CN117471303A - High-voltage isolating switch fault diagnosis method and terminal - Google Patents
High-voltage isolating switch fault diagnosis method and terminal Download PDFInfo
- Publication number
- CN117471303A CN117471303A CN202311502218.2A CN202311502218A CN117471303A CN 117471303 A CN117471303 A CN 117471303A CN 202311502218 A CN202311502218 A CN 202311502218A CN 117471303 A CN117471303 A CN 117471303A
- Authority
- CN
- China
- Prior art keywords
- curve
- interpolation
- power curve
- standard power
- isolating switch
- 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
- 238000000034 method Methods 0.000 title claims abstract description 74
- 238000003745 diagnosis Methods 0.000 title claims abstract description 43
- 230000006399 behavior Effects 0.000 claims abstract description 29
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 26
- 230000009471 action Effects 0.000 claims abstract description 25
- 238000012545 processing Methods 0.000 claims abstract description 20
- 230000011218 segmentation Effects 0.000 claims description 49
- 230000001174 ascending effect Effects 0.000 claims description 21
- 238000004364 calculation method Methods 0.000 claims description 13
- 238000005259 measurement Methods 0.000 claims description 11
- 238000004590 computer program Methods 0.000 claims description 8
- 230000001419 dependent effect Effects 0.000 claims description 5
- 230000008569 process Effects 0.000 abstract description 18
- 238000012544 monitoring process Methods 0.000 abstract description 15
- 230000000875 corresponding effect Effects 0.000 description 24
- 230000008859 change Effects 0.000 description 7
- 230000002159 abnormal effect Effects 0.000 description 6
- 238000001514 detection method Methods 0.000 description 6
- 238000004458 analytical method Methods 0.000 description 5
- 238000002955 isolation Methods 0.000 description 5
- 238000000691 measurement method Methods 0.000 description 5
- 230000009286 beneficial effect Effects 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 238000011160 research Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 239000011159 matrix material Substances 0.000 description 3
- 238000012552 review Methods 0.000 description 3
- 238000010845 search algorithm Methods 0.000 description 3
- 230000000087 stabilizing effect Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 238000003909 pattern recognition Methods 0.000 description 2
- 230000000630 rising effect Effects 0.000 description 2
- 101100136092 Drosophila melanogaster peng gene Proteins 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 240000005373 Panax quinquefolius Species 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 229910052956 cinnabar Inorganic materials 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 235000003642 hunger Nutrition 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000007257 malfunction Effects 0.000 description 1
- 238000005191 phase separation Methods 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 230000037351 starvation Effects 0.000 description 1
- 230000007847 structural defect Effects 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/327—Testing of circuit interrupters, switches or circuit-breakers
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention discloses a fault diagnosis method and terminal for a high-voltage isolating switch, wherein the fault diagnosis method comprises the following steps: acquiring a real-time power curve and a standard power curve of a high-voltage isolating switch; performing characteristic point interpolation processing on the real-time power curve to obtain an interpolation curve with the same length as the standard power curve; discretizing the interpolation curve, analyzing the similarity of the power curve in each discrete area and the corresponding discrete area on the standard power curve by adopting a long public subsequence algorithm, and obtaining a fault area with the similarity exceeding a threshold value; and determining the fault behaviors of the feature points corresponding to the fault areas. The invention can realize real-time monitoring of the action process of the isolating switch, ensure accurate completion of the opening and closing actions, effectively simplify the data processing flow and improve the fault diagnosis efficiency and applicability of the GIS isolating switch.
Description
Technical Field
The invention relates to the technical field of state monitoring and fault detection of primary equipment of power transformation, in particular to a fault diagnosis method and terminal of a high-voltage isolating switch.
Background
The gas-insulated fully-enclosed combined electrical apparatus (Gas Insulated Switchgear, abbreviated as GIS) is a key element for realizing the segmentation and isolation of electrical equipment in a power system. However, as a high-voltage isolating switch, a mechanical structure of the GIS isolating switch is easily damaged by assembly errors and long-term use, thereby causing a malfunction. The problem that the opening and closing are not in place is a great threat to the safety of a power system and personnel.
Because the GIS isolating switch has a closed structure, the opening and closing positions can be confirmed only through an opening and closing indication board linked with the operating mechanism. However, the switching-on/off indication board information cannot effectively judge the problem that switching-on/off is not in place caused by defects such as phase separation of a conductive part, breakage of a connecting rod and the like. Although some auxiliary detection means such as micro switches and gesture sensing exist to detect faults of an operating mechanism and a previous transmission path, the auxiliary detection means cannot meet the requirement of reliably judging the switching position and the switching position of the GIS. Therefore, a reliable GIS disconnecting switch switching position detection means is urgently needed in industry.
In order to solve the problem, in recent years, many researchers and engineers actively explore and develop various GIS isolating switch on-off position detection technologies. One of the main methods is to use advanced sensing technologies, such as driving motor power, vibration signal sensing, etc., to monitor the position and state of the internal elements of the GIS in real time. The sensors can accurately sense the opening and closing positions, and meanwhile, structural defects which can cause out-of-place problems can be monitored.
In addition to sensing technology, some methods based on image processing and pattern recognition are applied to the on-off position detection of the GIS isolating switch. Through installing the camera in the GIS, the image shot in real time is transmitted to the computer system, and intelligent analysis and judgment on the positions of the elements in the GIS can be realized by utilizing an image processing algorithm and a pattern recognition technology. The method has the characteristics of non-contact and strong real-time performance, and can accurately judge whether the separating and combining positions are correct.
However, in the power system, after the on-off action instruction of the isolating switch is issued, a key concern is whether the equipment completes the corresponding action, and the abnormal state of the on-off position of the isolating switch can be found in time so as to avoid possible running accidents. This is critical for safe and stable operation of the power system.
The isolating switch is used as a key element in a power system, and the opening and closing actions of the isolating switch are required to be accurate and reliable. After the opening and closing action instruction is issued, if the equipment fails to complete the corresponding action, abnormal circuit connection in the power system can be caused, and even serious accidents are caused. Therefore, the switching-on and switching-off positions of the isolating switch can accurately respond to the instruction, and the normal operation of the power system is ensured.
In addition, the abnormal state of the separation switch on/off position is also important to find timely. The isolating switch may be failed or damaged during operation, so that the switching position of the isolating switch cannot be switched accurately. If the power system is continued to operate in such an abnormal state, serious consequences such as disconnection, overload, short circuit, etc. may be caused, and even fire and explosion may be caused. Therefore, the abnormal states must be found in time, the operation is stopped in time, corresponding measures are taken, and the safety of the power system is ensured.
To achieve this, the existing domestic literature is as follows:
(1) A dynamic time warping based time series similarity measurement method (authors: yu Hongfei, zhang Jie) describes a Dynamic Time Warping (DTW) based time series similarity measurement method for processing one-dimensional curve data.
The disadvantages are as follows:
computational complexity: the Dynamic Time Warping (DTW) method has high computational complexity, and may require a long computational time, especially for long time sequences;
sensitivity to parameters: the DTW method needs to select proper parameters, such as bandwidth parameters, so that the effect on the result is large, and incorrect selection can lead to inaccurate similarity measurement;
Not applicable to large-scale data: the DTW method has a large demand for memory and computing resources, and is not well suited for large-scale time-series data.
(2) One-dimensional time series similarity measurement method studies (authors: zhang Yang, peng Hua), which studied one-dimensional time series similarity measurement methods, including shape-based methods and subsequence matching-based methods.
The disadvantages are as follows:
the breadth is limited: the paper may not go deep into the specific details and limitations of each approach, only providing overview information;
update possibility: the research in the field of time series similarity metrics is active and this overview may not cover the latest methods and advances.
(3) Research progress in time series similarity calculation (authors: ma Sheng man, cinnabar wave), which reviews the research progress in time series similarity calculation, covers a variety of methods for processing one-dimensional curve similarity.
The disadvantages are as follows:
lack of specific method descriptions: the paper may focus mainly on an overview of the progress of the study, without providing a detailed description of the method, and may require review of other specific literature to see details of the method;
insufficient experimental assessment: some reviews may not provide adequate experimental assessment to demonstrate the effectiveness of the various methods, which may require additional experimental research to verify.
(4) Dynamic time warping based time series similarity search algorithm (authors: yang Tianran, liu Guoping), which describes a dynamic time warping based time series similarity search algorithm for processing the similarity of one-dimensional curves.
The disadvantages are as follows:
parameter selection challenges: algorithms associated with DTW typically require the selection of parameters, such as bandwidth parameters, which may require different parameter settings for different data sets;
performance degradation: for very large time series databases, DTW-based search algorithms may suffer from reduced performance due to the high computational complexity.
In addition, there are also foreign documents: fastDTW Toward accurate dynamic time warping in linear time and space (by Stan Salvador and Philip Chan), which describes an algorithm for accelerating Dynamic Time Warping (DTW) to process the similarity of time series data; SAX Symbolic aggregate approximation (by Jesseica Lin, emmonn Keogh, li Wei, and Stefano Lonardi) describes a method of symbolizing time series data that can be used for curve similarity analysis; shape-based retrieval and analysis oftime series data (by Stefano Ratanamahatana and Eamonn Keogh), which investigated a Shape-based time series data retrieval and analysis method, covers the similarity of one-dimensional curves; time series motif discovery: A recent overview (by Dina Goldin and Pavel) ) This paper discusses methods of time series pattern discovery, including related techniques of curve similarity; locally Adaptive Dimensionality Reduction for Indexing Large Time Series Databases (by Jesin Zakaria, elkea. Rundensteiner, and Matthew o.ward), which describes a local adaptation method for down-scaling and indexing large time series databases to support curve similarity searches; matrix Profile I All Pairs Similarity Joins for Time Series: A Unifying View that Includes Motifs, discovery, and shape (by chip-Chia Michael Yeh, yan Zhu, liudmila Ulanova, nurjahan Begum, YIFIFEI Ding, hoangAnh Dau, diego Furtado Silva, abdullah Muen, emamnn J.Keogh), which describes a comprehensive time series similarity measurement method that can handle various features of one-dimensional curves, such as pattern, dissonance and shape; shape-aware Time Series Classificationwith Exponential Eigenfunction Bases (by Yan Zhu, chip-Chia MichaelYeh, zachary Zimmerman, kaveh Kamgar, and Eimonn J.Keogh) describes a time series classification method based on exponential characteristics, taking into account shape information of curves.
The above-mentioned disadvantages exist more or less in the foreign countries, so how to monitor the action process of the isolating switch in real time, ensure the accurate completion of the opening and closing actions, and respond to the abnormal situation in time gradually becomes a technical problem to be solved urgently.
Disclosure of Invention
The technical problems to be solved by the invention are as follows: the fault diagnosis method and the terminal for the high-voltage isolating switch can simplify the data processing flow, monitor the action process of the isolating switch in real time, ensure the accurate completion of the opening and closing actions, and improve the fault diagnosis efficiency and the applicability.
In order to solve the technical problems, the invention adopts the following technical scheme:
a fault diagnosis method for a high-voltage isolating switch comprises the following steps:
s1, acquiring a real-time power curve and a standard power curve of a high-voltage isolating switch;
s2, performing characteristic point interpolation processing on the real-time power curve to obtain an interpolation curve with the same length as the standard power curve;
s3, discretizing the interpolation curves, analyzing the similarity of the power curves in each discrete area and the corresponding discrete areas on the standard power curves by adopting a long public subsequence algorithm, and obtaining fault areas with the similarity exceeding a threshold value;
S4, determining the fault behaviors of the feature points corresponding to the fault areas.
In order to solve the technical problems, the invention adopts another technical scheme that:
a high voltage isolator fault diagnosis terminal comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the following steps when executing the computer program:
s1, acquiring a real-time power curve and a standard power curve of a high-voltage isolating switch;
s2, performing characteristic point interpolation processing on the real-time power curve to obtain an interpolation curve with the same length as the standard power curve;
s3, discretizing the interpolation curves, analyzing the similarity of the power curves in each discrete area and the corresponding discrete areas on the standard power curves by adopting a long public subsequence algorithm, and obtaining fault areas with the similarity exceeding a threshold value;
s4, determining the fault behaviors of the feature points corresponding to the fault areas.
The invention has the beneficial effects that: the power curve of the driving motor of the GIS isolating switch is used for fault diagnosis, and because the power change of the motor can reflect the dynamic behavior of the driving part, the fault point and the fault type of the fault point are judged, so that the real-time power curve and the standard power curve of the high-voltage isolating switch are obtained, the real-time power curve is interpolated to ensure that the lengths of the two curves which are compared are consistent, the similarity of the discretized interpolation curve and the standard power curve is analyzed through a long common subsequence algorithm, and a discrete area with larger similarity difference is found out, so that the characteristic point and the fault behavior corresponding to the fault are determined in the area, the real-time monitoring of the action process of the isolating switch is realized, the accurate completion of the switching action is ensured, the data processing flow is effectively simplified, and the fault diagnosis efficiency and the applicability of the GIS isolating switch are improved.
Drawings
FIG. 1 is a general flow chart of a fault diagnosis method for a high-voltage isolating switch according to an embodiment of the invention;
FIG. 2 is a graph showing the comparison of interpolation curves and their characteristic points with standard power curves and their characteristic points according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of a fault diagnosis terminal of a high-voltage isolating switch according to an embodiment of the present invention.
Description of the reference numerals:
1. a fault diagnosis terminal of a high-voltage isolating switch; 2. a memory; 3. a processor.
Detailed Description
In order to describe the technical contents, the achieved objects and effects of the present invention in detail, the following description will be made with reference to the embodiments in conjunction with the accompanying drawings.
Referring to fig. 1 and 2, a fault diagnosis method for a high-voltage isolating switch includes the steps of:
s1, acquiring a real-time power curve and a standard power curve of a high-voltage isolating switch;
s2, performing characteristic point interpolation processing on the real-time power curve to obtain an interpolation curve with the same length as the standard power curve;
s3, discretizing the interpolation curves, analyzing the similarity of the power curves in each discrete area and the corresponding discrete areas on the standard power curves by adopting a long public subsequence algorithm, and obtaining fault areas with the similarity exceeding a threshold value;
S4, determining the fault behaviors of the feature points corresponding to the fault areas.
From the above description, the beneficial effects of the invention are as follows: the power curve of the driving motor of the high-voltage isolating switch is used for fault diagnosis, and because the power change of the motor can reflect the dynamic behavior of the driving component, the fault point and the fault type of the fault point are helped to be judged, so that the real-time power curve and the standard power curve of the high-voltage isolating switch are obtained, the real-time power curve is interpolated to ensure that the lengths of the two curves which are compared are consistent, the similarity of the discretized interpolation curve and the standard power curve is analyzed through a long common subsequence algorithm, and a discrete area with larger similarity difference is found out, so that the characteristic point and the fault behavior corresponding to the fault are determined in the area, the real-time monitoring of the action process of the isolating switch is realized, the accurate completion of the switching action is ensured, the data processing flow is effectively simplified, and the fault diagnosis efficiency and the applicability of the high-voltage isolating switch are improved.
Further, the standard power curve is a power curve of a driving motor of the high-voltage isolating switch in a healthy state;
The real-time power curve is a power curve of a driving motor of the high-voltage isolating switch in real-time opening and closing actions;
the characteristic points comprise a starting point, an operating point, a stopping point, a peak point, a slope maximum point, a slope minimum point, an area maximum point and an area minimum point;
the step S2 specifically comprises the following steps:
interpolation of feature points in the real-time power curve using linear interpolation, at each of which two data points (x 1 ，y 1 ) And (x) 2 ，y 2 ) And connecting the two data points to obtain a straight line xy, and estimating the position x by using the straight line xy 1 And x 2 The value of x and the value of y are the same as the length of the standard power curve, and the linear interpolation formula is as follows:
where the x value represents the value of the independent variable and the y value represents the value of the dependent variable.
The above description shows that the real-time power curve can be obtained by monitoring the state of the driving motor of the high-voltage isolating switch in real time, so as to ensure the timeliness of fault monitoring, and the standard power curve is measured at the lower side of the high-voltage isolating switch in a healthy state, so that the fault position can be quickly found by comparing the similarity of the real-time power curve and the standard power curve; meanwhile, the positions of characteristic points representing the key behaviors of the high-voltage isolating switch driving motor, such as a starting point, an operating point, a stopping point, a peak point and the like, are selected on the real-time power curve to conduct linear interpolation, and the real-time power curve is interpolated to the same length as the standard power curve, so that the two curves can be conveniently compared in the subsequent steps.
Further, the steps S2 and S3 further include the steps of:
s23, matching the optimal correspondence between the interpolation curve and the standard power curve by adopting a mean square error method to obtain a mean square error minimum offset as follows:
wherein MSE represents mean square error, n is the length of the data set in the interpolation curve or the standard power curve, i E [1, n]，y 1，i The value of the ith data point of the 1 st data set in the standard power curve, y intrp,i Values for the i data point in the 2 nd data set in the interpolation curve or the standard power curve;
s24, re-marking the characteristic points matched with the standard power curve on an interpolation curve according to the minimum offset.
As can be seen from the above description, the best correspondence between the interpolation curve and the standard power curve can be found by the mean square error method, that is, an offset that minimizes the mean square error is found to correct the feature points marked on the interpolation curve, so as to solve the problem that smiling table change or delay may exist in the action of the high-voltage isolating switch, so that the position of the feature points on the real-time power curve may be offset.
Further, the step S3 specifically includes:
S31, discretizing the interpolation curve according to the slope to obtain an interpolation segmented curve comprising a plurality of segmented intervals, distributing labels or symbols for each segmented interval, discretizing the standard power curve according to the segmented intervals, and obtaining a standard power segmented curve comprising a plurality of segmented intervals;
s32, representing each section of curve in the interpolation segmentation curve and the standard power segmentation curve as a character string or sequence to obtain an interpolation sequence and a standard sequence;
s33, calculating the longest public subsequence between the interpolation sequence and the standard sequence by using a long public subsequence LCS algorithm;
s34, obtaining measurement of similarity of two curves by analyzing the ratio between the longest public subsequence and the length of each sequence in the interpolation sequence and the standard sequence;
s35, analyzing the characteristic points corresponding to the sequences with the measurement exceeding the threshold value in the segment section where the interpolation segment curve is located, and obtaining corresponding fault behaviors.
As can be seen from the above description, the interpolation curve and the standard power curve are discretized based on the slope of the interpolation curve, so that it is not difficult to know that the slope of the curve has three states of ascending, descending or stable, therefore, a plurality of ascending segmented sections, descending segmented sections and stable segmented sections which are unequal can be obtained by dividing the sections according to the slope, and each segmented section is marked by a label or a symbol, so that the subsequent determination of the segmented section process where the fault is located is more efficient; meanwhile, the interpolation segmentation curve and the standard power segmentation curve are serialized according to unequal segmentation intervals to obtain an interpolation sequence and a standard sequence, the similarity calculation between the two sequence curves can be carried out by adopting a long common subsequence LCS algorithm, and finally, the fault point and the fault behavior are analyzed according to the calculation result, so that the whole process is simple and efficient.
Further, the slope calculation in step S31 specifically includes:
for discrete data points in the interpolation piecewise curve, a difference method is used for estimating the slope, and the data point is set as y 1 ，y 2 ，y 3 ，…，y m The slope is obtained as follows:
slope j ＝y j+1 -y j (3)；
wherein j is E [1, m ], m is the total number of segments of the interpolation segmentation curve and the standard power curve;
for non-equally spaced data points on the x-axis, equation (3) above is replaced with:
according to the preset threshold value epsilon, if the slope of the segmented interval j The E is the ascending segmentation section;
if slope j Fall-down segmentA section;
if |slope j The section is a stable section if the grade is less than or equal to E;
the successive ascending, descending or stationary segmentation intervals are combined into one segmentation interval, resulting in an interpolated segmentation curve with a plurality of ascending, descending and/or stationary segmentation intervals.
As can be seen from the above description, by calculating the slope of the discrete data point of the interpolation curve and classifying, the merging is performed in the segment intervals where the slope of one continuous class is located, so that the subsequent process of serializing or serializing each segment interval can be effectively simplified, and the efficiency of fault diagnosis of the voltage isolating switch is further improved.
Referring to fig. 3, a high voltage isolating switch fault diagnosis terminal includes a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the processor implements the following steps when executing the computer program:
s1, acquiring a real-time power curve and a standard power curve of a high-voltage isolating switch;
s2, performing characteristic point interpolation processing on the real-time power curve to obtain an interpolation curve with the same length as the standard power curve;
s3, discretizing the interpolation curves, analyzing the similarity of the power curves in each discrete area and the corresponding discrete areas on the standard power curves by adopting a long public subsequence algorithm, and obtaining fault areas with the similarity exceeding a threshold value;
s4, determining the fault behaviors of the feature points corresponding to the fault areas.
From the above description, the beneficial effects of the invention are as follows: based on the same technical conception, the high-voltage isolating switch fault diagnosis method is matched with the high-voltage isolating switch fault diagnosis method, the high-voltage isolating switch fault diagnosis terminal is provided, the power curve of the driving motor of the high-voltage isolating switch is used for fault diagnosis, and the power change of the motor can reflect the dynamic behavior of the driving part, so that the fault point and the fault type of the motor can be judged, the real-time power curve and the standard power curve of the high-voltage isolating switch are obtained, the real-time power curve is interpolated to ensure that the lengths of the two curves which are compared are consistent, the similarity of the discretized interpolation curve and the standard power curve is analyzed through a long common subsequence algorithm, and a discrete area with larger similarity difference is found, so that the characteristic point and the fault behavior corresponding to the fault are determined in the area, the real-time monitoring of the action process of the isolating switch is realized, the accurate completion of the switching action is ensured, the data processing flow is effectively simplified, and the fault diagnosis efficiency and the applicability of the high-voltage isolating switch are improved.
Further, the standard power curve is a power curve of a driving motor of the high-voltage isolating switch in a healthy state;
the real-time power curve is a power curve of a driving motor of the high-voltage isolating switch in real-time opening and closing actions;
the characteristic points comprise a starting point, an operating point, a stopping point, a peak point, a slope maximum point, a slope minimum point, an area maximum point and an area minimum point;
the step S2 specifically comprises the following steps:
interpolation of feature points in the real-time power curve using linear interpolation, at each of which two data points (x 1 ，y 1 ) And (x) 2 ，y 2 ) And connecting the two data points to obtain a straight line xy, and estimating the position x by using the straight line xy 1 And x 2 The value of x and the value of y are the same as the length of the standard power curve, and the linear interpolation formula is as follows:
where the x value represents the value of the independent variable and the y value represents the value of the dependent variable.
The above description shows that the real-time power curve can be obtained by monitoring the state of the driving motor of the high-voltage isolating switch in real time, so as to ensure the timeliness of fault monitoring, and the standard power curve is measured at the lower side of the high-voltage isolating switch in a healthy state, so that the fault position can be quickly found by comparing the similarity of the real-time power curve and the standard power curve; meanwhile, the positions of characteristic points representing the key behaviors of the high-voltage isolating switch driving motor, such as a starting point, an operating point, a stopping point, a peak point and the like, are selected on the real-time power curve to conduct linear interpolation, and the real-time power curve is interpolated to the same length as the standard power curve, so that the two curves can be conveniently compared in the subsequent steps.
Further, the steps S2 and S3 further include the steps of:
s23, matching the optimal correspondence between the interpolation curve and the standard power curve by adopting a mean square error method to obtain a mean square error minimum offset as follows:
wherein MSE represents mean square error, n is the length of the data set in the interpolation curve or the standard power curve, i E [1, n]，y 1，i The value of the ith data point of the 1 st data set in the standard power curve, y intrp,i Values for the i data point in the 2 nd data set in the interpolation curve or the standard power curve;
s24, re-marking the characteristic points matched with the standard power curve on an interpolation curve according to the minimum offset.
As can be seen from the above description, the best correspondence between the interpolation curve and the standard power curve can be found by the mean square error method, that is, an offset that minimizes the mean square error is found to correct the feature points marked on the interpolation curve, so as to solve the problem that smiling table change or delay may exist in the action of the high-voltage isolating switch, so that the position of the feature points on the real-time power curve may be offset.
Further, the step S3 specifically includes:
S31, discretizing the interpolation curve according to the slope to obtain an interpolation segmented curve comprising a plurality of segmented intervals, distributing labels or symbols for each segmented interval, discretizing the standard power curve according to the segmented intervals, and obtaining a standard power segmented curve comprising a plurality of segmented intervals;
s32, representing each section of curve in the interpolation segmentation curve and the standard power segmentation curve as a character string or sequence to obtain an interpolation sequence and a standard sequence;
s33, calculating the longest public subsequence between the interpolation sequence and the standard sequence by using a long public subsequence LCS algorithm;
s34, obtaining measurement of similarity of two curves by analyzing the ratio between the longest public subsequence and the length of each sequence in the interpolation sequence and the standard sequence;
s35, analyzing the characteristic points corresponding to the sequences with the measurement exceeding the threshold value in the segment section where the interpolation segment curve is located, and obtaining corresponding fault behaviors.
As can be seen from the above description, the interpolation curve and the standard power curve are discretized based on the slope of the interpolation curve, so that it is not difficult to know that the slope of the curve has three states of ascending, descending or stable, therefore, a plurality of ascending segmented sections, descending segmented sections and stable segmented sections which are unequal can be obtained by dividing the sections according to the slope, and each segmented section is marked by a label or a symbol, so that the subsequent determination of the segmented section process where the fault is located is more efficient; meanwhile, the interpolation segmentation curve and the standard power segmentation curve are serialized according to unequal segmentation intervals to obtain an interpolation sequence and a standard sequence, the similarity calculation between the two sequence curves can be carried out by adopting a long common subsequence LCS algorithm, and finally, the fault point and the fault behavior are analyzed according to the calculation result, so that the whole process is simple and efficient.
Further, the slope calculation in step S31 specifically includes:
for discrete data points in the interpolation piecewise curve, a difference method is used for estimating the slope, and the data point is set as y 1 ，y 2 ，y 3 ，…，y m The slope is obtained as follows:
slope j ＝y j+1 -y j (3)；
wherein j is E [1, m ], m is the total number of segments of the interpolation segmentation curve and the standard power curve;
for non-equally spaced data points on the x-axis, equation (3) above is replaced with:
according to the preset threshold value epsilon, if the slope of the segmented interval j The E is the ascending segmentation section;
if slope j The < - ∈is a descending segmentation interval;
if |slope j The section is a stable section if the grade is less than or equal to E;
the successive ascending, descending or stationary segmentation intervals are combined into one segmentation interval, resulting in an interpolated segmentation curve with a plurality of ascending, descending and/or stationary segmentation intervals.
As can be seen from the above description, by calculating the slope of the discrete data point of the interpolation curve and classifying, the merging is performed in the segment intervals where the slope of one continuous class is located, so that the subsequent process of serializing or serializing each segment interval can be effectively simplified, and the efficiency of fault diagnosis of the voltage isolating switch is further improved.
The fault diagnosis method of the high-voltage isolating switch is used for monitoring whether fault behaviors occur in real time when the high-voltage isolating switch is in driving operation and performing corresponding fault diagnosis, and is specifically described below with reference to the embodiment.
Referring to fig. 1 to 2, a first embodiment of the present invention is as follows:
a fault diagnosis method for a high-voltage isolating switch, as shown in figure 1, comprises the following steps:
s1, acquiring a real-time power curve and a standard power curve of the high-voltage isolating switch.
In this embodiment, taking a GIS isolation switch as an example, since mechanical fault diagnosis of the GIS isolation switch is a complex problem, fault diagnosis using a power curve of a driving motor is a smart method. The power variation of the motor can reflect the dynamic dimensions of the drive components, which can be affected by mechanical failure. Therefore, by acquiring the real-time power curve of the isolating switch and comparing the real-time power curve with the standard power curve, possible faults can be effectively identified.
In this embodiment, the standard power curve is a power curve of a driving motor of the high-voltage isolation switch in a healthy state, and the real-time power curve is a power curve of the driving motor of the high-voltage isolation switch in a real-time opening and closing operation.
However, the real-time power curve changes correspondingly with each action of the driving motor, and the running time of the driving motor may be affected by mechanical faults, so that the lengths of the real-time power curve and the standard power curve may be different, and therefore, the lengths of the real-time power curve and the standard power curve need to be aligned, which is specifically as follows:
and S2, performing characteristic point interpolation processing on the real-time power curve to obtain an interpolation curve with the same length as the standard power curve. In this embodiment, the feature points include a start point, an operation point, a stop point, a peak point, a slope maximum point, a slope minimum point, an area maximum point, and an area minimum point.
In this embodiment, step S2 is specifically:
interpolation of feature points in a real-time power curve using linear interpolation, at each feature point two data points (x 1 ，y 1 ) And (x) 2 ，y 2 ) And connecting the two data points to obtain a straight line xy, and estimating the position x by using the straight line xy 1 And x 2 The value x and the value y between the two values are used for obtaining an interpolation curve with the same length as the standard power curve, and a linear interpolation formula is as follows:
where the x value represents the value of the independent variable and the y value represents the value of the dependent variable.
The real-time power curve is interpolated to the same length as the standard power curve using a linear difference method in order to ensure that the two curves can be conveniently compared in a subsequent step.
In this embodiment, the interpolation curve can be noted as y interp MATLAB is employed to test for linear differences, in which an interp1 function is used to perform one-dimensional interpolation, for which the syntax of the function is:
Y_interp＝interp1(X,Y,X_new,'linear')；
wherein:
x is the X coordinate of the original data.
Y is the Y coordinate of the raw data.
X new is the new X coordinate for interpolation.
Y_interpolation is the interpolation result at x_new.
Defining a standard power curve as a1 and a real-time power curve as a2, in order to interpolate the real-time power curve to the same length as the standard power curve:
the original x-coordinate is set to 1:length (a 2), which represents the index of the real-time power curve.
The original y-coordinate is set to the value of the real-time power curve, a2.
The new x-coordinate is set to linspace (1, length (a 2), length (a 1)) to interpolate the real-time power curve to the same length as the standard power curve.
Linear interpolation is performed using an interpolation 1 function.
interp_curve＝interp1(1:length(a2),y2,linspace(1,length(a2),length(y1)),'linear')；
Interp_curve is an interpolated curve having the same length as the standard power curve a1, as shown in FIG. 2.
S3, discretizing the interpolation curve, analyzing the similarity of the power curve in each discrete area and the corresponding discrete area on the standard power curve by adopting a long public subsequence algorithm, and obtaining a fault area with the similarity exceeding a threshold value.
S4, determining fault behaviors of the feature points corresponding to the fault areas.
In this embodiment, the power curve of the driving motor of the high-voltage isolating switch is used for fault diagnosis, because the power change of the motor can reflect the dynamic behavior of the driving component, thereby helping to judge the fault point and the fault type thereof, the real-time power curve and the standard power curve of the high-voltage isolating switch are obtained, the real-time power curve is interpolated to ensure that the lengths of the two curves which are compared are consistent, and then the similarity of the discretized interpolation curve and the standard power curve is analyzed through a long common subsequence algorithm, so as to find out the discrete area with larger similarity difference, thereby determining the feature point and the fault behavior corresponding to the fault in the area, realizing real-time monitoring of the action process of the isolating switch, ensuring the accurate completion of the switching action, effectively simplifying the data processing flow, and improving the fault diagnosis efficiency and applicability of the high-voltage isolating switch.
Meanwhile, the real-time power curve can be obtained by monitoring the state of the driving motor of the high-voltage isolating switch in real time, so that the timeliness of fault monitoring is ensured, and the standard power curve is measured at the side of the high-voltage isolating switch in a healthy state, so that the fault position can be quickly found by carrying out similarity comparison on the real-time power curve and the standard power curve; meanwhile, the positions of characteristic points representing the key behaviors of the high-voltage isolating switch driving motor, such as a starting point, an operating point, a stopping point, a peak point and the like, are selected on the real-time power curve to conduct linear interpolation, and the real-time power curve is interpolated to the same length as the standard power curve, so that the two curves can be conveniently compared in the subsequent steps.
Referring to fig. 1 and 2, a second embodiment of the invention is as follows:
in the first embodiment, the fault diagnosis method for a high-voltage isolating switch further includes the steps between steps S2 and S3:
in order to obtain the optimal correspondence between the real-time power curve and the standard power curve, S23, matching the optimal correspondence between the interpolation curve and the standard power curve by adopting a mean square error method to obtain the minimum deviation of the mean square error as follows:
wherein MSE represents mean square error, n is the length of the data set in the interpolation curve or standard power curve, i E [1, n]，y 1，i The value of the ith data point, y, of the 1 st data set in the standard power curve intrp,i Is the value of the i-th data point in the 2 nd data set in the interpolation curve or standard power curve.
S24, re-marking the characteristic points matched with the standard power curve on the interpolation curve according to the minimum offset.
The characteristic points are outside the characteristic points, and the characteristic points in the real-time power curve or the standard power curve can be outside the characteristic points. Other features, such as peaks, slopes, areas, etc., may also be extracted from the feature points, which may be related to mechanical failure, again without starvation.
Meanwhile, embodiments may also be built by modeling, if there is a large amount of sample data (power curves in normal and fault states), it may be considered to automatically identify faults using a machine learning model. For example, a support vector machine, random forest, or deep learning model may be used.
Real-time monitoring in order to detect the state of the GIS isolating switch in real time, the mean square error algorithm can be deployed into a real-time monitoring system.
In this embodiment, the best correspondence between the interpolation curve and the standard power curve can be found by the mean square error method, that is, an offset which minimizes the mean square error is found to correct the feature points marked on the interpolation curve, so as to solve the problem that smiling table change or delay may exist in the action of the high-voltage isolating switch, so that the position of the feature points on the real-time power curve may be offset.
The third embodiment of the invention is as follows:
in this embodiment, the step S3 specifically includes:
first, continuous curve data needs to be discretized.
S31, discretizing the interpolation curve according to the slope to obtain an interpolation segmented curve comprising a plurality of segmented sections, distributing labels or symbols for each segmented section, discretizing the standard power curve according to the segmented sections, and obtaining the standard power segmented curve comprising the segmented sections.
For example, each portion of the curve may be discretized by a slope: ascending, descending or stabilizing. Thus, an ascending paragraph may be marked U, a descending paragraph with D, and a stationary paragraph with S.
The slope calculation in step S31 specifically includes:
for discrete data points in the interpolation segmented curve, a difference method is used for estimating the slope, and the data point is set as y 1 ，y 2 ，y 3 ，…，y m The slope is obtained as follows:
slope j ＝y j+1 -y j (3)；
wherein j is [1, m ], and m is the total number of segments of the interpolation segmentation curve and the standard power curve.
For equally spaced data of the wanted axis, the difference of the above formula (3) is sufficient; for non-equally spaced data points on the x-axis, it is also necessary to divide the difference value by the spacing of x, i.e., replace equation (3) above with:
defining a threshold e, which may depend on the scale and accuracy of the data, is classified according to the slope value as follows:
if the slope of the segmented interval j The E is the ascending segmentation section;
if slope j The < - ∈is a descending segmentation interval;
if |slope j And the section is a stable section if the grade is less than or equal to E.
The successive ascending, descending, or stationary segmenting intervals are combined into one segmenting interval to obtain an interpolated segmenting curve having a plurality of ascending, descending, and/or stationary segmenting intervals, each segment being defined by its beginning and ending data points.
The slope of the discrete data points of the interpolation curve is calculated and classified, so that the subsequent process of carrying out character serialization or serialization on each segmented interval can be effectively simplified by merging the segmented intervals where the slopes of one continuous class are located, and the efficiency of fault diagnosis of the voltage isolating switch is further improved.
S32, representing each section of curve in the interpolation segmentation curve and the standard power segmentation curve as a character string or sequence to obtain an interpolation sequence and a standard sequence.
Such as "uuddsud.
S33, calculating the longest common subsequence between the interpolation sequence and the standard sequence by using a long common subsequence LCS algorithm.
The basic idea of the long common subsequence LCS algorithm is as follows:
a matrix L of size (n+1) x (n+1) is initialized, where two n are the lengths of the two sequences, respectively. All elements are initialized to 0.
For each pair of characters, if they are the same, the L value for that pair is the value of its upper left neighbor plus 1; otherwise, it is the maximum in its upper and left neighbors.
The maximum value in the matrix represents the length of the LCS.
S34, obtaining the measurement of the similarity of the two curves by analyzing the ratio between the longest common subsequence and the length of each sequence in the interpolation sequence and the standard sequence
S35, analyzing corresponding characteristic points of the sequence with the measurement exceeding the threshold value in the segment section where the interpolation segment curve is located, and obtaining corresponding fault behaviors.
In addition, the long common sub-LCS sequence algorithm itself may also provide which parts or features are common between the two curves.
In this embodiment, the interpolation curve and the standard power curve are discretized based on the slope of the interpolation curve, so that three states of rising, falling or stabilizing of the slope of the curve can be easily known, and therefore, a plurality of unequal rising segmented sections, falling segmented sections and stabilizing segmented sections can be obtained by dividing the slope, and each segmented section is marked by a label or a symbol, so that the subsequent determination of the segmented section process where the fault is located is more efficient; meanwhile, the interpolation segmentation curve and the standard power segmentation curve are serialized according to unequal segmentation intervals to obtain an interpolation sequence and a standard sequence, the similarity calculation between the two sequence curves can be carried out by adopting a long common subsequence LCS algorithm, and finally, the fault point and the fault behavior are analyzed according to the calculation result, so that the whole process is simple and efficient.
Referring to fig. 3, a fourth embodiment of the present invention is as follows:
A high voltage isolating switch fault diagnosis terminal 1, as shown in fig. 3, comprises a memory 2, a processor 3 and a computer program stored on the memory 2 and executable on the processor 3, the processor 3 implementing the steps of any of the above embodiments one to three when executing the computer program.
In summary, the fault diagnosis method and terminal for the high-voltage isolating switch provided by the invention have the following beneficial effects:
1. simplifying the data processing flow: the real-time power curve is interpolated and matched by mean square error, and then further analyzed by LCS, so that the whole process is more visual and simplified, and easy to understand and realize.
2. The MSE is used as a preliminary screening tool to quickly screen out the parts with higher similarity with the standard power curve by using the MSE matching, which helps to reduce the computational complexity and only the most relevant parts are subjected to deeper analysis.
3. Controllability and interpretation: parameters of the mean square error matching, such as thresholds, can be better controlled to accommodate different data conditions, and furthermore, MSE provides a direct numerical measure for ease of interpretation.
4. Relative computational efficiency: interpolation and MSE computation are generally more efficient than DTW method computation, especially for large data sets.
5. Applicability: the method is suitable for the condition of one-dimensional time sequence data, and is particularly suitable for the continuity characteristics of the data, such as a driving motor behavior curve.
The foregoing description is only illustrative of the present invention and is not intended to limit the scope of the invention, and all equivalent changes made by the specification and drawings of the present invention, or direct or indirect application in the relevant art, are included in the scope of the present invention.
Claims (10)
1. The fault diagnosis method for the high-voltage isolating switch is characterized by comprising the following steps of:
s1, acquiring a real-time power curve and a standard power curve of a high-voltage isolating switch;
s2, performing characteristic point interpolation processing on the real-time power curve to obtain an interpolation curve with the same length as the standard power curve;
s3, discretizing the interpolation curves, analyzing the similarity of the power curves in each discrete area and the corresponding discrete areas on the standard power curves by adopting a long public subsequence algorithm, and obtaining fault areas with the similarity exceeding a threshold value;
s4, determining the fault behaviors of the feature points corresponding to the fault areas.
2. The method for diagnosing a fault in a high-voltage isolating switch according to claim 1, wherein the standard power curve is a power curve of a driving motor of the high-voltage isolating switch in a healthy state;
The real-time power curve is a power curve of a driving motor of the high-voltage isolating switch in real-time opening and closing actions;
the characteristic points comprise a starting point, an operating point, a stopping point, a peak point, a slope maximum point, a slope minimum point, an area maximum point and an area minimum point;
the step S2 specifically comprises the following steps:
interpolation of feature points in the real-time power curve using linear interpolation, at each of the feature points,two data points (x 1 ，y 1 ) And (x) 2 ，y 2 ) And connecting the two data points to obtain a straight line xy, and estimating the position x by using the straight line xy 1 And x 2 The value of x and the value of y are the same as the length of the standard power curve, and the linear interpolation formula is as follows:
where the x value represents the value of the independent variable and the y value represents the value of the dependent variable.
3. The method for diagnosing a fault in a high voltage isolating switch as defined in claim 1, wherein the steps between the steps S2 and S3 further comprise the steps of:
s23, matching the optimal correspondence between the interpolation curve and the standard power curve by adopting a mean square error method to obtain a mean square error minimum offset as follows:
wherein MSE represents mean square error, n is the length of the data set in the interpolation curve or the standard power curve, i E [1, n ]，y 1，i The value of the ith data point of the 1 st data set in the standard power curve, y intrp,i Values for the i data point in the 2 nd data set in the interpolation curve or the standard power curve;
s24, re-marking the characteristic points matched with the standard power curve on an interpolation curve according to the minimum offset.
4. The method for diagnosing a fault of a high-voltage isolating switch according to claim 1, wherein the step S3 is specifically:
s31, discretizing the interpolation curve according to the slope to obtain an interpolation segmented curve comprising a plurality of segmented intervals, distributing labels or symbols for each segmented interval, discretizing the standard power curve according to the segmented intervals, and obtaining a standard power segmented curve comprising a plurality of segmented intervals;
s32, representing each section of curve in the interpolation segmentation curve and the standard power segmentation curve as a character string or sequence to obtain an interpolation sequence and a standard sequence;
s33, calculating the longest public subsequence between the interpolation sequence and the standard sequence by using a long public subsequence LCS algorithm;
s34, obtaining measurement of similarity of two curves by analyzing the ratio between the longest public subsequence and the length of each sequence in the interpolation sequence and the standard sequence;
S35, analyzing the characteristic points corresponding to the sequences with the measurement exceeding the threshold value in the segment section where the interpolation segment curve is located, and obtaining corresponding fault behaviors.
5. The method according to claim 4, wherein the slope calculation in step S31 is specifically:
for discrete data points in the interpolation piecewise curve, a difference method is used for estimating the slope, and the data point is set as y 1 ，y 2 ，y 3 ，…，y m The slope is obtained as follows:
slope j ＝y j+1 -y j (3)；
wherein j is E [1, m ], m is the total number of segments of the interpolation segmentation curve and the standard power curve;
for non-equally spaced data points on the x-axis, equation (3) above is replaced with:
according to the preset threshold value epsilon, if the slope of the segmented interval j >.e., then it is the ascending segmentA section;
if slope j The < - ∈is a descending segmentation interval;
if |slope j The section is a stable section if the grade is less than or equal to E;
the successive ascending, descending or stationary segmentation intervals are combined into one segmentation interval, resulting in an interpolated segmentation curve with a plurality of ascending, descending and/or stationary segmentation intervals.
6. A high voltage isolator fault diagnosis terminal comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the following steps when executing the computer program:
S1, acquiring a real-time power curve and a standard power curve of a high-voltage isolating switch;
s2, performing characteristic point interpolation processing on the real-time power curve to obtain an interpolation curve with the same length as the standard power curve;
s3, discretizing the interpolation curves, analyzing the similarity of the power curves in each discrete area and the corresponding discrete areas on the standard power curves by adopting a long public subsequence algorithm, and obtaining fault areas with the similarity exceeding a threshold value;
s4, determining the fault behaviors of the feature points corresponding to the fault areas.
7. The high-voltage isolating switch fault diagnosis terminal according to claim 6, wherein the standard power curve is a driving motor power curve of the high-voltage isolating switch in a healthy state;
the real-time power curve is a power curve of a driving motor of the high-voltage isolating switch in real-time opening and closing actions;
the characteristic points comprise a starting point, an operating point, a stopping point, a peak point, a slope maximum point, a slope minimum point, an area maximum point and an area minimum point;
the step S2 specifically comprises the following steps:
usingLinear interpolation interpolates feature points in the real-time power curve, at each of which two data points (x 1 ，y 1 ) And (x) 2 ，y 2 ) And connecting the two data points to obtain a straight line xy, and estimating the position x by using the straight line xy 1 And x 2 The value of x and the value of y are the same as the length of the standard power curve, and the linear interpolation formula is as follows:
where the x value represents the value of the independent variable and the y value represents the value of the dependent variable.
8. The high voltage isolating switch fault diagnosis terminal as defined in claim 6, wherein the steps between the steps S2 and S3 further comprise the steps of:
s23, matching the optimal correspondence between the interpolation curve and the standard power curve by adopting a mean square error method to obtain a mean square error minimum offset as follows:
wherein MSE represents mean square error, n is the length of the data set in the interpolation curve or the standard power curve, i E [1, n]，y 1，i The value of the ith data point of the 1 st data set in the standard power curve, y intrp,i Values for the i data point in the 2 nd data set in the interpolation curve or the standard power curve;
s24, re-marking the characteristic points matched with the standard power curve on an interpolation curve according to the minimum offset.
9. The high voltage isolating switch fault diagnosis terminal according to claim 6, wherein the step S3 is specifically:
S31, discretizing the interpolation curve according to the slope to obtain an interpolation segmented curve comprising a plurality of segmented intervals, distributing labels or symbols for each segmented interval, discretizing the standard power curve according to the segmented intervals, and obtaining a standard power segmented curve comprising a plurality of segmented intervals;
s32, representing each section of curve in the interpolation segmentation curve and the standard power segmentation curve as a character string or sequence to obtain an interpolation sequence and a standard sequence;
s33, calculating the longest public subsequence between the interpolation sequence and the standard sequence by using a long public subsequence LCS algorithm;
s34, obtaining measurement of similarity of two curves by analyzing the ratio between the longest public subsequence and the length of each sequence in the interpolation sequence and the standard sequence;
s35, analyzing the characteristic points corresponding to the sequences with the measurement exceeding the threshold value in the segment section where the interpolation segment curve is located, and obtaining corresponding fault behaviors.
10. The high voltage isolating switch fault diagnosis terminal according to claim 9, wherein the slope calculation in step S31 is specifically:
For discrete data points in the interpolation piecewise curve, a difference method is used for estimating the slope, and the data point is set as y 1 ，y 2 ，y 3 ，…，y m The slope is obtained as follows:
slope j ＝y j+1 -y j (3)；
wherein j is E [1, m ], m is the total number of segments of the interpolation segmentation curve and the standard power curve;
for non-equally spaced data points on the x-axis, equation (3) above is replaced with:
according to the preset threshold value epsilon, if the slope of the segmented interval j The E is the ascending segmentation section;
if slope j The < - ∈is a descending segmentation interval;
if |slope j The section is a stable section if the grade is less than or equal to E;
the successive ascending, descending or stationary segmentation intervals are combined into one segmentation interval, resulting in an interpolated segmentation curve with a plurality of ascending, descending and/or stationary segmentation intervals.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311502218.2A CN117471303A (en) | 2023-11-10 | 2023-11-10 | High-voltage isolating switch fault diagnosis method and terminal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311502218.2A CN117471303A (en) | 2023-11-10 | 2023-11-10 | High-voltage isolating switch fault diagnosis method and terminal |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117471303A true CN117471303A (en) | 2024-01-30 |
Family
ID=89623594
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311502218.2A Pending CN117471303A (en) | 2023-11-10 | 2023-11-10 | High-voltage isolating switch fault diagnosis method and terminal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117471303A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117972451A (en) * | 2024-03-28 | 2024-05-03 | 国网安徽省电力有限公司电力科学研究院 | GIS isolating switch switching position confirmation method |
-
2023
- 2023-11-10 CN CN202311502218.2A patent/CN117471303A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117972451A (en) * | 2024-03-28 | 2024-05-03 | 国网安徽省电力有限公司电力科学研究院 | GIS isolating switch switching position confirmation method |
| CN117972451B (en) * | 2024-03-28 | 2024-06-11 | 国网安徽省电力有限公司电力科学研究院 | GIS isolating switch switching position confirmation method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Telford et al. | Diagnosis of series DC arc faults—A machine learning approach | |
| Jadin et al. | Recent progress in diagnosing the reliability of electrical equipment by using infrared thermography | |
| Georgoulas et al. | Principal Component Analysis of the start-up transient and Hidden Markov Modeling for broken rotor bar fault diagnosis in asynchronous machines | |
| US9483049B2 (en) | Anomaly detection and diagnosis/prognosis method, anomaly detection and diagnosis/prognosis system, and anomaly detection and diagnosis/prognosis program | |
| Cheng et al. | A fusion prognostics method for remaining useful life prediction of electronic products | |
| CN117471303A (en) | High-voltage isolating switch fault diagnosis method and terminal | |
| Yu et al. | Risk-based fault detection using Self-Organizing Map | |
| CN111695620A (en) | Method and system for detecting and correcting abnormal data of time sequence of power system | |
| CN112149569A (en) | Voiceprint fault diagnosis method of transformer based on fuzzy C-means clustering algorithm | |
| CN116244279A (en) | High-voltage switch cabinet defect prediction method and system based on causal graph attention mechanism | |
| CN116500379A (en) | Accurate positioning method for voltage drop of STS device | |
| CN117150244B (en) | Intelligent power distribution cabinet state monitoring method and system based on electrical parameter analysis | |
| Mharakurwa | In‐Service Power Transformer Life Time Prospects: Review and Prospects | |
| CN116338449B (en) | Online testing method and system for switching characteristics of circuit breaker | |
| KR102284958B1 (en) | Partial discharge position estimation appratus and method | |
| CN115761405A (en) | Method for determining running state of high-voltage cable | |
| Tosyali et al. | Data-driven gantry health monitoring and process status identification based on texture extraction | |
| US20220253051A1 (en) | Method and assistance system for detecting an abnormal behaviour of a device | |
| Alawode et al. | Fuzzy logic and decision tree models for dissolved gas analysis in power transformer fault diagnosis | |
| CN117972451B (en) | GIS isolating switch switching position confirmation method | |
| CN110988663A (en) | Fault positioning method, device and equipment for gas insulated switchgear | |
| Zhou et al. | A New Validity Detection Method of Online Status Monitoring Data for Power Transformer | |
| CN117330952B (en) | Isolation switch fault diagnosis method based on persistent coherent analysis | |
| Okpo et al. | Induction Motor Fault Identification using Support Vector Machine | |
| CN117349710B (en) | Method for predicting breaking capacity of vacuum arc-extinguishing chamber, electronic equipment and medium |
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 |