CN106443238B - High-voltage equipment state evaluation method, online monitoring device evaluation method and device - Google Patents
High-voltage equipment state evaluation method, online monitoring device evaluation method and device Download PDFInfo
- Publication number
- CN106443238B CN106443238B CN201610757587.XA CN201610757587A CN106443238B CN 106443238 B CN106443238 B CN 106443238B CN 201610757587 A CN201610757587 A CN 201610757587A CN 106443238 B CN106443238 B CN 106443238B
- Authority
- CN
- China
- Prior art keywords
- spectrogram
- monitoring device
- voltage equipment
- distorted
- state
- 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
- 238000012806 monitoring device Methods 0.000 title claims abstract description 116
- 238000011156 evaluation Methods 0.000 title claims abstract description 36
- 238000000034 method Methods 0.000 claims abstract description 29
- 238000010219 correlation analysis Methods 0.000 claims 1
- 238000012544 monitoring process Methods 0.000 abstract description 15
- 230000008569 process Effects 0.000 abstract description 7
- 238000004458 analytical method Methods 0.000 description 25
- 239000012159 carrier gas Substances 0.000 description 16
- 238000005516 engineering process Methods 0.000 description 7
- 230000007547 defect Effects 0.000 description 6
- 238000012216 screening Methods 0.000 description 6
- 238000004587 chromatography analysis Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 238000009413 insulation Methods 0.000 description 4
- 230000008447 perception Effects 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 238000001228 spectrum Methods 0.000 description 4
- 238000013461 design Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 239000000463 material Substances 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 239000007789 gas Substances 0.000 description 2
- 238000009421 internal insulation Methods 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 238000004904 shortening Methods 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000009529 body temperature measurement Methods 0.000 description 1
- 238000005266 casting Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000002347 injection Methods 0.000 description 1
- 239000007924 injection Substances 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
Abstract
The invention discloses a high-voltage equipment state evaluation method, an on-line monitoring device evaluation method and a device, firstly, various acquired spectrograms are compared with an established typical spectrogram sample library, and the typical sample library at least comprises various 'distortion' type typical spectrograms; and then judging whether the acquired spectrogram belongs to a distorted spectrogram or an undistorted spectrogram, wherein the acquired spectrogram is related to the self fault of the online monitoring device when the acquired spectrogram belongs to the distorted spectrogram, and the acquired spectrogram is used for judging the running state of the high-voltage equipment when the acquired spectrogram belongs to the undistorted spectrogram. Through the process, the method and the device can improve the positive judgment rate of the running state evaluation of the high-voltage equipment, and have the function of monitoring and evaluating the running state of the on-line monitoring device.
Description
Technical Field
The invention belongs to the field of high-voltage equipment, and particularly relates to a high-voltage equipment state evaluation method, an on-line monitoring device evaluation method and an on-line monitoring device evaluation device.
Background
Based on the requirements of 'large operation' and 'large maintenance', accurate perception and reliability evaluation of the operation state and the control state of the equipment are one of the core contents of the intelligent high-voltage equipment. In the engineering construction of intelligent high-voltage equipment, the on-site state acquisition and evaluation analysis of high-voltage equipment information are realized through the mutual cooperation of a sensor, an intelligent component, a network and all levels of main stations, but in terms of the current practical application effect, the phenomena of poor equipment monitoring data availability, high false alarm rate, high alarm missing rate and the like generally exist, and the state maintenance guiding effect is not strong. Meanwhile, due to various reasons, a large amount of acquired device spectrogram data is not fully utilized, and resource waste is caused.
In the construction process of the intelligent transformer substation, the transformer oil chromatography monitoring, the high-voltage partial discharge monitoring and the high-voltage switch mechanism state monitoring technology are applied more and more widely, but actually, the application effect is unsatisfactory, and the method is mainly shown in the following aspects:
1) the monitoring device for host equipment running state perception is poor in reliability, low in effective alarm rate and the like, and the product maturity of the monitoring device needs to be further improved. As indicated by the 24 article in 2014 of the national grid, as long as 6 months in 2013, 28 electric power saving companies (including national grid operating companies) of the national grid share various transformer equipment online monitoring devices 30422. The method comprises 15 types of online monitoring devices including dissolved gas in oil, lightning arrester insulation online monitoring, ultrahigh frequency GIS partial discharge monitoring, infrared temperature measurement and the like. At present, equipment defects are discovered 2626 times by various types of devices, the defect discovery rate is 0.061 times/(bench year), the device failure rate is 0.124 times/(bench year), and the effective alarm rate of the devices is 7.7% ". Similarly, the spectrogram of the device has distortion due to the defects of the online monitoring device, such as application environment interference, an acquisition loop and a processing loop. FIG. 1 shows the spectrogram generated by the device for monitoring the dissolved gas in oil, wherein the spectrogram truly reflects the state of the host high-pressure equipment and the spectrogram of 'distortion' generated by the loop abnormality of the monitoring device are also provided. FIG. 2 shows a spectrogram generated by the GIS partial discharge online monitoring device, wherein the spectrogram truly reflects the state of host high-voltage equipment and a 'distortion' spectrogram influenced by environmental factors and installation factors is also provided.
