CN116819251A - High-voltage switch cabinet fault diagnosis method based on photoelectric monitoring - Google Patents
High-voltage switch cabinet fault diagnosis method based on photoelectric monitoring Download PDFInfo
- Publication number
- CN116819251A CN116819251A CN202310773708.XA CN202310773708A CN116819251A CN 116819251 A CN116819251 A CN 116819251A CN 202310773708 A CN202310773708 A CN 202310773708A CN 116819251 A CN116819251 A CN 116819251A
- Authority
- CN
- China
- Prior art keywords
- fault
- switch cabinet
- voltage switch
- discharge
- ultraviolet
- 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
- 238000003745 diagnosis Methods 0.000 title claims abstract description 32
- 238000012544 monitoring process Methods 0.000 title claims abstract description 32
- 238000000034 method Methods 0.000 title claims abstract description 27
- 238000013528 artificial neural network Methods 0.000 claims abstract description 12
- 238000004458 analytical method Methods 0.000 claims abstract description 8
- 230000009467 reduction Effects 0.000 claims abstract description 8
- 238000012549 training Methods 0.000 claims abstract description 7
- 238000000605 extraction Methods 0.000 claims abstract description 6
- 238000004806 packaging method and process Methods 0.000 claims description 20
- 238000001514 detection method Methods 0.000 claims description 17
- 238000012545 processing Methods 0.000 claims description 16
- 238000012806 monitoring device Methods 0.000 claims description 15
- 238000011156 evaluation Methods 0.000 claims description 11
- 238000001914 filtration Methods 0.000 claims description 6
- 238000005192 partition Methods 0.000 claims description 6
- 229910000831 Steel Inorganic materials 0.000 claims description 3
- 238000000354 decomposition reaction Methods 0.000 claims description 3
- 238000009413 insulation Methods 0.000 claims description 3
- 239000000463 material Substances 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 230000008054 signal transmission Effects 0.000 claims description 3
- 239000010959 steel Substances 0.000 claims description 3
- 238000012360 testing method Methods 0.000 description 5
- 239000012212 insulator Substances 0.000 description 3
- 239000000523 sample Substances 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 230000010349 pulsation Effects 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000008602 contraction Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 239000003292 glue Substances 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 230000002401 inhibitory effect Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000000825 ultraviolet detection Methods 0.000 description 1
Landscapes
- Testing Relating To Insulation (AREA)
Abstract
The invention relates to a high-voltage switch cabinet fault diagnosis method based on photoelectric monitoring, which comprises the following steps: collecting real-time and historical data of the running state of devices in the high-voltage switch cabinet; performing characteristic analysis, extraction, dimension reduction and standardization treatment on the acquired data, and accurately extracting ultraviolet discharge and infrared temperature characteristics; respectively constructing a self-adaptive fuzzy neural network aiming at the fault type and the fault grade, carrying out network parallel connection, taking ultraviolet discharge and infrared temperature characteristics as common input, setting different input weights, obtaining a network membership function and a fuzzy diagnosis rule through training the historical state characteristics, and respectively outputting a fault type diagnosis result and a fault grade assessment result; based on the parallel network structure, the ultraviolet discharge and infrared temperature characteristics acquired and processed in real time are input, and the fault type and grade assessment of the real-time monitoring information are output. Compared with the prior art, the invention has the advantages of accurate extraction of the temperature rise and partial discharge characteristics, accurate fault result classification and the like.
Description
Technical Field
The invention relates to the field of fault classification and evaluation of high-voltage switch cabinets, in particular to a high-voltage switch cabinet fault diagnosis method based on photoelectric monitoring.
Background
The high-voltage switch cabinet runs in the environment of high voltage, high current and strong magnetic field for a long time, and the phenomena of partial discharge and abnormal temperature rise in the cabinet are caused by insulation faults. The fault of the switch cabinet can be monitored and diagnosed in various modes such as infrared temperature, ultraviolet discharge pulse and the like. For detection arrangement and fault diagnosis of the running state of the switch cabinet, the simple arrangement of the acquisition system on the monitoring point positions, comprehensive analysis and fusion of multiple fault characteristics are striven for, so that the accuracy of fault assessment is improved, and the method has important practical significance for on-line monitoring of the high-voltage switch cabinet.
