CN112906729B - Fault distribution determination method, device and system of switch equipment - Google Patents
Fault distribution determination method, device and system of switch equipment Download PDFInfo
- Publication number
- CN112906729B CN112906729B CN201911229019.2A CN201911229019A CN112906729B CN 112906729 B CN112906729 B CN 112906729B CN 201911229019 A CN201911229019 A CN 201911229019A CN 112906729 B CN112906729 B CN 112906729B
- Authority
- CN
- China
- Prior art keywords
- fault
- distribution
- determining
- equipment
- information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000009826 distribution Methods 0.000 title claims abstract description 183
- 238000000034 method Methods 0.000 title claims abstract description 26
- 239000013598 vector Substances 0.000 claims abstract description 81
- 238000012549 training Methods 0.000 claims abstract description 38
- 238000013507 mapping Methods 0.000 claims abstract description 19
- 230000007547 defect Effects 0.000 claims description 37
- 238000012423 maintenance Methods 0.000 claims description 15
- 238000000354 decomposition reaction Methods 0.000 claims description 5
- 230000011218 segmentation Effects 0.000 claims description 5
- 239000011159 matrix material Substances 0.000 description 7
- 238000010586 diagram Methods 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 238000013145 classification model Methods 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
Abstract
The invention provides a fault distribution determining method, device and system of switching equipment. And then training the hidden dirichlet model based on the fault history vector so that the hidden dirichlet model outputs the distribution information of each switching device on the fault mode and the distribution information of each fault mode on the fault. And training a decomposer model based on the attribute vector and the distribution information of the switching equipment on the fault mode, and determining the mapping relation of the switching equipment on the fault mode. And inputting the attribute vector of the switching equipment with the fault distribution to be determined into the decomposer model, and determining the target distribution information of the switching equipment with the fault distribution to be determined on the fault mode. And determining the distribution information of the switching equipment with the fault distribution to be determined on the basis of the target distribution information and the distribution information of each fault mode on the fault. Therefore, the scheme can accurately predict the faults of the switching equipment.
Description
Technical Field
The present invention relates to the field of equipment protection technologies, and in particular, to a method, an apparatus, and a system for determining fault distribution of a switching device.
Background
The switching device is a control and protection device which plays a key role in the power system and directly affects the safe operation of the power grid. At present, the failure probability of the switching equipment is predicted in a data statistics mode and the like, so that the whole service life of the switching equipment is managed, the operation reliability of the switching equipment is improved, and the life cycle cost is reduced.
However, the mathematical statistics method takes all the historical data of one index of the switch operation data as a sample space, obtains a regression equation of six indexes through the mathematical statistics method, obtains a predicted value of the next time through the regression equation, and can not predict the failure of the equipment which never happens.
Therefore, how to provide a fault distribution determining method for a switching device, so as to implement accurate prediction of faults of the switching device, is a technical problem to be solved by those skilled in the art.
Disclosure of Invention
In view of this, the embodiment of the invention provides a fault distribution determining method of a switching device, which can accurately predict faults of the switching device.
In order to achieve the above object, the embodiment of the present invention provides the following technical solutions:
a fault distribution determination method for a switching device, comprising:
acquiring an attribute vector and a fault history vector of each switching device;
training an implicit dirichlet model based on the fault history vector, so that the implicit dirichlet model outputs distribution information of each switching device on a fault mode and distribution information of each fault mode on a fault;
training a decomposition machine model based on the attribute vector and the distribution information of the switching equipment on the fault mode, and determining the mapping relation of the switching equipment on the fault mode;
inputting an attribute vector of the switching equipment with the fault distribution to be determined into the decomposer model, and determining target distribution information of the switching equipment with the fault distribution to be determined on a fault mode;
and determining the distribution information of the switching equipment with the fault distribution to be determined on the basis of the target distribution information and the distribution information of each fault mode on the fault.
Optionally, the obtaining the attribute vector of each switching device includes:
acquiring the place, the equipment model and the operation time of occurrence of defects of each switch equipment;
segmenting the operation years until the defect occurs;
and determining the place, the equipment model and the running time of the segmented fault as the attribute vector of the switching equipment.
Optionally, the obtaining the fault history vector of each switching device includes:
acquiring equipment defect record information and maintenance operation information of each switching equipment;
classifying the equipment defect record information and the maintenance operation information according to the association relation between the preset fault number and the fault type, and determining the classified fault number as the fault history vector.
Optionally, training a decomposer model based on the attribute vector and distribution information of the switching device on the fault mode, and determining a mapping relationship of the switching device on the fault mode includes:
and taking the attribute vector as input information of the decomposer model, taking the distribution information of the switching equipment on the fault mode as output information of the decomposer model, and determining the mapping relation of the switching equipment on the fault mode.
Optionally, the determining, based on the target distribution information and the distribution information of each fault mode to the fault, the distribution information of the switching device to which the fault distribution is to be determined to the fault includes:
and determining the product of the target distribution information and the distribution information of each fault mode to faults as the distribution information of the switching equipment to be subjected to fault distribution determination to faults.
A fault distribution determining apparatus of a switching device, comprising:
the acquisition module is used for acquiring the attribute vector and the fault history vector of each switching device;
the first training module is used for training an implicit dirichlet allocation model based on the fault history vector so that the implicit dirichlet allocation model outputs the distribution information of each switching device to a fault mode and the distribution information of each fault mode to a fault;
the second training module is used for training a decomposer model based on the attribute vector and the distribution information of the switching equipment on the fault mode, and determining the mapping relation of the switching equipment on the fault mode;
the first determining module is used for inputting the attribute vector of the switching equipment with the fault distribution to be determined into the decomposer model and determining the target distribution information of the switching equipment with the fault distribution to be determined on the fault mode;
and the second determining module is used for determining the fault distribution information of the switching equipment with the fault distribution to be determined based on the target distribution information and the fault distribution information of each fault mode.
Optionally, the acquiring module includes:
an acquisition unit configured to acquire a location, an equipment model, an operation age up to a time when a defect occurs, equipment defect recording information, and maintenance operation information of each of the switching devices;
the dividing unit is used for dividing the operation years from the defect occurrence time to the defect occurrence time;
a first determining unit, configured to determine a location of the switching device, a device model, and a running period from when the defect occurs after segmentation as an attribute vector of the switching device;
and the second determining unit is used for classifying the equipment defect record information and the maintenance operation information according to the association relation between the preset fault number and the fault type, and determining the classified fault number as the fault history vector.
Optionally, the second training module includes:
the training unit is used for taking the attribute vector as input information of the decomposer model, taking the distribution information of the switching equipment on the fault mode as output information of the decomposer model, and determining the mapping relation of the switching equipment on the fault mode.
Optionally, the second determining module includes:
and a third determining unit, configured to determine that a product of the target distribution information and the distribution information of each fault mode to faults is distribution information of the switching device to faults of the fault distribution to be determined.
A fault distribution determining system of a switching device comprises any one of the fault distribution determining devices of the switching device.
Based on the above technical scheme, the embodiment of the invention provides a fault distribution determining method, device and system of switching equipment. And then training an implicit dirichlet model based on the fault history vector, so that the implicit dirichlet model outputs the distribution information of each switching device on a fault mode and the distribution information of each fault mode on a fault. And training a decomposition machine model based on the attribute vector and the distribution information of the switching equipment on the fault mode, and determining the mapping relation of the switching equipment on the fault mode. And inputting the attribute vector of the switching equipment with the fault distribution to be determined into the decomposer model, and determining the target distribution information of the switching equipment with the fault distribution to be determined on the fault mode. And determining the fault distribution information of the switching equipment of which the fault distribution is to be determined based on the target distribution information and the fault distribution information of each fault mode. Therefore, the scheme can accurately predict the faults of the switching equipment.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required to be used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only embodiments of the present invention, and that other drawings can be obtained according to the provided drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic flow chart of a fault distribution determining method of a switching device according to an embodiment of the present invention;
fig. 2 is a schematic flow chart of a fault distribution determining method of a switching device according to an embodiment of the present invention;
fig. 3 is a schematic flow chart of a fault distribution determining method of a switching device according to an embodiment of the present invention;
fig. 4 is a schematic flow chart of a fault distribution determining method of a switching device according to an embodiment of the present invention;
fig. 5 is a schematic flow chart of a fault distribution determining method of a switching device according to an embodiment of the present invention;
fig. 6 is a schematic structural diagram of a fault distribution determining apparatus of a switching device according to an embodiment of the present invention;
fig. 7 is a schematic architecture diagram of a fault distribution determining system of a switching device according to an embodiment of the present invention.
Detailed Description
Referring to fig. 1, fig. 1 is a fault distribution determining method of a switching device according to an embodiment of the present invention, including the steps of:
s11, acquiring attribute vectors and fault history vectors of each switch device;
specifically, an embodiment of the present invention provides a specific implementation manner for obtaining an attribute vector of each switching device, as shown in fig. 2, including:
s21, acquiring the place, the equipment model and the operation time limit of each switch equipment when the defect occurs;
s22, segmenting the operation years until the defect occurs;
s23, determining the place and the equipment model of the switch equipment and the running time after segmentation until the defect occurs as the attribute vector of the switch equipment.
In addition, as shown in fig. 3, the embodiment further provides a specific implementation manner of obtaining the fault history vector of each switching device, including:
s31, acquiring equipment defect record information and maintenance operation information of each switching equipment;
s32, classifying the equipment defect record information and the maintenance operation information according to the association relation between the preset fault number and the fault type, and determining the classified fault number as the fault history vector.
Specifically, in this embodiment, the operation and maintenance records of the device are captured from the power system management system, and the key contents to be captured include the location of the switching device, the device model, the operation period until the defect occurs, the defect record content of the device, and the maintenance operation information.
The switchgear is then segmented to the operational age at which the defect occurred. For example, the operation years of the switching device to the time of occurrence of the defect are divided into five stages of 0 to 5 years, 5 to 10 years, 10 to 15 years, 15 to 20 years, and more than 20 years. By segmentation, the original continuous data is converted into discrete category data.
And then, carrying out one-hot coding on the location, the equipment model and the segmented operation time limit of each switch equipment until the defect occurs, and connecting to obtain a one-dimensional vector attribute code for expressing the attribute of each switch equipment. Assuming that there are O different substations in the system, P types of switching devices and Q life segments, the attribute vector length obtained by each device in each time period is o+p+q. If M kinds of different devices are obtained in different time intervals, the shape of the attribute coding matrix of the whole devices in different time intervals is m× (o+p+q).
Illustratively, it is assumed that there are 3 substations, 2 devices, 5 life segments, where a certain device belongs to substation 1, device model 2. Then its one-hot vector at 0 to 5 years may be represented as [1, 0;0,1;1, 0], the one-hot vector at 5-10 years may be represented as [1, 0;0,1;0,1, 0], if the device has no operation records for a certain time interval, this encoding is not performed for this time interval.
In addition, since the accumulated operation records in the running of the power system, including the equipment defect record content and the maintenance operation content, are mostly text and descriptive data, the two contents need to be comprehensively considered to be converted into category data. In particular, the transfer of text data to category data may be manually noted or using a text classification model or the like.
Illustratively, the total number of the fault types obtained by the final labeling is assumed to be N, and all the fault types are numbered in sequence. For example, the classification results obtained are 37 kinds in total, and the failure numbers and the corresponding types are shown in table 1:
table 1 fault number and fault type comparison table
And (3) corresponding the obtained fault category data with the obtained equipment one-hot vector to obtain a corresponding data set in a (attribute vector: fault history vector) format. Also, the number of data sets should be M.
For example, suppose there are 3 substations, 2 devices, 5 life segments, where a certain device belongs to substation 1, device model 2. It has two failures of oil pump damage and contact failure in 0 to 5 years, and its attribute vector and failure history can be expressed as ([ 1, 0;0,1;1, 0]: [0,2 ]).
S12, training an implicit Dirichlet model based on the fault history vector, so that the implicit Dirichlet model outputs the distribution information of each switching device on a fault mode and the distribution information of each fault mode on faults;
that is, the present embodiment automatically extracts several failure modes of devices by training an LDA model (hidden dirichlet model), and obtains the distribution of each device in the training set to the failure modes, and the distribution of each failure mode to the failure.
It should be noted that the LDA model is a machine learning method applied to text mining, and is aimed at automatically extracting a determined number of topics from a document set, and obtaining a distribution of the topics from the document and a distribution of the topics from each word in the document, so as to play a role in approximating the content. In this solution, since the training samples are not documents but failure histories of the devices, the topics in the LDA are expressed as failure modes.
The whole M sample fault history vector sets are input into the LDA model and the model is trained. Assuming K fault modes are preset, through LDA model training, the model automatically extracts an implicit fault mode, and outputs two matrixes, namely a distribution matrix of (equipment: fault mode) and a distribution matrix of (fault mode: fault), wherein the shapes of the two matrixes are M multiplied by K and K multiplied by N respectively. Wherein the sum of each row of the two matrices is equal to 1. Generally, K is selected to be much smaller than the numbers of M and N, for example, K is selected to be 6, thereby playing a role in information compression.
S13, training a decomposition machine model based on the attribute vector and the distribution information of the switching equipment on the fault mode, and determining the mapping relation of the switching equipment on the fault mode;
specifically, as shown in fig. 4, the embodiment provides a specific implementation manner of training a decomposer model to determine a mapping relationship between the switching device and the fault mode based on the attribute vector and the distribution information of the switching device on the fault mode, including:
s41, taking the attribute vector as input information of the decomposer model, taking distribution information of the switching equipment on the fault mode as output information of the decomposer model, and determining the mapping relation of the switching equipment on the fault mode.
In this scheme, the attribute parameters of each device in the training set are used as input, the obtained distribution of the device to the fault mode is used as output to train the FM model, and the mapping relation between the device attribute to the device fault mode is obtained.
It should be noted that the nature of the FM method is a regression model that can adapt to the parameter estimation requirements under very high sparsity data inputs and can take into account the correlations of the parameters in different dimensions. The inventor considers that the equipment attribute coding in the scheme is performed by one-hot, the sparsity is large, and therefore the FM model is selected for training.
Specifically, the model is input into the obtained device attribute matrix, the shape of the model is M× (O+P+Q), the model output is the distribution matrix obtained through LDA (device: fault mode), the shape of the model is M×K, and the input of the FM model obtained through training in the embodiment is O+P+Q dimension, and the output is K dimension.
S14, inputting an attribute vector of the switching equipment with the fault distribution to be determined into the decomposer model, and determining target distribution information of the switching equipment with the fault distribution to be determined on a fault mode;
s15, determining the distribution information of the switch equipment with the fault distribution to be determined on the basis of the target distribution information and the distribution information of each fault mode on the fault.
Specifically, as shown in fig. 5, an embodiment of the present invention provides a specific implementation manner of determining fault distribution information of the switching device of the fault distribution to be determined based on the target distribution information and the fault distribution information of each fault mode, including:
s51, determining that the product of the target distribution information and the distribution information of each fault mode to faults is the distribution information of the switch equipment to be subjected to fault distribution determination to faults.
Schematically, in this embodiment, the attribute of the high-voltage switching device that needs to predict the fault distribution is input into the FM model, and the distribution of the device to be predicted on the fault mode is obtained first; and multiplying the distribution by the obtained fault mode-to-fault distribution matrix to finally obtain a fault distribution prediction result of the equipment.
And carrying out vector processing on the attribute of the high-voltage switch equipment needing to predict fault distribution to obtain one-hot attribute vector. The one-hot attribute vector is input into an FM model to obtain the distribution of the equipment to be predicted to different fault modes, wherein the shape of the distribution is 1 XK.
And multiplying the fault mode distribution of the equipment to be predicted by the obtained distribution matrix (fault mode: fault) to obtain the final occurrence probability prediction result of the equipment on different distributions. The two matrices are shaped as 1×k and k×n, respectively, the final result is 1×n, and the sum of the elements in the result vector should be equal to 1.
And finally, converting the prediction result into the expression of the fault type to the fault type according to the comparison table of the fault type and the corresponding number, and finally obtaining the distribution prediction result of the equipment to different fault types. Therefore, the scheme can accurately predict the faults of the switching equipment.
On the basis of the above embodiment, as shown in fig. 6, an embodiment of the present invention provides a fault distribution determining apparatus for a switching device, including:
an obtaining module 61, configured to obtain an attribute vector and a fault history vector of each switching device;
a first training module 62, configured to train an implicit dirichlet allocation model based on the fault history vector, so that the implicit dirichlet allocation model outputs distribution information of each of the switching devices for a fault mode and distribution information of each of the fault modes for a fault;
the second training module 63 is configured to train a decomposition machine model based on the attribute vector and distribution information of the switching device on the fault mode, and determine a mapping relationship of the switching device on the fault mode;
a first determining module 64, configured to input an attribute vector of a switching device of a fault distribution to be determined into the decomposer model, and determine target distribution information of the switching device of the fault distribution to be determined on a fault mode;
and a second determining module 65, configured to determine distribution information of the switching device to be determined for fault distribution on the basis of the target distribution information and the distribution information of each fault mode to faults.
Wherein, the acquisition module may include:
an acquisition unit configured to acquire a location, an equipment model, an operation age up to a time when a defect occurs, equipment defect recording information, and maintenance operation information of each of the switching devices;
the dividing unit is used for dividing the operation years from the defect occurrence time to the defect occurrence time;
a first determining unit, configured to determine a location of the switching device, a device model, and a running period from when the defect occurs after segmentation as an attribute vector of the switching device;
and the second determining unit is used for classifying the equipment defect record information and the maintenance operation information according to the association relation between the preset fault number and the fault type, and determining the classified fault number as the fault history vector.
In addition, the second training module may include:
the training unit is used for taking the attribute vector as input information of the decomposer model, taking the distribution information of the switching equipment on the fault mode as output information of the decomposer model, and determining the mapping relation of the switching equipment on the fault mode.
And the second determination module may include:
and a third determining unit, configured to determine that a product of the target distribution information and the distribution information of each fault mode to faults is distribution information of the switching device to faults of the fault distribution to be determined.
The working principle of the device is shown in the above method embodiments, and will not be repeated here.
In addition, the embodiment of the invention also provides a fault distribution determining system of the switch equipment, as shown in fig. 7, which comprises any one of the fault distribution determining devices of the switch equipment, and the working principle of the fault distribution determining device is shown in the embodiment of the device.
In summary, the invention provides a fault distribution determining method, device and system of a switching device, which firstly acquire attribute vectors and fault history vectors of each switching device. And then training the hidden dirichlet model based on the fault history vector so that the hidden dirichlet model outputs the distribution information of each switching device on the fault mode and the distribution information of each fault mode on the fault. And training a decomposer model based on the attribute vector and the distribution information of the switching equipment on the fault mode, and determining the mapping relation of the switching equipment on the fault mode. And inputting the attribute vector of the switching equipment with the fault distribution to be determined into the decomposer model, and determining the target distribution information of the switching equipment with the fault distribution to be determined on the fault mode. And determining the distribution information of the switching equipment with the fault distribution to be determined on the basis of the target distribution information and the distribution information of each fault mode on the fault. Therefore, the scheme can accurately predict the faults of the switching equipment.
In the present specification, each embodiment is described in a progressive manner, and each embodiment is mainly described in a different point from other embodiments, and identical and similar parts between the embodiments are all enough to refer to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant points refer to the description of the method section.
Those of skill would further appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative elements and steps are described above generally in terms of functionality in order to clearly illustrate the interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. The software modules may be disposed in Random Access Memory (RAM), memory, read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (8)
1. A fault distribution determining method for a switching device, comprising:
acquiring an attribute vector and a fault history vector of each switching device;
training an implicit dirichlet model based on the fault history vector, so that the implicit dirichlet model outputs the distribution information of each switching device to a fault mode and the distribution information of each fault mode to a fault;
training a decomposition machine model based on the attribute vector and the distribution information of the switching equipment on the fault mode, and determining the mapping relation of the switching equipment on the fault mode;
inputting an attribute vector of the switching equipment with the fault distribution to be determined into the decomposer model, and determining target distribution information of the switching equipment with the fault distribution to be determined on a fault mode;
based on the target distribution information and the distribution information of each fault mode to faults, determining the distribution information of the switching equipment to be determined to be distributed to faults, including: and determining the product of the target distribution information and the distribution information of each fault mode to faults as the distribution information of the switching equipment to be subjected to fault distribution determination to faults.
2. The fault distribution determining method of a switching device according to claim 1, wherein the obtaining an attribute vector of each switching device includes:
acquiring the place, the equipment model and the operation time of occurrence of defects of each switch equipment;
segmenting the operation years until the defect occurs;
and determining the place, the equipment model and the running time of the segmented fault as the attribute vector of the switching equipment.
3. The fault distribution determining method of a switching device according to claim 2, wherein the obtaining a fault history vector of each switching device includes:
acquiring equipment defect record information and maintenance operation information of each switching equipment;
classifying the equipment defect record information and the maintenance operation information according to the association relation between the preset fault number and the fault type, and determining the classified fault number as the fault history vector.
4. The method for determining a fault distribution of a switching device according to claim 1, wherein training a decomposer model based on the attribute vector and distribution information of the switching device on a fault mode, and determining a mapping relationship of the switching device on the fault mode comprises:
and taking the attribute vector as input information of the decomposer model, taking the distribution information of the switching equipment to the fault mode as output information of the decomposer model, and determining the mapping relation of the switching equipment to the fault mode.
5. A fault distribution determining apparatus of a switching device, comprising:
the acquisition module is used for acquiring the attribute vector and the fault history vector of each switching device;
the first training module is used for training an implicit dirichlet allocation model based on the fault history vector so that the implicit dirichlet allocation model outputs the distribution information of each switching device to a fault mode and the distribution information of each fault mode to faults;
the second training module is used for training a decomposer model based on the attribute vector and the distribution information of the switching equipment on the fault mode, and determining the mapping relation of the switching equipment on the fault mode;
the first determining module is used for inputting the attribute vector of the switching equipment with the fault distribution to be determined into the decomposer model and determining the target distribution information of the switching equipment with the fault distribution to be determined on the fault mode;
the second determining module is used for determining the fault distribution information of the switching equipment of which the fault distribution is to be determined based on the target distribution information and the fault distribution information of each fault mode; the second determining module includes: and a third determining unit, configured to determine that a product of the target distribution information and the distribution information of each fault mode to faults is distribution information of the switching device to faults of the fault distribution to be determined.
6. The fault distribution determining apparatus of a switching device according to claim 5, wherein the acquisition module includes:
an acquisition unit configured to acquire a location, an equipment model, an operation age up to a time when a defect occurs, equipment defect recording information, and maintenance operation information of each of the switching devices;
the dividing unit is used for dividing the operation years from the defect occurrence time to the defect occurrence time;
a first determining unit, configured to determine a location of the switching device, a device model, and a running period from when the defect occurs after segmentation as an attribute vector of the switching device;
and the second determining unit is used for classifying the equipment defect record information and the maintenance operation information according to the association relation between the preset fault number and the fault type, and determining the classified fault number as the fault history vector.
7. The fault distribution determination apparatus of a switching device according to claim 5, wherein the second training module comprises:
the training unit is used for taking the attribute vector as input information of the decomposer model, taking the distribution information of the switching equipment on the fault mode as output information of the decomposer model, and determining the mapping relation of the switching equipment on the fault mode.
8. A fault distribution determining system of a switching device, characterized by comprising a fault distribution determining apparatus of a switching device according to any of claims 5-7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911229019.2A CN112906729B (en) | 2019-12-04 | 2019-12-04 | Fault distribution determination method, device and system of switch equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911229019.2A CN112906729B (en) | 2019-12-04 | 2019-12-04 | Fault distribution determination method, device and system of switch equipment |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112906729A CN112906729A (en) | 2021-06-04 |
CN112906729B true CN112906729B (en) | 2024-01-26 |
Family
ID=76111106
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201911229019.2A Active CN112906729B (en) | 2019-12-04 | 2019-12-04 | Fault distribution determination method, device and system of switch equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112906729B (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102854461A (en) * | 2012-08-24 | 2013-01-02 | 中国电力科学研究院 | Probability forecasting method and system of switch equipment faults |
CN105930347A (en) * | 2016-04-05 | 2016-09-07 | 浙江远传信息技术股份有限公司 | Text analysis based power outage cause recognition system |
CN109697455A (en) * | 2018-11-14 | 2019-04-30 | 清华大学 | The method for diagnosing faults and device of distribution switchgear |
CN109936113A (en) * | 2019-03-29 | 2019-06-25 | 国网浙江省电力有限公司 | A kind of protection act intelligent diagnosing method and system based on random forests algorithm |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
GB201621434D0 (en) * | 2016-12-16 | 2017-02-01 | Palantir Technologies Inc | Processing sensor logs |
US11922377B2 (en) * | 2017-10-24 | 2024-03-05 | Sap Se | Determining failure modes of devices based on text analysis |
-
2019
- 2019-12-04 CN CN201911229019.2A patent/CN112906729B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102854461A (en) * | 2012-08-24 | 2013-01-02 | 中国电力科学研究院 | Probability forecasting method and system of switch equipment faults |
CN105930347A (en) * | 2016-04-05 | 2016-09-07 | 浙江远传信息技术股份有限公司 | Text analysis based power outage cause recognition system |
CN109697455A (en) * | 2018-11-14 | 2019-04-30 | 清华大学 | The method for diagnosing faults and device of distribution switchgear |
CN109936113A (en) * | 2019-03-29 | 2019-06-25 | 国网浙江省电力有限公司 | A kind of protection act intelligent diagnosing method and system based on random forests algorithm |
Non-Patent Citations (1)
Title |
---|
电力设备典型故障案例的文本挖掘;杜修明 等;高电压技术;第44卷(第04期);1078-1084 * |
Also Published As
Publication number | Publication date |
---|---|
CN112906729A (en) | 2021-06-04 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107944454B (en) | Machine learning-based semantic labeling method for transformer substation | |
CN114285728B (en) | Predictive model training method, traffic prediction device and storage medium | |
US20160148094A1 (en) | Computer implemented system for automating the generation of a business decision analytic model | |
CN112949715A (en) | SVM (support vector machine) -based rail transit fault diagnosis method | |
CN112818035B (en) | Network fault prediction method, terminal equipment and storage medium | |
Beghoura et al. | Green software requirements and measurement: random decision forests-based software energy consumption profiling | |
CN110968689A (en) | Training method of criminal name and law bar prediction model and criminal name and law bar prediction method | |
CN112906729B (en) | Fault distribution determination method, device and system of switch equipment | |
CN114151293A (en) | Fault early warning method, system, equipment and storage medium of fan variable pitch system | |
CN115964470B (en) | Method and system for predicting service life of motorcycle accessories | |
CN113920497B (en) | Nameplate recognition model training method, nameplate recognition method and related devices | |
CN109947728B (en) | Log file processing method and device | |
CN116008729A (en) | Power failure diagnosis early warning method, device, equipment and storage medium | |
CN110321425B (en) | Method and device for judging defect type of power grid | |
CN115269998A (en) | Information recommendation method and device, electronic equipment and storage medium | |
US11844134B1 (en) | Cell site repair part prediction machine learning system | |
Lee et al. | Analysis of EV charging load impact on distribution network using XAI technique | |
CN105335500A (en) | Covariant local feature aggregated image feature representation method | |
CN111611117A (en) | Hard disk fault prediction method, device, equipment and computer readable storage medium | |
CN113806750B (en) | File security risk detection method, training method, device and equipment of model | |
CN110807082A (en) | Quality spot check item determination method, system, electronic device and readable storage medium | |
CN111176928B (en) | Log processing method of embedded edge computing system | |
CN112183644B (en) | Index stability monitoring method and device, computer equipment and medium | |
Sima et al. | Detection framework for few-shot voltage disturbances in power systems by evaluation learning | |
CN117874508A (en) | Model training method, data extraction method and related equipment |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |