components are extracted to serve as vibration signal energy characteristic quantities;
and 4, step 4: combining the seven types of waveform characteristic quantities and the seven types of energy characteristic quantities of the transformer into a fourteen-dimensional composite characteristic vector;
and 5: reducing the dimension of the fourteen-dimensional composite feature vector into a two-dimensional principal component feature vector by using a principal component analysis method;
step 6: extracting a plurality of groups of vibration signal principal component characteristic vectors under a known state as training samples of the DBN network, and obtaining the corresponding relation between the characteristic vectors and the mechanical state of the transformer by the training samples through two stages of training of the DBN network;
and 7: and inputting the 2-dimensional principal component feature vector of the target transformer into the trained DBN network to obtain the mechanical state of the transformer.
2. The method for detecting the mechanical fault of the transformer based on the vibration signal composite eigenvector as claimed in claim 1, wherein six acquisition points of the vibration signal of the transformer are provided, which are respectively: a1, B2, C3, a4, B5 and C6.
3. The method for detecting the mechanical fault of the transformer based on the vibration signal composite eigenvector as claimed in claim 2, wherein the acquisition points A1, B2 and C3 of the vibration signal of the transformer are uniformly arranged on the top of the transformer box and respectively correspond to the positions of A, B, C three phases.
4. The method for detecting the mechanical fault of the transformer based on the vibration signal composite eigenvector as claimed in claim 2, wherein the acquisition points A4, B5 and C6 of the vibration signal of the transformer are uniformly arranged on the side surface of the transformer box body and respectively correspond to the positions of A, B, C three phases.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110496964.XA CN113191437A (en) | 2021-05-07 | 2021-05-07 | Transformer mechanical fault detection method based on vibration signal composite eigenvector |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110496964.XA CN113191437A (en) | 2021-05-07 | 2021-05-07 | Transformer mechanical fault detection method based on vibration signal composite eigenvector |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113191437A true CN113191437A (en) | 2021-07-30 |
Family
ID=76984430
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110496964.XA Pending CN113191437A (en) | 2021-05-07 | 2021-05-07 | Transformer mechanical fault detection method based on vibration signal composite eigenvector |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113191437A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103858343A (en) * | 2011-09-30 | 2014-06-11 | 高通Mems科技公司 | Cross-sectional dilation mode resonators and resonator-based ladder filters |
| CN106646096A (en) * | 2016-11-15 | 2017-05-10 | 国网四川省电力公司广安供电公司 | Transformer fault classification and identification method based on vibration analysis method |
| CN109374270A (en) * | 2018-09-19 | 2019-02-22 | 国网甘肃省电力公司电力科学研究院 | A GIS abnormal vibration analysis and mechanical fault diagnosis device and method |
| CN109374323A (en) * | 2018-09-29 | 2019-02-22 | 国网山西省电力公司阳泉供电公司 | Transformer mechanical fault detection method based on vibration signal index energy |
| CN109753951A (en) * | 2019-02-26 | 2019-05-14 | 河海大学 | An OLTC Fault Diagnosis Method Based on Instantaneous Energy Entropy and SVM |
| CN110082082A (en) * | 2019-04-28 | 2019-08-02 | 国网四川省电力公司南充供电公司 | A kind of GIS state identification method based on vibration signal Principal Component Analysis |
-
2021
- 2021-05-07 CN CN202110496964.XA patent/CN113191437A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103858343A (en) * | 2011-09-30 | 2014-06-11 | 高通Mems科技公司 | Cross-sectional dilation mode resonators and resonator-based ladder filters |
| CN106646096A (en) * | 2016-11-15 | 2017-05-10 | 国网四川省电力公司广安供电公司 | Transformer fault classification and identification method based on vibration analysis method |
| CN109374270A (en) * | 2018-09-19 | 2019-02-22 | 国网甘肃省电力公司电力科学研究院 | A GIS abnormal vibration analysis and mechanical fault diagnosis device and method |
| CN109374323A (en) * | 2018-09-29 | 2019-02-22 | 国网山西省电力公司阳泉供电公司 | Transformer mechanical fault detection method based on vibration signal index energy |
| CN109753951A (en) * | 2019-02-26 | 2019-05-14 | 河海大学 | An OLTC Fault Diagnosis Method Based on Instantaneous Energy Entropy and SVM |
| CN110082082A (en) * | 2019-04-28 | 2019-08-02 | 国网四川省电力公司南充供电公司 | A kind of GIS state identification method based on vibration signal Principal Component Analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112257530B (en) | Rolling bearing fault diagnosis method based on blind signal separation and support vector machine | |
| CN113159077A (en) | Three-phase inverter fault identification method based on hybrid convolutional neural network | |
| CN106841949B (en) | Method and device for monitoring stator insulation of three-phase asynchronous alternating current motor on line | |
| Ngaopitakkul et al. | Internal fault classification in transformer windings using combination of discrete wavelet transforms and back-propagation neural networks | |
| CN105547465A (en) | Transformer vibration signal winding state feature extraction method | |
| CN106932184A (en) | A kind of Diagnosis Method of Transformer Faults based on improvement hierarchical clustering | |
| CN102279358A (en) | MCSKPCA based neural network fault diagnosis method for analog circuits | |
| CN110082082B (en) | GIS state identification method based on vibration signal principal component analysis method | |
| CN110470465A (en) | A kind of breaker test method and system based on analysis of vibration signal | |
| CN109655266A (en) | A kind of Wind turbines Method for Bearing Fault Diagnosis based on AVMD and spectral coherence analysis | |
| CN110796120A (en) | Time domain feature-based circuit breaker mechanical fault XGboost diagnosis method | |
| CN112541524A (en) | BP-Adaboost multi-source information motor fault diagnosis method based on attention mechanism improvement | |
| CN117171544A (en) | Motor vibration fault diagnosis method based on multichannel fusion convolutional neural network | |
| CN111879514A (en) | Brushless direct current motor bearing fault diagnosis method based on ELM model | |
| Li et al. | Intelligent fault diagnosis of aeroengine sensors using improved pattern gradient spectrum entropy | |
| CN115273895A (en) | A Transformer Operating Status Monitoring Method and System Based on Wavelet Time-Frequency Entropy and RUSboost | |
| CN116956215A (en) | Fault diagnosis method and system for transmission system | |
| CN113009337B (en) | Interturn short circuit fault identification method for offshore wind power permanent magnet direct drive motor | |
| Xu et al. | CapsFormer: a novel bearing intelligent fault diagnosis framework with negligible speed change under small-sample conditions | |
| CN116089857A (en) | Transformer fault identification method based on CEEMDAN-DBN | |
| Cao et al. | A novel method for detection of wind turbine blade imbalance based on multi-variable spectrum imaging and convolutional neural network | |
| CN112986821B (en) | Fault diagnosis method for broken blade of variable pitch motor rotor of offshore wind turbine generator | |
| CN113191437A (en) | Transformer mechanical fault detection method based on vibration signal composite eigenvector | |
| CN109655711A (en) | Power distribution network internal overvoltage kind identification method | |
| CN117741347A (en) | A power quality disturbance detection method for distribution network containing distributed power sources |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210730 |
|
| RJ01 | Rejection of invention patent application after publication |