For the temperature and discharge data of the switch cabinet during operation, the temperature and discharge data are mostly collected by manual handheld equipment, so that the working intensity is high, and the monitoring efficiency is low; meanwhile, the problem of the internal space of the switch cabinet and the problem of strong electromagnetic environment during operation are considered, the independent hardware parts for data acquisition are fused on the premise of reducing the integrated size, and the hardware packaging system is designed, so that the acquisition system can be simply and effectively arranged on the detection point positions, and the accuracy of data acquisition is greatly influenced.
Aiming at the characteristics of large data volume of infrared and ultraviolet detection signals and harmonic interference, it is important to perform characteristic analysis and extraction on the sensing data. Because the acquisition card has high frequency and EMI interference, for the waveforms of high frequency, small pulse width and large amplitude range of discharge pulse, the time domain filtering method can not completely eliminate clutter and can cause the change of pulse amplitude, pulse width and phase characteristics; in the frequency domain method, the main frequency characteristic of discharge can not be extracted well for the mixed noise frequency, so that the fault characteristic is difficult to be highlighted.
At present, most of modes for on-line monitoring and fault diagnosis adopt single state characteristics, and because the state of a switch cabinet is not in one-to-one correspondence with fault characteristic variables, the fusion and comprehensive analysis of multiple types of fault characteristics are lacking, so that misjudgment and missed detection are easy to cause, and the accuracy of fault diagnosis is difficult to ensure. The method of self-adaptive training and classifying the multi-source signal characteristics by utilizing a single neural network ensures that each detection characteristic is constrained by irrelevant fuzzy rules, thereby increasing the redundancy of the network membership function and influencing the accuracy of network output.
Disclosure of Invention
The invention aims to provide a high-voltage switch cabinet fault diagnosis method based on photoelectric monitoring, which solves the problem of arrangement of a data acquisition system in a switch cabinet and realizes accurate classification and evaluation of operation faults of the switch cabinet.
The aim of the invention can be achieved by the following technical scheme:
a high-voltage switch cabinet fault diagnosis method based on photoelectric monitoring comprises the following steps:
s1, acquiring real-time and historical data of the running state of internal devices of a high-voltage switch cabinet by using a running state monitoring device of the high-voltage switch cabinet, wherein the running state monitoring device of the high-voltage switch cabinet comprises: the device comprises an acquisition card, an infrared sensor, an ultraviolet sensor, a power supply, a driving circuit and a filter circuit;
s2, carrying out feature analysis, extraction and dimension reduction and standardization processing on the data acquired in the S1, processing an original ultraviolet pulse based on a peak-valley detection positioning algorithm, processing an infrared temperature based on a wavelet denoising and sliding window filtering algorithm, and accurately extracting ultraviolet discharge and infrared temperature features;
s3, respectively constructing a self-adaptive fuzzy neural network aiming at the fault type and the fault grade, connecting the two networks in the parallel network structure in parallel by taking ultraviolet discharge and infrared temperature characteristics as common input, setting different weights of the two network input characteristics, obtaining a network membership function and a fuzzy diagnosis rule through training the historical state characteristics, and respectively outputting a fault type diagnosis result and a fault grade evaluation result;
s4, based on the parallel network structure of S3, inputting the ultraviolet discharge and infrared temperature characteristics acquired and processed in real time, judging through fuzzy rules and calculating membership, and outputting fault type and fault grade assessment results of real-time monitoring information.
Further, the high-voltage switch cabinet running state monitoring device is designed for minimizing independent compartments according to the size of each part, an integrated packaging structure is formed, and a shell of the integrated packaging structure is subjected to electromagnetic shielding by adopting steel materials.
Further, an integrated packaging structure of the high-voltage switch cabinet running state monitoring device is provided with a chute which is convenient for installing and detaching the sensor partition board inside the shell, and a collecting card signal transmission hole which is convenient for a signal line or a wireless module to access is arranged at the back of the shell.
Further, an integrated packaging structure of the high-voltage switch cabinet running state monitoring device is provided with an insulating fixing part at the bottom of the packaging structure, and the insulating fixing part is used for adsorbing and fixing the detection point and insulating the detection placement part.
Further, in the step S2, on-line data processing of the collected history and real-time data, such as infrared temperature waveforms and ultraviolet discharge waveforms, is performed to extract the characteristics of temperature average values, jump values, discharge pulse widths, frequencies and amplitudes, so as to realize dimension reduction from mass operation data to state characteristics.
Further, the processing of the original ultraviolet pulse based on the peak-to-valley detection positioning algorithm specifically comprises the following steps:
for the original ultraviolet pulse waveform, the peak of the discharge pulse and all the wave troughs containing interference signals are obtained by utilizing a peak searching algorithm, a plurality of wave troughs between adjacent discharge pulse peaks are processed by a peak-to-valley matching algorithm, so that the matched wave troughs shrink towards the average value of the interval wave troughs, and the interval positions of the matched wave troughs are obtained:
wherein val is a trough containing harmonic interference, peak is a precise discharge pulse crest, and a discharge pulse peak Gu Fuzhi and position information are obtained to obtain discharge pulse width, frequency and amplitude characteristics.
Further, the processing of the infrared temperature based on the wavelet denoising and sliding window filtering algorithm specifically comprises the following steps:
for the original infrared signal containing noise, adopting wavelet soft threshold to remove noise and filter high-frequency interference, for mother wavelet phi p (x) Performing wavelet continuous transformation, extracting and inhibiting high-frequency detail coefficients, setting the point to zero when the absolute value |omega| of the signal is smaller than a threshold lambda, otherwise, performing differential zero contraction on the absolute value |omega| and lambda:
wherein, |omega| is an absolute value, lambda is a threshold value, MAD is the median of the absolute value of the first-layer wavelet decomposition coefficient, and N is the signal length;
after the denoising is realized based on the wavelet denoising reconstruction signal, the outlier of the signal is removed by utilizing a sliding window median algorithm, so that the pulsation of the waveform is further reduced, and an accurate temperature average value and a jump value are obtained.
Further, the self-adaptive fuzzy neural network adopts a five-layer structure, wherein an input layer acquires node information; membership layer obtains fuzzy linguistic variables of input informationAnd calculate the membership degree For inputting x i J linguistic variables of (2); the rule layer puts forward fuzzy rules and calculates the fitness of each rule; the decision layer normalizes the fitness to reduce errors; the output layer obtains a network diagnosis result:
where y is the network output and represents the sum of the weighted averages of each rule, y j For each rule output, a j For the fitness of each rule,is a as j Is used for the normalization parameters of (a).
Further, the fault type diagnosis result comprises three classifications of no fault, temperature rise fault and partial discharge fault, and the fault grade evaluation result comprises three grades of good (I), general (II) and serious (III).
Further, the fault type and grade evaluation of the output real-time monitoring information are visually displayed through a three-dimensional view, specifically, three parameters including the fault grade, the height of the three-dimensional column and the color of the surface of the column are associated, an x-axis is set as the fault type, a y-axis is the fault grade, and a z-axis is the state index, so that the three-dimensional evaluation result of the state of the switch cabinet is displayed.
Compared with the prior art, the invention has the following beneficial effects:
(1) The invention integrates the hardware parts such as the sensors with independent data acquisition, designs a miniaturized packaging system, reduces electromagnetic interference between the system and the switch cabinet, and is simpler and more convenient to arrange in the monitoring point position in the switch cabinet.
(2) The invention provides the method for analyzing and extracting the characteristics of the infrared temperature and the ultraviolet pulse data under the running state of the switch cabinet, which is favorable for reducing the redundancy of the data and highlighting the fault characteristics and improves the accuracy of the later diagnosis.