2) The existing monitoring device focuses on data acquisition, only generates a spectrogram according to format requirements and sends the spectrogram to a post-stage main station for processing, spectrogram analysis on the device level is not deep enough, the spectrogram is only displayed when debugging and users have requirements, the correct early warning rate of the device is effectively improved, and the capabilities of spectrogram information mining and spectrogram recognition are enhanced.
3) The spectrogram generated by the online monitoring device contains abundant high-voltage host equipment state and monitoring device operation state information, and the phenomenon of spectrogram distortion caused by severe operation environment, device failure and aging of the online monitoring device sometimes occurs, so that the difficulty is brought to the fault of a post-level system discrimination device. Currently, the ability to identify "distorted" spectra is lacking, whether at the device level or at the background analysis level. During the operation of the intelligent high-voltage equipment, the phenomenon of spectrogram distortion caused by the change of the state or the operation condition of the monitoring device can occur.
Because the existing equipment has no 'distortion' spectrogram identification capability, the 'distortion' spectrograms are usually regarded as the operation characteristics of high-pressure host equipment, thereby causing false alarm.
Disclosure of Invention
The invention aims to provide a high-voltage equipment state evaluation method which is used for solving the problem of low correct judgment rate of high-voltage equipment operation state evaluation; and a state evaluation method of the high-voltage equipment monitoring device, which is used for solving the problem of monitoring and evaluating the self fault state of the on-line monitoring device and simultaneously solving the problem of false alarm of the high-voltage equipment.
In order to achieve the purpose, the invention discloses a spectrogram intelligent identification-based high-voltage equipment state evaluation method, which adopts the technical scheme that the method comprises the following steps:
s1, comparing various acquired spectrograms with an established typical spectrogram sample library, wherein the typical sample library at least comprises various 'distortion' typical spectrograms;
and S2, judging whether the acquired spectrogram belongs to a distorted spectrogram or an undistorted spectrogram, wherein the acquired spectrogram is related to the self fault of the online monitoring device when the acquired spectrogram belongs to the distorted spectrogram, and the acquired spectrogram is used for judging the running state of the high-voltage equipment when the acquired spectrogram belongs to the undistorted spectrogram.
For the acquired spectrogram, a clustering algorithm is used for selecting a 'distorted' spectrogram so as to separate a 'undistorted' spectrogram, or the clustering algorithm is used for selecting the 'distorted' spectrogram and the 'undistorted' spectrogram so as to divide the spectrogram into the 'distorted' spectrogram and the 'undistorted' spectrogram.
And when the acquired spectrogram belongs to a distorted spectrogram, the online monitoring device is used for sending a fault alarm signal of the online monitoring device, and simultaneously, the high-voltage equipment operation state alarm circuit or the high-voltage equipment control state alarm circuit is locked.
And the station level performs relevance analysis and reconfirmation on self fault alarm signals sent by the online monitoring device to prevent false alarm.
The invention discloses a spectrogram intelligent identification-based state evaluation method for a high-voltage equipment online monitoring device, which adopts the technical scheme that the method comprises the following steps:
s1, comparing various acquired spectrograms with an established typical spectrogram sample library, wherein the typical sample library comprises various 'distortion' typical spectrograms;
and S2, judging the acquired spectrogram, and when the acquired spectrogram belongs to a distorted spectrogram, sending a self fault alarm signal by the online monitoring device.
When the distortion spectrogram sends out a fault alarm signal of the online monitoring device, the high-voltage equipment operation state alarm loop or the high-voltage equipment control state alarm loop is locked.
The invention also provides a high-voltage equipment state evaluation device based on spectrogram intelligent identification, which comprises:
an alignment unit: the system is used for comparing various acquired spectrograms with a typical spectrogram sample library, wherein the typical spectrogram sample library at least comprises various 'distortion' typical spectrograms;
a judging unit: the method is used for judging whether the acquired spectrogram belongs to a distorted spectrogram or an undistorted spectrogram, is related to the self fault of the online monitoring device when the acquired spectrogram belongs to the distorted spectrogram, and is used for judging the running state of the high-voltage equipment when the acquired spectrogram belongs to the undistorted spectrogram.
The device also comprises a unit for sorting the acquired spectrogram by using a clustering algorithm to select a 'distorted' spectrogram so as to separate a 'non-distorted' spectrogram, or sorting the 'distorted' spectrogram and the 'non-distorted' spectrogram by using the clustering algorithm so as to divide the spectrogram into the 'distorted' spectrogram and the 'non-distorted' spectrogram.
The device also comprises a unit which is used for sending out a self fault alarm signal of the on-line monitoring device when the acquired spectrogram belongs to a distorted spectrogram and simultaneously locking the running state alarm loop of the high-voltage equipment or the control state alarm loop of the high-voltage equipment.
The device also comprises a unit for performing relevance analysis and reconfirmation on the self fault alarm signal sent by the on-line monitoring device at station level to prevent false alarm.
