CN116304957A - On-line identification method for monitoring state mutation of power supply and transformation equipment - Google Patents

On-line identification method for monitoring state mutation of power supply and transformation equipment Download PDF

Info

Publication number
CN116304957A
CN116304957A CN202310555317.0A CN202310555317A CN116304957A CN 116304957 A CN116304957 A CN 116304957A CN 202310555317 A CN202310555317 A CN 202310555317A CN 116304957 A CN116304957 A CN 116304957A
Authority
CN
China
Prior art keywords
mutation
data
power supply
liquid crystal
leaf node
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
Application number
CN202310555317.0A
Other languages
Chinese (zh)
Inventor
陈奇志
黄圣波
李胜
李妮锶
牟燕平
杨铁权
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Chengdu Jiaoda Guangmang Technology Co ltd
Original Assignee
Chengdu Jiaoda Guangmang Technology Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Chengdu Jiaoda Guangmang Technology Co ltd filed Critical Chengdu Jiaoda Guangmang Technology Co ltd
Priority to CN202310555317.0A priority Critical patent/CN116304957A/en
Publication of CN116304957A publication Critical patent/CN116304957A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/14—Fourier, Walsh or analogous domain transformations, e.g. Laplace, Hilbert, Karhunen-Loeve, transforms
    • G06F17/148—Wavelet transforms
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Electricity, gas or water supply
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S10/00—Systems supporting electrical power generation, transmission or distribution
    • Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications

Abstract

The invention relates to the technical field of power supply and transformation equipment monitoring, in particular to an online identification method for power supply and transformation equipment monitoring state mutation, which comprises the following steps: acquiring real-time monitoring data of equipment to represent original time sequence data of the equipment, and dividing the original time sequence data into a plurality of time sequence data; decomposing the time sequence data to obtain a difference component and a mean component; constructing a binary tree model based on the original time sequence, and searching the difference component and the mean component through the binary tree model to obtain mutation point positions and corresponding mutation time sequence data; and carrying out anomaly identification on the mutation time sequence data to obtain the mutation reason of the equipment. The invention can improve the accuracy of the on-line monitoring data of the power supply and transformation equipment and can accurately identify the source of the mutation data.

Description

On-line identification method for monitoring state mutation of power supply and transformation equipment
Technical Field
The invention relates to the technical field of power supply and transformation equipment monitoring, in particular to an online identification method for power supply and transformation equipment monitoring state mutation.
Background
With the rapid expansion of railway construction scale, the health status evaluation of most power supply and transformation equipment depends on a power failure preventive test, and on-line monitoring of the power supply and transformation equipment is required to detect and determine the running state of the equipment. At present, the quality of on-line monitoring data is low, the reasons for mutation of the monitoring data mainly include primary equipment abnormality, on-line monitoring equipment abnormality or environmental interference, and part of mutation data is not generated by primary equipment degradation and cannot be directly used for equipment operation and maintenance analysis. Therefore, the accuracy of identifying the mutation points only through the mean value or the trend is low, and the aim of cleaning the data cannot be achieved. Based on the above, in order to improve the accuracy and availability of the on-line monitoring data and accurately identify the source of the mutation data, we have designed an on-line identification method for the mutation of the monitoring state of the power supply and transformation equipment.
Disclosure of Invention
The invention aims to provide an on-line identification method for monitoring state mutation of power supply and transformation equipment, which improves the accuracy of on-line monitoring data of the power supply and transformation equipment and can accurately identify the source of the mutation data.
The embodiment of the invention is realized by the following technical scheme:
an online identification method for monitoring state mutation of power supply and transformation equipment, comprising the following steps:
acquiring real-time monitoring data of equipment to represent original time sequence data of the equipment, and dividing the original time sequence data into a plurality of time sequence data;
decomposing the time sequence data to obtain a difference component and a mean component;
constructing a binary tree model based on the original time sequence data, searching the difference component and the mean component through the binary tree model respectively, and obtaining mutation point positions and corresponding mutation time sequence data;
and carrying out anomaly identification on the mutation time sequence data to obtain the mutation reason of the equipment.
Optionally, the decomposing the time series data specifically includes decomposing the time series data through a Haar wavelet transform, and a calculation formula of the Haar wavelet transform is as follows:
Figure SMS_1
Figure SMS_2
wherein, the liquid crystal display device comprises a liquid crystal display device,
Figure SMS_3
for mean component>
Figure SMS_4
For the difference component +.>
Figure SMS_5
，/>
Figure SMS_6
Is a time sequence
Figure SMS_7
Length of->
Figure SMS_8
。
Optionally, the binary tree model is specifically divided into a TcA model and a TcD model, the difference component is searched through the TcD model, and the mean component is searched through the TcA model, wherein the specific steps of searching the difference component are as follows:
setting the difference component as non-leaf nodes, wherein each non-leaf node comprises a leaf node I
Figure SMS_9
And leaf node two->
Figure SMS_10
；
Based on
Figure SMS_11
Screening and searching non-leaf nodes through a TcD model, judging whether the searching is successful, and if not, carrying out statistical fluctuation searching on mean components through a TcA model, wherein +.>
Figure SMS_12
Is a threshold value; if yes, continuing the next step;
and judging whether searching to the last non-leaf node, if not, re-screening and searching the non-leaf node through a TcD model, and if so, outputting the non-leaf node.
Optionally, the specific step of searching the mean component for statistical fluctuation includes:
setting the mean component as non-leaf nodes, wherein each non-leaf node corresponds to the statistical fluctuation quantity
Figure SMS_13
、/>
Figure SMS_14
；
Based on the principle of statistical test of double-sample KS
Figure SMS_15
Screening the mean component by TcA model and determining whether the search is successful, wherein ∈>
Figure SMS_16
For confidence interval, ++>
Figure SMS_17
According to the significance level->
Figure SMS_18
Inquiring in a statistical table;
if not, judging that no mutation point exists, and ending;
if so, judging whether the last non-leaf node is searched, if not, screening and searching the mean component through a TcA model until the last non-leaf node is searched, outputting the non-leaf node, and if so, outputting the non-leaf node.
Optionally, the
Figure SMS_19
The calculation formula of (2) is as follows:
Figure SMS_20
Figure SMS_21
Figure SMS_22
wherein, the liquid crystal display device comprises a liquid crystal display device,
Figure SMS_24
，/>
Figure SMS_27
，/>
Figure SMS_29
，/>
Figure SMS_25
is a time series variable->
Figure SMS_26
，i∈[a,aM]，/>
Figure SMS_28
，/>
Figure SMS_30
Is a time series variable->
Figure SMS_23
，i∈[aM+1，c]；
The said
Figure SMS_31
The calculation formula of (2) is as follows:
Figure SMS_32
wherein, the liquid crystal display device comprises a liquid crystal display device,
Figure SMS_33
，/>
Figure SMS_34
。
optionally, the specific step of obtaining the mutation point position includes:
the last non-leaf node outputting the TcA model or the TcD model, wherein the two statistical fluctuation amounts corresponding to the non-leaf node
Figure SMS_35
、/>
Figure SMS_36
And characterizing the calculation formula of the mutation points through an empirical cumulative distribution function:
Figure SMS_37
wherein, the liquid crystal display device comprises a liquid crystal display device,
Figure SMS_40
is a variable->
Figure SMS_42
For real number set +.>
Figure SMS_44
Is->
Figure SMS_38
To be expressed as statistical fluctuation amount +.>
Figure SMS_41
，/>
Figure SMS_43
Is->
Figure SMS_45
To be expressed as statistical fluctuation amount +.>
Figure SMS_39
。
Optionally, the abnormality recognition is performed on the mutation time series data, specifically, the characteristic quantity recognized as the abnormality cause through the variation coefficient CV and the data offset rate CR.
Optionally, the calculation formula of the variation coefficient CV is:
Figure SMS_46
wherein, the liquid crystal display device comprises a liquid crystal display device,
Figure SMS_47
is the standard deviation of mutation interval->
Figure SMS_48
Is the mean value of the mutation interval.
Optionally, the calculation formula of the data offset rate CR is:
Figure SMS_49
Figure SMS_50
wherein, the liquid crystal display device comprises a liquid crystal display device,
Figure SMS_51
for the offset rate of the ith data, +.>
Figure SMS_52
For the current i-th data value, +.>
Figure SMS_53
For the offset of the initial data, +.>
Figure SMS_54
Is the offset rate of the last data.
The technical scheme of the embodiment of the invention has at least the following advantages and beneficial effects:
according to the embodiment of the invention, by establishing the binary tree model, whether the data mutation caused by environmental disturbance, abnormal monitoring equipment or the data mutation caused by primary equipment degradation is distinguished according to the feature analysis for the time sequence fragments with the mutation points, the running condition of the equipment can be truly reflected by the on-line monitoring data for equipment state analysis and fault diagnosis, the misjudgment rate of fault diagnosis is reduced, and the accuracy of equipment state analysis is improved.
Drawings
Fig. 1 is a logic flow diagram of an online identification method for monitoring a sudden change of a power supply and transformation device according to an embodiment of the present invention;
fig. 2 is a schematic flow chart of a binary tree model of a method for online identifying a mutation of a monitoring state of power supply and transformation equipment according to an embodiment of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. The components of the embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations.
As shown in fig. 1, fig. 1 is a logic flow diagram of an online identification method for monitoring a sudden change of a state of a power supply and transformation device according to an embodiment of the present invention.
In some embodiments, the present invention provides an online identification method for monitoring a sudden change in a power supply and transformation device, which can be applied to online monitoring of data change in a normal operation state of the power supply and transformation device, wherein the sudden change in data in the present embodiment is a phenomenon of "jumping" from one relatively stable state or stable continuous change trend to another stable state or stable continuous change trend.
In this embodiment, taking transformer oil chromatographic data as an example, there are many reasons for data distortion caused by unstable operation or interference of the monitoring device: the gas permeability of the oil-gas separation polymer film changes along with the temperature, the semiconductor gas sensor ages, the carrier gas is not replaced timely, and the chromatographic column is fixed and lost. The anomalies caused by these reasons are mostly in the field, and are characterized by various characteristics, low regularity, and numerical outliers, data steps and the like.
The mutation recognition of the embodiment mainly comprises two parts of mutation position recognition and mutation reason recognition, wherein the basic principle is that the position of a mutation point is firstly determined, and then the mutation reason is determined by analyzing the time slice data of the mutation point. The on-line monitoring data can be taken as a time sequence, a sliding window is adopted to intercept the data stream, wavelet transformation and double-sample KS inspection are combined to identify the mutation position, and then device abnormality identification is carried out on the interval data with abnormal mutation points, so that it can be understood that the whole mutation identification method is to determine the mutation position by utilizing the change condition of the time sequence, and the real-time on-line monitoring data of the power supply and transformation equipment is described herein, but not a specific quantity, for example, the CO2 of a transformer is used as a monitoring quantity.
Therefore, the present embodiment describes specific steps in combination with the above.
An online identification method for monitoring state mutation of power supply and transformation equipment, comprising the following steps:
acquiring real-time monitoring data of the equipment to represent original time series data of the equipment and dividing the original time series data into a plurality of time series data, wherein in the embodiment, the equipment is specifically set as primary equipment;
decomposing the time sequence data to obtain a difference component and a mean component;
constructing a binary tree model based on the original time sequence data, searching the difference component and the mean component through the binary tree model respectively, and obtaining mutation point positions and corresponding mutation time sequence data;
and carrying out anomaly identification on the mutation time sequence data to obtain the mutation reason of the equipment.
The present embodiment relates to the identification of mutation positions (i.e., mutation points):
the identification of the on-line monitoring mutation position is based on statistical analysis of time slices, the collected on-line monitoring data is a data stream with time characteristics, and the data stream needs to be preprocessed to ensure the identification efficiency and effect. It can be appreciated that the preprocessing includes several processing modes, and the details are not described in this embodiment.
In this embodiment, in order to ensure timeliness and recognition rate of mutation position recognition, the window length cannot be too long and too short, the window length l=16 is selected according to the comprehensive recognition rate and timeliness of the experimental result, and the analyzed time-series data segment is determined through a sliding window
Figure SMS_55
。
More specifically, the points identified in the initial sliding window are mutation points, the mutation points in the subsequent sliding window and the previous sliding window are taken as difference sets by mutation time, and repeated pushing of the mutation points can be avoided.
Decomposing a time series data segment into difference components of data by Haar wavelet transform
Figure SMS_56
And mean component
Figure SMS_57
The binary tree model (TcA model and TcD model) is built in combination with the original time sequence to search for mutation points, and the calculation formula of Haar wavelet transformation is as follows:
Figure SMS_58
Figure SMS_59
wherein, the liquid crystal display device comprises a liquid crystal display device,
Figure SMS_60
for mean component>
Figure SMS_61
For the difference component +.>
Figure SMS_62
，/>
Figure SMS_63
Is a time sequence
Figure SMS_64
Length of->
Figure SMS_65
。
Fig. 2 is a schematic flow chart of a binary tree model of a method for identifying a mutation of a monitoring state of power supply and transformation equipment according to an embodiment of the present invention.
And (3) searching for the fluctuation of the difference value based on the TcD model, namely, according to the fluctuation condition of the difference value reflecting data, if the absolute value of the difference value is larger, the fluctuation of the data is more severe, and considering that a mutation point possibly appears in the part with larger fluctuation, and searching for a selected path. Assume a non-leaf node
Figure SMS_66
There are two leaf nodes->
Figure SMS_67
,/>
Figure SMS_68
Corresponding to two time sequences, if
Figure SMS_69
Wherein->
Figure SMS_70
And selecting the non-leaf node with the highest mutation probability as a threshold value to search, and discarding the other two nodes until the last non-leaf node is searched. Otherwise, a second search rule is selected for searching, and the non-leaf nodes are the data of the penultimate row in fig. 2 as reference, and any non-leaf node can divide the time sequence X into two parts, such as cA1,2, and divide the X sequence into a left sequence { X1, X2, X3} and a right sequence { X4, X5, … xN }.
The second search rule is based on TcA model statistical fluctuation search, which is to compare the difference between two time series distribution functions according to the double-sample KS statistical test principle to judge whether mutation occurs or not, and to judge the non-leaf node
Figure SMS_71
Corresponding statistical fluctuation amount->
Figure SMS_72
、/>
Figure SMS_73
。
Figure SMS_74
The calculation formula of (2) is as follows:
Figure SMS_75
Figure SMS_76
Figure SMS_77
wherein, the liquid crystal display device comprises a liquid crystal display device,
Figure SMS_78
，/>
Figure SMS_82
，/>
Figure SMS_83
，/>
Figure SMS_80
is a time series variable->
Figure SMS_81
，i∈[a,aM]，/>
Figure SMS_84
，/>
Figure SMS_85
Is a time series variable->
Figure SMS_79
，i∈[aM+1，c]。
Figure SMS_86
The calculation formula of (2) is as follows:
Figure SMS_87
wherein, the liquid crystal display device comprises a liquid crystal display device,
Figure SMS_88
，/>
Figure SMS_89
。
if it is
Figure SMS_90
Wherein->
Figure SMS_91
For confidence interval, ++>
Figure SMS_92
According to the significance level->
Figure SMS_93
Inquiring in the statistical table, selecting the non-leaf node with the highest mutation probability to search, and discarding the other two nodes until the last non-leaf node is searched. If->
Figure SMS_94
And judging that the mutation points are temporarily absent.
Searching the last non-leaf node (the binary tree walks to the last non-leaf node according to the threshold judgment flow and the other nodes are discarded) through a TcA model or a TcD model, wherein at the moment, the two statistical fluctuation amounts corresponding to the non-leaf node
Figure SMS_97
、/>
Figure SMS_98
Can be made of->
Figure SMS_101
、/>
Figure SMS_95
Indicating (I)>
Figure SMS_100
、/>
Figure SMS_102
Are respectively->
Figure SMS_104
And
Figure SMS_96
is a cumulative distribution function of experience of (a). Due to->
Figure SMS_99
And->
Figure SMS_103
The largest statistical difference of (a) occurs before or after a signal jump, so the mutation should be expressed as:
Figure SMS_105
wherein, the liquid crystal display device comprises a liquid crystal display device,
Figure SMS_106
is a variable->
Figure SMS_109
For real number set +.>
Figure SMS_112
Is->
Figure SMS_108
To be expressed as statistical fluctuation amount +.>
Figure SMS_110
，/>
Figure SMS_111
Is->
Figure SMS_113
To be expressed as statistical fluctuation amount +.>
Figure SMS_107
。
And selecting a leaf node with the largest statistical fluctuation in the last non-leaf node as a mutation point.
Identification of the cause of the mutation:
in the present embodiment, the coefficient of variation CV and the data rate of deviation CR are used as feature amounts for abnormality recognition, and the coefficient of variation and the rate of deviation are calculated for time-series data in which a mutation point occurs.
The coefficient of variation CV is a statistical parameter for measuring the degree of dispersion of the observed value, the magnitude of the coefficient of variation is not only influenced by the degree of dispersion of the variable value, but also influenced by the average level of the variable, and the calculation formula of the coefficient of variation CV is as follows:
Figure SMS_114
wherein, the liquid crystal display device comprises a liquid crystal display device,
Figure SMS_115
is the standard deviation of mutation interval->
Figure SMS_116
Counting the operation data of the historical power supply and transformation equipment to obtain the minimum threshold value T of the variation coefficient of the monitoring data caused by the unstable operation of the on-line monitoring as the average value of the mutation interval, and if CV>T, the on-line monitoring shows pollution or aging in long-term use, and data deviation and miscellaneous peaks are caused by performance degradation.
For unstable operation of the monitoring device caused by short-time environmental interference, when the interference end data return to normal, the calculation formula of the data offset rate CR is as follows:
Figure SMS_117
if the short-term disturbance is caused by the environment, the product of the initial data offset rate and the end data offset rate of the time series is positive.
Figure SMS_118
Wherein, the liquid crystal display device comprises a liquid crystal display device,
Figure SMS_119
for the offset rate of the ith data, +.>
Figure SMS_120
For the current i-th data value, +.>
Figure SMS_121
For the offset of the initial data, +.>
Figure SMS_122
For the shift rate of the last data, the mutation point which does not satisfy the above condition is determined as a mutation caused by the degradation of the primary device.
The above is only a preferred embodiment of the present invention, and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims (9)

1. The online identification method for the monitoring state mutation of the power supply and transformation equipment is characterized by comprising the following steps of:
acquiring real-time monitoring data of equipment to represent original time sequence data of the equipment, and dividing the original time sequence data into a plurality of time sequence data;
decomposing the time sequence data to obtain a difference component and a mean component;
constructing a binary tree model based on the original time sequence data, searching the difference component and the mean component through the binary tree model respectively, and obtaining mutation point positions and corresponding mutation time sequence data;
and carrying out anomaly identification on the mutation time sequence data to obtain the mutation reason of the equipment.
2. The online identification method for the power supply and transformation equipment monitoring state mutation according to claim 1, wherein the time series data is decomposed, specifically through Haar wavelet transformation, and the calculation formula of the Haar wavelet transformation is as follows:
Figure QLYQS_1
Figure QLYQS_2
wherein, the liquid crystal display device comprises a liquid crystal display device,
Figure QLYQS_3
for mean component>
Figure QLYQS_4
For the difference component +.>
Figure QLYQS_5
，/>
Figure QLYQS_6
Is a time sequence
Figure QLYQS_7
Length of->
Figure QLYQS_8
。
3. The online identification method for the power supply and transformation equipment monitoring state mutation according to claim 1, wherein the binary tree model is specifically divided into a TcA model and a TcD model, a difference component is searched through the TcD model, and a mean component is searched through the TcA model, wherein the specific steps of searching the difference component are as follows:
setting the difference component as non-leaf nodes, wherein each non-leaf node comprises a leaf node I
Figure QLYQS_9
And the second leaf node
Figure QLYQS_10
；
Based on
Figure QLYQS_11
Screening and searching non-leaf nodes through a TcD model, judging whether the searching is successful, and if not, carrying out statistical fluctuation searching on mean components through a TcA model, wherein +.>
Figure QLYQS_12
Is a threshold value; if yes, continuing the next step;
and judging whether searching to the last non-leaf node, if not, re-screening and searching the non-leaf node through a TcD model, and if so, outputting the non-leaf node.
4. The online identification method for the mutation of the monitoring state of the power supply and transformation equipment according to claim 3, wherein the specific steps of searching the mean component for statistical fluctuation are as follows:
setting the mean component as non-leaf nodes, wherein each non-leaf node corresponds to the statistical fluctuation quantity
Figure QLYQS_13
、/>
Figure QLYQS_14
；
Based on the principle of statistical test of double-sample KS
Figure QLYQS_15
Screening the mean component by TcA model and determining whether the search is successful, wherein ∈>
Figure QLYQS_16
For confidence interval, ++>
Figure QLYQS_17
According to the significance level->
Figure QLYQS_18
Inquiring in a statistical table;
if not, judging that no mutation point exists, and ending;
if so, judging whether the last non-leaf node is searched, if not, screening and searching the mean component through a TcA model until the last non-leaf node is searched, outputting the non-leaf node, and if so, outputting the non-leaf node.
5. The power supply and transformation equipment monitoring state mutation online identification method according to claim 4, wherein the power supply and transformation equipment monitoring state mutation online identification method is characterized in that
Figure QLYQS_19
The calculation formula of (2) is as follows:
Figure QLYQS_20
Figure QLYQS_21
Figure QLYQS_22
wherein, the liquid crystal display device comprises a liquid crystal display device,
Figure QLYQS_24
，/>
Figure QLYQS_26
，/>
Figure QLYQS_28
，/>
Figure QLYQS_25
is a time series variable->
Figure QLYQS_27
，i∈[a,aM]，/>
Figure QLYQS_29
，/>
Figure QLYQS_30
Is a time series variable->
Figure QLYQS_23
，i∈[aM+1，c]；
The said
Figure QLYQS_31
The calculation formula of (2) is as follows:
Figure QLYQS_32
wherein, the liquid crystal display device comprises a liquid crystal display device,
Figure QLYQS_33
，/>
Figure QLYQS_34
。
6. the method for on-line identification of power supply and transformation equipment monitoring state mutation according to claim 5, wherein the specific acquisition step of the mutation point position is as follows:
the last non-leaf node outputting the TcA model or the TcD model, wherein the two statistical fluctuation amounts corresponding to the non-leaf node
Figure QLYQS_35
、/>
Figure QLYQS_36
And characterizing the calculation formula of the mutation points through an empirical cumulative distribution function:
Figure QLYQS_37
wherein, the liquid crystal display device comprises a liquid crystal display device,
Figure QLYQS_38
is a variable->
Figure QLYQS_41
For real number set +.>
Figure QLYQS_44
Is->
Figure QLYQS_40
To be expressed as statistical fluctuation amount +.>
Figure QLYQS_42
，/>
Figure QLYQS_43
Is->
Figure QLYQS_45
To be expressed as statistical fluctuation amount +.>
Figure QLYQS_39
。
7. The online identification method for the power supply and transformation equipment monitoring state mutation according to any one of claims 1-6, wherein the abnormality identification is performed on mutation time series data, specifically by using a mutation coefficient CV and a data offset rate CR as feature quantities for abnormality cause identification.
8. The online identification method for the power supply and transformation equipment monitoring state mutation according to claim 7, wherein the calculation formula of the variation coefficient CV is as follows:
Figure QLYQS_46
wherein, the liquid crystal display device comprises a liquid crystal display device,
Figure QLYQS_47
is the standard deviation of mutation interval->
Figure QLYQS_48
Is the mean value of the mutation interval.
9. The power supply and transformation equipment monitoring state mutation online identification method according to claim 8, wherein the calculation formula of the data offset rate CR is:
Figure QLYQS_49
Figure QLYQS_50
wherein, the liquid crystal display device comprises a liquid crystal display device,
Figure QLYQS_51
for the offset rate of the ith data, +.>
Figure QLYQS_52
For the current i-th data value, +.>
Figure QLYQS_53
For the offset rate of the initial data,
Figure QLYQS_54
is the offset rate of the last data.
CN202310555317.0A 2023-05-17 2023-05-17 On-line identification method for monitoring state mutation of power supply and transformation equipment Pending CN116304957A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310555317.0A CN116304957A (en) 2023-05-17 2023-05-17 On-line identification method for monitoring state mutation of power supply and transformation equipment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310555317.0A CN116304957A (en) 2023-05-17 2023-05-17 On-line identification method for monitoring state mutation of power supply and transformation equipment

Publications (1)

Publication Number Publication Date
CN116304957A true CN116304957A (en) 2023-06-23

Family

ID=86781872

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310555317.0A Pending CN116304957A (en) 2023-05-17 2023-05-17 On-line identification method for monitoring state mutation of power supply and transformation equipment

Country Status (1)

Country Link
CN (1) CN116304957A (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116610482A (en) * 2023-07-18 2023-08-18 山东理工大学 Intelligent monitoring method for operation state of electrical equipment
CN116756619A (en) * 2023-07-12 2023-09-15 常熟浦发第二热电能源有限公司 Equipment intelligent diagnosis method and system based on big data
CN117251804A (en) * 2023-11-17 2023-12-19 天津中电华利电器科技集团有限公司 Substation operation state monitoring data processing system and method

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103020478A (en) * 2012-12-28 2013-04-03 杭州师范大学 Method for checking reality of ocean color remote sensing product
CN108335467A (en) * 2018-01-05 2018-07-27 东华大学 A kind of fire on-line early warning and rapid analysis method based on catastrophe point detection
CN110399910A (en) * 2019-07-08 2019-11-01 东华大学 Fire abnormal point online test method based on sliding window and HWKS theoretical frame
CN110412215A (en) * 2019-07-08 2019-11-05 东华大学 The rapid detection method of exhaust emissions exception based on TSTKS algorithm
CN110522446A (en) * 2019-07-19 2019-12-03 东华大学 A kind of electroencephalogramsignal signal analysis method that accuracy high practicability is strong
CN112328961A (en) * 2020-11-04 2021-02-05 江苏海拓润达科技发展有限公司 On-line monitoring device quality evaluation system based on fault tree and Bayesian network
CN114018833A (en) * 2021-11-07 2022-02-08 福建师范大学 Method for estimating heavy metal content of soil based on hyperspectral remote sensing technology

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103020478A (en) * 2012-12-28 2013-04-03 杭州师范大学 Method for checking reality of ocean color remote sensing product
CN108335467A (en) * 2018-01-05 2018-07-27 东华大学 A kind of fire on-line early warning and rapid analysis method based on catastrophe point detection
CN110399910A (en) * 2019-07-08 2019-11-01 东华大学 Fire abnormal point online test method based on sliding window and HWKS theoretical frame
CN110412215A (en) * 2019-07-08 2019-11-05 东华大学 The rapid detection method of exhaust emissions exception based on TSTKS algorithm
CN110522446A (en) * 2019-07-19 2019-12-03 东华大学 A kind of electroencephalogramsignal signal analysis method that accuracy high practicability is strong
CN112328961A (en) * 2020-11-04 2021-02-05 江苏海拓润达科技发展有限公司 On-line monitoring device quality evaluation system based on fault tree and Bayesian network
CN114018833A (en) * 2021-11-07 2022-02-08 福建师范大学 Method for estimating heavy metal content of soil based on hyperspectral remote sensing technology

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116756619A (en) * 2023-07-12 2023-09-15 常熟浦发第二热电能源有限公司 Equipment intelligent diagnosis method and system based on big data
CN116756619B (en) * 2023-07-12 2023-11-21 常熟浦发第二热电能源有限公司 Equipment intelligent diagnosis method and system based on big data
CN116610482A (en) * 2023-07-18 2023-08-18 山东理工大学 Intelligent monitoring method for operation state of electrical equipment
CN116610482B (en) * 2023-07-18 2023-10-17 山东理工大学 Intelligent monitoring method for operation state of electrical equipment
CN117251804A (en) * 2023-11-17 2023-12-19 天津中电华利电器科技集团有限公司 Substation operation state monitoring data processing system and method
CN117251804B (en) * 2023-11-17 2024-04-19 天津中电华利电器科技集团有限公司 Substation operation state monitoring data processing system and method

Similar Documents

Publication Publication Date Title
CN116304957A (en) On-line identification method for monitoring state mutation of power supply and transformation equipment
CN111353482B (en) LSTM-based fatigue factor recessive anomaly detection and fault diagnosis method
CN110197588B (en) Method and device for evaluating driving behavior of large truck based on GPS track data
CN105718876B (en) A kind of appraisal procedure of ball-screw health status
CN112232447B (en) Construction method of complete sample set of power equipment state monitoring data
CN104764869B (en) Transformer gas fault diagnosis and alarm method based on multidimensional characteristics
CN112284440B (en) Sensor data deviation self-adaptive correction method
CN106447210A (en) Distribution network equipment health degree dynamic diagnosis method involving credibility evaluation
CN115982602B (en) Photovoltaic transformer electrical fault detection method
CN111738308A (en) Dynamic threshold detection method for monitoring index based on clustering and semi-supervised learning
CN111665066B (en) Equipment fault self-adaptive upper and lower early warning boundary generation method based on convolutional neural network
CN111458149B (en) Method and system for predicting performance and service life of rolling bearing
CN111579731A (en) Early transformer defect early warning method based on combined model
CN111504647A (en) AR-MSET-based performance degradation evaluation method for rolling bearing
CN113918642A (en) Data filtering, monitoring and early warning method based on power Internet of things equipment
CN113342784A (en) Database design method for risk assessment of main transformer equipment of power grid
CN110108474B (en) Online monitoring and evaluating method and system for operation stability of rotary machine
CN115375026A (en) Method for predicting service life of aircraft engine in multiple fault modes
CN114509620A (en) Transformer fault diagnosis method based on monitoring data
Aburaghiega et al. Enhancement of power transformer state of health diagnostics based on fuzzy logic system of DGA
CN111314110B (en) Fault early warning method for distributed system
CN116189802A (en) Transformer fault early warning method based on gas concentration time sequence data
CN114167282B (en) Motor fault diagnosis and degradation trend prediction system
CN110308713A (en) A kind of industrial process failure identification variables method based on k neighbour reconstruct
CN115659271A (en) Sensor abnormality detection method, model training method, system, device, 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
RJ01 Rejection of invention patent application after publication

Application publication date: 20230623