(3) According to the invention, two self-adaptive fuzzy neural networks are connected in parallel, the input weights with different impact ratios are set by using the same characteristic information, and the judgment of the temperature rise and the partial discharge faults of the switch cabinet and the classification of fault levels are comprehensively and accurately carried out.
Drawings
Fig. 1 is a view of a high-voltage switch cabinet operation state monitoring device and a hardware package, wherein (a) is a view of the hardware package, and (b) is a schematic view of the device structure;
FIG. 2 is a flow chart of a fault diagnosis method of the high-voltage switch cabinet of the invention;
FIG. 3 is a parallel adaptive fuzzy neural network structure based on weight input;
FIG. 4 is a diagram showing an example three-dimensional display of a switch cabinet fault diagnosis result in one embodiment;
the reference numerals in the drawings are: i is a packaging shell, II is a sensor partition board, III is a shell front cover plate, and IV is a shell rear cover plate; the device comprises a device main switch 1, an infrared and ultraviolet sensor probe and a discrete switch 2, a sensor partition board mounting chute 3, a power module compartment 4, a voltage conversion module compartment 5, an infrared sensor driving circuit compartment 6, an ultraviolet sensor driving circuit compartment 7, a filter circuit compartment 8, a boosting module compartment 9 and an acquisition card signal line or wireless module access port 10.
Detailed Description
The invention will now be described in detail with reference to the drawings and specific examples. The present embodiment is implemented on the premise of the technical scheme of the present invention, and a detailed implementation manner and a specific operation process are given, but the protection scope of the present invention is not limited to the following examples.
The embodiment first provides a high tension switchgear running state monitoring device, includes: the system comprises an acquisition card, an infrared sensor, an ultraviolet phototube, an internal UPS power supply, a driving circuit and a filter circuit, wherein each dimension parameter is measured for each part, and a three-dimensional modeling software is utilized for minimizing independent compartment design to form an integrated packaging structure.
In one embodiment, as shown in fig. 1, the packaging part comprises a packaging shell I, a sensor baffle ii, a shell front cover plate iii and a shell rear cover plate iv; the device comprises a device main switch 1, an infrared and ultraviolet sensor probe and discrete switch 2, a sensor partition board mounting chute 3, a power module compartment 4, a voltage conversion module compartment 5, an infrared sensor driving circuit compartment 6, an ultraviolet sensor driving circuit compartment 7, a filter circuit compartment 8, a boosting module compartment 9 and an acquisition card signal line or wireless module access port 10.
The shell of the integrated packaging structure adopts steel materials to carry out electromagnetic shielding of the monitoring device and the switch cabinet.
The integrated packaging structure is provided with a chute 3 which is convenient for the installation and disassembly of the sensor partition board inside the shell, a collecting card signal transmission hole 10 which is convenient for the access of a signal wire or a wireless module is arranged at the back of the shell, and an insulating fixing part which is used for adsorbing, fixing a detection point position and realizing insulation with the detection placement position is arranged at the bottom of the package. In this embodiment, the insulating fixing member is a glue, and is disposed at four corners of the bottom of the package.
The monitoring device packages all discrete components, the collection and transmission of infrared and ultraviolet signals of devices in the switch cabinet can be independently realized without an external power supply, when the packaging system is adsorbed on a detection point, the main switch is responsible for the on-off of an internal voltage rising and dropping circuit and a driving circuit, the discrete switches respectively control the use of infrared and ultraviolet sensor probes, and the PC end receives and stores state sensing data of the collection packaging system through a wireless serial port or a USB line.
The embodiment also provides a fault diagnosis method of the high-voltage switch cabinet based on photoelectric monitoring, which is shown in fig. 2 and comprises the following steps:
s1, acquiring real-time and historical data of the running state of the internal device of the high-voltage switch cabinet by using the running state monitoring device of the high-voltage switch cabinet.
According to the insulator equipment in the high-voltage switch cabinet, after the power supply, the sensor and the modules are connected, the acquisition card is set to acquire the temperature and the discharge information of the insulator on-line monitoring with the frequency of 25kHZ, and the monitoring information is transmitted to the computer end through the acquisition card signal wire or the wireless module, so that the real-time data monitoring and the historical information storage are realized.
S2, carrying out feature analysis, extraction and dimension reduction and standardization processing on the data acquired in the S1, processing an original ultraviolet pulse based on a peak-valley detection positioning algorithm, processing an infrared temperature based on a wavelet denoising and sliding window filtering algorithm, and accurately extracting ultraviolet discharge and infrared temperature features.
And extracting the characteristics of the infrared temperature waveform and the ultraviolet discharge waveform from the historical and real-time data acquired by the acquisition system to extract the characteristics of a temperature average value, a mutation value, a discharge pulse width, a discharge pulse frequency, a discharge pulse amplitude and the like, thereby realizing the dimension reduction from mass monitoring operation data to a state characteristic diagnosis network.
For the original infrared detection waveform containing noise, a wavelet soft threshold denoising method is adopted to carry out the primary wavelet phi p (x) Performing wavelet continuous transformation, decomposing an original signal at different resolution levels to obtain a smooth signal and a detail signal, extracting and suppressing a high-frequency detail coefficient, setting the point to zero when the absolute value |omega| of the signal is smaller than a threshold lambda, otherwise, performing differential shrinkage towards zero between the absolute value |omega| and lambda:
wherein, MAD is the median of the absolute value of the first-layer wavelet decomposition coefficient, the embodiment takes Gaussian noise standard deviation adjustment coefficient equal to 0.6745, and N is the signal length. After the reconstruction signal realizes noise reduction, a sliding window is set, and the median in the window replaces the sliding window median algorithm of the current data to remove outliers, so that the pulsation of waveforms is further reduced, and the accurate temperature average value and jump value characteristics are obtained.
For the original ultraviolet pulse waveform, the peak of the discharge pulse and all the wave troughs containing interference signals are obtained by utilizing a peak searching algorithm, a plurality of wave troughs between adjacent discharge pulse peaks are processed by a peak-to-valley matching algorithm, so that the matched wave troughs shrink towards the average value of the interval wave troughs, and the interval positions of the matched wave troughs are obtained:
wherein val is a trough containing harmonic interference, peak is a precise discharge pulse crest, and the peak Gu Fuzhi and position information of the discharge pulse are obtained, so that the characteristics of discharge pulse width, frequency, amplitude and the like are obtained.
S3, respectively constructing a self-adaptive fuzzy neural network aiming at the fault type and the fault grade, connecting the self-adaptive fuzzy neural network in parallel, taking ultraviolet discharge and infrared temperature characteristics as common input by two networks in a parallel network structure, setting different weights of the input characteristics of the two networks, obtaining a network membership function and a fuzzy diagnosis rule through training the historical state characteristics, and respectively outputting a fault type diagnosis result and a fault grade evaluation result.
The adaptive fuzzy neural network adopts a five-layer structure, as shown in fig. 3, an input layer acquires an input feature vector x= [ x ] 1 ,x 2 ,L,x n ] T The method comprises the steps of carrying out a first treatment on the surface of the Membership layer obtains fuzzy linguistic variables of input information For inputting x i Is defined as +.>The rule layer sets the fuzzy rule if x 1 Is->Is->Is->And calculates the corresponding fitness of each rule>The decision layer normalizes the fitness to reduce errors; the output layer performs a weighted average of the sum of each rule:
wherein y is network output, y j For each rule to be output in correspondence with,is a as j Is used for the normalization parameters of (a). The structure forms nonlinear mapping of input and output variables, realizes modeling of input-output data pairs in iteration by using a back propagation algorithm, calculates an optimal weight of a membership function in a network by a local approximation mode, and realizes network construction.
The parallel self-adaptive fuzzy neural network improves the diagnosis precision, and in the temperature rise and discharge test of the pollution insulator, the influence of input characteristics on the output result is related by analyzing test data and test phenomena. For the temperature rise fault, the temperature is obviously increased and simultaneously discharge information is accompanied; for partial discharge faults, the discharge signals are obvious and the temperature is only reflected as small vibration, so that the input weights of the infrared and ultraviolet characteristics in fault type and grade evaluation networks are respectively set as Kt by using experimental empirical values 1 =0.45,Kp 1 =0.55、Kt 2 =0.2,Kp 2 =0.8。
S4, based on the parallel network structure of S3, inputting the ultraviolet discharge and infrared temperature characteristics acquired and processed in real time, judging through fuzzy rules and calculating membership, and outputting fault type and fault grade assessment results of real-time monitoring information.
And (3) combining the temperature rise and discharge physical appearance of the test object, extracting infrared and ultraviolet signal characteristics through a data processing method in S2, defining data of a plurality of groups of input characteristic and output index nonlinear mapping relations of output fault type indexes alpha=0, 1,2 and fault grade ranges 0 < beta < 1, and training an optimal weight value of a membership function in the simulation network in S3 by using the manual setting as a training set and a testing set to obtain a complete parallel network for fault type division and grade assessment of the infrared and ultraviolet signals.
Where α=0, 1,2 is defined as three types of failure, abnormal temperature rise, partial discharge, β e (0,0.4), β e (0.4,0.7), β e (0.7,1) is defined as three failure levels of severe (III), general (II), good (I).
In one embodiment, for outputting the fault diagnosis result, the fault grade index may be related to the three-dimensional column height parameter and the column surface color parameter, and assuming that each output state index β is inversely related to the three-dimensional column height parameter, so that the fault severity is in direct proportion to the column height parameter, and the column height parameter is mapped by using the RGB color gradient function to perform column filling and color smoothing processing, setting the x axis as the fault type, the y axis as the fault grade, and the z axis as the state index, and obtaining a three-dimensional gradient histogram to represent the state index, so as to draw and display the three-dimensional evaluation result of the switch cabinet state, as shown in fig. 4.
The foregoing describes in detail preferred embodiments of the present invention. It should be understood that numerous modifications and variations can be made in accordance with the concepts of the invention by one of ordinary skill in the art without undue burden. Therefore, all technical solutions which can be obtained by logic analysis, reasoning or limited experiments based on the prior art by a person skilled in the art according to the inventive concept shall be within the scope of protection defined by the claims.
Claims (10)
1. The high-voltage switch cabinet fault diagnosis method based on photoelectric monitoring is characterized by comprising the following steps of:
s1, acquiring real-time and historical data of the running state of internal devices of a high-voltage switch cabinet by using a running state monitoring device of the high-voltage switch cabinet, wherein the running state monitoring device of the high-voltage switch cabinet comprises: the device comprises an acquisition card, an infrared sensor, an ultraviolet sensor, a power supply, a driving circuit and a filter circuit;
s2, carrying out feature analysis, extraction and dimension reduction and standardization processing on the data acquired in the S1, processing an original ultraviolet pulse based on a peak-valley detection positioning algorithm, processing an infrared temperature based on a wavelet denoising and sliding window filtering algorithm, and accurately extracting ultraviolet discharge and infrared temperature features;
s3, respectively constructing a self-adaptive fuzzy neural network aiming at the fault type and the fault grade, connecting the two networks in the parallel network structure in parallel by taking ultraviolet discharge and infrared temperature characteristics as common input, setting different weights of the two network input characteristics, obtaining a network membership function and a fuzzy diagnosis rule through training the historical state characteristics, and respectively outputting a fault type diagnosis result and a fault grade evaluation result;
s4, based on the parallel network structure of S3, inputting the ultraviolet discharge and infrared temperature characteristics acquired and processed in real time, judging through fuzzy rules and calculating membership, and outputting fault type and fault grade assessment results of real-time monitoring information.
2. The high-voltage switch cabinet fault diagnosis method based on photoelectric monitoring according to claim 1, wherein the high-voltage switch cabinet running state monitoring device is designed for minimizing independent compartments according to the size of each part, an integrated packaging structure is formed, and a shell of the integrated packaging structure is subjected to electromagnetic shielding by adopting steel materials.
3. The fault diagnosis method for the high-voltage switch cabinet based on photoelectric monitoring according to claim 2, wherein an integrated packaging structure of the high-voltage switch cabinet running state monitoring device is provided with a chute which is convenient for installing and detaching a sensor partition board inside a shell, and a signal transmission hole of an acquisition card which is convenient for a signal line or a wireless module to access is arranged on the back of the shell.
4. The method for diagnosing faults of the high-voltage switch cabinet based on photoelectric monitoring according to claim 2, wherein an integrated packaging structure of the high-voltage switch cabinet running state monitoring device is provided with an insulating fixing part at the bottom of the packaging, and the insulating fixing part is used for adsorbing and fixing a detection point position and realizing insulation with a detection placement position.
5. The method for diagnosing faults of the high-voltage switch cabinet based on photoelectric monitoring according to claim 1, wherein in the step S2, on-line data processing of infrared temperature waveforms and ultraviolet discharge waveforms is carried out on collected historical and real-time data so as to extract temperature average values, jump values, discharge pulse width, frequency and amplitude characteristics and achieve dimension reduction from mass operation data to state characteristics.
6. The method for diagnosing faults of a high-voltage switch cabinet based on photoelectric monitoring according to claim 5, wherein the processing of the original ultraviolet pulse based on the peak-to-valley detection and positioning algorithm is specifically as follows:
for the original ultraviolet pulse waveform, a peak searching algorithm is utilized to obtain a discharge pulse peak and all wave troughs, and a plurality of wave troughs between adjacent discharge pulse peaks are processed through a peak-to-valley matching algorithm:
wherein val is a trough containing harmonic interference, peak is a precise discharge pulse crest, and a discharge pulse peak Gu Fuzhi and position information are obtained to obtain discharge pulse width, frequency and amplitude characteristics.
7. The method for diagnosing faults of the high-voltage switch cabinet based on photoelectric monitoring as claimed in claim 5, wherein the method for processing the infrared temperature based on the wavelet denoising and sliding window filtering algorithm specifically comprises the following steps:
for the original infrared signal containing noise, adopting wavelet soft threshold denoising to filter high-frequency interference:
wherein, |omega| is an absolute value, lambda is a threshold value, MAD is the median of the absolute value of the first-layer wavelet decomposition coefficient, and N is the signal length;
and removing signal outliers by utilizing a sliding window median algorithm based on wavelet denoising results to obtain a temperature average value and a jump value.
8. The high-voltage switch cabinet fault diagnosis method based on photoelectric monitoring according to claim 1, wherein the self-adaptive fuzzy neural network adopts a five-layer structure, and an input layer acquires node information; the membership layer fuzzifies the input information; the rule layer calculates the fitness of the data corresponding to each rule; the decision layer normalizes the fitness to reduce errors; the output layer obtains a network diagnosis result:
where y is the network output and represents the sum of the weighted averages of each rule, y j For each rule output, a j For the fitness of each rule,is a as j Is used for the normalization parameters of (a).
9. The method for diagnosing the fault of the high-voltage switch cabinet based on the photoelectric monitoring according to claim 8, wherein the fault type diagnosis result comprises three categories of no fault, temperature rise fault and partial discharge fault, and the fault grade evaluation result comprises three grades of good, normal and serious.
10. The method for diagnosing the faults of the high-voltage switch cabinet based on the photoelectric monitoring is characterized in that the fault type and grade assessment of output real-time monitoring information are visually displayed through a three-dimensional view, specifically, three parameters including fault grade, three-dimensional column height and column surface color are associated, an x-axis is set to be the fault type, a y-axis is the fault grade, a z-axis is set to be a state index, and a three-dimensional assessment result of the state of the switch cabinet is displayed.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310773708.XA CN116819251A (en) | 2023-06-27 | 2023-06-27 | High-voltage switch cabinet fault diagnosis method based on photoelectric monitoring |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310773708.XA CN116819251A (en) | 2023-06-27 | 2023-06-27 | High-voltage switch cabinet fault diagnosis method based on photoelectric monitoring |
Publications (1)
Publication Number | Publication Date |
---|---|
CN116819251A true CN116819251A (en) | 2023-09-29 |
Family
ID=88121584
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310773708.XA Pending CN116819251A (en) | 2023-06-27 | 2023-06-27 | High-voltage switch cabinet fault diagnosis method based on photoelectric monitoring |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116819251A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN118068132A (en) * | 2024-04-17 | 2024-05-24 | 国网山西省电力公司太原供电公司 | Cable anomaly identification method and system based on time-frequency analysis |
CN118134290A (en) * | 2024-05-07 | 2024-06-04 | 国网山西省电力公司运城供电公司 | Photovoltaic array running state evaluation and fault diagnosis method based on improved ANFIS |
-
2023
- 2023-06-27 CN CN202310773708.XA patent/CN116819251A/en active Pending
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN118068132A (en) * | 2024-04-17 | 2024-05-24 | 国网山西省电力公司太原供电公司 | Cable anomaly identification method and system based on time-frequency analysis |
CN118134290A (en) * | 2024-05-07 | 2024-06-04 | 国网山西省电力公司运城供电公司 | Photovoltaic array running state evaluation and fault diagnosis method based on improved ANFIS |
CN118134290B (en) * | 2024-05-07 | 2024-07-23 | 国网山西省电力公司运城供电公司 | Photovoltaic array running state evaluation and fault diagnosis method based on improved ANFIS |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN116819251A (en) | High-voltage switch cabinet fault diagnosis method based on photoelectric monitoring | |
US11408797B2 (en) | GIL fault on-line monitoring system based on vibration signals and support vector machine | |
CN106840637B (en) | Based on the GIS mechanical oscillation signal Time-Frequency Analysis Method for improving HHT algorithm | |
CN112307950B (en) | Detail feature quantity extraction and intelligent analysis method for GIS vibration defect identification | |
CN111679166A (en) | Switch cabinet partial discharge fault multi-source information fusion detection early warning system and method based on wireless transmission technology | |
CA2455457A1 (en) | Wire fault detection | |
CN112986870A (en) | Distributed power transformer winding state monitoring method and system based on vibration method | |
CN113484700A (en) | Switch cabinet partial discharge detection method based on indoor intelligent inspection robot | |
Gao et al. | Fault diagnosis method of rolling bearings based on adaptive modified CEEMD and 1DCNN model | |
CN208076016U (en) | A kind of GIL On-line Faults monitoring system based on vibration signal support vector machines | |
CN117634905A (en) | Power grid business data feature extraction method | |
CN111638047A (en) | GIS equipment mechanical fault diagnosis system and method | |
CN112329825A (en) | Transformer mechanical fault diagnosis method based on information dimension division and decision tree lifting | |
CN110969238A (en) | Method and device for calibrating electricity consumption data | |
CN114034375B (en) | Ultra-high voltage transmission line noise measurement system and method | |
CN115758112A (en) | Online identification method for transient disturbance of electric power | |
CN115356605A (en) | Method for monitoring running state of power distribution switch control equipment | |
CN115291156A (en) | Online detection system and detection method for error characteristics of voltage transformer | |
Bansal et al. | Automatic fault detection in a class of Wheatstone bridge-based transducer using ANNs in Verilog HDL | |
Sun et al. | A Novel Fault Diagnosis of GIS Partial Discharge Based on Improved Whale Optimization Algorithm | |
CN106405341A (en) | Method used for reducing partial discharge ultrahigh frequency signal | |
CN112114215A (en) | Transformer aging evaluation method and system based on error back propagation algorithm | |
Hussain et al. | An adaptive denoising algorithm for online condition monitoring of high-voltage power equipment | |
Wu et al. | Design and implementation of pulse diagnosis instrument based on wavelet transform | |
CN113358930B (en) | Harmonic wave test system, device and method based on signal deviation |
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 |