The invention also provides a device for evaluating the state of the high-voltage equipment on-line monitoring device based on spectrogram intelligent identification, which comprises:
s1, a comparison unit: the system is used for comparing various acquired spectrograms with an established typical spectrogram sample library, wherein the typical sample library comprises various 'distortion' typical spectrograms;
s2, a judging unit: and the online monitoring device is used for judging the acquired spectrogram, and when the spectrogram belongs to a distorted spectrogram, the online monitoring device sends out a self fault alarm signal.
The device also comprises a unit for locking the high-voltage equipment running state alarm loop or the high-voltage equipment control state alarm loop when the distortion spectrogram sends out a fault alarm signal of the online monitoring device.
The invention has the beneficial effects that:
according to the method for evaluating the state of the high-voltage equipment based on spectrogram intelligent identification, the distorted spectrogram is removed on the basis of the existing method, the operation state of the high-voltage equipment is evaluated only by the undistorted spectrogram, the correct judgment rate of the operation state evaluation of the high-voltage equipment is improved, and the problem of high false alarm rate of the high-voltage equipment in engineering application can be effectively solved.
Because the high-voltage equipment has no spectrogram distinguishing capability, the high-voltage equipment working platform is connected with the server platform, the spectrogram can be effectively identified by using a computer intelligent algorithm, and the intelligent sensing level of the state of the power transmission and transformation station equipment is improved.
The spectrogram characteristics are associated with the states of the online monitoring device and the power equipment by using a clustering algorithm, so that the fault judgment of the online monitoring device and the state analysis of the power equipment can be effectively realized.
The method for evaluating the state of the high-voltage equipment on-line monitoring device based on spectrogram identification realizes the function of monitoring and evaluating the running state of the on-line monitoring device.
The fault alarm signal sent by the on-line monitoring device is respectively used as the locking condition of the alarm loop of the running state and the control state of the high-voltage equipment, so that the signal transmitted by the on-line monitoring device can not be transmitted to the high-voltage equipment, and the phenomenon of mistaken alarm of the running state of the high-voltage equipment is avoided.
Drawings
FIG. 1-a is a non-distorted spectrogram generated by an oil chromatography on-line monitoring device;
FIG. 1-b is a chart of the over-low temperature and carrier gas under-pressure spectrum generated by the oil chromatography on-line monitoring device;
FIG. 1-c is an electromagnetic interference and grounding failure spectrogram generated by an oil chromatography on-line monitoring device;
FIG. 1-d is a repeated sample injection spectrogram generated by the oil chromatography on-line monitoring device;
FIG. 2-a is a "non-distorted" spectrogram of a built-in sensor generated by the partial discharge online monitoring device;
FIG. 2-b is a "non-distorted" spectrogram of a built-in sensor generated by the partial discharge online monitoring device;
FIG. 2-c is a diagram of an external sensor casting empty offset spectrum;
FIG. 2-d is a spectrogram generated by a surrounding device;
fig. 3 is a flowchart of a high-voltage equipment state evaluation method based on spectrogram intelligent identification.
Detailed Description
The embodiment of the high-voltage equipment state evaluation method based on spectrogram intelligent identification, disclosed by the invention, comprises the following steps of:
in the first embodiment, as shown in fig. 3, the method for evaluating the state of the high-voltage device based on spectrogram intelligent recognition specifically includes the following steps:
establishing a typical spectrogram sample library, and carrying out support spectrogram source analysis:
on the basis of the research of the spectrogram generating technology of the online monitoring device, various distortion spectrograms of the online monitoring device under various working conditions are collected and sorted out by carrying out system analysis on the aspects of the design, materials and application environment of monitored products and combining with the field application and field test technology of the products, as shown in figures 1 and 2, characteristic parameters for spectrogram mode identification have frequency, steepness, deviation, peak-appearing position, asymmetry and the like, and a reference template is provided for effectively identifying the non-distortion spectrogram and the distortion spectrogram.
Screening two categories of a distorted spectrogram and an undistorted spectrogram:
based on a typical sample library template, the sample library template comprises various distorted spectrograms, the currently acquired spectrograms are classified into distorted spectrogram feature cluster classes represented by sample library samples through a clustering algorithm based on division, objects in the same cluster have higher similarity, and the objects in different clusters have larger difference, so that the spectrogram is classified into which cluster class, namely, the situation represented by the typical sample in the cluster class can be reflected, then the distorted spectrograms are picked out, and the spectrograms which are not classified into the distorted class are classified into non-distorted spectrogram classes, so that the non-distorted spectrogram is sorted out; for example, if the spectrogram v0 is similar to the spectrogram v1 of an oil chromatographic device in a typical sample library with too low temperature and under-pressure carrier gas, the spectrogram is directly classified as a "distorted" spectrogram, and similarly, if the spectrogram v0 is similar to the spectrogram v2 of an oil chromatographic device with poor grounding or spectrogram v3 of an oil chromatographic device with repeated feeding, the spectrogram is also directly classified as a "distorted" spectrogram. The distortion spectrogram is related to the application environment and the loop of the on-line monitoring device, can be used for diagnosing the operation fault of the on-line monitoring device and determining the fault type of the on-line monitoring device, does not truly reflect the operation state of the high-voltage equipment and cannot be used for analyzing and evaluating the operation state of the high-voltage equipment, and the non-distortion spectrogram is related to the actual state of the high-voltage equipment and can be used for analyzing the operation state or the control state of the high-voltage equipment. In addition, other algorithms with clustering functions can be used for screening the distorted spectrogram.
The method is used for analyzing and processing the undistorted spectrogram, truly reflects the running state of the host equipment through the spectrogram screened by the preceding stage, can realize the correlation of the internal insulation state of the high-voltage equipment by combining a neural network, an SVM algorithm and the like according to the fault spectrogram characteristics of the host equipment, and realizes the real-time perception of the running state and the control state of the host equipment.
Further optimization of this example, analysis of the "distorted" spectrum:
the method is characterized in that a distorted spectrogram type spectrogram is analyzed and processed, correlation between spectrogram characteristics and internal defects of a state monitoring device is realized through a clustering algorithm mainly by referring to a distorted spectrogram typical sample, a fault alarm signal and a fault analysis guidance suggestion of the online monitoring device are further given, the generated spectrogram reflecting the running state of the high-voltage equipment and related monitoring parameters are not real reflections of the actual running state of the equipment, and an error result is caused if the distorted spectrogram type spectrogram is used for evaluating the state of the high-voltage equipment. If the correlation degree of a certain distortion type spectrogram of the transformer oil chromatogram online monitoring device and the sample characteristics of the over-low temperature and carrier gas under-pressure spectrogram is high, and the clustering algorithm identifies the same type, an alarm signal of the over-low temperature and carrier gas under-pressure of the transformer oil chromatogram online monitoring device is given to remind an operator to process; and meanwhile, the self fault alarm signal mark of the on-line monitoring device is used as a locking condition of the alarm loop of the running state or the control state of the high-voltage equipment, so that the phenomenon of mistakenly alarming the running state of the high-voltage equipment is avoided. For example, in order to prevent false alarm, the transformer insulation alarm module is locked by the alarm signal of 'low temperature and carrier gas under-voltage' of the transformer oil chromatogram on-line monitoring device, so as to prevent the error alarm of high temperature or discharge in the transformer of the host equipment by taking the working condition of carrier gas under-voltage of the oil chromatogram monitoring device.
The embodiment is further optimized to carry out system-level alarm discrimination and improve the correct alarm rate
Due to the current situation that the existing state monitoring device has large product quality difference and low correct alarm rate, the station level also utilizes the spectrogram analysis result to carry out relevance analysis and reconfirmation on the alarm information sent by the online monitoring device, thereby preventing false alarm.
In addition, when the spectrogram analysis indicates that the high-voltage equipment is abnormal and no device early warning signal exists, the possibility of missing alarm can be reduced by shortening a monitoring period or a multi-parameter information fusion algorithm and the like.
Example two: specifically, as shown in fig. 3, the method for evaluating the state of the high-voltage device based on spectrogram intelligent identification specifically comprises the following steps:
establishing a typical spectrogram sample library, and carrying out support spectrogram source analysis:
on the basis of the research of the spectrogram generating technology of the online monitoring device, various 'distortion' spectrograms and 'non-distortion' spectrograms under various working conditions of the online monitoring device are collected and sorted out by performing system analysis on the aspects of the design, materials and application environment of monitored products and combining with the field application and field test technology of the products, as shown in fig. 1 and 2, characteristic parameters for spectrogram mode identification have frequency, steepness, deviation, peak-appearing positions, asymmetry and the like, and a reference template is provided for effectively identifying the 'non-distortion' spectrograms and the 'distortion' spectrograms.
Screening two categories of a distorted spectrogram and an undistorted spectrogram:
based on a typical sample library template, the sample library template comprises various 'distorted' spectrograms, the currently acquired spectrograms are classified into 'distorted' spectrogram characteristics represented by sample library samples and 'undistorted' spectrogram characteristic clusters through a clustering algorithm, objects in the same cluster have higher similarity, the objects in different clusters have larger difference, so that the spectrogram is classified into which cluster, namely, the situation represented by the typical sample in the cluster can be reflected, and the 'distorted' spectrogram and the 'undistorted' spectrogram are picked out, so that the spectrograms are divided into the 'distorted' spectrogram and the 'undistorted' spectrogram; for example, if the spectrogram v0 is similar to the spectrogram v1 of an oil chromatographic device in a typical sample library with too low temperature and under-pressure carrier gas, the spectrogram is directly classified as a "distorted" spectrogram, and similarly, if the spectrogram v0 is similar to the spectrogram v2 of an oil chromatographic device with poor grounding or spectrogram v3 of an oil chromatographic device with repeated feeding, the spectrogram is also directly classified as a "distorted" spectrogram. The distortion map is related to the application environment and the loop of the on-line monitoring device, can be used for diagnosing the operation fault of the on-line monitoring device and determining the fault type of the on-line monitoring device, does not truly reflect the operation state of the high-voltage equipment and cannot be used for analyzing and evaluating the operation state of the high-voltage equipment, and the non-distortion map is related to the actual state of the high-voltage equipment and can be used for analyzing the operation state or the control state of the high-voltage equipment. In addition, other algorithms with clustering functions can be used for screening the distorted spectrogram.
Step (3) "distortion" spectrogram analysis:
the method is characterized in that a distorted spectrogram type spectrogram is analyzed and processed, correlation between spectrogram characteristics and internal defects of a state monitoring device is realized through a clustering algorithm mainly by referring to a distorted spectrogram typical sample, a fault alarm signal and a fault analysis guidance suggestion of the online monitoring device are further given, the generated spectrogram reflecting the running state of the high-voltage equipment and related monitoring parameters are not real reflections of the actual running state of the equipment, and an error result is caused if the distorted spectrogram type spectrogram is used for evaluating the state of the high-voltage equipment. If the correlation degree of a certain distortion type spectrogram of the transformer oil chromatogram online monitoring device and the sample characteristics of the over-low temperature and carrier gas under-pressure spectrogram is high, and the clustering algorithm identifies the same type, an alarm signal of the over-low temperature and carrier gas under-pressure of the transformer oil chromatogram online monitoring device is given to remind an operator to process; and meanwhile, the self fault alarm signal mark of the on-line monitoring device is used as a locking condition of the alarm loop of the running state or the control state of the high-voltage equipment, so that the phenomenon of mistakenly alarming the running state of the high-voltage equipment is avoided. For example, in order to prevent false alarm, the transformer insulation alarm module is locked by the alarm signal of 'low temperature and carrier gas under-voltage' of the transformer oil chromatogram on-line monitoring device, so as to prevent the error alarm of high temperature or discharge in the transformer of the host equipment by taking the working condition of carrier gas under-voltage of the oil chromatogram monitoring device.
Step (4) "non-distortion" spectrogram analysis
The method is used for analyzing and processing the undistorted spectrogram, truly reflects the running state of the host equipment through the spectrogram screened by the preceding stage, can realize the correlation of the internal insulation state of the high-voltage equipment by combining a neural network, an SVM algorithm and the like according to the fault spectrogram characteristics of the host equipment, and realizes the real-time perception of the running state and the control state of the host equipment.
Step (5) of system-level alarm discrimination is carried out, and the correct alarm rate is improved
Due to the current situation that the existing state monitoring device has large product quality difference and low correct alarm rate, the station level also utilizes the spectrogram analysis result to carry out relevance analysis and reconfirmation on the alarm information sent by the online monitoring device, thereby preventing false alarm.
In addition, when the spectrogram analysis indicates that the high-voltage equipment is abnormal and no device early warning signal exists, the possibility of missing alarm can be reduced by shortening a monitoring period or a multi-parameter information fusion algorithm and the like.
The embodiment of the invention relates to a spectrogram intelligent identification-based state evaluation method for a high-voltage equipment online monitoring device, which comprises the following steps:
establishing a typical spectrogram sample library, and carrying out support spectrogram source analysis:
on the basis of the research of the spectrogram generating technology of the online monitoring device, various distortion spectrograms of the online monitoring device under various working conditions are collected and sorted out by carrying out system analysis on the aspects of the design, materials and application environment of monitored products and combining with the field application and field test technology of the products, as shown in figures 1 and 2, characteristic parameters for spectrogram mode identification have frequency, steepness, deviation, peak-appearing position, asymmetry and the like, and a reference template is provided for effectively identifying the non-distortion spectrogram and the distortion spectrogram.
Screening two categories of a distorted spectrogram and an undistorted spectrogram:
based on a typical sample library template, the sample library template comprises various distorted spectrograms, the currently acquired spectrograms are classified into distorted spectrogram feature cluster classes represented by sample library samples through a clustering algorithm, the distorted spectrograms are picked out, the objects in the same cluster have higher similarity, the objects in different clusters have larger difference, so that the spectrogram is classified into which cluster class, the condition represented by the typical sample in the cluster class can be reflected, and the spectrograms which are not classified into the distorted class are classified into undistorted spectrogram classes, so that the undistorted spectrograms are classified; for example, if the spectrogram v0 is similar to the spectrogram v1 of an oil chromatographic device in a typical sample library with too low temperature and under-pressure carrier gas, the spectrogram is directly classified as a "distorted" spectrogram, and similarly, if the spectrogram v0 is similar to the spectrogram v2 of an oil chromatographic device with poor grounding or spectrogram v3 of an oil chromatographic device with repeated feeding, the spectrogram is also directly classified as a "distorted" spectrogram. The distortion map is related to the application environment and the loop of the on-line monitoring device, can be used for diagnosing the operation fault of the on-line monitoring device and determining the fault type of the on-line monitoring device, does not truly reflect the operation state of the high-voltage equipment and cannot be used for analyzing and evaluating the operation state of the high-voltage equipment, and the non-distortion map is related to the actual state of the high-voltage equipment and can be used for analyzing the operation state or the control state of the high-voltage equipment. In addition, other algorithms with clustering functions can be used for screening the distorted spectrogram.
Step (3) "distortion" spectrogram analysis:
the method is characterized in that a distorted spectrogram type spectrogram is analyzed and processed, correlation between spectrogram characteristics and internal defects of a state monitoring device is realized through a clustering algorithm mainly by referring to a distorted spectrogram typical sample, a fault alarm signal and a fault analysis guidance suggestion of the online monitoring device are further given, the generated spectrogram reflecting the running state of the high-voltage equipment and related monitoring parameters are not real reflections of the actual running state of the equipment, and an error result is caused if the distorted spectrogram type spectrogram is used for evaluating the state of the high-voltage equipment. If the correlation degree of a certain distortion type spectrogram of the transformer oil chromatogram online monitoring device and the sample characteristics of the over-low temperature and carrier gas under-pressure spectrogram is high, and the clustering algorithm identifies the same type, an alarm signal of the over-low temperature and carrier gas under-pressure of the transformer oil chromatogram online monitoring device is given to remind an operator to process; and meanwhile, the self fault alarm signal mark of the on-line monitoring device is used as a locking condition of the alarm loop of the running state or the control state of the high-voltage equipment, so that the phenomenon of mistakenly alarming the running state of the high-voltage equipment is avoided. For example, in order to prevent false alarm, the transformer insulation alarm module is locked by the alarm signal of 'low temperature and carrier gas under-voltage' of the transformer oil chromatogram on-line monitoring device, so as to prevent the error alarm of high temperature or discharge in the transformer of the host equipment by taking the working condition of carrier gas under-voltage of the oil chromatogram monitoring device.
The invention also provides a high-voltage equipment state evaluation device based on spectrogram intelligent identification, which comprises a comparison unit and a judgment unit, wherein the comparison unit is used for comparing various acquired spectrograms with a typical spectrogram sample library, and the typical sample library at least comprises various 'distortion' typical spectrograms; the judgment unit is used for judging whether the acquired spectrogram belongs to a distorted spectrogram or an undistorted spectrogram, is related to the fault of the online monitoring device when the acquired spectrogram belongs to the distorted spectrogram, and is used for judging the running state of the high-voltage equipment when the acquired spectrogram belongs to the undistorted spectrogram.
The above evaluation apparatus is actually a functional module framework, and each unit thereof is a process or a program corresponding to steps (1) to (2) in the first embodiment of the above evaluation method and steps (1) to (5) in the second embodiment. Therefore, the evaluation device will not be described in detail. The evaluation device can operate in the high-voltage equipment as a program, can improve the positive judgment rate of the evaluation of the operating state of the high-voltage equipment, and avoids the problem of high false alarm rate of the high-voltage equipment in engineering application.
The invention also provides a state evaluation device of the high-voltage equipment on-line monitoring device based on spectrogram intelligent identification, which comprises a comparison unit and a judgment unit, wherein the comparison unit is used for comparing various acquired spectrograms with an established typical spectrogram sample library, and the typical sample library comprises various 'distortion' typical spectrograms; the judging unit is used for judging the acquired spectrogram, and when the acquired spectrogram belongs to a distorted spectrogram, the online monitoring device sends out a self fault alarm signal.
The above evaluation device is actually a functional module framework, and each unit thereof is a process or program corresponding to steps (1) to (3) of the above-mentioned high-voltage equipment state evaluation method based on spectrogram intelligent recognition. Therefore, the evaluation device will not be described in detail. The evaluation device can be operated in the high-voltage equipment on-line monitoring device as a program, can enable the on-line monitoring device to detect and evaluate the self operation state, and can avoid false alarm of the high-voltage equipment.
Claims (10)
1. A method for evaluating the state of a high-voltage device, characterized by the steps of:
s1, comparing various acquired spectrograms with an established typical spectrogram sample library, wherein the typical spectrogram sample library at least comprises various 'distortion' typical spectrograms;
s2, judging whether the acquired spectrogram belongs to a distorted spectrogram or an undistorted spectrogram, wherein when the acquired spectrogram belongs to the distorted spectrogram, the acquired spectrogram is related to the fault of the online monitoring device, and when the acquired spectrogram belongs to the undistorted spectrogram, the acquired spectrogram is used for judging the running state of the high-voltage equipment;
the distortion spectrogram is related to an application environment and a self loop of the on-line monitoring device, is used for diagnosing the self operation fault of the on-line monitoring device and determining the fault type of the on-line monitoring device, does not truly reflect the operation state of the high-voltage equipment, cannot be used for analyzing and evaluating the operation state of the high-voltage equipment, and the non-distortion spectrogram is related to the actual state of the high-voltage equipment and can be used for analyzing the operation state or the control state of the high-voltage equipment.
2. The method as claimed in claim 1, wherein the acquired spectrogram is divided into a "distorted" spectrogram by selecting a "distorted" spectrogram by a clustering algorithm, thereby separating a "non-distorted" spectrogram, or a "distorted" spectrogram and a "non-distorted" spectrogram by a clustering algorithm, thereby separating the spectrogram into a "distorted" spectrogram and a "non-distorted" spectrogram.
3. The method for evaluating the state of the high-voltage equipment according to claim 1 or 2, wherein when the acquired spectrogram belongs to a distorted spectrogram, the method is used for sending a fault alarm signal of the online monitoring device, and simultaneously, the operation state alarm loop or the control state alarm loop of the high-voltage equipment is locked.
4. The method for evaluating the state of the high-voltage equipment according to claim 3, wherein the station level performs correlation analysis and reconfirmation on of the self-fault alarm signal sent by the on-line monitoring device to prevent false alarm.
5. A state evaluation method for an online monitoring device of high-voltage equipment is characterized by comprising the following steps:
s1, comparing various acquired spectrograms with an established typical spectrogram sample library, wherein the typical spectrogram sample library comprises various 'distortion' typical spectrograms;
s2, judging the acquired spectrogram, and when the acquired spectrogram belongs to a distorted spectrogram, sending a self fault alarm signal by the online monitoring device;
the distortion spectrogram is related to an application environment and a self loop of the on-line monitoring device, is used for diagnosing the self operation fault of the on-line monitoring device and determining the fault type of the on-line monitoring device, does not truly reflect the operation state of the high-voltage equipment, cannot be used for analyzing and evaluating the operation state of the high-voltage equipment, and the non-distortion spectrogram is related to the actual state of the high-voltage equipment and can be used for analyzing the operation state or the control state of the high-voltage equipment.
6. The method as claimed in claim 5, wherein the high voltage device operation state alarm circuit or the high voltage device control state alarm circuit is locked when the distortion spectrogram emits the fault alarm signal of the on-line monitoring device.
7. A high-voltage device condition evaluation apparatus, characterized by comprising:
an alignment unit: the system is used for comparing various acquired spectrograms with a typical spectrogram sample library, wherein the typical spectrogram sample library at least comprises various 'distortion' typical spectrograms;
a judging unit: the device is used for judging whether the acquired spectrogram belongs to a distorted spectrogram or an undistorted spectrogram, is related to the self fault of the online monitoring device when belonging to the distorted spectrogram, and is used for judging the running state of the high-voltage equipment when belonging to the undistorted spectrogram;
the distortion spectrogram is related to an application environment and a self loop of the on-line monitoring device, is used for diagnosing the self operation fault of the on-line monitoring device and determining the fault type of the on-line monitoring device, does not truly reflect the operation state of the high-voltage equipment, cannot be used for analyzing and evaluating the operation state of the high-voltage equipment, and the non-distortion spectrogram is related to the actual state of the high-voltage equipment and can be used for analyzing the operation state or the control state of the high-voltage equipment.
8. The apparatus as claimed in claim 7, further comprising a unit for sorting the acquired spectrogram by a clustering algorithm to obtain a "distorted" spectrogram, thereby separating the "undistorted" spectrogram, or by a clustering algorithm to obtain a "distorted" spectrogram and a "undistorted" spectrogram, thereby separating the spectrogram into a "distorted" spectrogram and a "undistorted" spectrogram.
9. The device for evaluating the state of the high-voltage equipment according to claim 7 or 8, further comprising a unit for sending a fault alarm signal of the on-line monitoring device when the acquired spectrogram belongs to a distorted spectrogram, and simultaneously locking the operation state alarm loop or the control state alarm loop of the high-voltage equipment.
10. A state evaluation device of a high-voltage equipment on-line monitoring device is characterized by comprising:
s1, a comparison unit: the system is used for comparing various acquired spectrograms with an established typical spectrogram sample library, wherein the typical spectrogram sample library comprises various 'distortion' typical spectrograms;
s2, a judging unit: the online monitoring device is used for judging the acquired spectrogram, and when the spectrogram belongs to a distorted spectrogram, the online monitoring device sends out a self fault alarm signal;
the distortion spectrogram is related to an application environment and a self loop of the on-line monitoring device, is used for diagnosing the self operation fault of the on-line monitoring device and determining the fault type of the on-line monitoring device, does not truly reflect the operation state of the high-voltage equipment, cannot be used for analyzing and evaluating the operation state of the high-voltage equipment, and the non-distortion spectrogram is related to the actual state of the high-voltage equipment and can be used for analyzing the operation state or the control state of the high-voltage equipment.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610757587.XA CN106443238B (en) | 2016-08-27 | 2016-08-27 | High-voltage equipment state evaluation method, online monitoring device evaluation method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610757587.XA CN106443238B (en) | 2016-08-27 | 2016-08-27 | High-voltage equipment state evaluation method, online monitoring device evaluation method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106443238A CN106443238A (en) | 2017-02-22 |
| CN106443238B true CN106443238B (en) | 2020-02-21 |
Family
ID=58090822
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610757587.XA Active CN106443238B (en) | 2016-08-27 | 2016-08-27 | High-voltage equipment state evaluation method, online monitoring device evaluation method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106443238B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| RU2749338C1 (en) * | 2020-10-21 | 2021-06-08 | Федеральное государственное автономное образовательное учреждение высшего образования «Дальневосточный федеральный университет» (ДВФУ) | Method for monitoring technical condition of electric power equipment |
| CN117435947A (en) * | 2023-12-20 | 2024-01-23 | 山东和兑智能科技有限公司 | Lightning arrester state monitoring system and method |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101726674A (en) * | 2008-10-30 | 2010-06-09 | 武汉华瑞测控科技有限公司 | Method for on-line assessment of insulation on-line monitoring system of substation |
| CN102023278A (en) * | 2010-07-16 | 2011-04-20 | 华北电力大学 | Method and device for identifying partial discharge interference of GIS equipment |
| CN104375064A (en) * | 2014-11-07 | 2015-02-25 | 深圳供电局有限公司 | Partial discharge on-line monitoring alarm confidence coefficient analysis method and device |
| CN204188753U (en) * | 2014-07-07 | 2015-03-04 | 国网吉林省电力有限公司检修公司 | A kind of dry reactor office is placed on line monitoring system |
| CN104459483A (en) * | 2014-11-10 | 2015-03-25 | 国网甘肃省电力公司刘家峡水电厂 | 330 kV power cable line partial discharge monitoring method and device |
| US8996323B1 (en) * | 2011-06-30 | 2015-03-31 | Amazon Technologies, Inc. | System and method for assessing power distribution systems |
| CN104914362A (en) * | 2015-05-26 | 2015-09-16 | 深圳供电局有限公司 | Insulation state monitoring system and method of SF6 gas transformer |
| CN105512474A (en) * | 2015-12-02 | 2016-04-20 | 国网山东省电力公司电力科学研究院 | Transformer state monitoring data anomaly detection method |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TW201217800A (en) * | 2010-10-20 | 2012-05-01 | Chroma Ate Inc | for precisely measuring spectrum response of solar cell to provide correct energy conversion efficiency value |
| KR20130034330A (en) * | 2011-09-28 | 2013-04-05 | 한국전력공사 | Method and system for a detecting malfunction of power system map-board |
-
2016
- 2016-08-27 CN CN201610757587.XA patent/CN106443238B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101726674A (en) * | 2008-10-30 | 2010-06-09 | 武汉华瑞测控科技有限公司 | Method for on-line assessment of insulation on-line monitoring system of substation |
| CN102023278A (en) * | 2010-07-16 | 2011-04-20 | 华北电力大学 | Method and device for identifying partial discharge interference of GIS equipment |
| US8996323B1 (en) * | 2011-06-30 | 2015-03-31 | Amazon Technologies, Inc. | System and method for assessing power distribution systems |
| CN204188753U (en) * | 2014-07-07 | 2015-03-04 | 国网吉林省电力有限公司检修公司 | A kind of dry reactor office is placed on line monitoring system |
| CN104375064A (en) * | 2014-11-07 | 2015-02-25 | 深圳供电局有限公司 | Partial discharge on-line monitoring alarm confidence coefficient analysis method and device |
| CN104459483A (en) * | 2014-11-10 | 2015-03-25 | 国网甘肃省电力公司刘家峡水电厂 | 330 kV power cable line partial discharge monitoring method and device |
| CN104914362A (en) * | 2015-05-26 | 2015-09-16 | 深圳供电局有限公司 | Insulation state monitoring system and method of SF6 gas transformer |
| CN105512474A (en) * | 2015-12-02 | 2016-04-20 | 国网山东省电力公司电力科学研究院 | Transformer state monitoring data anomaly detection method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106443238A (en) | 2017-02-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103645425B (en) | High-voltage cable insulation defect partial discharge on-line monitoring diagnosis method | |
| CN103728429B (en) | On-line water quality monitoring method and Monitoring systems | |
| CN106056879B (en) | A kind of ammeter remote monitoring system based on intelligent recognition | |
| CN115982602B (en) | Photovoltaic transformer electrical fault detection method | |
| CN107037062B (en) | Coaxial cable material microwave characteristic on-line monitoring and diagnosing device and method | |
| CN112611936A (en) | Distribution network transformer fault dynamic detection and classification system based on edge calculation | |
| CN111007365A (en) | Ultrasonic partial discharge identification method and system based on neural network | |
| CN110866616A (en) | Fault early warning method and device for secondary equipment of transformer substation | |
| CN106443238B (en) | High-voltage equipment state evaluation method, online monitoring device evaluation method and device | |
| CN112987696A (en) | Regional power distribution network equipment management platform and operation method thereof | |
| CN113848471B (en) | Intelligent fault positioning method and system for relay protection system | |
| CN109406923A (en) | A kind of power supply system of train earth leakage failure prediction method and device | |
| CN114460519A (en) | On-site and terminal fusion management system and method based on power transformer inspection | |
| CN109298285A (en) | A kind of identification of distribution network cable initial failure and early warning system and method based on transient disturbance | |
| CN106646106B (en) | Electric network fault detection method based on outlier's detection technology | |
| CN103163402B (en) | Relay protection device state monitoring apparatus based on secondary circuit and monitoring method | |
| CN110488125B (en) | Method and system for realizing active early warning processing based on state information quantity of circuit breaker | |
| KR101984257B1 (en) | Cloud service based big data analysing system and method therein | |
| CN114859160A (en) | Method and system for detecting electromagnetic interference of hydroelectric generating set, electronic equipment and storage medium | |
| CN109765878A (en) | The aided analysis method of new-energy automobile CAN bus network failure | |
| CN112446320B (en) | Fuzzy recognition logic gate multiple complementary partial discharge distinguishing device and method | |
| CN113672658B (en) | Power equipment online monitoring error data identification method based on complex correlation coefficient | |
| CN203178422U (en) | Partial discharge measuring system | |
| CN213069198U (en) | Source range neutron detector fault diagnosis device | |
| CN114118166A (en) | Transformer fault diagnosis system and method based on intelligent frequency spectrum monitoring |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |