CN111026624A - Fault prediction method and device of power grid information system - Google Patents
Fault prediction method and device of power grid information system Download PDFInfo
- Publication number
- CN111026624A CN111026624A CN201911093204.3A CN201911093204A CN111026624A CN 111026624 A CN111026624 A CN 111026624A CN 201911093204 A CN201911093204 A CN 201911093204A CN 111026624 A CN111026624 A CN 111026624A
- Authority
- CN
- China
- Prior art keywords
- maintenance data
- fault
- model
- maintenance
- prediction
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3447—Performance evaluation by modeling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3055—Monitoring arrangements for monitoring the status of the computing system or of the computing system component, e.g. monitoring if the computing system is on, off, available, not available
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3058—Monitoring arrangements for monitoring environmental properties or parameters of the computing system or of the computing system component, e.g. monitoring of power, currents, temperature, humidity, position, vibrations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3466—Performance evaluation by tracing or monitoring
- G06F11/3476—Data logging
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Quality & Reliability (AREA)
- General Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Computer Hardware Design (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Supply And Distribution Of Alternating Current (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a fault prediction method and a device of a power grid information system, wherein the method comprises the following steps: acquiring a first-level operation and maintenance data index in a first-level time period from the operation and maintenance log according to the set first-level time period; dividing the first-level operation and maintenance data indexes according to a second-level time period to obtain a plurality of groups of second-level operation and maintenance data indexes, and processing a plurality of operation and maintenance data indexes of the same time node to obtain a model value for representing the characteristics of the plurality of corresponding operation and maintenance data indexes under the time node; and training the prediction model by using the obtained multiple model values and multiple actually obtained fault values as a training set, predicting by using the trained prediction model, and reporting the fault if the operation and maintenance data index of the future time node is in the fault value range. By adopting the scheme, the prediction of the fault condition of the node at the future time is realized, so that the pressure of real-time operation and maintenance on operation and maintenance personnel is relieved, and meanwhile, the loss caused by the fault is avoided.
Description
Technical Field
The invention relates to the field of data processing, in particular to a fault prediction method and device of a power grid information system.
Background
The data processing capacity in the power grid information system is huge, and meanwhile, the influence of the power grid information system on various industries is huge. Therefore, the method is very important for operation and maintenance work of the power grid information system.
The operation and maintenance scheme in the prior art usually adopts manual operation and maintenance, script tool operation and maintenance, flow tool operation and maintenance and automatic operation and maintenance. Wherein, manual operation and maintenance refers to all the operation and maintenance problems, and is basically completed by manual operation. The operation and maintenance of the script tool means that the reproducible operation and maintenance operation is realized by shell scripts, and the Perl, Ruby, Python and other programming languages can be used for writing the operation and maintenance script tool. And (4) operation and maintenance of the process tool, namely, a plurality of script tools are connected in series by using the process, and the script execution result is verified at the same time. The automatic operation and maintenance means that a series of operation and maintenance analysis tools are used for visually monitoring the system state, so that the stable operation of the system is guaranteed, and the intelligent scheduling system resources meet the requirements of stable and efficient operation of the system.
However, in the prior art, the operation and maintenance scheme is very dependent on manual operation, and meanwhile, due to the fact that the operation and maintenance scheme belongs to real-time operation and maintenance, the maintenance pressure on operation and maintenance personnel is high, when a system fault occurs, if the experience and technical level of the operation and maintenance personnel do not reach the standard, the fault recovery time limit is too long, and great loss is easily caused to the industry depending on a power grid information system.
Disclosure of Invention
The purpose of the invention is as follows: the invention aims to provide a fault prediction method and a fault prediction device for a power grid information system.
The technical scheme is as follows: the embodiment of the invention provides a fault prediction method of a power grid information system, which comprises the following steps: acquiring an operation and maintenance log from the power grid information system; acquiring a primary operation and maintenance data index in a primary time period from the operation and maintenance log according to the set primary time period; dividing the first-level operation and maintenance data indexes according to a second-level time period to obtain a plurality of groups of second-level operation and maintenance data indexes, and processing a plurality of operation and maintenance data indexes of the same time node in the plurality of groups of second-level operation and maintenance data indexes obtained by division to obtain model values for representing the characteristics of the plurality of operation and maintenance data indexes corresponding to the time node; and training the prediction model by using the obtained multiple model values and the actually obtained multiple fault values as a training set, and predicting by using the trained prediction model.
Specifically, the primary operation and maintenance data index includes operation and maintenance data indexes of a plurality of time nodes within the primary time period, and time intervals between the time nodes are the same.
Specifically, the operation and maintenance data index includes at least one of the following: CPU, memory, storage utilization rate, I/O bandwidth, network delay, bandwidth, server temperature, fan revolution, ambient temperature, and humidity.
Specifically, the model value is calculated by using the following formula:
wherein d represents the number of groups of the secondary operation and maintenance data indexes obtained by division, and xiAnd d, representing the operation and maintenance data indexes at the same time node in the d groups of secondary operation and maintenance data indexes, wherein the value x when the minimum value is obtained in f (x) is a model value.
Specifically, the obtained multiple model values are fitted to obtain a model curve corresponding to the model values, the model curve and fault model values corresponding to the model curve on time nodes are used as training sets, and a prediction model constructed by a long-short term memory network algorithm is trained.
Specifically, the fault model value is obtained by processing a plurality of fault values at the same time node, and is used for representing characteristics of the plurality of corresponding fault values at the time node.
Specifically, if the operation and maintenance data index under the future time node is determined to belong to the fault value range according to the prediction result, the fault report is carried out.
Specifically, the prediction model is used for calculating past time nodes, and if the calculated operation and maintenance data indexes belong to a fault value range and the operation and maintenance data indexes corresponding to the actual time nodes belong to a non-fault value range, feedback training is performed on the prediction model.
The embodiment of the invention also provides a fault prediction device of a power grid information system, which comprises the following steps: a first acquisition unit, a second acquisition unit, a processing unit and a prediction unit, wherein: the first obtaining unit is used for obtaining an operation and maintenance log from the power grid information system; the second obtaining unit is used for obtaining a primary operation and maintenance data index in a primary time period from the operation and maintenance log according to the set primary time period; the processing unit is used for dividing the first-level operation and maintenance data indexes according to a second-level time period to obtain a plurality of groups of second-level operation and maintenance data indexes, and processing a plurality of operation and maintenance data indexes of the same time node in the plurality of groups of second-level operation and maintenance data indexes obtained by division to obtain model values for representing the characteristics of the plurality of corresponding operation and maintenance data indexes under the time node; and the prediction unit is used for training the prediction model by taking the obtained multiple model values and the actually obtained multiple fault values as a training set, predicting by using the trained prediction model, and reporting the fault if the operation and maintenance data indexes of the future time node are in the fault value range.
Specifically, the processing unit is further configured to calculate the model value by using the following formula:
wherein d represents the number of groups of the secondary operation and maintenance data indexes obtained by division, and xiRepresenting the operation and maintenance data index of the same time node in the d groups of second-level operation and maintenance data indexes, wherein the x value is the minimum value when f (x) is obtainedA model value.
Has the advantages that: compared with the prior art, the invention has the following remarkable advantages: by acquiring the operation and maintenance data indexes, processing the operation and maintenance data indexes to obtain model values representing the characteristics of the operation and maintenance data indexes, and training the prediction model by using the model values as a training set, the prediction of the fault condition of a future time node is realized, so that the pressure of real-time operation and maintenance on operation and maintenance personnel is relieved, and meanwhile, the loss caused by the fault is avoided.
Drawings
Fig. 1 is a schematic flow chart of a fault prediction method of a power grid information system provided in an embodiment of the present invention;
FIG. 2 is a graph of a model curve fit to model values in an embodiment of the present invention;
FIG. 3 is a schematic diagram of a determination range of a prediction model obtained by training according to the model curve determined in FIG. 2 according to an embodiment of the present disclosure;
fig. 4 is a schematic structural diagram of a fault prediction apparatus of a power grid information system provided in an embodiment of the present invention.
Detailed Description
The technical scheme of the invention is further explained by combining the attached drawings.
Referring to fig. 1, a schematic flow chart of a fault prediction method of a power grid information system according to an embodiment of the present invention is shown, where the method includes specific steps, and the detailed description is given below with reference to the specific steps.
And step S101, acquiring an operation and maintenance log from the power grid information system.
In specific implementation, the operation and maintenance data indexes recorded in the operation of the power grid information system are recorded in the operation and maintenance log.
And step S102, acquiring a primary operation and maintenance data index in a primary time period from the operation and maintenance log according to the set primary time period.
In a specific implementation, the primary time period may be set by a user according to an actual application scenario.
In the embodiment of the present invention, the primary operation and maintenance data index includes operation and maintenance data indexes of a plurality of time nodes within the primary time period, and time intervals between the time nodes are the same.
For example, in units of minutes, every 1 minute is taken as a time node, that is, every 1 minute, the operation and maintenance data index at the corresponding time node is obtained.
In an embodiment of the present invention, the operation and maintenance data index includes at least one of the following: CPU, memory, storage utilization rate, I/O bandwidth, network delay, bandwidth, server temperature, fan revolution, ambient temperature, and humidity.
In a specific implementation, the number of the types of the operation and maintenance data indexes may be one, or may be multiple. When the number of the operation and maintenance data index types is multiple, corresponding data processing can be performed on each type, and weighting is performed according to the prediction result of each type to obtain the final prediction result.
Step S103, dividing the first-level operation and maintenance data indexes according to a second-level time period to obtain multiple groups of second-level operation and maintenance data indexes, and processing multiple operation and maintenance data indexes of the same time node in the multiple groups of second-level operation and maintenance data indexes obtained through division to obtain model values for representing the characteristics of the multiple corresponding operation and maintenance data indexes under the time node.
In a specific implementation, a secondary time period refers to a length of time used to divide the primary time period. For example, the primary time period is 30 days, the secondary time period is 1 day, and every 1 minute is a time node. The corresponding secondary operation and maintenance data indicators in each set of secondary time periods include 1440 operation and maintenance data indicators within 1 day.
In particular, in order to further improve the accuracy of the prediction result, the training data for training the prediction model needs to have a higher correlation. Therefore, a plurality of operation and maintenance data indexes of the same time node in the secondary operation and maintenance data indexes can be processed to obtain a model value. For example, the primary time period is 30 days, the secondary time period is 1 day, and every 1 minute is a time node. And then, 30 groups of secondary operation and maintenance data indexes are obtained in total, 30 operation and maintenance data indexes of the time node at 9 points in the secondary time period are obtained, the 30 operation and maintenance data indexes are obtained in total, and the model value used for representing the characteristics of the 30 corresponding operation and maintenance data indexes at the time node at 9 points is obtained. Since each group of secondary operation and maintenance data indexes corresponds to 1440 time nodes, 1440 model values can be finally obtained.
In the embodiment of the invention, the model value is calculated by adopting the following formula:
wherein d represents the number of groups of the secondary operation and maintenance data indexes obtained by division, and xiAnd d, representing the operation and maintenance data indexes at the same time node in the d groups of secondary operation and maintenance data indexes, wherein the value x when the minimum value is obtained in f (x) is a model value.
In specific implementation, a plurality of operation and maintenance data indexes corresponding to each time node may be calculated, and finally, a model value corresponding to each time node may be obtained.
And step S104, training the prediction model by using the obtained multiple model values and the actually obtained multiple fault values as a training set, and predicting by using the trained prediction model.
In the embodiment of the invention, if the operation and maintenance data index under the future time node is determined to belong to the fault value range according to the prediction result, the fault report is carried out.
In the embodiment of the invention, the obtained multiple model values are fitted to obtain the model curves corresponding to the model values, the model curves and the fault model values corresponding to the model curves on the time nodes are used as training sets, and the prediction model constructed by the long-short term memory network algorithm is trained.
In the embodiment of the invention, the fault model value is obtained by processing a plurality of fault values under the same time node and is used for representing the characteristics of the plurality of corresponding fault values under the time node.
In specific implementation, for example, a fault value is first obtained, and a plurality of fault values at the same time node are processed to obtain a corresponding fault model value.
In specific implementation, after training is completed, the prediction model can predict the operation and maintenance data index at a future time node, and because the training set includes both a model value and a fault value, after the operation and maintenance data index at the future time node is obtained through prediction, the prediction model can judge whether the operation and maintenance data index at the future time node belongs to a fault value range, if so, a fault may occur in the power grid information system at the future time node, and correspondingly, fault reporting is performed to avoid the occurrence of the fault.
In the embodiment of the invention, the prediction model is used for calculating the past time nodes, and if the operation and maintenance data index obtained by calculation belongs to the fault value range and the operation and maintenance data index corresponding to the actual time node belongs to the non-fault value range, the feedback training is carried out on the prediction model.
In specific implementation, after the training of the prediction model is completed, the prediction accuracy of the prediction model can be judged through the operation and maintenance data indexes under the existing time nodes, and the feedback training is performed on the prediction model, so that the accuracy of the prediction result is further improved.
Fig. 2 is a graph of a model curve obtained by fitting model values according to an embodiment of the present invention. The curve in fig. 2 is obtained from the actually obtained operation and maintenance data index, and when the actually obtained operation and maintenance data index changes, the model curve also changes correspondingly.
In a specific implementation, the model curve in fig. 2 is obtained for a time node every 1 minute, based on a primary time period of 30 days and a secondary time period of 1 day. Where the ordinate is the numerical value of the model value and the abscissa is the time node. The multiple curves are fit to discrete model values by using quadratic polynomials, cubic polynomials or even sextic polynomials, and the higher the degree of the polynomials is, the better the fitting effect on the model is. When the number of functions is 5, the correlation of the fitting function has reached 0.9769. The time taken to calculate the fitting function only needs 1.5 s. When the number of times of the function is increased to 6 times, although the correlation coefficient of the fitting function is higher, the consumed calculation time reaches 8.43 s. In an actual production environment, the accuracy of fitting of a proper disclaimer function can be selected to replace the great improvement of the system operation efficiency. A polynomial function of degree 5 can therefore be selected in this scenario as the applied model curve.
Fig. 3 is a schematic diagram of a judgment range of the prediction model obtained by training according to the model curve determined in fig. 2 in the embodiment of the present invention.
In specific implementation, the prediction model is an operation and maintenance data index two classifier obtained by training by using a model value and a fault value. The point inside the two lines as shown in fig. 3 is a point that does not cause a system failure, and the outside point is a failure point. And inputting the parameters into the classifier function, if the return value is greater than 0, determining that the parameters are on the inner side of the line and are normal points. Otherwise, if the return value is less than or equal to 0, the fault point is on the line or outside the line. And (4) bringing the model value and the actual value into a prediction model for calculation, wherein the return state values are the same. The returned state is proved to be consistent. The system is in a normal running state, and the actual system running state is reflected well by the predicted value. The operation and maintenance data index in the interval does not cause system failure. By using the verification of the model value, a certain deviation can be found between the actual value and the predicted value, but the deviations are in a reasonable range, and the prediction model can better reflect the current running state of the system.
Referring to fig. 4, it is a schematic structural diagram of a fault prediction apparatus 40 of a power grid information system provided in an embodiment of the present invention, specifically including: a first acquisition unit 401, a second acquisition unit 402, a processing unit 403, and a prediction unit 404, wherein:
the first obtaining unit 401 is configured to obtain an operation and maintenance log from the power grid information system;
the second obtaining unit 402 is configured to obtain a first-level operation and maintenance data index in a first-level time period from the operation and maintenance log according to the set first-level time period;
the processing unit 403 is configured to divide the first-stage operation and maintenance data index into multiple groups of second-stage operation and maintenance data indexes according to a second-stage time period, and process multiple operation and maintenance data indexes of a same time node in the multiple groups of second-stage operation and maintenance data indexes obtained by division to obtain a model value for characterizing features of the multiple corresponding operation and maintenance data indexes under the time node;
the prediction unit 404 is configured to train the prediction model using the obtained plurality of model values and the actually obtained plurality of fault values as a training set, and predict the prediction model using the trained prediction model.
In the embodiment of the present invention, the primary operation and maintenance data index includes operation and maintenance data indexes of a plurality of time nodes within the primary time period, and time intervals between the time nodes are the same.
In an embodiment of the present invention, the operation and maintenance data index includes at least one of the following: CPU, memory, storage utilization rate, I/O bandwidth, network delay, bandwidth, server temperature, fan revolution, ambient temperature, and humidity.
In this embodiment of the present invention, the processing unit 403 may be further configured to calculate the model value by using the following formula:
wherein d represents the number of groups of the secondary operation and maintenance data indexes obtained by division, and xiAnd d, representing the operation and maintenance data indexes at the same time node in the d groups of secondary operation and maintenance data indexes, wherein the value x when the minimum value is obtained in f (x) is a model value.
In this embodiment of the present invention, the prediction unit 404 may be further configured to fit the obtained multiple model values to obtain a model curve corresponding to the model values, and train the prediction model constructed by the long-term and short-term memory network algorithm by using the model curve and the fault model value corresponding to the model curve on the time node as a training set.
In this embodiment of the present invention, the prediction unit 404 may be further configured to process the fault model value by using a plurality of fault values under the same time node, and is used to characterize characteristics of the plurality of corresponding fault values under the time node.
In this embodiment of the present invention, the prediction unit 404 may be further configured to perform a fault report if it is determined that the operation and maintenance data indicator at the future time node belongs to the fault value range according to the prediction result.
In this embodiment of the present invention, the prediction unit 404 may be further configured to calculate a past time node by using the prediction model, and perform feedback training on the prediction model if the calculated operation and maintenance data index belongs to a fault value range and the operation and maintenance data index corresponding to the actual time node belongs to a non-fault value range.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting the same, and although the present invention is described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: modifications and equivalents may be made to the embodiments of the invention without departing from the spirit and scope of the invention, which is to be covered by the claims.
Claims (10)
1. A fault prediction method of a power grid information system is characterized by comprising the following steps:
acquiring an operation and maintenance log from the power grid information system;
acquiring a primary operation and maintenance data index in a primary time period from the operation and maintenance log according to the set primary time period;
dividing the first-level operation and maintenance data indexes according to a second-level time period to obtain a plurality of groups of second-level operation and maintenance data indexes, and processing according to a plurality of operation and maintenance data indexes of the same time node in the plurality of groups of second-level operation and maintenance data indexes obtained by division to obtain model values for representing the characteristics of the plurality of operation and maintenance data indexes corresponding to the time node;
and training the prediction model by using the obtained multiple model values and the actually obtained multiple fault values as a training set, and predicting by using the trained prediction model.
2. The grid information system fault prediction method according to claim 1, wherein the primary operation and maintenance data indicators comprise operation and maintenance data indicators at a plurality of time nodes within the primary time period, and a time interval between each time node is the same.
3. The grid information system fault prediction method of claim 2, wherein the operation and maintenance data indicators comprise at least one of: CPU, memory, storage utilization rate, I/O bandwidth, network delay, bandwidth, server temperature, fan revolution, ambient temperature, and humidity.
4. The method of predicting the failure of the grid information system according to claim 1, wherein the model value is calculated using the following formula:
wherein d represents the number of groups of the secondary operation and maintenance data indexes obtained by division, and xiAnd d, representing the operation and maintenance data indexes at the same time node in the d groups of secondary operation and maintenance data indexes, wherein the value x when the minimum value is obtained in f (x) is a model value.
5. The method for predicting the fault of the grid information system according to claim 4, wherein training the prediction model using the obtained plurality of model values and the actually obtained plurality of fault values as a training set comprises:
and fitting the obtained multiple model values to obtain a model curve corresponding to the model values, and training a prediction model constructed by the long-short term memory network algorithm by taking the model curve and the fault model values corresponding to the model curve on the time nodes as a training set.
6. The method according to claim 5, wherein the fault model value is obtained by processing a plurality of fault values at the same time node, and is used for characterizing the corresponding plurality of fault values at the time node.
7. The method of predicting faults of a grid information system according to claim 5, further comprising, after the predicting using the trained predictive model:
and if the operation and maintenance data index under the future time node is determined to belong to the fault value range according to the prediction result, carrying out fault report.
8. The method according to claim 5, wherein the prediction model is used to calculate past time nodes, and if the calculated operation and maintenance data index belongs to a fault value range and the operation and maintenance data index corresponding to the actual time node belongs to a non-fault value range, the prediction model is subjected to feedback training.
9. A failure prediction apparatus for a grid information system, comprising: a first acquisition unit, a second acquisition unit, a processing unit and a prediction unit, wherein:
the first obtaining unit is used for obtaining an operation and maintenance log from the power grid information system;
the second obtaining unit is used for obtaining a primary operation and maintenance data index in a primary time period from the operation and maintenance log according to the set primary time period;
the processing unit is used for dividing the first-level operation and maintenance data indexes according to a second-level time period to obtain a plurality of groups of second-level operation and maintenance data indexes, and processing a plurality of operation and maintenance data indexes of the same time node in the plurality of groups of second-level operation and maintenance data indexes obtained by division to obtain model values for representing the characteristics of the plurality of corresponding operation and maintenance data indexes under the time node;
and the prediction unit is used for training the prediction model by taking the obtained multiple model values and the actually obtained multiple fault values as a training set, predicting by using the trained prediction model, and reporting the fault if the operation and maintenance data indexes of the future time node are in the fault value range.
10. The grid information system fault prediction device of claim 9, wherein the processing unit is further configured to calculate the model value using the following equation:
wherein d represents the number of groups of the secondary operation and maintenance data indexes obtained by division, and xiAnd d, representing the operation and maintenance data indexes at the same time node in the d groups of secondary operation and maintenance data indexes, wherein the value x when the minimum value is obtained in f (x) is a model value.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911093204.3A CN111026624B (en) | 2019-11-11 | 2019-11-11 | Fault prediction method and device of power grid information system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911093204.3A CN111026624B (en) | 2019-11-11 | 2019-11-11 | Fault prediction method and device of power grid information system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111026624A true CN111026624A (en) | 2020-04-17 |
CN111026624B CN111026624B (en) | 2023-06-02 |
Family
ID=70205201
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201911093204.3A Active CN111026624B (en) | 2019-11-11 | 2019-11-11 | Fault prediction method and device of power grid information system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111026624B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113607413A (en) * | 2021-08-26 | 2021-11-05 | 上海航数智能科技有限公司 | Bearing component fault monitoring and predicting method based on controllable temperature and humidity |
CN117791856A (en) * | 2023-12-20 | 2024-03-29 | 武汉人云智物科技有限公司 | Power grid fault early warning method and device based on inspection robot |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102331543A (en) * | 2011-06-23 | 2012-01-25 | 上海市安全生产科学研究所 | Support vector machine based fault electric arc detection method |
CN108197748A (en) * | 2018-01-08 | 2018-06-22 | 国网江苏省电力有限公司 | A kind of flow of the people Forecasting Methodology based on mind evolutionary |
CN108375715A (en) * | 2018-03-08 | 2018-08-07 | 中国电力科学研究院有限公司 | A kind of distribution network line fault risk day prediction technique and system |
WO2019137052A1 (en) * | 2018-01-11 | 2019-07-18 | 华为技术有限公司 | Method and device for network operation and maintenance |
-
2019
- 2019-11-11 CN CN201911093204.3A patent/CN111026624B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102331543A (en) * | 2011-06-23 | 2012-01-25 | 上海市安全生产科学研究所 | Support vector machine based fault electric arc detection method |
CN108197748A (en) * | 2018-01-08 | 2018-06-22 | 国网江苏省电力有限公司 | A kind of flow of the people Forecasting Methodology based on mind evolutionary |
WO2019137052A1 (en) * | 2018-01-11 | 2019-07-18 | 华为技术有限公司 | Method and device for network operation and maintenance |
CN108375715A (en) * | 2018-03-08 | 2018-08-07 | 中国电力科学研究院有限公司 | A kind of distribution network line fault risk day prediction technique and system |
Non-Patent Citations (2)
Title |
---|
杨丽君等: "基于多代理***的智能配电网分散协调恢复策略", 《电工电能新技术》 * |
杨航等: "基于Storm的局部放电信号集合经验模态分解", 《计算机工程与应用》 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113607413A (en) * | 2021-08-26 | 2021-11-05 | 上海航数智能科技有限公司 | Bearing component fault monitoring and predicting method based on controllable temperature and humidity |
CN117791856A (en) * | 2023-12-20 | 2024-03-29 | 武汉人云智物科技有限公司 | Power grid fault early warning method and device based on inspection robot |
Also Published As
Publication number | Publication date |
---|---|
CN111026624B (en) | 2023-06-02 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108375715B (en) | Power distribution network line fault risk day prediction method and system | |
CN113240289B (en) | Power grid dispatching control system operation state evaluation method and system | |
CN113570138B (en) | Method and device for predicting residual service life of equipment of time convolution network | |
JP7053152B2 (en) | Systems and methods for optimizing recommended inspection intervals | |
CN110633194B (en) | Performance evaluation method of hardware resources in specific environment | |
CN111026624B (en) | Fault prediction method and device of power grid information system | |
CN114066262A (en) | Method, system and device for estimating cause-tracing reasoning of abnormal indexes after power grid dispatching and storage medium | |
CN114862275B (en) | Storage logistics system reliability assessment method and system based on digital twin model | |
CN111400850B (en) | Equipment fault analysis method, device, equipment and storage medium | |
CN113988325A (en) | Power system fault early warning method and device, terminal equipment and storage medium | |
CN113742993A (en) | Method, device, equipment and storage medium for predicting life loss of dry-type transformer | |
CN115453356A (en) | Power equipment running state monitoring and analyzing method, system, terminal and medium | |
CN117318069B (en) | Power supply system fault self-healing method and system | |
CN111080484A (en) | Method and device for monitoring abnormal data of power distribution network | |
US12007751B2 (en) | Methods for smart gas data management, internet of things systems, and storage media | |
CN111367255A (en) | Performance evaluation test system and method for multi-variable control system | |
CN116566839A (en) | Communication resource quality evaluation system for power enterprises | |
CN106650154B (en) | Nuclear power conventional island equipment reliability monitoring method based on thermal performance index | |
CN116305741A (en) | Updating method and device for digital twin model of power equipment | |
CN115809805A (en) | Power grid multi-source data processing method based on edge calculation | |
CN114696328A (en) | Power line loss analysis method, system and storage medium | |
CN114970311A (en) | Method for establishing remote module life prediction model and life prediction method | |
CN114418237A (en) | Distribution network power supply safety capability evaluation standard quantification method, system, equipment and medium | |
CN113807743A (en) | Power grid dispatching automation software reliability assessment method and system | |
CN114414938B (en) | Dynamic response method and system for power distribution network faults |
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 |